VSEBINA – CONTENTS
IZVIRNI ZNANSTVENI ^LANKI – ORIGINAL SCIENTIFIC ARTICLES
Improvement of the casting of special steel with a wide solid-liquid interface
Izbolj{anje ulivanja posebnega jekla s {irokim intervalom trdno-teko~e
T. Mauder, J. Stetina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Non-traditional non-destructive testing of the alkali-activated slag mortar during the hardening
Netradicionalno neporu{no preizku{anje z alkalijami aktivirane malte med strjevanjem
L. Topoláø, P. Rypák, K. Tim~aková-[amárková, L. Pazdera, P. Rovnaník . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Multi-walled carbon nanotubes effect in polypropylene nanocomposites
Vpliv ve~stenskih ogljikovih nanocevk v nanokompozitih iz polipropilena
C. E. Ban, A. Stefan, I. Dinca, G. Pelin, A. Ficai, E. Andronescu, O. Oprea, G. Voicu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Experimental and numerical study of hot-steel-plate flatness
Eksperimentalni in numeri~ni {tudij ravnosti vro~ih plo{~ iz jekla
J. Hrabovský, M. Pohanka, P. J. Lee, J. H. Kang . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Investigation of hole effects on the critical buckling load of laminated composite plates
Preiskava vpliva luknje na kriti~no upogibno obremenitev laminiranih kompozitnih plo{~
A. Kurºun, E. Topal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Corrosion of the refractory zirconia metering nozzle due to molten steel and slag
Korozija ognjeodporne cirkonske dozirne {obe s staljenim jeklom in `lindro
K. Wiœniewska, D. Madej, J. Szczerba. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Effects of an epoxy-resin-fiber substrate for a -shaped microstrip antenna
Vpliv z vlakni oja~ane epoksi podlage pri -obliki mikrotrakaste antene
Md. M. Islam, M. R. I. Faruque, M. Tariqul Islam, H. Arshad . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
X-ray radiography of AISI 4340-2205 steels welded by friction welding
Rentgenski pregled jekel AISI 4340-2205, varjenih s trenjem
U. Caligulu, M. Yalcinoz, M. Turkmen, S. Mercan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Thermodynamic properties and microstructures of different shape-memory alloys
Termodinami~ne lastnosti in mikrostruktura razli~nih zlitin z oblikovnim spominom
L. Gomid`elovi}, E. Po`ega, A. Kostov, N. Vukovi}, D. @ivkovi}, D. Manasijevi} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
The relationship between thermal treatment of serpentine and its reactivity
Odvisnost med toplotno obdelavo serpentina in njegovo aktivnostjo
G. Su~ik, A. Szabóová, L’. Popovi~, D. Hr{ak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Deformations and velocities during the cold rolling of aluminium alloys
Deformacija in hitrosti pri hladnem valjanju aluminijevih zlitin
M. Mi{ovi}, N. Tadi}, M. Ja}imovi}, M. Janji} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Prediction of the chemical non-homogeneity of 30MnVS6 billets with genetic programming
Napovedovanje nehomogenosti kemijske sestave pri gredicah 30MnVS6 s pomo~jo genetskega programiranja
M. Kova~i~, D. Novak. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Effect of the TiBN coating on a HSS drill when drilling the MA8M Mg alloy
Vpliv TiBN prevleke na HSS svedru pri vrtanju MA8M Mg zlitine
F. Karaca, B. Aksakal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Application of the Taguchi method to select the optimum cutting parameters for tangential cylindrical grinding of AISI D3
tool steel
Uporaba Taguchi metode za izbiro optimalnih parametrov odrezavanja pri tangencialnem cilindri~nem bru{enju orodnega jekla AISI D3
C. Ozay, H. Ballikaya, V. Savas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Effects of friction-welding parameters on the morphological properties of an Al/Cu bimetallic joint
Vpliv parametrov tornega varjenja na morfolo{ke lastnosti Al/Cu bimetalnega spoja
V. D. Mila{inovi}, R. V. Radovanovi}, M. D. Mila{inovi}, B. R. Gligorijevi} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
ISSN 1580-2949
UDK 669+666+678+53 MTAEC9, 50(1)1–162(2016)
MATER.
TEHNOL.
LETNIK
VOLUME 50
[TEV.
NO. 1
STR.
P. 1–162
LJUBLJANA
SLOVENIJA
JAN.–FEB.
2016
Characterisation of the mechanical and corrosive properties of newly developed glass-steel composites
Karakterizacija mehanskih in korozijskih lastnosti novo razvitih kompozitov steklo-jeklo
O. Lyubimova, E. Gridasova, A. Gridasov, G. Frieling, M. Klein, F. Walther . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Phase analysis of the slag after submerged-arc welding
Analiza faz v `lindri pri oblo~nem varjenju pod pra{kom
M. Prijanovi~ Tonkovi~, J. Lamut . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Optimizing the parameters for friction welding stainless steel to copper parts
Optimiranje parametrov pri tornem varjenju nerjavnega jekla na bakrene dele
M. Sahin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
WEDM cutting of Inconel 718 nickel-based superalloy: effects of cutting parameters on the cutting quality
WEDM rezanje nikljeve superzlitine Inconel 718: vpliv parametrov rezanja na kvaliteto rezanja
U. Çaydaº, M. Ay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
STROKOVNI ^LANKI – PROFESSIONAL ARTICLES
Influence of dredged sediment on the shrinkage behavior of self-compacting concrete
Vpliv izkopanih sedimentov na kr~enje samozgo{~evalnega betona
N. E. Bouhamou, F. Mostefa, A. Mebrouki, K. Bendani, N. Belas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Study of the properties and hygrothermal behaviour of alternative insulation materials based on natural fibres
[tudij lastnosti in higrotermalno obna{anje alternativnih izolacijskih materialov na osnovi naravnih vlaken
J. Zach, M. Reif, J. Hroudová . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Prediction of the elastic moduli of chicken-feather-reinforced PLA and a comparison with experimental results
Napovedovanje modulov elasti~nosti PLA, oja~anega s pi{~an~jim perjem in primerjava z eksperimentalnimi rezultati
U. Özmen, B. Okutan Baba . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Composites based on inorganic matrices for extreme exposure conditions
Kompoziti z anorgansko osnovo za izpostavitev ekstremnim razmeram
A. Dufka, T. Melichar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
The effect of EO and steam sterilization on the mechanical and electrochemical properties of titanium Grade 4
Vpliv EO in sterilizacije s paro na mehanske in elektrokemijske lastnosti titana Grade 4
M. Basiaga, W. Walke, Z. Paszenda, A. Kajzer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Influence of the carbide-particle spheroidisation process on the microstructure after the quenching and annealing of
100CrMnSi6-4 bearing steel
Vpliv procesa sferoidizacije karbidnih delcev na mikrostrukturo jekla 100CrMnSi6-4 za le`aje po kaljenju in popu{~anju
J. Dlouhy, D. Hauserova, Z. Novy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
2 Materiali in tehnologije / Materials and technology 50 (2016) 1,
T. MAUDER, J. STETINA: IMPROVEMENT OF THE CASTING OF SPECIAL STEEL WITH ...
3–6
IMPROVEMENT OF THE CASTING OF SPECIAL STEEL WITH A
WIDE SOLID-LIQUID INTERFACE
IZBOLJ[ANJE ULIVANJA POSEBNEGA JEKLA S [IROKIM
INTERVALOM TRDNO-TEKO^E
Tomas Mauder, Josef Stetina
Brno University of Technology, Faculty of Mechanical Engineering, Technicka 2, 616 69 Brno, Czech Republic
mauder@fme.vutbr.cz, stetina@fme.vutbr.cz
Prejem rokopisa – received: 2014-07-29; sprejem za objavo – accepted for publication: 2015-03-03
doi:10.17222/mit.2014.122
In the last few years, steelmakers have been facing a significant decrease in the steel demand caused by the global economic
crisis. Positive economic results have mostly been reached in the steel factories that have focused on special steel production
with higher product capabilities, such as higher strength grades, steel design for acidic environments, steel for the offshore
technology, etc. These steels must keep the mechanical properties, such as the resistance to rapture, compression strength,
stress-strain properties, etc., within strict limits. The numerical calculations and optimization of casting parameters were pro-
vided. The results show the recommended casting parameters and differences between the examined steel and classic
low-carbon steels.
Keywords: continuous casting, experimental measurement, numerical simulation, optimization
Proizvajalci jekel se zadnja leta spopadajo z zmanj{evanjem povpra{evanja po jeklu, kar je posledica globalne ekonomske krize.
Pozitivne ekonomske rezultate so dosegle `elezarne, ki so se osredinile na proizvodnjo posebnih jekel in z ve~jimi proizvodnimi
zmogljivostmi, kot so visokotrdnostna jekla, jekla za delo v kislem okolju, jekla za naftne plo{~adi itd. Ta jekla morajo v ozkih
intervalih obdr`ati svoje mehanske lastnosti, kot so pretrg, tla~na trdnost, raztezek pri nategu itd. Izvr{eni so bili izra~uni in
optimizacija postopka ulivanja. Rezultati ka`ejo predlagane parametre litja in razlike med preiskovanim jeklom in navadnim
malooglji~nim jeklom.
Klju~ne besede: kontinuirano litje, eksperimentalne meritve, numeri~na simulacija, optimizacija
1 INTRODUCTION
Continuous casting (CC) of steel, as an industrialized
method of solidification processing, has a relatively short
history of only about 60 years. In fact, the CC ratio in the
world of steel industry now reaches more than 95 % of
crude-steel output (Figure 1).1–3 Through the years, the
product quality, production efficiency, operating safety
and casting of special steels and alloys have increased.
Today, nearly all steel grades can be produced and
productivity goals exceed by far those envisaged in the
1960s/1970s; the limits of casting-section sizes have
been increased to support new steel-grade developments
like thick high-strength steel plates or to realize new pro-
cess routes like the direct link between the casting and
rolling steps.4
In the early 1990s, continuous casting was an estab-
lished and already matured technology. The production
was focused on cost reductions through higher casting
speeds, a better utilization of energy, the optimization of
equipment performance and a reduction of the main-
tenance expenses using the equipment with a longer
lifetime. The key factor, which made continuous casting
the "main-stream technology", was the continuous inno-
vation. From the metallurgical point of view, the state-
of-the-art continuous casters have the features that
enable strand treatment through special cooling and
soft-reduction technologies.5 Sophisticated process mo-
dels allow an online process simulation and closed-loop
control to further optimize the product quality and pro-
ductivity goals.
Today, steelmakers in the European Union are facing
a significantly decreasing steel demand caused by the
global economic crisis. Figure 1 shows the crude-steel
production progress in the EU between 2005 and 2012
Materiali in tehnologije / Materials and technology 50 (2016) 1, 3–6 3
UDK 669.18:621.74.047:519.61/.64 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)3(2016)
Figure 1: Crude-steel production in EU2
Slika 1: Proizvodnja surovega jekla v EU2
promulgated by The World Steel Association.2 The
economic crisis in 2008 caused a deep slump in the steel
production in the EU. Positive economic results were
mostly reached by the steel factories that focused on
special steel production with higher product capabilities,
such as higher-strength grades, steel plates for barrel
boilers, steel design for acidic environments and steel for
the offshore technology. The production of special steel
is the only prospective for the EU to keep the
competitiveness with the Asian market.
Steel grades for acidic environments and for the off-
shore technology must keep the mechanical properties,
such as the resistance to rapture, compression strength,
stress-strain properties and so on, within strict limits. A
breakdown situation caused by a low quality of steel
could have a catastrophic effect on the material and
human losses. The defects of the steel for acidic envi-
ronments have been known more than 50 years. Despite
that, the world oil, gas and engineering companies still
make huge efforts to improve the operations in acidic
environments and to avoid critical situations. The me-
chanical properties for these grades of steel are specified
by the European Standard EN 10020. Their production,
unlike the classic low-carbon steel grades, requires a
special treatment like the soft reduction, electromagnetic
stirring, different cooling conditions, etc., to avoid crack
defects.
The casting of special steel (C0.18, Ni0.04, V0.004,
N0.003 w/%) was performed by the steelmaker Vitkovice
Steel, a. s. A macroscopic examination (the Baumann
method) shows many defects in the final quality of the
steel, such as high porosity, centerline segregation and
cracks.
This paper deals with the results of a numerical
simulation of the temperature field and optimization of a
casting process by analyzing the casting parameters and
their influences on the quality of the steel.6
2 DATA FROM THE MACROGRAPHY
The testing set contains twenty-five samples from
four heats (casting sequences). With a macroscopic test,
we evaluated the cracks in the transverse and longitu-
dinal directions, the centerline segregation according to
SMS DEMAG, the segregation index, the discontinuity
and the lack of homogeneity.
The chemical composition of the steel is in Table 1
and the macrography results are shown in Table 2 and
Figures 2 to 5.
Table 1: Chemical composition of the examined steel in mass frac-
tions, w/%
Tabela 1: Kemijska sestava preiskovanega jekla v masnih dele`ih,
w/%
C Si Mn P Cr
0.18 0.38 1.49 0.021 0.07
S Ni Mo Cu Al
0.004 0.05 0.017 0.028 0.029
Nb Ti V Ca
0.001 0.003 0.004 0.003
Table 2: Macrostructure results
Tabela 2: Ocena makrostrukture
H
ea
t
(c
as
ti
ng
se
qu
en
ce
)
Macrostructure
S
ur
fa
ce
cr
ac
ks
C
ol
um
na
r
cr
ac
ks
C
en
te
rl
in
e
cr
ac
ks
M
id
w
ay
cr
ac
ks
T
ra
ns
ve
rs
e
cr
ac
ks
S
eg
re
ga
ti
on
in
de
x
S
M
S
D
E
M
A
G
26 393 6 106 23 106 5 3 2
26 394 5 107 23 107 5 2 2
26 395 6 103 23 105 5 1-2 2
26 396 5 108 22 107 5 2-3 3
occurrence of defects Centerline segregation, cracks, localdiscontinuity
From the results, it is obvious that the quality of the
steel is not sufficient and has to be improved.
3 NUMERICAL SIMULATIONS
The simulation of the continuous-casting process is
based on a transient numerical model of the temperature
field. This model was specially modified to simulate the
real casting machine operated in Vitkovice Steel, a.s. The
T. MAUDER, J. STETINA: IMPROVEMENT OF THE CASTING OF SPECIAL STEEL WITH ...
4 Materiali in tehnologije / Materials and technology 50 (2016) 1, 3–6
Figure 2: Steel sample from heat 26 393
Slika 2: Vzorec jekla iz taline 26 393
Figure 3: Steel sample from heat 26 394
Slika 3: Vzorec jekla iz taline 26 394
model represents a unique combination of numerical mo-
deling and a large number of experimental measure-
ments. Its results are validated with long-time measure-
ments made during the real casting process. A detailed
description of the numerical model can be found in7.
Thermophysical properties are calculated by the IDS
solidification package. The results for the examined steel
are in Figure 6.
The casting parameters, such as the casting speed, the
pouring temperature, the heat removal from the mold, the
cooling intensity in the secondary cooling zone, etc., for
the numerical simulation were taken from the real
measurement data for heats 26 393–26 396. The results
from the simulation of heat 26 393 are in Figures 7 and
8.
The numerical simulation reveals that the exanimated
steel is characterized by a long mushy zone in compa-
rison with the classic low-carbon steels. The metallur-
gical length reaches 20.08 m and the mushy zone is
almost 10 m long. For the classic low-carbon steels, the
mushy zone is proximately 4–6 m long. The inner
quality of steel is also influenced by the position of the
metallurgical length. The steel should be fully solidified
in close distance to the caster unbending point. The ca-
ster operating in Vitkovice Steel, a.s., has the unbending
point located 12.6 m from the meniscus. So, the defects
can also be caused by the long distance between the
position of the metallurgical length and the unbending
point. These rules8 together with the method for the opti-
mum cooling9 give a set of conditions for the optimiza-
tion of the casting parameters.
4 RESULTS AND DISCUSSION
According to the optimization criteria, the numerical
model and fuzzy-regulation algorithm6 were used to
calculate new casting parameters for the exanimated
steel. In order to get the metallurgical-length position
between 12–15 m, the casting speed has to decrease to
89 % of the original speed. The cooling intensity for the
particular cooling circuit increases, on average, to
T. MAUDER, J. STETINA: IMPROVEMENT OF THE CASTING OF SPECIAL STEEL WITH ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 3–6 5
Figure 8: Distribution of liquid and solid steel
Slika 8: Porazdelitev teko~ega in trdnega jekla
Figure 6: Thermophysical properties of the examined steel
Slika 6: Fizikalno-termi~ne lastnosti preiskovanega jekla
Figure 5: Steel sample from heat 26 396
Slika 5: Vzorec jekla iz taline 26 396
Figure 7: Temperature distribution before optimization
Slika 7: Porazdelitev temperature pred optimizacijoFigure 4: Steel sample from heat 26 395
Slika 4: Vzorec jekla iz taline 26 395
13.4 %. The results of the numerical simulation and opti-
mization are in Figures 9 and 10.
The result was given to the Vitkovice Steel, a.s., as a
casting recommendation for this type of steels. Despite
the fact that the first macroscopic results (Figure 11)
show a quality improvement, more experimental
measurements and numerical simulations are required in
order to get a general view of the behavior of the
examined steel.
5 CONCLUSION
Numerical modeling and optimization will play an
increasing role in the future improvements to the conti-
nuous casting of steel. A combination of the optimiza-
tion algorithm based on fuzzy logic with the numerical
model of the temperature field improves the casting
parameters. With the optimum casting parameters, a
better final quality of cast steel can be achieved. The
algorithm is very general and its calculations can be used
for any steel grade and caster geometry.
Acknowledgement
This work is an output of the research and scientific
activities of NETME Centre, the regional R&D center
built with the financial support from the Operational
Programme Research and Development for Innovations
within the project NETME Centre (New Technologies
for Mechanical Engineering), Reg. No. CZ.1.05/2.1.00/
01.0002 and, in the follow-up sustainability stage,
supported through NETME CENTRE PLUS (LO1202)
with the financial means from the Ministry of Education,
Youth and Sports under the "National Sustainability
Programme I".
6 REFERENCES
1 J. P. Birat et al., The Making, Shaping and Treating of Steel: Casting
Volume, 11th edition, AISE Steel Foundation, Pittsburgh, PA 2003,
1000
2 Crude steel production, The World Steel Association [online],
Brussels, Belgium 2013 [cit. 2013-03-27], http://www.world-
steel.org/
3 C. A. Däcker et al., The History of Mould Slag Films Downwards
the Mould and How it Affects Heat Flux and Shell Growth in Conti-
nuous Casting of Steels, Proceedings of the METEC InSteelCon
2011, Düsseldorf 2011, 8
4 A. Flick, Ch. Stoiber, Trends in Continuous Casting of Steel – Yes-
terday, Today and Tomorrow, Proceedings of the METEC InSteelCon
2011, Düsseldorf 2011, 8
5 Y. H. Chang et al., Development and Application of Dynamic
Secondary Cooling and Dynamic Soft Reduction Control for Slab
Castings, Proceedings of the METEC InSteelCon 2011, Düsseldorf
2011, 6
6 T. Mauder, C. Sandera, J. Stetina, A Fuzzy-Based Optimal Control
Algorithm for a Continuous Casting Process, Mater. Tehnol., 46
(2012) 4, 325–328
7 T. Mauder, C. Sandera, J. Stetina, Optimal control algorithm for con-
tinuous casting process by using fuzzy logic, Steel Res. Int., 86
(2015) 7, 785–798, doi:10.1002/srin.201400213
8 G. S. Jansto, Steelmaking and Continuous Casting Process
Metallurgy Factors Influencing Hot Ductility Behavior of Niobium
Bearing Steels, Proceedings of the METAL 2013, Brno, 2013, 32–39
9 A. A. Ivanova, V. A. Kapitanov, A. V. Kuklev, Method of calculating
the optimum parameters for the air-mist cooling of a continuous-cast
slab, Metallurgist, 56 (2012), 173–179, doi:10.1007/s11015-012-
9555-2
T. MAUDER, J. STETINA: IMPROVEMENT OF THE CASTING OF SPECIAL STEEL WITH ...
6 Materiali in tehnologije / Materials and technology 50 (2016) 1, 3–6
Figure 11: Steel sample from the casting after optimization
Slika 11: Vzorec litega jekla po optimizaciji
Figure 10: Distribution of liquid and solid steel
Slika 10: Porazdelitev teko~ega in trdnega jekla
Figure 9: Temperature distribution after optimization
Slika 9: Porazdelitev temperature po optimizaciji
L. TOPOLÁØ et al.: NON-TRADITIONAL NON-DESTRUCTIVE TESTING OF THE ALKALI-ACTIVATED SLAG ...
7–10
NON-TRADITIONAL NON-DESTRUCTIVE TESTING OF THE
ALKALI-ACTIVATED SLAG MORTAR DURING THE HARDENING
NETRADICIONALNO NEPORU[NO PREIZKU[ANJE Z
ALKALIJAMI AKTIVIRANE MALTE MED STRJEVANJEM
Libor Topoláø, Peter Rypák, Kristýna Tim~aková-[amárková, Lubo{ Pazdera,
Pavel Rovnaník
Brno University of Technology, Faculty of Civil Engineering, Veveri 331/95, 602 00 Brno, Czech Republic
topolar.l@fce.vutbr.cz
Prejem rokopisa – received: 2014-07-30; sprejem za objavo – accepted for publication: 2015-01-30
doi:10.17222/mit.2014.130
This paper reports the results of the measurements of alkali-activated slag mortars made during the hardening and drying of
specimens. The alkali-activated slag is a material with a great potential for practical use. The main drawback of this material is
its high level of autogenous and, especially, drying shrinkage, which causes a deterioration in the mechanical properties. The
aim of this paper is to present the effects of the treatment method for mortars and the curing time on the microstructures of the
alkali-activated slag mortars. The knowledge of the microstructure/performance relationship is the key to a true understanding
of the material behaviour. The results obtained in the laboratory are useful for understanding the various stages of the
micro-cracking activity during the hardening process in quasi-brittle materials such as alkali-activated slag mortars and for their
extension to field applications. Non-destructive acoustic-analysis methods – the impact-echo method as a traditional method and
the acoustic-emission method as a non-traditional method for civil engineering – were used for the experiment. The principle of
the impact-echo method is based on analysing the response of an elastic-impulse-induced mechanical wave. Acoustic emission
is the term for the noise emitted by materials and structures when they are subjected to stress. The types of stress can be
mechanical, thermal or chemical. Ultrasound testing and the loss in mass were used as complementary methods for the tested
samples.
Keywords: acoustic-emission method, loss in mass, impact-echo method, ultrasound testing, alkali-activated slag mortars
^lanek obravnava rezultate meritev med strjevanjem in su{enjem vzorcev malt, aktiviranih z alkalijami. Z alkalijami aktivirana
`lindra je material, ki ima velik potencial za prakti~no uporabo. Glavna pomanjkljivost tega materiala je, da ima sam po sebi, {e
posebno pa pri su{enju, velik skr~ek, ki povzro~i poslab{anje mehanskih lastnosti. Namen tega ~lanka je predstavitev vpliva
metode obdelave malte in ~asa strjevanja na mikrostrukturo z alkalijami aktivirane malte. Razumevanje odvisnosti med
mikrostrukturo in zmogljivostjo je klju~ za pravilno razumevanje vedenja materiala. Rezultati, dobljeni v laboratoriju, so
koristni za razumevanje razli~nih stopenj nastajanja mikrorazpok med procesom strjevanja kvazikrhkega materiala, kot je malta
z `lindro, aktivirano z alkalijami, in za njihov prenos na gradbi{~e. Neporu{ne analizne metode z akusti~no emisijo in metoda
udarec – odmev kot tradicionalne ter metoda akusti~ne emisije kot netradicionalna metoda v gradbeni{tvu, so bile uporabljene
pri preizkusu. Princip metode udarec – odmev temelji na analizi odgovora elasti~nega impulznega mehanskega vala. Akusti~na
emisija je izraz za hrup, ki ga oddajata material in zgradba, ko sta izpostavljena napetosti. Napetosti so lahko mehanske,
termi~ne ali kemijske. Preiskava z ultrazvokom in izguba mase sta bili uporabljeni kot komplementarni metodi pri preizkusnih
vzorcih.
Klju~ne besede: metoda akusti~ne emisije, izguba mase, metoda udarec – odmev, preiskava z ultrazvokom, malta z `lindro,
aktivirano z alkalijami
1 INTRODUCTION
Alkali-activated aluminosilicate materials represent
an alternative to ordinary Portland-cement-based mate-
rials, reducing the impact of the building industry on the
environment and exhibiting new superior properties.
Alkali-activated slag (AAS) is based on granulated
blast-furnace slag that can be activated by alkali hydro-
xides, carbonates or, most preferably, by silicates.1 The
type and dosage of the activator as well as the way of the
curing have significant effects on the hydration course
and final mechanical properties.2 The major disadvantage
of AAS is an increased shrinkage during the hardening
period, caused by both the autogenous and drying
shrinkage, which finally results in a volume contraction,
micro-cracking and deterioration of tensile and bending
properties.3
The impact-echo method (IE) is a type of the non-
destructive testing method. A short-term mechanical
impact, generated by tapping a hammer against the
surface of a concrete structure, produces low-frequency
stress waves which propagate into the structure.4 A wave
generated in this way propagates through the specimen
structure and reflects from the defects located in the
volume of specimen or on its surface. Surface displace-
ments caused by the reflected waves are recorded by a
transducer located adjacent to the impact.5 The signal is
digitized via an analogue/digital data system and trans-
mitted to a computer’s memory. This signal describes the
transient local vibrations, caused by the mechanical-
wave multiple reflections inside the structure. The
dominant frequencies of these vibrations give an account
of the condition of the structure that the waves pass
through.6
Materiali in tehnologije / Materials and technology 50 (2016) 1, 7–10 7
UDK 691:620.179.1 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)7(2016)
Acoustic emission (AE) is the term for the noise
emitted by materials and structures when they are sub-
jected to stress. The types of stresses can be mechanical,
thermal or chemical. This emission is caused by a rapid
release of the energy within a material due to the events
such as a crack formation and its subsequent extension
occurring under the applied stress, generating transient
elastic waves that can be detected by piezoelectric sen-
sors. The acoustic-emission system allows us to monitor
the changes in the material behaviour over a long time
and without moving one of its components, i.e., sensors.
This makes the technique quite unique along with the
ability to detect the crack propagation occurring not only
on the surface but also deep inside the material. The
acoustic-emission method is considered to be a "passive"
non-destructive technique, because it usually identifies
defects while they develop during the test.7
Ultrasonic testing is the name given to the study and
application of ultrasound, which is a sound too high to
be detected by the human ear, i.e., of the frequencies
greater than about 18 kHz. Ultrasonic waves have a wide
variety of applications. For example, ultrasound with
high intensity is used for cutting, cleaning and destroy-
ing a tissue in medicine. For the non-destructive testing
(NDT), ultrasound with a lower intensity is used. An
ultrasonic inspection can be used for a flaw detection/
evaluation, dimensional measurements, a material cha-
racterization and more. Ultrasonic testing (UT) is based
on the propagation of low-amplitude waves through a
material, measuring the time of travel or detecting any
change in the intensity over a given distance. Applica-
tions include distance gauging, flaw detection and
parameter measurement (such as the elastic modulus and
the grain size), all relating to the material structure.8
2 EXPERIMENTAL PART
2.1 Material
The mixture consisted of 450 g of fine-grained
granulated blast-furnace slag [tramberk 380 (a specific
surface area of 380 m2 kg–1), 180 g of sodium silicate
(water glass) with a modulus of 1.6, 1350 g of silica sand
and 95 mL of water. The AAS slurry was poured into
steel moulds (40 mm × 40 mm × 160 mm) to set and
after 24 h the samples were demoulded and immersed in
water for another (2, 6 and 27) d before the testing.
2.2 Experimental set-up
For the impact-echo method, a short mechanical
impulse (a hammer blow) was applied to the surface of a
specimen during the test and detected by means of a
piezoelectric sensor (Figure 1). The impulse reflected
from the surface and also from the micro-cracks and
defects present on the specimen under investigation. The
resonance frequency created in this way was determined
by means of a frequency analysis. Dominant frequencies
could be determined from the response signal by means
of the fast Fourier transform. A MIDI piezoelectric
sensor was used to pick up the response and the respec-
tive impulses were directed into the input of a TiePie
engineering oscilloscope, two-channel Handyscope HS3
with a resolution of 16 bits.
The initiation of cracks during the hardening was
monitored with the method of acoustic emission. AE
signals were detected by measuring equipment DAKEL
XEDO with four channels (Figure 2). The AE sensors
(type IDK09) were attached to the surface with beeswax.
The change in the mass during the hardening was
measured using equipment QuantumX with a Z6
bending-beam load cell for the maximum mass of 50 kg
by HBM.
Measuring equipment PUNDIT (portable ultrasonic
non-destructive digital indicating tester) Plus was used
for the ultrasonic testing. For the testing speed of the
sound through the mortar specimens, the coefficient of
variation for the repeated measurements at the same
location was 2 %. The accuracy of the pulse velocity was
L. TOPOLÁØ et al.: NON-TRADITIONAL NON-DESTRUCTIVE TESTING OF THE ALKALI-ACTIVATED SLAG ...
8 Materiali in tehnologije / Materials and technology 50 (2016) 1, 7–10
Figure 2: Photography of the acoustic-emission measurement
Slika 2: Posnetek meritve akusti~ne emisije
Figure 1: Photography of the impact-echo measurement
Slika 1: Posnetek meritve udarec – odmev
a direct function of the accuracy of the measured
distance between the transducer faces. The PUNDIT
instruments have a transit time resolution of 0.1 s. All
the measurements were carried out for 336 h (14 d)
immediately after the specimens were pulled out of the
immersion water.
3 RESULTS AND DISCUSSION
To evaluate the crack formation during spontaneous
drying, we focused on the activity of AE with respect to
the most used parameter, which is the number of signals
overshooting the pre-set threshold. The diagrams in
Figures 3 to 5 show the dependence of the number of
overshoots and the loss of mass versus the time of
measurement. It was assumed that the number of micro-
cracks could be inferred from the AE activity. Unfortu-
nately, the AE signals originate not only from the crack
formation but also from the process of water evaporation.
However, most of the AE activity was observed within
the first 24 h of spontaneous drying, which corresponds
to approximately 50 % loss in mass. Therefore, at the
beginning the AE signals could be attributed to both the
drying process and the crack formation, whereas after 24
h of drying the observed signals corresponded mainly to
the formation of microcracks. The highest number of
overshoots during the remaining time of the measure-
ment was detected for the specimen that was cured in
water for 2 d (Figure 3). The reason for such a diffe-
rence in comparison with the specimens cured for 6 d
and 27 d arise from a shorter hydration time. Three days
after the mixing, the hydration process was still not
complete and the weak basic structure was not able to
bear a heavy stress; therefore, the AAS matrix was more
susceptible to the cracking caused by drying shrinkage.
To evaluate the signals with the impact-echo method
the fast Fourier transform was used. The modification of
the dominant frequency during the drying process is dis-
played in Figure 6. The results show that the frequencies
decreased from the initial values of 10.10 kHz, 10.86
kHz, 12.12 kHz to the steady values of 5.90 kHz, 7.50
kHz, 8.44 kHz, respectively, for the specimens cured in
water for (2, 6 and 27) d, respectively. Similarly, the
ultrasonic velocity decreased from the initial values of
(3760, 3960, 4450) m s–1 to the steady values of (2010,
2470, 2970) m s–1, respectively, for the specimens
immersed in water for (2, 6 and 27) d, respectively (Fig-
ure 7). The pulse cannot travel across the material/air
interface, but it is able to travel from the transmitter to
the receiver by diffraction at the crack edge. As the travel
path is longer than the distance between the transducers,
the apparent pulse velocity is lower than through the
sound material.
L. TOPOLÁØ et al.: NON-TRADITIONAL NON-DESTRUCTIVE TESTING OF THE ALKALI-ACTIVATED SLAG ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 7–10 9
Figure 6: Change in the dominant frequency over time
Slika 6: Spreminjanje prevladujo~e frekvence s ~asom
Figure 4: Results for the specimen cured in water for 6 d
Slika 4: Rezultati vzorca, ki se je 6 d utrjeval v vodi
Figure 3: Results for the specimen cured in water for 2 d
Slika 3: Rezultati vzorca, ki se je 2 d utrjeval v vodi
Figure 5: Results for the specimen cured in water for 27 d
Slika 5: Rezultati vzorca, ki se je 27 d utrjeval v vodi
The frequencies as well as the ultrasonic velocity
exhibit a decreasing trend of the values. The initial, quite
steep drop corresponds to the evaporation of the water
absorbed in the pore system, leaving air voids that do not
transmit ultrasonic signals. The further decrease
connected with the crack formation caused by drying
shrinkage is bit more moderate. These results are in very
good accordance with the acoustic-emission measure-
ments. Higher absolute values of the ultrasonic velocity
for the specimens cured for longer times are associated
with a denser and more compact structure.
The comparison in the mass loss for variously cured
specimens is given in Figure 8. The loss in mass was
calculated relative to the steady state after spontaneous
drying. The relative mass of the steady state was set to 1.
The specimens cured for 27 d in water lost only 28 % of
mass, whereas the specimens immersed in the water bath
for 6 d and 2 d decreased their mass by 41 % and 46 %,
respectively. It can be assumed that the AAS specimens
that were not completely hydrated were more porous
and, hence, contained higher amounts of the evaporable
water.
4 CONCLUSIONS
The paper deals with the use acoustic non-destructive
methods for monitoring the alkali-activated slag mortars
during the process of drying and hardening. Volume
variations in the alkali-activated slag mortars are
connected with autogenous and drying shrinkage. The
loss in mass observed during the setting and hardening of
AAS is a result of the drying process. The rate of the
moisture release is in good accordance with the number
of signals detected with the AE method. The changes in
the dominant frequency towards lower values detected
with the impact-echo method for all three specimens are
visible and there is also a trend of a decrease in the
ultrasonic velocity, indicating that a large number of new
inhomogeneities appeared in the tested specimens during
their storage in air. It is assumed that most of these
changes can be attributed to the crack formation; there-
fore, it can be concluded that the main process leading to
a deterioration of the AAS binder is the drying shrink-
age.
Acknowledgement
This paper was elaborated with the financial support
of the Czech Science Foundation project CSF No.
13-09518S and the Ministry of Education, Youth and
Sports of the Czech Republic under the "National
Sustainability Programme I" (project No. LO1408
AdMaS UP), as an activity of the regional Centre
AdMaS (Advanced Materials, Structures and Tech-
nologies).
5 REFERENCES
1 C. Shi, P. V. Krivenko, D. Roy, Alkali-Activated Cements and Con-
cretes, Taylor & Francis, Oxon, UK 2006, doi:10.4324/
9780203390672
2 C. Shi, R. L. Day, Some factors affecting early hydration characte-
ristics of alkali-slag cements, Cem. Concr. Res., 26 (1996), 439–447,
doi:10.1016/S0008-8846(96)85031-9
3 M. A. Cincotto, A. A. Melo, E. L. Repetto, Effect of different activa-
tors type and dosages and relation with autogenous shrinkage of
activated blast furnace slag cement, Proceedings of the 11th
International Congress on the Chemistry of Cement, Durban, South
Africa, 2003, 1878–1888
4 M. Sansalone, N. J. Carino, Impact-Echo: A Method for Flaw Detec-
tion in Concrete Using Transient Stress Waves, National Bureau of
Standards, Gaithersburg, Maryland, 1986, NBSIR 86-3452
5 I. Pl{ková, Z. Chobola, M. Matysík, Assessment of ceramic tile frost
resistance by means of the frequency inspection method, Cera-
mics-Silikáty, 55 (2011) 2, 176–182
6 M. T. Liang, P. J. Su, Detection of Corrosion Damage of Rebar in
Concrete Using Impact-Echo Method, Cem. Concr. Res., 31 (2001),
1427–1436, doi:10.1016/S0008-8846(01)00569-5
7 Ch. U. Grosse, M. Ohtsu, Acoustic Emission Testing, Springer-Ver-
lag, Berlin 2008, doi:10.1007/978-3-540-69972-9
8 J. Blitz, G. Simpson, Ultrasonic Methods of Non-Destructive Test-
ing, Springer-Verlag, New York, LLC 1991
L. TOPOLÁØ et al.: NON-TRADITIONAL NON-DESTRUCTIVE TESTING OF THE ALKALI-ACTIVATED SLAG ...
10 Materiali in tehnologije / Materials and technology 50 (2016) 1, 7–10
Figure 8: Change in the loss in mass in relative units over time
Slika 8: Spreminjanje izgube mase v relativnih enotah s ~asom
Figure 7: Change in the ultrasonic velocity over time
Slika 7: Spreminjanje hitrosti ultrazvoka s ~asom
C. E. BAN et al.: MULTI-WALLED CARBON NANOTUBES EFFECT IN POLYPROPYLENE NANOCOMPOSITES
11–16
MULTI-WALLED CARBON NANOTUBES EFFECT IN
POLYPROPYLENE NANOCOMPOSITES
VPLIV VE^STENSKIH OGLJIKOVIH NANOCEVK
V NANOKOMPOZITIH IZ POLIPROPILENA
Cristina-Elisabeta Ban1,2, Adriana Stefan1, Ion Dinca1, George Pelin1,2,
Anton Ficai2, Ecaterina Andronescu2, Ovidiu Oprea2, Georgeta Voicu2
1National Institute for Aerospace Research "Elie Carafoli" Bucharest, Materials Unit, 220 Iuliu Maniu Blvd, 061126 Bucharest, Romania
2University Politehnica of Bucharest, Faculty of Applied Chemistry and Materials Science, 1-7 Polizu St., 011061 Bucharest, Romania
ban.cristina@incas.ro
Prejem rokopisa – received: 2014-07-31; sprejem za objavo – accepted for publication: 2015-02-06
doi:10.17222/mit.2014.142
The paper presents a study concerning thermoplastic nanocomposites having polypropylene as the matrix and different contents
of carboxyl-functionalized multi-walled carbon nanotubes as the nanofiller. The materials are obtained by melt compounding
the nanofiller powder and polymer pellets through the extrusion process followed by injection molding into specific-shape
specimens. The materials are evaluated in terms of mechanical properties such as the tensile and flexural strengths and moduli,
the thermal stability under load (the heat deflection temperature) and the thermal-behavior properties using a TG-DSC analysis.
The fracture cross-section is analyzed using FTIR spectroscopy and SEM microscopy to evaluate the bulk characteristics of the
materials. The results show positive effects of the nanofiller addition to the thermoplastic polymer on the mechanical strength
and modulus of the materials during flexural and tensile tests, while in the case of the thermal stability under load, the nanofiller
addition has a minor influence on the heat-deflection-temperature values.
Keywords: polypropylene, melt mixing, carbon nanotubes, mechanical properties, thermal resistance
^lanek predstavlja {tudijo termoplasti~nih nanokompozitov s polipropilensko osnovo in razli~no vsebnostjo s karboksilom
obdelanih, ve~stenskih ogljikovih nanocevk kot nanopolnilom. Materiali so bili dobljeni iz taline, sestavljene iz prahu
nanopolnila in peletov polimerov, s postopkom ekstruzije, ki mu je sledilo tla~no litje vzorcev. Materiali so ocenjeni glede
mehanskih lastnosti, kot so natezna in upogibna trdnost in moduli, toplotne stabilnosti pri obremenitvi (deformacijska toplota)
in toplotne zna~ilnosti z uporabo TG-DSC analize. Za oceno zna~ilnosti osnovnega materiala je bil analiziran prelom s pomo~jo
FTIR spektroskopije in SEM mikroskopije. Rezultati ka`ejo pozitivne u~inke vpliva dodatka nanopolnila termoplasti~nemu
polimeru na mehansko trdnost in module materiala pri upogibnem in nateznem preizkusu, medtem ko ima dodatek nanopolnila
manj{i vpliv na vrednosti deformacijske toplote pri obremenitvi.
Klju~ne besede: polipropilen, me{anje taline, ogljikove nanocevke, mehanske lastnosti, toplotna obstojnost
1 INTRODUCTION
Polymer nanocomposites found applications in a
wide variety of fields, from microelectronics to aero-
space.1 Carbon nanotube-based polymer composites
combine the good processability of the matrix with the
remarkable functional properties of these nanofillers.
Multi-walled-carbon-nanotube (MWCNT) filled isotactic
polypropylene (PP) nanocomposites can be obtained
through several processing methods, such as melt mix-
ing, solution casting and in-situ polymerization, among
them, melt mixing having some major advantages as it
combines high speed and simplicity with the absence of
solvents and contaminants.2 For the production of these
nanocomposites, a double-screw extruder is a more
appropriate device than a single-screw extruder.3 The
formation of a filler network structure (Figure 1)
depends on several parameters, e.g., the concentration or
dispersion states of the nanotubes in the matrix. Carbon
nanotubes have a tendency to form agglomerates that
lead to a decrease in the surface area, consequently
hindering the structure formation.4 The screw speed is a
strong factor that influences the dispersion of the carbon
nanotubes in polypropylene; too high a speed rate can
generate a mechanical degradation of the final nanocom-
posite as a high shear stress can affect the nanotubes
structure, while too low a speed rate may be insufficient
for an aggregate disentanglement.4
Achieving a good dispersion is influenced also by the
surface optimization between the two phases. The
MWCNT-matrix interfacial-adhesion enhancement can
be obtained by modifying the MWCNT surface through
the non-covalent functionalization that maintains the
nanotube structure or the covalent functionalization, such
as acid treatment creating carboxyl and hydroxyl groups
on the surface, that enhances the load transfer to the
matrix.5 There are studies showing that properties
enhancements are achieved for smaller nanotubes con-
tents and moderate acid-treatment times.5,6
The study presents the characterization of isotactic
polypropylene filled with carboxyl-functionalized
MWCNT obtained through the simple melt-extrusion
technique. The results show an improvement in the
tensile and flexural strengths and moduli when adding
Materiali in tehnologije / Materials and technology 50 (2016) 1, 11–16 11
UDK 678.7:620.3:66.017 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)11(2016)
the nanofiller. In the case of a higher MWCNT loading
(w(MWCNT) = 4 %), the nanocomposites present a
higher stiffness, decomposition temperature and thermal
stability under load, while at a lower loading
(w(MWCNT) = 2 %) they exhibit a better mechanical
strength.
2 EXPERIMENTAL SECTION
2.1 Materials
The matrix was an isotactic polypropylene (TIPP-
LEN H 949 purchased from Basplast SRL) of the homo-
polymer type with a flow index of 45. The nanofiller was
carboxyl-functionalized multi-walled carbon nanotubes
of a 95 % purity (Chengdu Organic Chemicals Co. Ltd,
outer diameter: 10–20 nm, COOH content: w = 2.56 %,
length: 10–30 μm, specific surface area: 233 m2/g,
density: ~2.1 g/cm3).
2.2 Nanocomposites synthesis
The nanocomposite samples were obtained by direct
melt compounding using a twin-screw extruder (Leistritz
LSM 3034 with a 34 mm screw diameter). The poly-
propylene (PP) pellets and nanofiller powder were mixed
at a gradual temperature increase on the ten heating areas
of the extruder, with a temperature profile between
150–170 °C and at a screw speed rate of 220 min–1. The
filaments were cooled in water, chopped, dried and
injected at 165–185 °C into the specimens with a spe-
cific shape. Samples of pure PP, 2 and 4 % of mass frac-
tions of MWCNT (relative to the matrix) were obtained,
as higher contents favor agglomeration and lead to
economically non-viable materials.
2.3 Testing and characterization
The nanocomposites were subjected to a spectro-
scopy analysis (Thermo iN10 MX, mid-infrared FTIR
microscope/ATR mode) and scanning electron micro-
scopy (SEM – HITACHI S2600N microscope), in the
fracture cross-section to highlight the nanofiller pre-
sence. Tensile and flexural tests were performed using an
INSTRON 5982 machine, on a minimum of 5 specimens
per test, according to SR EN ISO 527-27 at a tensile rate
of 50 mm/min, on 1A-type specimens and SR EN ISO
1788 at a test speed of 2 mm/min, for conventional
deflection on rectangular specimens. The thermal-degra-
dation behavior was followed by TG-DSC (Netzsch TG
449C STA Jupiter) heating at 10 K/min, from 25–900 °C,
under a dried-air flow of 10 mL/min. The HDT thermal
stability under load was evaluated using Qualitest HDT1
according to SR EN ISO 75, using a 2 °C/min heating
rate and the standard deflection of 0.34 mm at a flexural
stress of 1.8 MPa. The density was calculated as the ratio
between mass and volume; the volume was measured
using the displacement method.
3 RESULTS AND DISCUSSION
3.1 FTIR spectroscopy
Figure 2 presents the spectra of the isotactic PP in
comparison with the samples nanofilled with 2 and 4 %
of mass fractions of MWCNT. All the spectra present the
characteristic peaks of PP: –CH2 and –CH3 stretching vibra-
tions (2800–2950 cm–1) 9, –CH3 and –CH2 bending
(1376, 1456 cm–1) 10,11, and C–CH3 stretching (841 cm–1) 10.
Calculating the ratio of the absorption bands at 998
and 973 cm–1 (A998/A973) 12 characteristic for the pure
isotactic PP, the isotacticity index for the PP used in this
study was 77.8 %.
There are some differences between the MWCNT
samples due to the matrix-nanofiller interaction.13
Between 1000-1100 cm–1 the signals are more intense,
probably due to the C=O bonding from the COOH
C. E. BAN et al.: MULTI-WALLED CARBON NANOTUBES EFFECT IN POLYPROPYLENE NANOCOMPOSITES
12 Materiali in tehnologije / Materials and technology 50 (2016) 1, 11–16
Figure 2: FTIR spectra of the polypropylene-based nanocomposites
Slika 2: FTIR-spektri nanokompozitov na osnovi polipropilena
Figure 1: Exfoliated nanocomposite formation during the polymer-melt mixing
Slika 1: Nastanek ekspandiranega nanokompozita med me{anjem taline polimera
functionalization.14 The increase in the signal intensity at
1078 cm–1 can be due to the stretching vibration of
C-O.15 Minor differences appear at 1500–1750 cm–1,
more visible in the 4 % of mass fractions of MWCNT-
COOH samples; the weak peak at approximately 1738
cm–1 can be due to the C=O stretching vibration from the
COOH group15 and the modifications at 1550 cm–1 to the
OH groups in C-OH from the nanotubes treatment.16
Minor differences between the FTIR spectra can be
due to the low MWCNT content or a weak connection
with the polymer, probably because of the non-polar
nature of the polypropylene.
3.2 SEM electronic microscopy
SEM highlights the sample morphology at different
magnifications. The cross-section is strongly influenced
by the nanofiller, presenting visible edges and cracks
prior to the tensile-test fracture. The fracture area of the
samples with w(MWCNT) = 2 % is rougher, more like
the simple PP than the ones with w(MWCNT) = 4 %,
due to the lower nanofiller content. At a 500× magni-
fication, a good dispersion of MWCNT can be noticed
(Figure 3). In PP/2 % MWCNT there are some pores,
probably as impurities from the carbon nanotubes. In the
case of PP/4 % MWCNT there are no visible pores, but
there are some non-uniform areas, most likely due to the
higher nanofiller content that increases the agglomera-
tion tendencies. In the case of PP/4 % MWCNT, frac-
ture-initiation sites can be observed (Figure 4).
3.3 Mechanical testing
Figure 5 presents the stress-strain curves of the
replicas of the mediated specimens corresponding to the
samples. Figure 6 presents the load-extension evolution,
illustrating that during the flexural test the materials
mainly exhibit the same behavior; in the first stage,
PP/2 % MWCNT followed the PP trend, then its line was
the same as the one for PP/4 % MWCNT.
Table 1 presents the summary of the mechanical and
heat-deflection results. The nanocomposites exhibit
C. E. BAN et al.: MULTI-WALLED CARBON NANOTUBES EFFECT IN POLYPROPYLENE NANOCOMPOSITES
Materiali in tehnologije / Materials and technology 50 (2016) 1, 11–16 13
Figure 3: SEM images of PP-based nanocomposites compared to simple PP, at 500× magnification
Slika 3: SEM-posnetki nanokompozitov na osnovi PP v primerjavi z enostavnim PP, pov. 500×
Figure 5: Stress-strain curves corresponding to PP-based nanocom-
posites
Slika 5: Krivulje napetost – raztezek nanokompozitov na osnovi PP
Figure 4: SEM images of PP-based nanocomposites at 1500× mag-
nification
Slika 4: SEM-posnetka nanokompozitov na osnovi PP, pov. 1500×
superior characteristics compared to the bare PP; the
increase is more significant for the Young’s modulus,
while the density remains low.
Table 1: Results for mechanical and thermal stability under load tests
Tabela 1: Rezultati mehanske in toplotne stabilnosti pri obremenitvi
Sample PP PP/MWCNT-COOH (2 %)
PP/MWCNT-
COOH (4 %)
Density, g/cm3 0.87 0.91 0.93
Tensile stress at
tensile strength, MPa 35.34 37.56 37.38
Young’s modulus,
MPa 2026.1 2307 2413.71
Flexure stress at
tensile strength, MPa 32.51 35.42 33.59
Young’s flexure
strain, MPa 1439 1547.66 1575.19
HDT, °C 66.9 69.2 70.3
Adding 2 % of mass fractions of MWCNT-COOH
generated an increase in the tensile modulus of about
15 % while 4 % of mass fractions of led to a 20 %
increase compared to the bare PP. The flexural modulus
of the w = 4 % sample presented an increase by 10 %,
proving that adding MWCNT-COOH leads to stiffer
materials.
For the tensile and flexural strengths, the increase
values are in the range of 5–9 % for both the 2 and 4 %
samples, with slightly higher values for the 2 % sample.
This fact can be due to the higher content of the
nanofiller, which favors agglomerations, an issue that
was probably not overcome with the mechanical disper-
sion using the extrusion. At the 2 % loading, there is a
significant increase in the elongation at break, while at
the 4 % loading, both the strength and the elongation at
break decrease compared to the 2 % loading, indicating a
decrease in the ductility.17 At the higher MWCNT con-
tent filler-filler agglomerates are likely to act as stress-
concentrating sites17 as observed in the SEM images,
resulting in lower strength values. The number of the
stress-concentration sites increases with the nanotubes
content, leading to a decrease in the elongation.18
The mechanical properties show that the composite
failure is dependent on the nanofiller content.
3.4 HDT thermal stability under load
The thermal stability under load is in concordance
with the other test results. There is a minor increase in
HDT for the MWCNT sample, from 66.9 to 69.2 °C for
the 2 % loading and 70.3 °C for the 4 % loading, which
is confirmed by the DSC curves of the PP-based mate-
rials that present no change in the thermal behavior up to
200 °C.
3.5 Thermal degradation behavior
The TG curves from Figure 7a indicate that the ma-
terials are thermally stable up to approximately 200–220
°C; after this temperature they undergo thermal-degra-
dation processes.
Table 2 summarizes the TG-DSC results, showing
that a MWCNT-COOH addition leads to an increase in
the initial temperature of the degradation process (Tonset)
on the TG curve of approximately 65 °C. The 50 %
weight loss is generally considered to be an indicator of
the structural destabilization19, which occurs up to
C. E. BAN et al.: MULTI-WALLED CARBON NANOTUBES EFFECT IN POLYPROPYLENE NANOCOMPOSITES
14 Materiali in tehnologije / Materials and technology 50 (2016) 1, 11–16
Figure 7: a) TG and b) DSC curves for PP/MWCNT nanocomposites
compared to simple PP
Slika 7: a) TG- in b) DSC-krivulje za PP/MWCNT kompozite v pri-
merjavi z enostavnim PP
Figure 6: Load-extension evolution during the flexural testing of
PP-based nanocomposites
Slika 6: Obna{anje obremenitev – raztezek med upogibnim preizku-
som nanokompozitov na osnovi PP
approximately 330 °C for PP, up to 370 °C for PP/2 %
MWCNT and 390 °C for PP/4 % MWCNT. The end of
the degradation process (Tendset) is shifted towards higher
values as the MWCNT content increases.
Table 2: TG-DSC analysis results
Tabela 2: Rezultati TG-DSC analiz
Sample Tonset,°C
Tendset,
°C
Tmelting,
°C
Tdecomp1,
°C
Tdecomp2,
°C
PP 263.3 390 167.9 279.2 410/461.5/496.7
PP/2 %
MWCNT 326.5 408.7 166.8 362.6
441.4/487.4/
546.4
PP/4 %
MWCNT 327.2 441.2 167.3 365.2
471.4/500.9/
561.4
In the 25–200 °C region, an endothermic effect is
registered on the DSC curve (Figure 7b), associated
with the polymer melting process. As melting is a
physical process, the effect is not accompanied by the
weight loss. The maximum rate of the endothermic effect
is reached at approximately the same temperature (167
°C) for all the samples, showing that the MWCNT
addition does not induce changes in the melting tempe-
rature, but the intensities of the corresponding peaks are
lower for the MWCNT samples.
In the regions above 200 °C, exothermic effects
appear, associated with the decomposition processes that
occur in two main stages, mainly between 200–400 °C
and 400–600 °C, divided into several secondary exother-
mic effects, accompanied by the weight loss. The re-
corded effects for the PP-based materials are similar, but
because of their different compositions, the peak inten-
sities differ.
In the 200–400 °C region, the maximum weight loss
occurs for all the samples, being approximately 94 %.
On the DSC curve, the peaks corresponding to the exo-
thermic effects are shifted towards higher values; the
peak for PP appears at 279 °C, for PP/2 % MWCNT it
appears at 363 °C and for PP/4 % MWCNT it appears at
365 °C, but with a lower intensity for the MWCNT
samples. The increase in the decomposition temperature
of the MWCNT nanocomposites could be due to the
barrier effect generated by the nanotubes, when they are
well dispersed into the matrix, hindering oxygen
diffusion and retarding the thermo-oxidative degradation
of polypropylene.20,21 MWCNT acts as the protective
agent against a thermal degradation of polypropylene.
Between 400–600 °C, the exothermic effects are
accompanied by a weight change of approximately 6 %
for all the samples. Also, in this region, the exothermic
effects recorded on the DSC curves of PP/MWCNT
appear at higher temperatures and with higher intensities,
with the difference increasing with the MWCNT content.
This shows that the protective effect is more pronounced
at higher temperatures and higher MWCNT contents.
The residual mass is extremely low (up to w = 0.5 %
for all the samples), proving that although the
PP/MWCNT nanocomposites burn slower than the bare
PP, they burn nearly completely, indicating that the
eventual flame-retardancy properties of these materials
are probably due to the chemical and physical processes
in the condensed phase rather than the gas phase.22
The TG-DSC analysis results prove that the addition
of MWCNT improves the decomposition temperature of
the polypropylene nanocomposites and, consequently,
the thermal-degradation resistance.
4 CONCLUSIONS
The study presents a characterization of an isotactic
polypropylene filled with carboxyl-functionalized
MWCNT prepared through the simple and quick way of
the melt-extrusion technique. The results show an
improvement in the mechanical strength and modulus,
thermal stability under load as well as decomposition
temperature when adding the nanofiller. In the case of
the higher MWCNT loading (w(MWCNT) = 4 %), the
nanocomposites exhibited higher stiffness, decompo-
sition temperature and thermal stability under load, while
in the case of the lower loading (w(MWCNT) = 2 %) the
nanocomposites exhibited better mechanical strengths.
The nanocomposites maintained their low-density
advantages showing a minor increase when adding the
nanofiller. The results prove that the thermoplastic
polymers loaded with carbon nanotubes might be a new
class of light and strong composites that could find
applications in a large variety of fields.
Further compatibilization of the polypropylene ma-
trix by grafting it with different agents such as maleic
anhydride, methylstyrene23 and copolymers based on
maleic anhydride24 can lead to nanocomposites with even
higher mechanical and thermal properties, due to an
increased matrix-filler adhesion.
Acknowledgments
This work was funded by the Romanian Ministry of
Education through the PN-II-PT-PCCA-168/2012 project
"Hybrid composite materials with thermoplastic matrices
doped with fibres and disperse nano-fillings for materials
with special purposes" and by the Sectoral Operational
Programme "Human Resources Development 2007–
2013" of the Ministry of European Funds through the
Financial Agreement POSDRU/159/1.5/S/132397.
5 REFERENCES
1 F. Hussain, M. Hojjati, M. Okamoto, R. E. Gorga, J. Compos. Mater.,
40 (2006) 17, 1511–1565, doi:10.1177/0021998306067321
2 E. Logakis, E. Pollatos, Ch. Pandis, V. Peoglos, I. Zuburtikudis, C.
G. Delides, A. Vatalis, M. Gjoka, E. Syskakis, K. Viras, P. Pissis,
Compos. Sci. Technol., 70 (2010) 2, 328–335, doi:10.1016/
j.compscitech.2009.10.023
3 A. Szentes, G. Horvath, Cs. Varga, Hungarian Journal of Industrial
Chemistry, 38 (2010) 1, 67–70
C. E. BAN et al.: MULTI-WALLED CARBON NANOTUBES EFFECT IN POLYPROPYLENE NANOCOMPOSITES
Materiali in tehnologije / Materials and technology 50 (2016) 1, 11–16 15
4 T. Y. Hwang, H. J. Kim, Y. Ahn, J. W. Lee, Korea-Australia Rheo-
logy Journal, 22 (2010) 2, 141–148
5 D. Bikiaris, A. Vassilou, K. Chrissafis, K. M. Paraskevopoulos, A.
Jannakoudakis, A. Docoslis, Polym. Degrad. Stab., 93 (2008) 5,
952–967, doi:10.1016/j.polymdegradstab.2008.01.033
6 S. P. Bao, S. C. Tjong, Mater. Sci. Eng. A, 458 (2007) 1–2, 508–516,
doi:10.1016/j.msea.2007.08.050
7 European Standard SR EN ISO 527-2: Determination of tensile
properties of plastics, Test conditions for moulding and extrusion
plastics, 2000
8 European Standard SR EN ISO 178: Plastics, Determination of
flexural properties, 2003
9 F. Ozmihci, Polypropylene – Natural Zeolite Composite Films,
Dissertation Thesis, Materials Science and Engineering Department,
Izmir Institute of Technology, 1999, p. 48
10 I. Karacan, H. Benli, The use of infrared-spectroscopy technique for
the structural characterization of isotactic polypropylene fibres,
Journal of Textile & Apparel, 21 (2011) 2, 116–123
11 G. Parthasarthy, M. Sevegney, R. M. Kannan, J. Polym. Sci., Part B:
Polym. Phys., 40 (2002) 22, 2539–2551, doi:10.1002/polb.10304
12 A. R. Horrocks, J. A. D’souza, J. Appl. Polym. Sci., 42 (1991) 1,
243–261, doi:10.1002/app.1991.070420129
13 L. V. Diyakon, O. P. Dmytrenko, N. P. Kulish, Yu. I. Prylutskyy, Yu.
E. Grabovskiy, N. M. Belyy, S. A. Alekseev, A. N. Alekseev, Yu. I.
Sementsov, N. A. Gavrylyuk, V. V. Shlapatskaya, L. Valkunas, R.
Ritter, P. Scharff, Functional Materials, 15 (2008) 2, 169–174
14 V. T. Le, C. L. Ngo, Q. T. Le, T. T. Ngo, D. N. Nguyen, M. T. Vu,
Adv. Nat. Sci.: Nanosci. Nanotechnol, 4 (2013) 3, 1–5, doi:10.1088/
2043-6262/4/3/035017
15 S. C. Her, C. Y. Lai, Materials, 6 (2013) 6, 2274–2284, doi:10.3390/
ma6062274
16 C. R. Biswal, K. Mishra, P. L. Nayak, Synthesis and Characterization
of Modified Multi-Walled Carbon Nanotubes Filled Thermoplastic
Natural Rubber Composite, Middle-East Journal of Scientific
Research, 18 (2013) 2, 168–176, doi:10.5829/idosi.mejsr.2013.18.
2.12431
17 S. A. Girei, S. P. Thomas, M. A. Atieh, K. Mezghani, S. K. De, S.
Bandyopadhyay, A. Al-Juhani, J. Thermoplast. Compos. Mater., 25
(2012) 3, 333–350, doi:10.1177/0892705711406159
18 S. R. Katti, B. K. Sridhara, L. Krishnamurthy, G. L. Shekar, Mecha-
nical Behaviour of MWCNT Filled Polypropylene Thermoplastic
Composites, Indian Journal of Advances in Chemical Science, 2
(2014), 6–8
19 E. M. Sabri, O. Emel, Fibres & Textiles in Eastern Europe, 21 (2013)
2, 22–27
20 F. Avalos-Belmontes, L. F. Ramos-deValle, E. Ramirez-Vargas, S.
Sanchez-Valdes, J. Mendez-Nonel, R. Zitzumbo-Guzman, Journal of
Nanomaterials, 2012 (2012), 1-8, doi:10.1155/2012/406214
21 N. Khelidj, X. Colin, L. Audouin, J. Verdu, C. Monchy-Leroy, V.
Prunier, Polym. Degrad. and Stab., 91 (2006) 7, 1593–1597,
doi:10.1016/j.polymdegradstab.2005.09.011
22 T. Kashiwagi, E. Grulke, J. Hilding, R. Harris, W. Awad, J. Douglas,
Macromol. Rapid Commun., 23 (2002) 13, 761–765
23 E. Manias, A. Touny, L. Wu, K. Strawhecker, B. Lu, T. C. Chung,
Chem. Mater., 13 (2001) 10, 3516–3523, doi:10.1021/cm0110627
24 A. Szentes, C. Varga, G. Horváth, L. Bartha, Z. Kónya, H. Haspel, J.
Szél, Á. Kukovecz, eXPRESS Polymer Letters, 6 (2012) 6, 494–502,
doi:10.3144/expresspolymlett.2012.52
C. E. BAN et al.: MULTI-WALLED CARBON NANOTUBES EFFECT IN POLYPROPYLENE NANOCOMPOSITES
16 Materiali in tehnologije / Materials and technology 50 (2016) 1, 11–16
J. HRABOVSKÝ et al.: EXPERIMENTAL AND NUMERICAL STUDY OF HOT-STEEL-PLATE FLATNESS
17–21
EXPERIMENTAL AND NUMERICAL STUDY OF
HOT-STEEL-PLATE FLATNESS
EKSPERIMENTALNI IN NUMERI^NI [TUDIJ RAVNOSTI VRO^IH
PLO[^ IZ JEKLA
Jozef Hrabovský1, Michal Pohanka1, Pil Jong Lee2, Jong Hoon Kang2
1Heat Transfer and Fluid Flow Laboratory, Faculty of Mechanical Engineering, Brno University of Technology, Technická 2, 616 69 Brno,
Czech Republic
2POSCO, Rolling Technology & Process Control Research Group, 1, Geodong-dong, Nam-gu, Pohang, Gyeongbuk 790-785, Korea
hrabovsky@fme.vutbr.cz
Prejem rokopisa – received: 2014-07-31; sprejem za objavo – accepted for publication: 2015-02-10
doi:10.17222/mit.2014.153
One aspect of the steel-product quality is the flatness of a steel plate. This is one of the reasons why it is important to describe
and understand the process of steel-plate deformation during the cooling process. The temperature distribution has the largest
impact on the deformation of a strip. The temperature distribution is affected by the cooling process. The cooling homogeneity
or inhomogeneity is the most important factor influencing the final flatness of a cooled strip or steel plate. Inhomogeneous
cooling can lead to large differences in the thermal distribution inside the material and also to high deformations. The cooling
homogeneity is mainly influenced by the water distribution in the cooling section. The goal of this paper is to experimentally
and numerically study and describe the deformation process of a hot steel plate during the cooling process. To meet these goals,
experimental measurements of a cooled steel plate were carried out and the boundary conditions and temperature field were
obtained. Based on this data, two numerical models were created. The first numerical model focused on the cooling process, the
thermal-field simulation and the input-data preparation for the next step. In the next step, the second numerical model was
generated using the finite-element method and an analysis of the structure and deformation of the steel plate was simulated. The
description of the shape deformation of cooled steel plates should lead to an improved flatness of final products.
Keywords: cooling process, deformation, flatness, numerical simulation
Eden od vidikov kvalitete izdelka iz jekla je ravnost jeklene plo{~e. To je eden od razlogov, zakaj je potrebno opisati in razumeti
postopek deformiranja plo{~e med njenim ohlajanjem. Razporeditev temperature ima glavni vpliv na deformacijo traku. Na
razporeditev temperature vpliva proces ohlajanja. Homogenost ali nehomogenost hlajenja je najpomembnej{i faktor, ki vpliva na
kon~no ravnost ohlajenega traku ali plo{~e. Nehomogeno hlajenje lahko povzro~i velike razlike v razporeditvi toplote znotraj
materiala in tudi velike deformacije. Na homogenost hlajenja najbolj vpliva razporeditev vode v podro~ju hlajenja. Namen tega
~lanka je eksperimentalni in numeri~ni {tudij ter opis procesa deformacije vro~e jeklene plo{~e med procesom hlajenja. Za
dosego namena so bile eksperimentalno izmerjene hlajene jeklene plo{~e in dobljeni so bili mejni pogoji in temperaturno polje.
Na osnovi teh podatkov sta bila postavljena dva numeri~na modela. Prvi numeri~ni model je bil osredoto~en na proces hlajenja,
simulacijo temperaturnega polja in pripravo vhodnih podatkov za naslednji korak. V naslednjem koraku je bil generiran drugi
numeri~ni model s pomo~jo metode kon~nih elementov in simulirana je bila analiza strukture ter deformacije jeklene plo{~e.
Opis deformacije oblike ohlajanih jeklenih plo{~ naj bi omogo~il bolj{o ravnost kon~nih proizvodov.
Klju~ne besede: proces hlajenja, deformacija, ravnost, numeri~na simulacija
1 INTRODUCTION
The flatness of a hot steel plate or strip is an import-
ant issue of heat treatment in the steel industry. POSCO
Korea is also dealing with this problem. The flatness of a
final product has a significant impact on the quality and
price of the strip. The flatness of a steel plate is mainly
affected by the water distribution and cooling homo-
geneity during the quenching process. Two important
non-homogeneous types of water distribution can occur.
The first type of non-homogeneity is caused by different
intensities of the cooling from the top and bottom sides
of a steel plate. The second type of non-homogeneity
during a cooling process is created by different water
distributions in the center and corners of a steel plate.
Both of these types can occur simultaneously and pro-
duce problems with regard to the flatness of a steel
plate1,2. Non-homogeneous cooling also affects other
aspects such as material properties, phase changes,
residual stresses and so on3–5. The combination of all the
aspects has a significant impact on the final quality of a
steel plate and, therefore, the cooling process must be
studied. The description of the cooling process can be
performed using several parameters, including the cool-
ing intensity, the distribution of the heat-transfer coeffi-
cient (HTC) and so on6,7. Experimental measurements
were carried out at defined conditions to describe the
cooling process and non-homogeneity identification
during the quenching. The HTC distribution was ob-
tained from the results and a model of the steel-surface
thermal field was prepared. The inverse problem with a
sequential estimation of the time and varying boundary
conditions was used to simulate the cooling process and
time-dependent boundary conditions8. The calculated
thermal field of the steel plate was used in the next
numerical simulation. This numerical simulation was
Materiali in tehnologije / Materials and technology 50 (2016) 1, 17–21 17
UDK 519.61/.64:536.413:62-415 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)17(2016)
focused on steel-plate deformations and an analysis of
the flatness. The numerical simulation was based on the
finite-element method in Ansys, a commercial software
program.
2 EXPERIMENTAL MEASUREMENT OF
COOLING HOMOGENEITY
The cooling process under defined conditions was
experimentally measured using a laboratory apparatus
prepared in the Heat Transfer and Fluid Flow Laboratory.
The experimental apparatus is shown in Figure 1. The
experimental measurements were focused on the study of
the cooling non-homogeneity in the longitudinal and
transverse directions of the test plate and also of the
differences in the water distribution from the top and
bottom sides.
A test plate equipped with three thermocouple
sensors was used to investigate the heat transfer in the
longitudinal and transverse directions. A schematic of
the experimental measurement is depicted in Figure 2.
The thermocouples recorded the temperature history dur-
ing the cooling process and, using an inverse analysis,
the surface temperatures and heat-transfer coefficients
were computed. Both the upper and lower surfaces were
investigated.
The measurements were performed for several
conditions. The examples of the experimental conditions
are presented in Table 1. For the cooling from the top, a
pool with an adjustable height was used to set the proper
water layer.
Table 1: Examples of the measurement conditions
Tabela 1: Primeri pogojev pri meritvah
Surface Spray distance(mm)
Flow rate
(m3/min)
Water layer
(mm)
Velocity
(m/s)
Top 400 3.0 54 0.8
Top 400 4.5 65 0.8
Bottom 50 3.0 – 0.8
Bottom 50 5.3 – 0.8
The temperature history recorded during the experi-
ment was used to compute the time-dependent boundary
conditions. These boundary conditions were computed
using the sequential estimation of the time-varying
boundary conditions and the HTCs were obtained using
the inverse solution of the problem for each measure-
ment. Using the applied calculation methodology, the
HTC distribution on the steel plate was obtained. The
evaluated data for the full temperature range of the
experiments performed for the upper and lower cooling
are shown in Figure 3. These figures represent the HTC
distribution as a function of the surface temperature and
the position.
J. HRABOVSKÝ et al.: EXPERIMENTAL AND NUMERICAL STUDY OF HOT-STEEL-PLATE FLATNESS
18 Materiali in tehnologije / Materials and technology 50 (2016) 1, 17–21
Figure 3: HTC distribution for: a) upper and b) bottom cooling
Slika 3: HTC-razporeditev za hlajenje: a) zgoraj in b) spodaj
Figure 2: Schematic illustration of the experimental measurement
Slika 2: Shemati~en prikaz izvedbe meritev
Figure 1: Experimental apparatus
Slika 1: Eksperimentalna naprava
3 NUMERICAL SIMULATION
The numerical simulation was prepared in two steps.
In the first step, a numerical model of the thermal field
was prepared. In this step, FDM methods were consi-
dered as the solver for the heat conduction. The second
step was focused on the preparation of the numerical
model for the structural analysis of the steel plate. The
second numerical model was based on the FEM.
3.1 Temperature-field simulation
A cooling test plate was simulated to verify the HTC
values obtained with the laboratory measurements. The
selected discretization equations obtained using the FDM
have a clear physical meaning – they are not simply a
formal mathematical approximation. The derived
equations represent the conservation principles (mainly
the energy conservation) for each control volume and the
resulting numerical solution correctly satisfies the con-
servation over the whole calculation domain. As the
conductivities of the neighboring volumes may be
different due to the different temperature, there may be a
discontinuity of the slope (dT/dx) at the control volume
boundary. Whenever there is a need to get the tempera-
ture at a control volume boundary (e.g., during the inter-
polation of the temperature across the control volume), a
detailed equation should be used together with the
temperature at the node.
When the mass density or specific heat becomes
more dependent on the temperature, special care must be
paid to the time integration: if the time step is too large,
the latent heat of the phase change may be neglected. In
the case of a phase change, the properties are highly
dependent on the temperature. This problem can be
eliminated by solving it using the enthalpy instead of the
specific heat and the mass density. The approach that
was used and overcomes this problem was inspired by
the apparent-heat-capacity method.
The boundary conditions were prepared from the data
presented in the previous section (Figure 3). The data
was considered as a function of three variables: HTC
(length, Tsurface, width). An example of the applied boun-
dary conditions at a time of 32 s for the upper cooling of
the test plate is shown in Figure 4.
The computed temperature histories for the entire
cooling process are shown in Figure 5. This figure
indicates a large temperature difference between the
surface and the center of the plate in the middle of the
plate. The applied method for the temperature-field
simulation agrees with the plant measurements.
3.2 Simulation of deformation
This section contains the structural analysis of the
steel plate focusing on the deformations. The measured
thermal loading produced by the simulated cooling pro-
cess was described in the previous chapter. The purpose
of this calculation was to assess the impact of the cooling
process on the deformation of the steel plate produced
experimentally.
The calculation methodology for the structural
analysis of the FE model consists of three steps. In the
first step, the 3D finite-element model was created with
respect to the real dimensions and conditions. In the
second step, the calculated temperature field, the struc-
tural boundary conditions and the initial conditions were
applied. In the final step, the analysis of the structure and
deformation of the steel plate was calculated. Two types
of loads were considered for the prepared FE model: the
experimentally measured thermal field as the thermal
load and the mechanical loads represented by the move-
ment of the plate. The analysis of the steel-plate defor-
mation due to different conditions was carried out for
two cases. In the first case, the specimen was tested
under measured cooling conditions, and the second case
was the same as the first case (the same cooling con-
ditions) but without considering the gravity. The case
without the gravity acceleration was performed to obtain
the true deformation due to the cooling process.
J. HRABOVSKÝ et al.: EXPERIMENTAL AND NUMERICAL STUDY OF HOT-STEEL-PLATE FLATNESS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 17–21 19
Figure 5: Center and surface temperature histories in the middle of
the plane
Slika 5: Temperaturna zgodovina sredine in povr{ine na sredini
povr{ine
Figure 4: Temperature profile (different scales on the axes)
Slika 4: Profil temperature (razli~na merila po oseh)
The thermal loads were FDM analyzed and applied
as the body loads of the steel plate. The temperature was
evaluated at the center of the steel plate along a defined
path (the symmetry plane) on the top and bottom sides.
The evaluation of the temperature distribution along the
defined path is shown in Figure 6. The temperature
distributions (Figure 6) show that at the beginning (14 s)
and at the end of the cooling (100.5 s), the top and
bottom temperatures are almost equal. At a time of
19.5 s it is clear that the cooling is less intense on the top
and that the bottom temperatures are lower.
Different values of the temperature on both sides of
the steel plate lead to different stresses and deformations.
The presented thermal field was used for the structural
analysis and the deformation of the steel plate was cal-
culated. The final vertical deformation at the end of the
cooling reached 2.8 mm. The maximum vertical defor-
mation is located at the corner of the steel plate.
The deformation of the steel plate along the path at
the corner and along the path at the center was evaluated
(Figure 7). The comparison of these two locations
confirmed two reasons for the maximum deformation in
the corner. The first reason is the non-homogeneous ther-
mal field at the corner. The second reason is that the
corner is free and a deformation is possible. In the
center, the steel plate is constrained by the symmetry (in
reality the steel plate would continue). The gravitational
acceleration contributes to a decrease in the total
deformation.
For the second case, the same type of cooling as for
the first case was used and the gravitational acceleration
was not considered. The final vertical deformation at the
end of the cooling reached 596 mm. The maximum ver-
tical deformation is located symmetrically at the
beginning of the steel plate. The deformations of the
steel plate in this case are mainly affected by the thermal
distribution and the thermal gradients.
The presented deformation results confirm that the
more intense cooling from the bottom side of the steel
plate leads to the final deformation on the lower-cooling
side of the steel plate (the top side). This effect can be
explained with the following mechanism. On the bottom
side, which is cooled more intensely, the temperature
drops quickly and the stiffness (K) represented by the
modulus of elasticity (E) of the bottom layer is changed
to a higher value. In the middle part of the steel plate, the
temperature is still high and the stiffness is low (the
modulus of elasticity at a higher temperature). On the top
side of the steel plate, the temperature also decreases but
not as intensely as on the bottom side. The stiffness of
the top layer is lower than the stiffness of the bottom
layer.
The combination of the temperature and the stiffness
distribution of defined layers leads to different directions
of the deflection (U) in each layer. The highest deflection
in the axial direction of the steel plate is in the middle
layer due to the higher temperature and thermal expan-
J. HRABOVSKÝ et al.: EXPERIMENTAL AND NUMERICAL STUDY OF HOT-STEEL-PLATE FLATNESS
20 Materiali in tehnologije / Materials and technology 50 (2016) 1, 17–21
Figure 6: Evaluated temperatures through the path at defined time
points
Slika 6: Ocenjene temperature skozi sredino, v dolo~enih ~asovnih
trenutkih
Figure 8: Schematic illustration of the steel-plate deformation mecha-
nism
Slika 8: Shemati~en prikaz mehanizma deformacije plo{~e iz jekla
Figure 7: Evaluation of the total vertical deformation at a time of
100.5 s
Slika 7: Ocena skupne vertikalne deformacije pri ~asu 100,5 s
sion. On the bottom side of the steel plate, deflection
occurs in the opposite direction than in the middle layer
due to intense cooling. At the top layer, the lower cool-
ing intensity leads to a deflection in the same direction as
at the bottom layer, but the deflection is lower than on
the bottom side. These deflection distributions through
the thickness of the steel plate generate a deformation in
the intensely cooled side direction.
Non-homogeneous cooling on the bottom and top
sides of the steel plate cause a time shift in the tempe-
rature distribution on both sides. The time shift in the
temperatures on both sides causes a deflection inverted
from the bottom and top layers, changing the final defor-
mation of the steel plate in the direction of the actual
cooled side (the top side). A schematic illustration of the
deformation of the steel mechanism due to non-homo-
geneous cooling is depicted in Figure 8. The elastic-
plastic material properties of steel are considered in the
numerical model and, therefore, a permanent deforma-
tion occurs.
4 CONCLUSION
The analyses presented in this paper focused on the
study of non-homogeneous cooling and its impact on the
deformation of a steel plate. Several numerical models of
steel plates were prepared. The first model computed
time-dependent temperature fields. Plant measurements
were simulated using this model. The results obtained
from the simulation agree with data obtained during the
plant measurements. The second numerical model
focused on the cooling process of the steel plate and the
impact of thermal fields on the final deformations of the
steel plate. The FE simulation of the cooling process
showed the impact of the non-homogeneity in the
thermal field on the final deformations. The simulations
confirmed that the plate is bent towards the side with the
higher cooling intensity in the initial cooling stage;
however, in the later stages, the plate is bent towards the
opposite side, with the lower cooling intensity.
Acknowledgement
The paper presented was supported through project
CZ.1.07/2.3.00/30.0005 of the Brno University of
Technology.
This work is an output of the research and scientific
activities of the NETME Centre, regional R&D center
built with the financial support from the Operational
Programme Research and Development for Innovations
within the project NETME Centre (New Technologies
for Mechanical Engineering), Reg. No. CZ.1.05/2.1.00/
01.0002 and, in the follow-up sustainability stage,
supported through NETME CENTRE PLUS (LO1202)
by the financial means from the Ministry of Education,
Youth and Sports under the National Sustainability
Programme I.
5 REFERENCES
1 X. Wang, Q. Yang, A. He, Calculation of thermal stress affecting
strip flatness change during run-out table cooling in hot steel strip
rolling, Journal of Materials Processing Technology, 207 (2008),
130–146, doi:10.1016/j.jmatprotec.2007.12.076
2 Y. J. Jung, G. T. Lee, C. G. Kang, Coupled thermal deformation
analysis considering strip tension and with/without strip crown in
coiling process of cold rolled strip, Journal of Materials Processing
Technology, 130–131 (2002), 195–201, doi:10.1016/S0924-0136
(02)00705-7
3 H. N. Han, J. K. Lee, H. J. Kim, Y. S. Jin, A model for deformation,
temperature and phase transformation behavior of steel on run-out
table in hot strip mill, Journal of Materials Processing Technology,
128 (2002), 216–225, doi:10.1016/S0924-0136(02)00454-5
4 X. Wang, F. Li, Q. Yang, A. He, FEM analysis for residual stress
prediction in hot rolled steel strip during the run-out table cooling,
Applied Mathematical Modelling, 37 (2013), 586–609, doi:10.1016/
j.apm.2012.02.042
5 S. Serajzadeh, Prediction of temperature distribution and phase
transformation on the run-out table in the process of hot strip rolling,
Applied Mathematical Modelling, 27 (2003), 861–875, doi:10.1016/
S0307-904X(03)00085-4
6 M. Chabicovsky, M. Raudensky, Experimental Investigation of a
Heat Transfer Coefficient, Mater. Tehnol., 47 (2013) 3, 395–398
7 M. Raudensky, Heat Transfer Coefficient Estimation by Inverse Con-
duction Algorithm, International Journal of Numerical Methods for
Heat and Fluid Flow, 3 (1993) 3, 257–266, doi:10.1108/eb017530
8 M. Pohanka, P. Kotrbacek, Design of cooling units for heat
treatment, In: F. Czerwinski (Ed.), Heat Treatment – Conventional
and Novel Applications, InTech, 2012, 1–20, doi:10.5772/50492
J. HRABOVSKÝ et al.: EXPERIMENTAL AND NUMERICAL STUDY OF HOT-STEEL-PLATE FLATNESS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 17–21 21

A. KURªUN, E. TOPAL: INVESTIGATION OF HOLE EFFECTS ON THE CRITICAL BUCKLING LOAD ...
23–27
INVESTIGATION OF HOLE EFFECTS ON THE CRITICAL
BUCKLING LOAD OF LAMINATED COMPOSITE PLATES
PREISKAVA VPLIVA LUKNJE NA KRITI^NO UPOGIBNO
OBREMENITEV LAMINIRANIH KOMPOZITNIH PLO[^
Ali Kurºun, Ersin Topal
Department of Mechanical Engineering, Hitit University, 19030 Çorum, Turkey
alikursun@hitit.edu.tr
Prejem rokopisa – received: 2014-08-01; sprejem za objavo – accepted for publication: 2015-03-04
doi:10.17222/mit.2014.164
In this study, the effects of the hole diameter on the buckling behavior of glass-fiber-reinforced, laminated composite
rectangular plates were investigated both experimentally and numerically. As the test specimens, E-glass/epoxy symmetric-ply
composites with eight layers were manufactured using the hand lay-up technique and drilled with different hole diameters
ranging from 10 to 40 mm. The laminated composite plates were arranged with different symmetric orientation angles such as
[(0°/90°)2]s, [(–15°/15°)2]s, [(–30°/30°)2]s and [(–45°/45°)2]s. The experimental critical-buckling loads of the plates were found
by clamping the bottom and upper edges and then these results were compared with the results obtained with the numerical
analysis. The determination of the critical buckling loads for the laminated composite plates with different hole diameters was
performed using the ANSYS 12.1® finite-element-analysis software. The numerical analysis showed good agreement with the
experimental results for different hole diameters and fiber orientation angles. It was concluded that the critical buckling loads
strongly depend on the diameter of the hole and fiber orientation angles.
Keywords: buckling, glass fibers, finite element method (FEM)
V tej {tudiji je bil eksperimentalno in numeri~no preiskovan vpliv premera luknje na obna{anje pri upogibu s steklenimi vlakni
laminirane kompozitne {tirioglate plo{~e. Kompozitni vzorci so bili izdelani s simetri~nim polaganjem E-steklo/epoksi, z
osmimi plastmi, z uporabo tehnike ro~nega polaganja in vrtanja lukenj razli~nega premera od 10 do 40 mm. Laminirane
kompozitne plo{~e so bile izdelane z razli~nimi, simetri~no orientiranimi koti vlaken, [(0°/90°)2]s, [(–15°/15°)2]s, [(–30°/30°)2]s
in [(–45°/45°)2]s. Eksperimentalna kriti~na upogibna obremenitev plo{~ je bila ugotovljena z vpetjem spodnjega in zgornjega
roba, rezultati pa so bili primerjani z rezultati numeri~ne analize. Dolo~itev kriti~ne upogibne obremenitve za laminirane kom-
pozitne plo{~e z luknjami z razli~nim premerom, je bila izvr{ena z uporabo programa za analizo kon~nih elementov ANSYS
12.1®. Numeri~na analiza je pokazala dobro ujemanje z rezultati preizkusov, za razli~ne premere lukenj in orientacijske kote
vlaken. Ugotovljeno je, da je kriti~na upogibna obremenitev mo~no odvisna od premera luknje in od kota orientacije vlaken.
Klju~ne besede: upogibanje, steklena vlakna, metoda kon~nih elementov (FEM)
1 INTRODUCTION
Fiber-reinforced composite plates are widely used in
many types of engineering applications such as the aero-
space, automotive and marine industries, as well as in
medical prosthetic devices, electronic circuit boards and
sports equipment because of certain properties such as
high specific stiffness and strength. A plate which is
subjected to an axial compressive load ought to remain
stable. If, in spite of a small addition of an axial or lateral
disturbance applied to a plate, it remains to be in equili-
brium, then the plate is said to be stable. If a small
additional disturbance results in a large response and the
plate does not return to its original equilibrium configu-
ration, the plate is said to be unstable.
The magnitude of the compressive axial load, at
which the plate becomes unstable is called the critical
buckling load. If the load is increased beyond this critical
load, it results in a large deflection and the plate seeks
another equilibrium configuration. Thus, the load at
which a plate becomes unstable is of practical import-
ance in the design of structural elements. One of the
important issues is the prediction of the critical buckling
loads of composite materials. Here we determine the cri-
tical buckling loads of the laminated composite rectan-
gular plates with a cylindrical hole using experimental
and numerical methods.
The buckling behavior of symmetric cross-ply and
angle-ply laminated flat composite columns was des-
cribed in1, investigating the effects of the column
thickness and wideness, the orientation angles, the fillet
radius and the modulus ratios on the critical buckling
load with the finite-element method (FEM) based on the
first-order shear-deformation theory (FSDT). Arman et
al.2 investigated the effect of the delamination around a
single circular hole on the critical buckling load of
woven-fabric-laminated composite plates. Besides the
critical buckling load, they determined the critical
delamination diameter. A buckling analysis of a woven-
glass-polyester and boron/epoxy laminated composite
plate containing a circular/elliptical hole given in3,4,
respectively, was done using the FEM.
Aydogdu5 studied the thermal-buckling behavior of
cross-ply laminated beams, subjected to different boun-
Materiali in tehnologije / Materials and technology 50 (2016) 1, 23–27 23
UDK 519.61/.64:66.017:620.174:666.189.2 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)23(2016)
dary conditions. He used the Ritz method to obtain the
critical buckling temperatures. Kim and Lee6 developed
fine-beam elements named as 2-, 3- and 4-noded isopara-
metric beam elements to analyze the lateral buckling of
the shear-deformable laminated composites. Jung and
Han7 presented an approach for the initial buckling beha-
vior of laminated composite plates and shells subjected
to the combined in-plane loading. Laurin et al.8 deve-
loped an approach for the multiscale-failure behavior for
various stacking sequences. The buckling response of
laminated composite structures with a delamination was
investigated by Lee and Park9.
Matsunaga10 used the method of the power series
expansion of continuous-displacement components to
come up with the two-dimensional global-deformation
theory for the thermal buckling of laminated composites
and sandwich plates. The thermal-buckling behavior of
laminated hybrid-composite plates containing a hole,
subjected to a uniform temperature was studied by
ªahin11. The thermal-buckling mode shapes of laminated
composites with varying fiber orientations, stacking se-
quences and E1/E2 ratios are studied in12. The thermo-
mechanical-buckling response of laminated composites
and sandwich plates was investigated by Wu and Chen13.
The post- and thermal-buckling behavior of lami-
nated composite beams with temperature-dependent
material properties is given in14. Dash and Singh15 deve-
loped an isoparametric nonlinear finite-element method
for the buckling and post-buckling of laminated com-
posite plates. The buckling behavior of a composite
structure based on the generalized differential quadrature
rule (GDQR) and Rayleigh-Ritz (R-R) method is presen-
ted in16,17. Topal and Uzman18 investigated the optimi-
zation of the critical failure mode of simply supported
laminated composite plates under in-plane static loads.
Shufrin et al.19 generated a semi-analytical formulation
based on the total energy minimization and the iterative
extended Kantorovich approach for the buckling analysis
of symmetrically laminated composite plates. Malekza-
deh and Shojaee20 studied the buckling response of
carbon-nanotube-reinforced, quadrilateral laminated
plates. Özben21 presented the critical buckling load for
laminated composite plates using the FEM and analytical
methods. The effects of the hole location and diameter
on the lateral-buckling response of woven-fabric-lami-
nated composite cantilever beams were determined
experimentally and numerically in22.
In this study, the effects of the hole diameter and
fiber orientations on the buckling behavior of glass-
fiber-reinforced, laminated composite rectangular plates
were investigated experimentally and numerically under
compression. Firstly, the test specimens for different
fiber orientation angles were prepared with and without
the holes. The test specimens made of glass/epoxy con-
sisted of eight symmetric plies. A uniformly distributed
load was applied to the upper edges of the plate speci-
mens and critical buckling loads were determined
experimentally and numerically. The results obtained
with the experimental and numerical methods were in
good agreement. The results showed that the critical
buckling loads strongly depend on the diameter of the
hole and the fiber orientation angles.
2 MATERIALS AND METHODS
2.1 Test specimens
For the experiments, laminated composite plates con-
sisting of eight layers of e-glass woven fabrics with
various fiber orientation angles such as [(0°/90°)2]s,
[(–15°/15°)2]s, [(–30°/30°)2]s and [(–45°/45°)2]s were
manufactured. The fiber volume fraction and the nominal
thickness of the composite were approximately 55 % and
2 mm, respectively. For the matrix material, the CY225
epoxy resin and hardener HY225 were used. The curing
process was implemented at 120 °C for 3 h under a
pressure of 0.25 MPa.23 Then, the composite was cooled
to room temperature. The prepared plates were cut to the
dimensions of 100 × 100 mm and drilled in the center
with different hole diameters such as 0 (without a hole),
10, 20, 30 and 40 mm respectively. Table 1 shows the
mechanical properties of the laminated composites.
Table 1: Mechanical properties of the composites plate
Tabela 1: Mehanske lastnosti kompozitnih plo{~
E1 = E2
(GPa) E3 (GPa)
G12 = G13 =
G23 (GPa)
v12 v13 = v23
28.500 17.100 7.840 0.148 0.088
2.2 Experimental procedure of the critical buckling
load
The buckling tests were done in the compression
mode on a Shimadzu AG-100 kN test machine. The
buckling-test apparatus designed by Arman et al.2 con-
sists of two clamped edges at the bottom and top and two
free edges as the boundary conditions as shown in
Figure 1. As a result of the clamped boundary condition
of the buckling-test apparatus that holds the specimen at
the bottom and top edges, each having a 10 mm
A. KURªUN, E. TOPAL: INVESTIGATION OF HOLE EFFECTS ON THE CRITICAL BUCKLING LOAD ...
24 Materiali in tehnologije / Materials and technology 50 (2016) 1, 23–27
Figure 1: Buckling-test apparatus2
Slika 1: Naprava za upogibni preizkus2
clamping length, the height of the test specimens were
determined to be 80 mm as shown in Figure 1.
Three identical specimens were tested for each hole
diameter and all the specimens were subjected to an
axial compressive load until the first buckling mode was
reached as shown in Figure 2. The laminated plate
became unstable as the first buckling mode was reached.
The magnitude of the axial compressive load, at which
the plate becomes unstable, is called the critical buckling
load. The other buckling modes were not studied in this
paper.
The test results were taken as a text file on the data
card of the test machine. Then the graphs of the load-
displacement variations were created for each specimen.
A MATLAB® code was written for the upheaval of the
slope of the load-displacement curve. The critical buckl-
ing load was determined accordingly and shown in Fig-
ure 3. In this figure, Pcr is the critical buckling load.
2.3 Finite-element buckling model
A three-dimensional finite-element analysis of the
composite plates with a single circular hole was per-
formed using the commercial finite-element software
ANSYS® 12.1. The model consists of eight layers with
dimensions of 100 × 100 × 0.25 mm for each layer and
the 2 mm total thickness of the plate. The diameter of the
hole changes from 0 (without hole) to 10, 20, 30 and 40
mm. Additionally, the fiber orientation changes from
[(0°/90°)2]s to [(–15°/15°)2]s, [(–30°/30°)2]s and
[(–45°/45°)2]s. Therefore, a total of 20 models were
constructed. After the modeling of the composite plates
with a circular hole, the number of layers, the element
type, the material properties of the laminated composite
plates, the fiber orientation angles and the thickness of
each layer were introduced to the finite-element pro-
gram. The element type was a SHELL layered element
having six degrees of freedom at each node: the trans-
lations in the nodal x, y and z directions and the rotations
about the nodal x, y and z axes. Finally, for the analysis,
the unit pressure was applied to one of the clamped
edges, the model was meshed and the program was run.
3 RESULTS AND DISCUSSION
The critical-buckling-load results were obtained both
experimentally and numerically for the [(0°/90°)2]s,
[(–15°/15°)2]s, [(–30°/30°)2]s and [(–45°/45°)2]s fiber
orientations for all the buckling-test specimens having 0
(without hole), 10, 20, 30 and 40 mm hole diameters,
respectively. The experimental results for each hole
diameter are presented in Figure 4. In this figure, three
identical experimental graphs and their average values
are shown.
It is clear that all the repeated experimental tests for
each group with various hole-diameter values show a very
similar buckling behavior as reported in Figures 4a to 4e.
The average values of all the results obtained expe-
rimentally and numerically are given in Table 2 in order
to compare them in terms of the critical buckling load. In
the same table, P*cr refers to the critical buckling load for
the specimen without a hole.
Table 2: Critical buckling loads for numerical and experimental stu-
dies
Tabela 2: Kriti~na upogibna obremenitev pri numeri~ni {tudiji in pri
preizkusu
Hole
dia-
meter
(mm)
Critical buckling loads (Pcr, P*cr) (N)
Fiber orientations (°)
[(0°/90°)2]s [(–15°/15°)2]s[(–30°/30°)2]s[(–45°/45°)2]s
Exp. Num. Exp. Num. Exp. Num. Exp. Num.
0 4867.24497.04347.34296.34000.74023.03660.13853.0
10 4310.54261.54277.74155.43934.13891.83540.83734.4
20 4194.74073.53896.23970.93700.63709.13140.33550.0
30 3710.93758.93703.33666.93434.53424.42830.73272.2
40 3326.33365.33117.83279.23034.33050.52280.92905.4
In addition, the experimental and numerical results
showing the variation in the critical buckling load versus
the hole diameter were presented in Figure 5. It is clear
A. KURªUN, E. TOPAL: INVESTIGATION OF HOLE EFFECTS ON THE CRITICAL BUCKLING LOAD ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 23–27 25
Figure 3: Determination of the experimental critical buckling load
Slika 3: Eksperimentalno dolo~anje kriti~ne upogibne obremenitve
Figure 2: View of the buckling-test process: a) before buckling and b)
after buckling
Slika 2: Izgled upogibnega preizkusa: a) pred upogibanjem, b) po
upogibanju
from this figure that the experimental and numerical
results are consistent with each other. Figures 5 and 6
were automatically scaled starting at buckling loads of
1000 N and 2800 N, respectively, in order to clarify the
difference between of the results.
From Table 2 and Figure 5, it is concluded that the
critical buckling load for all the types of fiber orientation
angle has the highest value for the specimens without a
hole and the lowest value for the specimens having a 40
mm hole diameter.
Figure 6 shows the variation in the critical buckling
load versus the hole diameter in terms of fiber orien-
tation. The critical buckling load for the same hole dia-
meter is maximum at the orientation angle of [(0°/90°)2]s
and it decreases with the order of [(–15°/15°)2]s,
[(–30°/30°)2]s and [(–45°/45°)2]s as reported in Figure 6.
And also, the critical buckling load gradually decreases,
while the hole diameter increases for all the types of
fiber orientation.
A. KURªUN, E. TOPAL: INVESTIGATION OF HOLE EFFECTS ON THE CRITICAL BUCKLING LOAD ...
26 Materiali in tehnologije / Materials and technology 50 (2016) 1, 23–27
Figure 4: Experimental critical buckling loads for: a) without a hole, b) 10 mm hole diameter, c) 20 mm hole diameter, d) 30 mm hole diameter
and e) 40 mm hole diameter
Slika 4: Eksperimentalne kriti~ne upogibne obremenitve pri: a) brez luknje, b) premer luknje 10 mm, c) premer luknje 20 mm, d) premer luknje
30 mm in e) premer luknje 40 mm
Figure 6: Numerical results for the critical buckling load versus the
hole diameter for all types of fiber orientation
Slika 6: Numeri~ni rezultati za kriti~no upogibno obremenitev v od-
visnosti od premera luknje, pri vseh orientacijah vlaken
Figure 5: Comparison between the experimental and numerical
results for fiber orientations: a) [(0°/90°)2]s, b) [(–15°/15°)2]s, c)
[(–30°/30°)2]s and d) [(–45°/45°)2]s
Slika 5: Primerjava rezultatov preizkusov in numeri~nih rezultatov pri
orientaciji vlaken: a) [(0°/90°)2]s, b) [(–15°/15°)2]s, c) [(–30°/30°)2]s
in d) [(–45°/45°)2]s
4 CONCLUSION
In this study, the buckling response of the composite
plates with symmetric orientation angles such as
[(0°/90°)2]s, [(–15°/15°)2]s, [(–30°/30°)2]s and
[(–45°/45°)2]s with a single circular hole were examined
by employing an experimental study and a numerical
analysis performed with the finite-element technique.
The conclusions that can be made in the contribution are
as follows:
• The critical buckling loads are strongly dependent on
the hole size for all the types of fiber orientation.
• The maximum values of the critical buckling load
were obtained for the specimens with the [(0°/90°)2]s
orientation angle.
• The buckling load decreases exponentially with a de-
crease in the fiber orientation angle of the composite
material.
5 REFERENCES
1 H. Akbulut, O. Gundogdu, M. Þengül, Finite Elements in Analysis
and Design, 46 (2010) 12, 1061–1067, doi:10.1016/j.finel.2010.
07.004
2 Y. Arman, M. Zor, S. Aksoy, Composites Science and Technology,
66 (2006) 15, 2945–2953, doi:10.1016/j.compscitech.2006.02.014
3 M. Aydin Komur, F. Sen, A. Ataþ, N. Arslan, Advances in Engineer-
ing Software, 41 (2010) 2, 161–164, doi:10.1016/j.advengsoft.2009.
09.005
4 D. Ouinas, B. Achour, Composites Part B: Engineering, 55 (2013),
575–579, doi:10.1016/j.compositesb.2013.07.011
5 M. Aydogdu, Composites Science and Technology, 67 (2007) 6,
1096–1104, doi:10.1016/j.compscitech.2006.05.021
6 N. I. Kim, J. Lee, International Journal of Mechanical Sciences, 68
(2013), 246–257, doi:10.1016/j.ijmecsci.2013.01.023
7 W. Y. Jung, S. C. Han, Composite Structures, 109 (2014), 119–129,
doi:10.1016/j.compstruct.2013.10.055
8 F. Laurin, N. Carrere, J. F. Maire, Composite Structures, 80 (2007) 2,
172–182, doi:10.1016/j.compstruct.2006.04.074
9 S. Y. Lee, D. Y. Park, International Journal of Solids and Structures,
44 (2007) 24, 8006–8027, doi:10.1016/j.ijsolstr.2007.05.023
10 H. Matsunaga, Composite Structures, 68 (2005) 4, 439–454,
doi:10.1016/j.compstruct.2004.04.010
11 Ö. S. Þahin, Composites Science and Technology, 65 (2005) 11–12,
1780–1790, doi:10.1016/j.compscitech.2005.03.007
12 L. C. Shiau, S. Y. Kuo, C. Y. Chen, Composite Structures, 92 (2010)
2, 508–514, doi:10.1016/j.compstruct.2009.08.035
13 Z. Wu, W. Chen, International Journal of Mechanical Sciences, 49
(2007) 6, 712–721, doi:10.1016/j.ijmecsci.2006.10.006
14 A. R. Vosoughi, P. Malekzadeh, Ma. R. Banan, Mo. R. Banan, Inter-
national Journal of Non-Linear Mechanics, 47 (2012) 3, 96–102,
doi:10.1016/j.ijnonlinmec.2011.11.009
15 P. Dash, B. N. Singh, Mechanics Research Communications, 46
(2012), 1–7, doi:10.1016/j.mechrescom.2012.08.002
16 M. Darvizeh, A. Darvizeh, R. Ansari, C. B. Sharma, Composite
Structures, 63 (2004) 1, 69–74, doi:10.1016/s0263-8223(03)00133-8
17 Y. Tang, X. Wang, International Journal of Mechanical Sciences, 53
(2011) 2, 91–97, doi:10.1016/j.ijmecsci.2010.11.005
18 U. Topal, Ü. Uzman, Thin-Walled Structures, 45 (2007) 7–8,
660–669, doi:10.1016/j.tws.2007.06.002
19 I. Shufrin, O. Rabinovitch, M. Eisenberger, Composite Structures, 82
(2008) 4, 521–531, doi:10.1016/j.compstruct.2007.02.003
20 P. Malekzadeh, M. Shojaee, Thin-Walled Structures, 71 (2013),
108–118, doi:10.1016/j.tws.2013.05.008
21 T. Özben, Computational Materials Science, 45 (2009) 4,
1006–1015, doi:10.1016/j.commatsci.2009.01.003
22 E. Eryiðit, M. Zor, Y. Arman, Composites Part B: Engineering, 40
(2009) 2, 174–179, doi:10.1016/j.compositesb.2008.07.005
23 A. Kursun, M. Senel, Experimental Techniques, 37 (2013) 6, 41–48,
doi:10.1111/j.1747-1567.2011.00738.x
A. KURªUN, E. TOPAL: INVESTIGATION OF HOLE EFFECTS ON THE CRITICAL BUCKLING LOAD ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 23–27 27

K. WIŒNIEWSKA et al.: CORROSION OF THE REFRACTORY ZIRCONIA METERING NOZZLE ...
29–32
CORROSION OF THE REFRACTORY ZIRCONIA METERING
NOZZLE DUE TO MOLTEN STEEL AND SLAG
KOROZIJA OGNJEODPORNE CIRKONSKE DOZIRNE [OBE
S STALJENIM JEKLOM IN @LINDRO
Klaudia Wiœniewska, Dominika Madej, Jacek Szczerba
AGH University of Science and Technology, Faculty of Materials Science and Ceramics, Department of Ceramics and Refractories,
Al. A. Mickiewicza 30, 30-059 Kraków, Poland
kwis@agh.edu.pl
Prejem rokopisa – received: 2014-08-08; sprejem za objavo – accepted for publication: 2015-03-03
doi:10.17222/mit.2014.188
This paper presents a study on the phase composition and microstructure changes of a sintered Mg-stabilized zirconia metering
nozzle exposed to the corrosive effect of the molten steel and slag in a tundish for continuous casting for 30 h. A macroscopic
observation of the corroded material showed cracks and two zones that were distinguished with respect to colour. An X-ray
diffraction analysis showed that the dark layer was richer in stabilized ZrO2 than the light layer. After the corrosion test, the
nozzle had higher contents of MgO, SiO2, CaO, Al2O3 and Fe2O3 than the reference sample as evidenced by an X-ray
fluorescence analysis. Moreover, during the corrosion process, liquid steel and slag infiltrated the zirconia material, which was
confirmed with a SEM investigation. Along the hot face of the metering nozzle, the grains of zirconium oxide recrystallized
with a high-temperature structure of ZrO2 and dissolved MgO and CaO derived from the slag, stabilizing this phase.
Keywords: partially stabilized ZrO2, metering nozzle, corrosive effect of molten steel, refractories
^lanek predstavlja {tudijo fazne sestave in mikrostrukturnih sprememb sintrane, z Mg stabilizirane cirkonske dozirne {obe,
izpostavljene korozijskim vplivom staljenega jekla in `lindre po 30 urnem delovanju v vmesni ponovci pri postopku
kontinuirnega litja. Makroskopsko opazovanje korodiranega materiala je pokazalo razpoke in dve podro~ji, ki sta se razlikovali
po barvi. Rentgenska difrakcijska analiza je pokazala, da je bilo temnej{e podro~je bogatej{e s stabiliziranim ZrO2, kot pa
svetlej{e podro~je. Rentgenska fluorescen~na analiza je pokazala, da ima {oba po korozijskem preizkusu, v primerjavi z
referen~nim vzorcem, vi{jo vsebnost MgO, SiO2, CaO, Al2O3 in Fe2O3. Poleg tega sta, med procesom korozije, teko~e jeklo in
`lindra prodirala v cirkonijev material, kar je bilo potrjeno s SEM preiskavo. Vzdol` vro~ega ~ela dozirne {obe so zrna
cirkonijevega oksida rekristalizirala v visoko temperaturno strukturo ZrO2, raztopila MgO in CaO, ki izvirata iz `lindre in
stabilizirala to fazo.
Klju~ne besede: delno stabiliziran ZrO2, dozirna {oba, korozijski vpliv staljenega jekla, ognjevarna gradiva
1 INTRODUCTION
Partially stabilized ZrO2 (PSZ) materials are widely
used as refractories owing to their high refractoriness,
high corrosion resistance and good thermal-shock resis-
tance. In the metal industry, the most popular are Mg-
and Ca-stabilized zirconia and Y-stabilized zirconia,
applied in the production of thermal-barrier coatings. It
is commonly known that during the corrosion process of
PSZ due to molten steel and slag, a cubic or tetragonal
phase is destabilized and transformed into monoclinic
zirconia, related to the microcrack formation1. The
destabilization process is associated with the reaction of
SiO2 from slag with CaO and MgO from the zirconia
material2,3. Zirconium oxide is used for metering nozzles
in the continuous casting of steel2,4,5. A metering nozzle
is mounted to the bottom of a tundish and used to control
the flow of molten steel. The working temperature of a
metering nozzle is up to 1600 °C.
The main factors causing the damage of the metering
nozzle are the erosion of the liquid flowing steel, the
infiltration of the molten steel and slag into the refrac-
tory material, the slag composition and the rapid tem-
perature changes2. The chemical composition of the slag
is based on the MgO-CaO-SiO2-Al2O3 system, but the
proportion between the components depends on the type
of the produced steel.
In this paper the phase composition and microstruc-
ture changes of a sintered Mg-stabilized zirconia meter-
ing nozzle, exposed to the corrosive effect of the molten
low-carbon steel and slag in a tundish for continuous
casting for 30 h were studied.
2 EXPERIMENTAL METHOD
The procedure of this experiment included the
following steps: preparing a sintered Mg-stabilized
zirconia metering nozzle under industrial conditions, a
corrosion test in a tundish for continuous steel casting
and a post-experiment study of the zirconia metering
nozzle.
The metering nozzle was exposed to a corrosive envi-
ronment for 30 h. The factors influencing the corrosion
of the nozzle were the molten steel and slag. After the
corrosion test, the corroded nozzle (designated as CN)
Materiali in tehnologije / Materials and technology 50 (2016) 1, 29–32 29
UDK 666.76:620.193 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)29(2016)
was studied and compared with the reference sample
(designated as RS), which was a zirconia metering
nozzle before serving. For the XRF, XRD and SEM/EDS
investigations, the samples were taken from the middle
of the heights of the metering nozzles.
The X-ray fluorescence (XRF) method was used to
determine the contents of oxides such as MgO, SiO2,
CaO, Al2O3 and Fe2O3. An Axios mAX wavelength dis-
persive X-ray fluorescence spectrometer from PANaly-
tical was used for this study. The crystalline phases were
identified with X-ray diffraction using Cu-K radiation
on a FPM Seifert XRD7 diffractometer with the Bragg-
Brentano geometry, while the quantitative composition
was determined with the Rietveld method. The dark zone
(designated as CN-D) and the light zone (designated as
CN-L) of the corroded nozzle were examined separately.
The microstructure was observed using a scanning elec-
tron microscope (SEM) coupled with an energy-disper-
sive X-ray system (EDS). The ultra-high-definition
NOVA NANO SEM 200 was used for this purpose.
3 RESULTS
During a macroscopic observation of the corroded
material, cracks and two zones were distinguished with
respect to colour. The dark zone was strongly corroded
due to molten steel and slag (the hot-zone refractory
material) while the light zone remain unchanged. Figure
1 shows a photograph of the metering nozzle exposed to
the corrosive effect of the molten steel.
The XRF analysis confirmed that the main compo-
nent of the nozzle was zirconium oxide; the content of
ZrO2 was about 90 % in both CN and RS samples. More-
over, the XRF study revealed content changes in the
oxides such as MgO, SiO2, CaO, Al2O3 and Fe2O3. After
the corrosion test, sample CN had higher contents of
these oxides, which is illustrated in Figure 2.
The XRD analysis with the Rietveld quantitative-
phase analysis showed that the dark layer (CN-D) was
richer in the stabilized ZrO2 than the light layer (CN-L)
or the RS sample. In the CN-D sample, the content of the
stabilized zirconium oxide was 11.5 %, while in the
K. WIŒNIEWSKA et al.: CORROSION OF THE REFRACTORY ZIRCONIA METERING NOZZLE ...
30 Materiali in tehnologije / Materials and technology 50 (2016) 1, 29–32
Figure 4: Microstructure of test sample RS
Slika 4: Mikrostruktura preizkusnega vzorca RS
Figure 1: Refractory zirconia metering nozzle exposed to the corro-
sive effect of the molten steel and slag
Slika 1: Ognjevarna cirkonska dozirna {oba, izpostavljena korozijske-
mu vplivu staljenega jekla in `lindre
Figure 2: Contents of the secondary oxides in test samples RS and CN
(XRF analysis)
Slika 2: Vsebnost sekundarnih oksidov v preizkusnih vzorcih RS in
CN (XRF-analiza)
Figure 3: XRD patterns of test samples RS, CN-L and CN-D
Slika 3: Rentgenogrami preizkusnih vzorcev RS, CN-L in CN-D
CN-L sample, it was 1 % and in the RS sample, it was
3.3 %. The XRD results are shown in Figure 3.
Figures 4 to 6 show the microstructures of test sam-
ples RS, CN-L and CN-D, respectively. The microstruc-
ture of the CN-L sample is similar to the reference
sample. The grains of zirconium oxide are surrounded by
a silicate phase. The microstructure of sample CN-D
shows micro-fracturing along the hot-face of the nozzle.
The crystallites of zirconium oxide are bigger than in
samples CD-L and RS.
4 DISCUSSION
In the reference sample, the main crystalline phases
were monoclinic ZrO2 and Mg-stabilized zirconium
oxide, which were confirmed with the XRD analysis.
The SEM/EDS investigation showed that the secondary
phase was forsterite (Mg2SiO4). However, the content of
forsterite was too low to be detected on the XRD pattern.
After the operation, there were two zones in the nozzle, a
corroded and an uncorroded zone. The XRD analysis
with the Rietveld method confirmed that test samples
CN-L and RS were not significantly different in their
phase composition. The biggest changes in the phase
composition were found in the corroded zone (CN-D),
where the content of stabilized zirconia was the highest.
The obtained results are not in agreement with the
earlier results presented in2,3,5. The stabilization process
of zirconium oxide can be explained with the dissolution
of MgO and CaO derived from the slag in zirconia at a
high temperature. The XRF and SEM/EDS investigations
confirmed that, at the temperature of the corrosion test,
the molten slag and steel penetrated the zirconia nozzle.
As can be seen in Figure 6, the microstructure of
CN-D was changed. The fine grains of ZrO2 recry-
stallized at the corrosion-test temperature, which was
observed as an increase in their dimension. The high-
temperature phase of zirconium oxide was stabilized due
to MgO and CaO, which dissolved in ZrO2. The
micro-fractures along the hot face of the metering nozzle
and the macroscopic cracks present in the corroded
nozzle could be attributed to the rapid temperature
changes during the corrosion test.
K. WIŒNIEWSKA et al.: CORROSION OF THE REFRACTORY ZIRCONIA METERING NOZZLE ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 29–32 31
Figure 6: Microstructure of test sample CN-D obtained with the EDS analysis; Spot 1 – zirconium oxide, Spot 2 – slag, Spot 3 – iron
Slika 6: Mikrostruktura preizkusnih vzorcev CN-D z EDS-analizo; to~ka 1 – cirkonov oksid, to~ka 2 – `lindra, to~ka 3 – `elezo
Figure 5: Microstructure of test sample CN-L obtained with the EDS analysis; Spot 1 – zirconium oxide, Spot 2 – forsterite
Slika 5: Mikrostruktura preizkusnega vzorca CN-L z EDS-analizo; to~ka 1 – cirkonov oksid, to~ka 2 – forsterit
5 CONCLUSION
The results of the corrosion of a refractory zirconia
metering nozzle tested under industrial conditions are re-
ported. During the service in a continuous-casting
tundish, the molten steel and slag infiltrated the zirconia
nozzle. A rapid change in the temperature had a signifi-
cant influence on the formation of cracks and micro-
fractures.
Along the hot face of the zirconia nozzle, the grains
of ZrO2 recrystallized with a high-temperature structure.
The oxides present in the slag dissolved in zirconia and
stabilized the ZrO2 phase.
Acknowledgments
This work was supported by grant no. INNOTECH-
K2/IN2/16/181920/NCBR/13 of the National Centre for
Research and Development.
6 REFERENCES
1 A. Sibil, T. Douillard, C. Cayron, N. Godin, M. R’mili, G. Fantozzi,
Microcracking of high zirconia refractories after tm phase
transition during cooling: An EBSD study, Journal of the European
Ceramic Society, 31 (2011), 1525–1531, doi:10.1016/j.jeurceramsoc.
2011.02.033
2 E. Volceanov, A. Abagiu, M. Becherescu, A. Volceanov, P. Nita, R.
Trusca, F. Mihalache, Development of Zirconia Composite Ceramics
and Study on Their Corrosion Resistance up to 1600 °C, Key
Engineering Materials, 264–268 (2004), 1739–1742, doi:10.4028/
www.scientific.net/KEM.264-268.1739
3 Y. Hemberger, C. Berthold, K. G. Nickel, Wetting and Corrosion of
Yttria Stabilized Zirconia by Molten Slags, Journal of the European
Ceramic Society, 32 (2012), 2859–2866, doi:10.1016/j.jeurceramsoc.
2011.12.005
4 A. H. Bui, S. C. Park, I. S. Chung, H. G. Lee, Dissolution Behavior
of Zirconia Refractories During Continuous Casting of Steel, Metals
and Materials International, 12 (2006) 5, 435–440, doi:10.1007/
BF03027711
5 M. O. Suk, J. H. Park, Corrosion Behaviors of Zirconia Refractory
by CaO-SiO2-MgO-CaF2 Slag, Journal of the American Ceramic So-
ciety, 92 (2009), 717–723, doi:10.1111/j.1551-2916.2008.02905.x
K. WIŒNIEWSKA et al.: CORROSION OF THE REFRACTORY ZIRCONIA METERING NOZZLE ...
32 Materiali in tehnologije / Materials and technology 50 (2016) 1, 29–32
Md. M. ISLAM et al.: EFFECTS OF AN EPOXY-RESIN-FIBER SUBSTRATE FOR A -SHAPED MICROSTRIP ANTENNA
33–37
EFFECTS OF AN EPOXY-RESIN-FIBER SUBSTRATE FOR A
-SHAPED MICROSTRIP ANTENNA
VPLIV Z VLAKNI OJA^ANE EPOKSI PODLAGE PRI -OBLIKI
MIKROTRAKASTE ANTENE
Md. Moinul Islam1, Mohammad R. I. Faruque1, Mohammad Tariqul Islam2,
Haslina Arshad3
1Centre for Space Science (ANGKASA), Kompleks Penyelidikan Building, Universiti Kebangsaan, Malaysia
2Department of Electrical, Electronic & Systems Engineering, Universiti Kebangsaan, Malaysia
3Centre of Artificial Intelligence Technology, Universiti Kebangsaan Malaysia, 43600 UKM, Bangi, Selangor D. E., Malaysia
mmoiislam@yahoo.com; rashedgen@yahoo.com; titareq@yahoo.com; has@ftsm.ukm.my
Prejem rokopisa – received: 2014-08-18; sprejem za objavo – accepted for publication: 2015-03-09
doi:10.17222/mit.2014.206
A -shaped microstrip antenna using an epoxy-resin-fibre substrate is presented. The proposed antenna consists of a circular
slot and two rectangular slots printed on a dielectric resin-fibre substrate and is excited by a 50- microstrip transmission line.
A commercially available, high-frequency structural simulator (HFSS) based on the finite-element method (FEM) was used in
this investigation. The nearly omni-directional and bidirectional radiation pattern exhibited average gains of 3.12 dBi and 5.44
dBi for the lower band and upper band, respectively. The effects of epoxy-resin-fiber are discussed through comparisons of
different substrate materials.
Keywords: epoxy resin-fibre, microstrip line, -shaped
Predstavljena je mikrotrakasta antena -oblike na epoksi podlagi, oja~ani z vlakni. Predlagana antena sestoji iz kro`ne re`e in
dveh pravokotnih re`, natiskanih na dielektri~ni podlagi iz smole z vlakni in sta vzbujani s 50  mikrotrakastim vodnikom. V tej
preiskavi je bil uporabljen komercialno razpolo`ljiv visoko frekven~ni strukturni simulator (HFSS), ki temelji na metodi
kon~nih elementov (FEM).V spodnjem in zgornjem pasu je vsesmerno sevanje kazalo 3,12 dBi, dvosmerno sevanje pa 5,44 dBi.
Vpliv vlaken za oja~anje je bil prikazan s primerjavo razli~nih materialov podlage.
Klju~ne besede: vlakna za oja~anje epoksija, mikrotrakasta linija, -oblika
1 INTRODUCTION
The microstrip patch antenna plays an important role
as a harbinger in wireless communication systems and is
now being used to address the changing demands of
future antenna technology. The microstrip patch antenna
has been extensively used in wireless communication
systems, because they are conformal, have a low profile,
are easy to fabricate with integrated circuits (ICs), and
enable easy integration with array and electronic compo-
nents. Many researchers have an interest in designing
microstrip antennas and still face a major challenge to
implement these applications. Currently, various types of
antennas have been proposed to face the increasing
requirements for a modern bearable wireless communi-
cation device that has the capability of consolidating
more than one communication system into a single
module.1–6
In7, a rectangular slot antenna was proposed for
dual-frequency operation. Su et al.8 presented a printed
dipole antenna using U-slot arms to enable dual-band
operation. Suh and Chang9 reported a low-cost micro-
strip dipole antenna for wireless communications. In10, a
PIFA antenna with a U-slot was presented for dual-band
operation. Lin et al.11 mentioned a dual-loop antenna for
use with a 2.4/5 GHz Wireless LAN. A monopole an-
tenna with double-T was presented in12 for 2.4/5.2-GHz
WLAN operations. A planar antenna was investigated
with bandwidth enhancement for X-band applications13.
An E-shaped patch antenna of wideband circularly
polarized was presented for wireless applications14. A
Compact 5.5-GHz Band-Rejected UWB Antenna was
proposed using Complementary Split Ring Resonators15.
A new double L-shaped multiband patch antenna was
presented on a polymer resin material substrate16.
In this study, a -shaped microstrip antenna was
designed on a 1.6-mm-thick epoxy-resin-fibre substrate
material. The downlink frequency range is from 4.74
GHz to 4.87 GHz and the uplink frequency range is from
8.42 GHz to 8.73 GHz. The results will be discussed in
detail with a parametric study.
2 ANTENNA GEOMETRY AND PARAMETRIC
STUDY
High-Frequency Structure Simulator (HFSS), a com-
mercially available Ansoft package, is a powerful and
efficient three-dimensional (3D) full-wave simulation
software that solves EM equations through the sub-
division of a large problem into easy constituent units
Materiali in tehnologije / Materials and technology 50 (2016) 1, 33–37 33
UDK 66.017:678.686 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)33(2016)
and then consolidating the solution as a matrix of
simultaneous equations for the complete problem space,
which provides a numerical solution to Maxwell’s
equations using the FEM. Hence, HFSS was used in this
study.
The geometry of the proposed antenna is shown in
Figure 1. The antenna comprises three conducting slots
on the patch and two on the ground. A circular slot and
two similar lateral rectangular slots are on the patch and
two rectangular slots are on the ground of the proposed
antenna. The two rectangular slots are of equal length LP
and width WP. R is the radius of the circular slot. The
design procedure begins with the radiating patch, along
with the substrate, the ground plane and a feed line. The
antenna was printed on a FR4 substrate with 1.6 mm in
thickness that exhibits a relative permittivity of 4.60, a
relative permeability of 1, and a dielectric loss tangent of
0.02. One circular slot and two rectangular slots are cut
from the rectangular copper patch. Another two
rectangular slots are also cut from the ground plane. In
this manner, the proposed slotted circle patch antenna is
produced. Two resonant frequencies of 4.51 GHz and
8.35 GHz are obtained by adjusting the length and width
of the slots of the proposed antenna. Here, a microstrip
line is used to feed the RF signal into the proposed
antenna. The Sub Miniature version A (SMA) connector
is used at the end of antenna feeding line for the input
RF signal.
Finally, the optimal dimensions were determined as
follows: L = 40 mm, Lp = 12 mm, Lg = 40 mm, Ls = 4
mm, R = 12 mm, W = 40 mm, Wp = 4 mm, Wg = 40 mm,
Ml =17 mm, and Mw = 6 mm.
The epoxy-resin-fibre consists of reinforcing insula-
tion material. There are various types of epoxy-resin-
fibre material for use as an antenna substrate. In this
study, we have used FR4 as the epoxy-resin-fibre sub-
strate material that is impregnated with thermoset resin.
FR4 has superior mechanical and dielectric properties,
good moisture/heat resistance, stable electrical perfor-
mance at high temperature, good flatness and a smooth
surface. FR4 is widely used to produce printed-circuit
boards (PCBs).
The length, width, VSWR, return loss of the patch
antenna can be calculated from Equations (1) to (6)
presented in17, where L and W are the length and width
of the patch, respectively, c is the velocity of light, r is
the dielectric constant of substrate, h is the thickness of
the substrate, f0 is the target centre frequency, e is the
effective dielectric constant and  is the radiation
coefficient:
W
c
f
=
+
2
1
20
 r
(1)
L
c
f
l= −
2
2
0  r
Δ (2)
  e r r= + + − +
⎛
⎝
⎜ ⎞
⎠
⎟1
2
1
1
2
1 1
10
( ) ( )
h
W
(3)
Δl h
W h
W h
=
+ +
− +
0 412
03 08
0 258 08
.
( . )( / . )
( . )( / . )


e
e
(4)
VSWR =
+
−
1
1


(5)
Return loss = −
⎛
⎝
⎜
⎞
⎠
⎟10
1
lg

(6)
Md. M. ISLAM et al.: EFFECTS OF AN EPOXY-RESIN-FIBER SUBSTRATE FOR A -SHAPED MICROSTRIP ANTENNA
34 Materiali in tehnologije / Materials and technology 50 (2016) 1, 33–37
Figure 2: Return loss of simulation using different substrate materials
Slika 2: Povratne izgube pri simulaciji z uporabo razli~nih materialov
podlage
Figure 1: Proposed antenna: a) top view and b) bottom view
Slika 1: Predlagana antena: a) pogled iz vrha in b) pogled iz dna
Table 1: Dielectric properties of the different substrates
Tabela 1: Dielektri~ne lastnosti razli~nih podlag
Material Permittivity Loss Tangent
Teflon (tm) 2.1 0.01
RT/Duroid 5870 2.33 0.0023
Epoxy resin-fiber (Proposed) 4.66 0.02
Al2O3 9.8 0.0009
RT/Duroid 6010 10.2 0.0023
The return losses determined by the simulation using
different substrate materials are shown in Figure 2. No
resonance was found on the lower band when we used
the high-permittivity materials of Duroid 6010, and
Al2O3 ceramic and the low permittivity materials of
Duroid 5870, and Teflon as a substrate. Finally, FR4 was
used in the proposed design as the epoxy-resin-fiber
substrate material and two strong resonances were
achieved, with the desired bandwidth and high gain. The
10-dB bandwidths of 130 MHz from 4.74 GHz to 4.87
GHz and of 310 MHz from 8.42 GHz to 8.73 GHz were
achieved. The dielectric properties of the materials are
listed in Table 1.
A parametric study was performed to observe the
effects of the proposed antenna parameters. In particular,
the effects of the different parameters on the return loss
were observed.
Figure 3 shows the return loss of the simulation for
different values of R. The simulation includes L = 40 mm,
Lp = 12 mm, Lg = 40 mm, Ls = 4 mm, W = 40 mm, Wp =
4 mm, Wg = 40 mm, Ml = 17 mm, and Mw = 6 mm with
the different values of R. The graph indicates that better
coupling is obtained for the upper band using the value
of radius as 11 mm and 13 mm. By using R = 12 mm, the
desired dual-band operation was obtained, with a better
coupling on both the lower and upper bands.
The return loss of the simulation for different values
of Ml is shown in Figure 4. The simulation includes L =
40 mm, Lp = 12 mm, Lg = 40 mm, Ls = 4 mm, R = 12
mm, W = 40 mm, Wp = 4 mm, Wg = 40 mm and Mw = 6
mm with Ml. The results presented in the graph clearly
indicate that improved coupling was achieved at the
upper band when using the value of Ml as 40 mm. As a
result, the optimized value is 40 mm.
Figure 5 shows the return loss simulation for diffe-
rent values of Mw. The simulation includes L = 40 mm,
Lp = 12 mm, Lg = 40 mm, Ls = 4 mm, R = 12 mm, W = 40
mm, Wp = 4 mm, Wg = 40 mm, Ml =17 mm, and with the
different values of Mw. The width of the microstrip line
has a greater effect on the coupling for both the lower-
and upper-band frequencies. This coupling can be
achieved when Mw is 6 mm.
3 RESULTS AND DISCUSSION
The gain of the proposed antenna is shown in Figure
6. An average gain of 3.12 dBi is achieved with the first
resonance of 4.80 GHz and 5.44 dBi is achieved with the
second resonance of 8.57 GHz. In addition, the gain for
the upper band is greater than that for the lower band.
Figure 7 shows the radiation efficiency of the proposed
Md. M. ISLAM et al.: EFFECTS OF AN EPOXY-RESIN-FIBER SUBSTRATE FOR A -SHAPED MICROSTRIP ANTENNA
Materiali in tehnologije / Materials and technology 50 (2016) 1, 33–37 35
Figure 5: Return loss of the simulation using different values of Mw
Slika 5: Povratne izgube pri simulaciji z uporabo razli~nih vrednosti
Mw
Figure 3: Return loss of the simulation using different values of R
Slika 3: Povratne izgube pri simulaciji z uporabo razli~nih vrednosti R
Figure 4: Return loss of the simulation using different values of Ml
Slika 4: Povratne izgube pri simulaciji z uporabo razli~nih vrednosti Ml
antenna. The average lower-band efficiency is 75.18 %
and the higher-band efficiency is 81.35 %, i.e., the
lower-band radiation efficiency is smaller than the
higher-band efficiency.
The radiation pattern of the proposed antenna is
shown in Figure 8 for: a) 4.80 GHz at the E-plane, b)
4.80 GHz at the H-plane, c) 8.57 GHz at the E-plane, d)
8.57 GHz at the H-plane. The E and E	 fields indicate
the cross-polar and co-polar components, respectively.
The effect of cross-polarization in the radiation pattern is
due to the lower microstrip antenna. The cross
polarization effect is higher in the H-plane for both
resonances. When the frequency increases, the effect
increases, enabling simple interpretation from the
radiation pattern. Moreover, nearly omni-directional and
symmetrical radiation patterns were attained along both
the E-plane and the H-plane. The same radiation pattern
was found to exist over the C and X-bands. The obtained
radiation patterns indicate that the proposed antenna
delivers linear polarization, where the level of
cross-polarization is lower than that of the
co-polarization in all of the simulated radiation patterns.
When the radiation pattern of a microstrip antenna is
symmetric and omni-directional, it provides some
reasonable benefits. One benefit is that the resonance
does not shift for different directions, so a large amount
of stable power is in the direction of the broadside beam.
Another advantage is that the radiation pattern is more
reliable on the operational bands.
Figure 9 shows the current distribution of the pro-
posed antenna for: a) 4.51 GHz and b) 8.35 GHz. A large
amount of current flows through the feeding line. The
Md. M. ISLAM et al.: EFFECTS OF AN EPOXY-RESIN-FIBER SUBSTRATE FOR A -SHAPED MICROSTRIP ANTENNA
36 Materiali in tehnologije / Materials and technology 50 (2016) 1, 33–37
Figure 9: Current distribution at: a) 4.80 GHz and b) 8.57 GHz
Slika 9: Razporeditev toka pri: a) 4,80 GHz in b) 8,57 GHz
Figure 7: Radiation efficiency of the proposed antenna
Slika 7: U~inkovitost sevanja predlagane antene
Figure 8: Radiation pattern of the proposed antenna: a) 4.80 GHz at
the E-plane, b) 4.80 GHz at the H-plane, c) 8.57 GHz at the E-plane
and d) 8.57 GHz at the H-plane
Slika 8: Sevalni diagram predlagane antene: a) 4,80 GHz v ravnini E,
b) 4,80 GHz v ravnini H, c) 8,57 GHz v ravnini E in d) 8,57 GHz v
ravnini H
Figure 6: Gain of the proposed antenna
Slika 6: Izkoristek predlagane antene
electric field was initiated at this point. The current
distribution is more stable in the lower band than in the
upper band. The creation of the electric field near the
slots is reasonable. As a result, the excitation is strong
over all parts of the antenna for both the lower and upper
bands.
4 CONCLUSION
A -shaped microstrip antenna using an epoxy-
resin-fibre substrate was proposed in this paper. A simple
-shaped patch was proposed to miniaturize the antenna
and to obtain a new operating band resonant mode. An
antenna structure was designed, simulated and finally
characterized. The -shaped resonator, compact size,
stable nearly omni-directional and directional radiation
patterns, low cross-polarization, and efficiency with
improved bandwidth and higher gain make the proposed
-shaped dual band antenna a candidate for use in
C-band and X-band applications.
Acknowledgements
This work was supported by Universiti Kebangsaan
Malaysia under grant, Arus Perdana, No. AP-2013-011.
5 REFERENCES
1 W. C. Liu, C. M. Wu, N. C. Chu, A compact CPW-fed slotted patch
antenna for dual-band operation, IEEE Antenna Wirel. Propag. Lett.,
9 (2010), 110–113, doi:10.1109/LAWP.2010.2044135
2 M. M. Islam, M. R. I. Faruque, W. Hueyshin, J. S. Mandeep, T.
Islam, A double inverted F-shape patch antenna for dual-band ope-
ration, International Journal of Antennas and Propagation, 2014
(2014), 8, doi:10.1155/2014/791521
3 C. P. Hsieh, T. C. Chiu, C. H. Lai, Compact dual-band slot antenna at
the corner of the ground plane, IEEE Trans. Antennas Propag., 57
(2009), 3423–3426, doi:10.1109/TAP.2009.2027348
4 M. M. Islam, M. T. Islam, M. R. I. Faruque, Dual-band operation of a
microstrip patch antenna on a duroid 5870 substrate for Ku- and
K-bands, The Scientific World Journal, 2013 (2013), 10,
doi:10.1155/2013/378420
5 C. M. Wu, Dual-band CPW-fed cross-slot monopole antenna for
WLAN operation, IET Microw. Antennas Propag., 1 (2007),
542–546, doi:10.1049/iet-map:20050116
6 M. R. I. Faruque, M. T. Islam, B. Yatim, M. A. M. Ali, Analysis of
the effects of metamaterials on the radio-frequency electromagnetic
fields in the human head and hand, Mater. Tehnol., 47 (2013) 1,
129–133
7 J. W. Wu, H. M. Hsiao, J. H. Lu, S. H. Chang, Dual broad band de-
sign of rectangular slot antenna for 2.4 and 5 GHz wireless, Electron.
Lett., 40 (2004), 1461–1463, doi:10.1049/el: 20046873(410)
8 C. M. Su, H. T. Chen, K. L. Wong, Printed dual band dipole antenna
with U-slot arms for 2.4/5.2 GHz WLAN operation, Electr. Lett., 38
(2002), 1308–1309, doi:10.1049/el:20020919
9 Y. H. Suh, K. Chang, Low cost microstrip fed dual frequency printed
dipole antenna for wireless communications, Electr. Lett., 36 (2000),
1177–1179, doi:10.1049/el:20000880
10 D. Nashhat, H. A. Elsadek, H. Ghali, Dual band reduced size PIFA
antenna with U-slot for blue tooth and WLAN operations, Proc. of
IEEE Antennas Propagation International Symposium, 2 (2003),
962–965, doi:10.1109/APS.2003.1219395
11 C. C. Lin, G. Y. Lee, K. L. Wong, Surface mount dual loop antenna
for 2.4/5 GHz WLAN operations, Electr. Lett., 39 (2003),
1302–1304, doi:10.1049/el:20030845
12 Y. L. Kuo, K. L. Wong, Printed double-T monopole antenna for
2.4/5.2 GHz dual band WLAN operations, IEEE Trans. Antennas
Propag., 51 (2003), 2187–2192, doi:10.1109/TAP.2003.816391
13 M. R. I. Faruque, M. M. Islam, M. T. Islam, Investigation of a planar
antenna with bandwidth enhancement for X-band applications,
Electronics World, 120 (2014), 12–16
14 A. Khidre, K. F. Lee, F. Yang, A. Eisherbeni, Wideband circularly
polarized E-shaped patch antenna for wireless applications, IEEE
Antennas and Propag. Magazine, 52 (2010), 219–229, doi:10.1109/
MAP.2010.5687547
15 M. M. Islam, M. R. I. Faruque, M. T. Islam, A Compact 5.5 GHz
Band-Rejected UWB Antenna Using Complementary Split Ring
Resonators, The Scientific World Journal, 2014 (2014), 10,
doi:10.1155/2014/528489
16 M. H. Ullah, M. T. Islam, J. S. Mandeep, N. Misran, A new double
L-shaped multiband patch antenna on a polymer resin material
substrate, Applied Physics A, 110 (2013), 199–205, doi:10.1007/
s00339-012-7114-0
17 I. J. Bahl, P. Bhartia, Microstrip Antennas, 2nd Edn., Artech House,
Boston, London 1980
Md. M. ISLAM et al.: EFFECTS OF AN EPOXY-RESIN-FIBER SUBSTRATE FOR A -SHAPED MICROSTRIP ANTENNA
Materiali in tehnologije / Materials and technology 50 (2016) 1, 33–37 37

U. CALIGULU et al.: X-RAY RADIOGRAPHY OF AISI 4340-2205 STEELS WELDED BY FRICTION WELDING
39–45
X-RAY RADIOGRAPHY OF AISI 4340-2205 STEELS WELDED BY
FRICTION WELDING
RENTGENSKI PREGLED JEKEL AISI 4340-2205, VARJENIH
S TRENJEM
Ugur Caligulu1, Mahmut Yalcinoz2, Mustafa Turkmen3, Serdar Mercan4
1Firat University, Faculty of Technology, Dept. of Met. and Materials Eng., Elazig, Turkey
2Firat University, Faculty of Technical Education, Dept. of Metallurgy Education, Elazig, Turkey
3Kocaeli University, Hereke Vocational School, 41800 Kocaeli, Turkey
4Cumhuriyet University, Faculty of Technology, Dept. of Mechatronic Eng., Sivas, Turkey
ucaligulu@firat.edu.tr
Prejem rokopisa – received: 2014-08-25; sprejem za objavo – accepted for publication: 2015-03-04
doi:10.17222/mit.2014.211
In this study, X-ray radiographic tests of friction-welded AISI 4340-AISI 2205 steels were investigated. AISI 4340 tempered
steel and AISI 2205 duplex stainless steel, each of 12 mm diameter, were used to fabricate the joints. The friction-welding tests
were carried out using a direct-drive-type friction-welding machine for different parameters. After this process, the radiographic
tests of the welded joints were examined by X-ray diffraction. The experimental results indicated that the AISI 4340 tempered
steel could be joined to the AISI 2205 duplex stainless steel using the friction-welding technique and for achieving a weld with
sufficient strength. The result of the radiographic tests indicated that by increasing the rotation speed, the friction pressure and
the forging pressure, the amount of flash increased for all the specimens. In contrast, when increasing the friction time the
amount of flash decreased. The best properties for steels AISI 4340-2205 were observed for the specimens welded at a rotation
speed of 2200 min–1, a friction pressure of 40 MPa, a forging pressure of 80 MPa, a friction time of 6 s and a forging time of 3 s.
Keywords: AISI 4340, AISI 2205, friction welding, radiographic test
V tej {tudiji so bili rentgensko pregledani spoji jekel AISI 4340-AISI 2205, zvarjeni s trenjem. Spoji so bili izdelani s popu{~e-
nim AISI 4340 jeklom in AISI 2205 dupleks nerjavnim jeklom, vsako premera 12 mm. Preizkus varjenja s trenjem je bil izvr{en
z uporabo neposredno gnanega stroja za varjenje s trenjem, pri razli~nih parametrih. Po tem postopku so bili zvarjeni spoji
pregledani z rentgenom. Rezultati preizkusov so pokazali, da je z varjenjem s trenjem mogo~e spojiti AISI 4340 popu{~eno
jeklo in AISI 2205 dupleks nerjavno jeklo in da je mogo~e dobiti zvar z zadostno trdnostjo. Rezultati rentgenskih preiskav so
pokazali, da pri nara{~ajo~i hitrosti vrtenja, tornega tlaka in tlaka pri kovanju dele` zmeh~anega roba nara{~a pri vseh vzorcih.
Nasprotno pa se pri podalj{anju ~asa trenja zmanj{a koli~ina zmeh~anega roba. Najbolj{e lastnosti pri jeklih AISI 4340-2205 so
bile opa`ene pri vzorcih, varjenih z rotacijo 2200 min–1, tornim tlakom 40 MPa, kova{kim tlakom 80 MPa, ~asom trenja 6 s in
~asom kovanja 3 s.
Klju~ne besede: AISI 4340, AISI 2205, varjenje s trenjem, rentgensko testiranje
1 INTRODUCTION
Duplex stainless steel (DSS) is well known for its
excellent strength and corrosion resistance. However,
joining DSS plates by fusion welding causes a sig-
nificant reduction in the mechanical properties, because
of microstructure changes during weld solidification. It
is essential to maintain the characteristics of the weld
zone to use DSS in servicing highly critical environ-
ments, such as ocean-mining machinery, oil and gas pipe
lines, desalination plants and chemical tankers of ships,
etc. DSS has ferrite () and austenite (
) in approxi-
mately equal proportions, which possess body centered
cubic (BCC) and face centered cubic structure (FCC),
respectively.1 During the controlled alloying process of
the DSS, under equilibrium conditions, ferrite-promoting
elements (Cr, Mo, Mn, W, Nb, Si, Ti and V) will con-
centrate by diffusing in the ferrite. At the same time,
austenite-promoting elements (Ni, C, N, Co and Cu) will
concentrate by diffusing in the austenite phases. This
gives the formation of a dual-phase microstructure.2,3 But
the welding of DSS forces the microstructure to remain
in an excessive ferritic nature, because of the higher
amounts of ferrite promoting elements in its chemical
composition, and also due to a faster cooling rate.
Austenite usually nucleates in the temperature range
1200–900 °C. During cooling, the weld zone remains in
this temperature range for a very short period of time,
i.e., from 4 s to 15 s. Thus, the arc energy and filler
metal composition play a major role in the microstruc-
tural stability after welding.4
Tempered types of steel are machinery manufactured
steels with and without alloy, whose chemical composi-
tions, especially in terms of carbon content, are suitable
for hardening and which show high toughness under a
specific tensile strength at the end of the tempering
process. Tempered types of steel, due to their superior
mechanical properties, acquired at the end of the temper-
ing process, are used in a wide range of areas, including
the manufacture of parts such as various machine and
engine parts, forging parts, various screws, nuts and stud
Materiali in tehnologije / Materials and technology 50 (2016) 1, 39–45 39
UDK 669.14:621.791.1:620.179.152 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)39(2016)
bolts, crank shafts, shafts, control and drive components,
piston rods, various shafts, gears. For this reason, tem-
pered steels are the type of steel used and produced at
the highest rate after unalloyed steels and construction
steels. These steels constitute the most important part of
the machinery-manufacturing steels. Generally, such
steels are used for the production of fitting, axle shaft,
the shaft and the gear.5–9
Friction welding is a solid-state joining process that
can be used to join a number of different metals. The
process involves making welds in which one component
is moved relative to, and in pressure contact with, the
mating component to produce heat at the faying surfaces.
Softened material begins to extrude in response to the
applied pressure, creating an annular upset. Heat is
conducted away from the interfacial area for forging to
take place. The weld is completed by the application of a
forge force during or after the cessation of the relative
motion. The joint undergoes hot working to form a
homogenous, full surface, high-integrity weld. Friction
welding is the only viable method in this field to
overcome the difficulties encountered in the joining of
dissimilar materials with a wide variety of physical
characteristics. The advantages of this process are,
among others, no melting, high reproducibility, short
production time and a low energy input.10–19
Welding technology is commonly used in many
areas. Because it is aimed to provide high and constant
quality in manufacturing sector and in products, the
importance of non-destructive is the testing methods in
quality-control strategies. Accordingly, the non-destruc-
tive testing of welded joints has become a part of the
total quality system.20,21 Being one of the most important
parts of quality control, non-destructive material testing
method is the complementary part of the manufacturing.
The non-destructive method is the common name for
testing methods through which the static and dynamic
information about the materials are obtained by testing
the materials without damaging them. Thanks to the
non-destructive testing method, defects such as cracks
occurred during manufacturing or after used for a while,
space in internal structure, edge reduction, etc., are de-
tected (Table 1).
The methods applied in non-destructive testing are
visual testing, liquid-penetrant testing, eddy-current
testing, magnetic particle inspection, ultrasonic inspec-
tion and radiographic inspection.22
High-energy electromagnetic waves may penetrate
into many materials. The radiation penetrating a specific
material may affect the radiation-sensitive films that are
put on the other side of the material. After the develop-
ment of the films, the image of the inside of the material
is seen. This image occurs because of the spaces in the
material or thickness/density changes. This method is
called radiographic testing, which is one of the oldest
methods of nondestructive testing and has been in use for
approximately five decades. Among the advantages of
this method, compared to other methods, such as ultra-
sound tests, is the formation of an internal šphotograph’
of the material, which no other method is able to achieve.
Various radiation sources may be used in radiographic
testing. The radiographic testing of weld bead or casting
pieces using X-rays or gamma source is one of the most
important uses for this inspection method. The energy
gap of the X-ray used in industrial radiography is gene-
rally between 50 kV and 350 kV. The beam energy varies
according to the type and thickness of the material. In
order to get precise results from the testing, it has to be
done in accordance with the standards. These standards
are determined by considering the type of the material
and/or the type of product. There are also application
standards together with the standards according to which
the acceptance levels are determined. The testing is done
by determining the standards suitable for the features of
the product. Radiographic testing is generally applied
according to the EN 1435 or EN 12517 standards.23–25
The radiography method is applied to ferromagnetic,
non-ferromagnetic metals and other all materials.
Because X-ray provides the opportunity to analyze the
microstructure of the materials without making any
damage, it is widely used in non-destructive testing. Via
X-ray or gamma ray, thickness changes, structural
U. CALIGULU et al.: X-RAY RADIOGRAPHY OF AISI 4340-2205 STEELS WELDED BY FRICTION WELDING
40 Materiali in tehnologije / Materials and technology 50 (2016) 1, 39–45
Table 1: Non-destructive testing experiments in industrial areas37
Tabela 1: Neporu{ne preiskave na podro~ju industrijskih preiskav37
Practice Area Function Application Examples
Research and
Development
Structural evaluation of materials, Comparison of
production and assembly methods and evaluation
findings.
Examination of fatigue and microstructure of metals
and the detection of cracks in the welding seam.
Production
Control
Method
Determination of the variable production method and
to control.
Radiographic and ultrasonic thickness measuring
method and determination of the manufacturing
parameters.
Quality
Control
Defective parts and the detection of abnormalities,
Manufacturing assembly defects, place and method of
evaluation.
Poor adhesion, cracking in welding, metal in the
non-uniform pores and the determination of material
defects.
During the
service
evaluation
Wear and use during the early identification of
abnormalities.
Corrosion in pipes and location of warehouses and
detection, Variety of early-warning systems in
vehicles.
changes, inner defects, montage details can be
determined.26–30
The conventional method of inspection requires that
the radiographic images are first-rate and are conse-
quently controlled by international standards. However,
radiographic inspection by inspectors is done subjec-
tively and requires great experience, keenness of vision
and knowledge of the techniques employed, and yet even
when done adequately, interpretation errors occur
whether it be the non-detection of a defect present or the
incorrect classification of a detected defect.25,31–33 The
advantages of the radiography method may be seen as
follows, the result is shown with an image, permanent
records that may be seen outside of the test area can be
obtained, the sensitivity is shown on every film and it
may be applied to any kind of material. As for the dis-
advantages, they may be sorted as follows, it is not
suitable for thick pieces, may be harmful to health, direct
calorie is needed for two-dimensional faults, the film
needs posing and showing, is not suitable for automa-
tion, surface defects, and it does not give information
about the depth of the defect under the surface. The
equipment that is used is rather expensive in comparison
with other methods and at most it needs careful work
concerning the radiation safety.34
In the present paper, X-ray radiographic testing of
AISI 4340-2205 steels welded by friction welding were
investigated.
2 MATERIALS AND METHODS
AISI 4340 tempered steel and AISI 2205 duplex
stainless steel of 12 mm diameter were used to fabricate
the joints in this study. Table 2 illustrates the chemical
compositions of the base metals. The friction-welding
tests were carried out using a direct-drive-type friction-
welding machine. Table 3 has the mechanical properties
and Table 4 has the physical properties of AISI 4340 and
AISI 2205 steels. Table 5 illustrates the experimental
conditions. The experimental set-up is shown in Figure
1.17
Table 4: Physical properties of copper and low carbon steel
Tabela 4: Fizikalne lastnosti bakra in malo oglji~nega jekla
Materials (10-6)

(W/m °C)

(nΩ m)
E
(GPa)
AISI 4340
AISI 2205 14.7 19 85 200
: Thermal Expansion Coefficient (20-800 °C)
: Thermal Conductive (20 °C)
: Electrical Resistance (20 °C)
E: Elastic modulus (20 °C)
After the friction-welding procedure the specimens
were divided into sections transversely in order to
investigate the microstructural variations from the centre
to the outside of the weld. Transverse sections were pre-
pared, and then grinding and polishing with 3-μm dia-
mond paste were made in order to conduct a metallo-
graphic examination of the joined materials. The
specimens were etched in a chemical solution for AISI
4340 2 % HNO3 + 98 % alcohol and in a solution for
AISI 2205 25 % HNO3 + 75 % pure water to conduct the
microstructural examination (7.5 V + 30 s) The micro-
structures of the joints were observed using light
microscopy (LM), the energy-dispersive spectroscopy
(EDS) and X-ray diffraction (XRD). Microhardness
measurements were taken under a load of 50 g. The
tensile tests were conducted at room temperature with
10–2 mm s–1 cross-head rate.
In controlling the weld seam, as the thickness is 4
mm, radiographic testing was chosen from among the
U. CALIGULU et al.: X-RAY RADIOGRAPHY OF AISI 4340-2205 STEELS WELDED BY FRICTION WELDING
Materiali in tehnologije / Materials and technology 50 (2016) 1, 39–45 41
Table 2: Chemical compositions of test materials
Tabela 2: Kemijska sestava preizkusnih materialov
Materials
Alloy Elements (w/%)
C Mn Si P S Cr Mo Ni N Cu
AISI 4340 0.4 0.8 0.3 0.035 0.040 0.9 0.3 2.00 - -
AISI 2205 0.018 1.686 0.309 0.026 0.003 22.333 3.379 4.932 0.191 0.097
Table 3: Mechanical properties of copper and low carbon steel
Tabela 3: Mehanske lastnosti bakra in malo oglji~nega jekla
Materials Tensile Strength(MPa)
Yield Strength 0.2 %
(MPa)
Elongation
(%)
Microhardness
(HV)
AISI 4340 659 400 20.98 201
AISI 2205 956 620 20 328
Figure 1: Experimental set-up17
Slika 1: Eksperimentalni sestav17
non-destructive methods and an X-ray tube was chosen
as the radiation source. The principle can be seen in Fig-
ure 2.35,36
Table 5: The process parameters used in the friction welding
Tabela 5: Parametri procesa, uporabljeni pri varjenju s trenjem
Sample
no
Welding parameters
Rotating
speed
(min–1)
Friction
pressure
(MPa)
Forging
pressure
(MPa)
Friction
time
(s)
Forging
time
(s)
Axial
short-
ening
(mm)
S1 2200 30 60 6 3 4.00
S2 2200 30 60 10 5 4.37
S3 2100 30 60 6 3 3.15
S4 2100 30 60 10 5 3.41
S5 2000 30 60 6 3 2.10
S6 2000 30 60 10 5 2.80
S7 2200 40 80 6 3 4.50
S8 2200 40 80 10 5 4.78
S9 2100 40 80 6 3 3.44
S10 2100 40 80 10 5 3.96
S11 2000 40 80 6 3 2.20
S12 2000 40 80 10 5 2.60
Tests TS 5127 and EN 1435 were applied, according
to the standards, in class B and in a type that will cover
the area affected by the weld and the heat (Figure 3).
The X-ray tension that was chosen according to the
thickness of the material was 130 kV to image (Figure
4).
The X-ray device, Rigaku Radioflex–300EGS3 type,
having the capacity of 300 kV was used (Figures 5a and
5b).
U. CALIGULU et al.: X-RAY RADIOGRAPHY OF AISI 4340-2205 STEELS WELDED BY FRICTION WELDING
42 Materiali in tehnologije / Materials and technology 50 (2016) 1, 39–45
Figure 5: a) Rigaku mark Radioflex-300EGS3-type device and b)
control panel
Slika 5: a) Naprava vrste Rigaku Radioflex-300EGS3 in b) kontrolna
plo{~a
Figure 4: The deep-penetration thickness and material as a function
up to 500 kV for the X-ray device and to determine the voltage plot
graphic
Slika 4: Debelina globine penetracije pri materialu v odvisnosti od
napetosti do 500 kV za rentgensko napravo in za dolo~anje diagrama
napetost debelina
Figure 3: Test preparation for plane wall and one wall
Slika 3: Priprava preizkusa za ravno steno
Figure 6: Film, penetremeter, stenciling pattern and beam setting
Slika 6: Film, merilec globine penetracije, vzorec {ablone, nastavitev
snopa
Figure 2: Working principle of radiographic test35,36
Slika 2: Princip delovanja rentgenskega preizkusa35,36
C4 type, 100 × 240 mm2 Kodak film was used. Front
and back lead screens with a thickness of 0.125 mm
were used. The weld seam applied to the material with
the thickness of 4 mm was filmed by sending the beam
to the pose diagram for 48 s. The distance between the
X-ray device and film was 600 mm. The placement of
the film is shown in Figure 6.37,38
3 RESULTS AND DISCUSSION
The flash obtained was symmetric, which indicated
plastic deformation on both the rotating and upsetting
(reciprocating) side. The integrity of the joints was eva-
luated for the friction-welded joints. The friction-pro-
cessed joints were sectioned perpendicular to the bond
line and observed through an optical microscope. It is
clear that there were no cracks and voids in the weld
interface. From the microstructural observations, the
microstructures formed in the interface zone during or
after FW processes, there are three distinct zones across
the specimens identified as unaffected zone (UZ),
deformed zone (DZ) and transformed and recrystallized
fully plastic deformed zone (FPDZ).39 Typical grain
refinement occurred in the DZ region by the combined
effect of the thermal and mechanical stresses (Figure 7).
A typical micrograph showing the different morpho-
logies of the microstructure at different zones of the
friction-processed joint is shown in Figure 7.
According to the International Institute of Welding,
welding defects and the explanations of the radiographic
images were defined as in Table 6.37,38 The films are
placed into the viewer shown in Figure 8 and the image
is evaluated according to Table 6. It was determined that
the most common welding defects shown in Table 6
have a lack of penetration according to the definitions of
welding defects and the radiographic images (D) (S2). In
other samples the defects were not shown.
In Figure 9 the radiographic testing images of all the
samples are shown. The experimental results indicated
that AISI 4340 tempered steel could be joined to AISI
2205 duplex stainless steel using the friction-welding
technique and for achieving a weld with sufficient
strength. The result of the radiographic tests indicated
that by increasing the rotation speed, friction pressure
and forging pressure the amount of flash increased in all
the specimens. In contrast, when increasing the friction
time the amount of flash decreased. The best properties
of the AISI 4340-2205 steels were observed for the spe-
cimens welded at a rotation speed of 2200 min–1, a
friction pressure of 40 MPa, a forging pressure of 80
U. CALIGULU et al.: X-RAY RADIOGRAPHY OF AISI 4340-2205 STEELS WELDED BY FRICTION WELDING
Materiali in tehnologije / Materials and technology 50 (2016) 1, 39–45 43
Table 6: Definition of weld defects and radiographic image37,38
Tabela 6: Definicije napak zvarov in rentgenskih posnetkov37,38
A: Gas gaps
Aa: Porosity
Ab: Gas bubbles
Description * Because the captured gas bubbles are formed.
* Gas channels or long gaps
Radiographic
Image
* Sharp black shadows around the circle.
* Sharp black depending on the round or the long shadows of the error change.
B: Slag
Ba: Slag
Bb: Slag errors
Description * Slag or other foreign materials during the welding.
* Captured within gaps slag or foreign matter.
Radiographic
Image
* Dark shadows or random shapes.
* Continuous dark lines parallel to the seam edge welding.
C. Insufficient
Welding
Description Between the main material source material during welding seam merger due to lack
of two-dimensional error.
Radiographic
Image
Sharp-edged thin dark line.
D. Insufficient Deep
Penetration
Description Merger at the root welding of the lack of sewing filled fully with the welding or
root.
Radiographic
Image
The middle of the dark seam continuous or discrete line welding.
E. Cracks
Ea: Vertical Cracks
Eb: Horizontal
Cracks
Description Local tensile strength of metal exceeded.
Radiographic
Image
Flat thin dark line.
F. Swelter Channel Description Welding material on the surface along the seam formed channel or groove.
Radiographic
Image
Weldings are spread wide and dark line along the seam.
Figure 7: Regions in which occurred microstructural changes39
Slika 7: Podro~ja, kjer se pojavijo spremembe v mikrostrukturi39
MPa, a friction time of 6 s and a forging time of 3 s
(Figure 9).
4 CONCLUSIONS
In this study, X-ray radiographic testing of AISI 4340
tempered steel and AISI 2205 duplex stainless steel
welded with friction welding were investigated. The
following results were obtained.
• Friction-welding experiments were carried out using
a direct-drive-type friction-welding machine accord-
ing to Table 5. This study concluded that the AISI
4340 tempered steel could be joined succesfully to
AISI 2205 duplex stainless steel using the friction-
welding technique. The best joining was seen in
number S7. It was clear that the joining decreased in
the other samples.
• Comprehensive microstructural investigations for the
AISI 4340-2205 steels’ friction-welded joints re-
vealed that there were different regions at the welding
interface, the wideness of fully plasticized deformed
zone (FPDZ) decreases when rotational speed and
friction pressure increase.
• The larger microstructural changes take place in the
HAZs. An increase in the contraction of the samples
was observed after increasing the friction-welding
rotation speeds. The width of the HAZ is mainly
affected by the friction time and rotation speed. This
infers that the width and formation of HAZs that
occurred as a result of the reactions taking place at
the welding interface have an adverse effect on the
mechanical strength and, consequently, the quality of
the friction-welded joints.
• It has been determined that the most common weld-
ing defects shown in Table 6 have a lack of pene-
tration according to the definitions of welding defects
and the radiographic images (D) (Sample No: 2). In
other samples there were no defects. The result of the
radiographic tests indicated that by increasing the
rotation speed, the friction pressure and the forging
pressure the amount of flash increased in all the
specimens. In contrast, when increasing the friction
time the amount of flash decreased. The best proper-
ties of were AISI 4340-2205 steels observed for the
specimens welded at a rotation speed of 2200 min–1, a
friction pressure of 40 MPa, a forging pressure of 80
MPa, a friction time of 6 s and a forging time of 3 s
(Figure 9).
• The highest deformation was always for the AISI
4340 tempered steel side and in all samples the
original structure was preserved in the undeformed
region.
5 REFERENCES
1 ASM Handbook on welding, vol. 6, ASM International Publisher,
1993, 471–481
2 J. Charles, Duplex stainless steel, A Review, Proc. 7th Duplex Int.
Conf. & Expo, Grado, Italy, 2007
3 J. C. Lippold, D. J. Kotecki, Welding metallurgy and weldability of
stainless steels, Wiley-Interscience, 2005
4 A. V. Jebaraj, L. Ajaykumar, Microstructure analysis and the influ-
ence of shot peening on stress corrosion cracking resistance of
duplex stainless steel welded joints, Indian Journal of Engineering &
Materials Sciences, 21 (2014), 155–167
5 S. D. Meshram, T. Mohandas, R. G. Madhusudhan, Friction welding
of dissimilar pure metals, Journal of Materials Processing Tech-
nology, 184 (2008), 330–337, doi:10.1016/j.jmatprotec.2006.11.123
6 P. Sathiya, S. Aravindan, A. Noorul Haq, Some experimental
investigations on friction welded stainless steel joints, Materials and
Design, 29 (2008), 1099–1109, doi:10.1016/j.matdes.2007.06.006
7 S. Celik, I. Ersozlu, Investigation of the mechanical properties and
microstructure of friction welded joints between AISI 4140 and AISI
1050 steels, Mater. Design, 30 (2009), 970–976,
doi:10.1016/j.matdes.2008.06.070
8 http://www.celmercelik.com.
9 M. Sahin, H. E. Akata, An experimental study on friction welding of
medium carbon and austenitic stainless steel components, Indust.
Lubricat. Tribol., 56 (2004) 2, 122–129, doi:10.1108/
00368790410524074
10 M. Sahin, Evaluation of the joint interface properties of auste-
nitic-stainless steels (AISI 304) joined by friction welding, Materials
and Design, 28 (2007), 2244–2250, doi:10.1016/j.matdes.2006.05.
031
11 I. Kirik, N. Ozdemir, Weldability and joining characteristics of AISI
420/AISI 1020 steels using friction welding, International Journal of
Materials Research, 104 (2013) 8, 769–775, doi:10.3139/146.110917
12 N. Ozdemir, F. Sarsilmaz, A. Hascalik, Effect of rotational speed on
the interface properties of friction-welded AISI 304L to 4340 steel,
Mater. Des., 28 (2007), 301–307, doi:10.1016/j.matdes.2005.06.011
13 Welding handbook, Welding Processes, Volume 2, Eighth edition,
American Welding Society Inc., Miami 1997, 739–761
14 R. E. Chalmers, The Friction Welding Advantage, Manufacturing
Engineering, 126 (2001), 64–65
U. CALIGULU et al.: X-RAY RADIOGRAPHY OF AISI 4340-2205 STEELS WELDED BY FRICTION WELDING
44 Materiali in tehnologije / Materials and technology 50 (2016) 1, 39–45
Figure 9: Radiographic test photographs of samples (S1 – S12)
Slika 9: Rentgenski posnetki vzorcev (S1 – S12)
Figure 8: Film-examination device (Viewer)
Slika 8: Naprava za pregled filma
15 N. Ozdemir, Investigation of the mechanical properties of friction-
welded joints between AISI 304L and AISI 4340 steel as a function
rotational speed, Materials Letters, 55 (2005), 2504–2509,
doi:10.1016/j.matlet.2005.03.034
16 D. E. Spindler, What Industry Needs to Know about Friction
Welding, Welding Journal, (1994), 37–42
17 I. Kirik, N. Ozdemir, U. Caligulu, Effect of particle size and volume
fraction of the reinforcement on the microstructure and mechanical
properties of friction welded MMC to AA 6061 aluminum alloy,
Kovove Mater., 51 (2013) 4, 221–227, doi:10.4149/km_2013_4_221
18 I. Kirik, N. Ozdemir, F. Sarsilmaz, Microstructure and Mechanical
Behaviour of Friction Welded AISI 2205/AISI 1040 Steel Joints,
Materials Testing, 54 (2012) 10, 683–687, doi:10.3139/120.110379
19 M. B. Uday, M. N. Ahmad Fauzi, H. Zuhailawati, A. B. Ismail,
Advances in friction welding process: a review, Science and
Technology of Welding and Joining, 15 (2010) 7, 534–558,
doi:10.1179/136217110X12785889550064
20 M. Taskin, U. Caligulu, M. Türkmen, X-Ray Tests of AISI 430 and
304 Stainless Steels and AISI 1010 Low Carbon Steel Welded by
CO2 Laser Beam Welding, Materials Testing, 53 (2011) 11–12,
741–747, doi:10.3139/120.110283
21 H. Dikbas, U. Caligulu, M. Taskin, M. Türkmen, X-Ray Radio-
graphy of Ti6Al4V Welded by Plasma Tungsten Arc (PTA) Welding,
Materials Testing, 55 (2013) 3, 197–202, doi:10.3139/120.110426
22 http://makina.ktu.edu.tr/static/lab_foy/lab21.doc
23 http://www.wtndt.metu.edu.tr/ndt/tr/node/8
24 http://asalmakina.com/anasayfam.asp?sid=6&pid=4
25 R. R. da Silva, L. P. Calôba, M. H. S. Siqueira, J. M. A. Rebello,
Pattern recognition of weld defects detected by radiographic test,
NDT&E International, 37 (2004) 6, 461–470, doi:10.1016/j.ndteint.
2003.12.004
26 T. Tekiz, The Non-destructive Testings, ITU Faculty of Mechanical
Engineering, Istanbul, 1984
27 M. Albayrak, The Control and Inspection of the Welding Seams,
IGDAS, 1997
28 http://www.ndt-ed.org
29 http://www.wtndt.metu.edu.tr
30 TS EN 444, TS EN 462 Standards, 1994
31 K. Aoki, Y. Suga, Intelligent image processing for abstraction and
discrimination of defect image in radiographic film, Proceedings of
the Seventh International Offshore and Polar Engineering Con-
ference, Honolulu, USA, 1997, 527
32 A. Kehoe, G. A. Parker, Image processing for industrial radiographic
inspection: image enhancement, Br J NDT, 32 (1990) 4, 183–190
33 Y. Cherfa, Y. Kabir, R. Drai, X-rays image segmentation for NDT of
welding defects, 7th European Conference on Non Destructive
Testing, Copenhagen, 1998, 2782
34 C. R. Clayton, K. G. Martin, Conf. Proceedings High Nitrogen
Steels, The Institute of Metals, Lille, 1989, 256
35 S. Ekinci, The Evaluation of the Welding Seam Errors with Digital
Radiographic Methods, The Atom Energy Foundation of Turkey,
Istanbul
36 The Certificate of Material Testing Knowledge, Eregli Iron and Steel
Plants T.A.S
37 N. Ozakin, H. Baycik, The Radiographic Inspection of the Welding
Seam of the Body of Ship, The 4th Iron–Steel Congress, Karabuk,
2007, 289
38 A. Topuz, The Non-destructive Inspections, YTU, Istanbul, 1993
39 S. Mercan, N. Ozdemir, A Couple of AISI 2205/AISI 1020 Material
Combination with Friction Welding Method, NWSA-Technological
Applied Sciences, 2A0080, 8 (2013) 2, 18–34, doi:10.12739/
NWSA.2013.8.2.2A0080
U. CALIGULU et al.: X-RAY RADIOGRAPHY OF AISI 4340-2205 STEELS WELDED BY FRICTION WELDING
Materiali in tehnologije / Materials and technology 50 (2016) 1, 39–45 45

L. GOMID@ELOVI] et al.: THERMODYNAMIC PROPERTIES AND MICROSTRUCTURES ...
47–53
THERMODYNAMIC PROPERTIES AND MICROSTRUCTURES OF
DIFFERENT SHAPE-MEMORY ALLOYS
TERMODINAMI^NE LASTNOSTI IN MIKROSTRUKTURA
RAZLI^NIH ZLITIN Z OBLIKOVNIM SPOMINOM
Lidija Gomid`elovi}1, Emina Po`ega1, Ana Kostov1, Nikola Vukovi}2,
Dragana @ivkovi}3, Dragan Manasijevi}3
1Mining and Metallurgy Institute, Zeleni bulevar 35, 19210 Bor, Serbia
2University of Belgrade, Faculty of Mining and Geology, \u{ina 7, 11000 Belgrade, Serbia
3University of Belgrade, Technical Faculty, VJ 12, 19210 Bor, Serbia
lgomidzelovic@yahoo.com
Prejem rokopisa – received: 2014-08-26; sprejem za objavo – accepted for publication: 2015-02-06
doi:10.17222/mit.2014.212
The results of a thermodynamic-properties calculation conducted using a general solution model (GSM) and an experimental
investigation of the microstructures of different shape-memory alloys (SMAs) are presented in this paper. The investigated
alloys belong to ternary systems Cu-Al-Zn and Cu-Mn-Ni and to quaternary system Ni-Cu-Fe-Mn. The examinations were
conducted using light microscopy (LM) and scanning electron microscopy with energy-dispersive X-ray spectrometry
(SEM-EDX).
Keywords: thermodynamics, shape-memory alloys, microstructure, LM, SEM-EDX
V tem ~lanku so predstavljeni rezultati termodinami~nih izra~unov lastnosti, ki so bili izvr{eni z uporabo splo{nega modela
re{itev (GSM) in eksperimentalne preiskave mikrostrukture razli~nih zlitin z oblikovnim spominom (SMAs). Preiskovane zlitine
pripadajo ternarnim sistemom Cu-Al-Zn in Cu-Mn-Ni in kvaternarnem sistemu Ni-Cu-Fe-Mn. Preiskave so bile izvedene s
pomo~jo svetlobne mikroskopije (LM), z vrsti~no elektronsko mikroskopijo (SEM) in z rentgensko energijsko disperzijsko
spektrometrijo (EDX).
Klju~ne besede: termodinamika, zlitine z oblikovnim spominom, mikrostruktura, LM, SEM-EDX
1 INTRODUCTION
Shape-memory materials are able to recover their
original shape after being distorted, at the presence of the
right stimulus. These materials include: a) shape-me-
mory alloys, b) shape-memory polymers, c) shape-me-
mory composites and newly developed d) shape-memory
hybrids1.
The shape-memory effect was first discovered for a
gold-cadmium alloy in the 1930s, but this type of
behavior of materials did not attracted the attention of
the researchers until 1960s, when a significant recover-
able strain was observed for a Ni-Ti alloy, enabling
commercial applications.
Shape-memory alloys (SMAs) are characterized by
unique properties (pseudoelasticity and shape-memory
effect), which enable them to "remember" their original
shapes. These alloys are used as activators, changing
their shapes, positions and other mechanical characte-
ristics in a response to a variation in the temperature and
electromagnetic field.
SMAs can be classified, in accordance with the
alloying metals, into:
1. Alloys based on nickel (Ti-Ni, Ni-Mn-Ga)
2. Alloys based on copper (Cu-Zn-Al, Cu-Zn-Si, Cu-
Zn-Sn, Cu-Zn-Ga, Cu-Zn-Mn, Cu-Zn-Al-Ni, Cu-Zn-
Al-Mn, Cu-Al-Ni, Cu-Al-Be, Cu-Al-Mn)
3. Alloys based on iron (Fe-Mn, Fe-Ni-C, Fe-Mn-Cr,
Fe-Mn-Si, Fe-Ni-Nb, Fe-Co-Ni-Ti)
4. Alloys based on noble metals (Au-Cd, Au-Ag, Pt-Al,
Pt-Ga, Pt-Ti, Pt-Cr)
5. Exotic alloys (In-Te, In-Cd, V-Nb)2.
The interest in SMAs is continuously increasing as
new areas of application are discovered. Today, SMAs
are used in different areas such as civil engineering3,4, the
production of microsystems5, medicine6–8, earthquake
technologies9–11 and robotics12,13.
The first copper-based SMA to be commercially
exploited was the Cu-Al-Zn alloy and the shape-memory
alloys from this ternary system typically contain mass
fractions of w(Zn) = 15–30 % and w(Al) = 3–7 %.
Cu-Mn-Ni shape-memory alloys are magnetic, but
some of their properties (like the brittleness) limit their
applications, so the alloying elements like gallium, iron
or aluminum are added to an alloy in order to achieve
satisfying characteristics.
The objective of this work is to provide some new
information about the thermodynamics and microstruc-
tures of selected shape-memory alloys.
Materiali in tehnologije / Materials and technology 50 (2016) 1, 47–53 47
UDK 536.7:66.017 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)47(2016)
2 EXPERIMENTAL WORK
The characterization of the selected shape-memory
alloys was done using light microscopy and a SEM-EDX
analysis. The samples were obtained from the industrial
production. The composition, shape and production
method of the investigated samples are given in Table 1.
The samples were used as prepared (no annealing).
The microstructural analysis of the investigated
samples was performed with light microscopy (LM),
using a Reichert MeF2 microscope (a magnification of
up to 500×) and a SEM-EDX analysis performed on a
JEOL JSM-6610LV scanning electron microscope (a
magnification of up to 300000×) coupled with an Oxford
Instruments, X-Max 20 mm2 SDD, energy-dispersive
X-ray spectrometer (an accelerating voltage of 20 kV
and a beam current of 1.25 nA). Prior to the
metallographic analysis, the surfaces of the polished
samples were etched with an appropriate etching
solution (Table 2) in order to reveal the structures of the
investigated alloys.
Table 2: Solutions used for sample etching
Tabela 2: Raztopine, uporabljene za jedkanje vzorca
Sample Etching solution
A1 HCl+H2O2+H2O
A2 FeCl3+HCl+H2O
A3 FeCl3+HCl+H2O
A4 FeCl3+HCl+H2O
3 THEORETICAL FUNDAMENTALS
Among many available methods for calculating the
thermodynamic properties of a ternary system based on
the information about the constitutive binary systems,
Chou’s general solution model (GSM)14,15 proved to be
the most reasonable in all respects, overcoming the inhe-
rent defects of the traditional symmetrical and asymme-
trical geometric models. This model breaks down the
boundaries between symmetrical and asymmetrical sys-
tems and generalizes various situations; the accuracy of
the calculation was also proven with practical
examples16,17.
Recently, a new, improved version of the general
solution model based on the Redlich-Kister parameters
was presented by Zhang and Chou18. As the older version
of GSM required a series of integration processes, which
significantly complicated the calculation, and a large
number of real systems can be approximately fit using a
Redlich-Kister polynomial, a new formalism, based on
the binary Redlich-Kister-type parameters, was pre-
sented.
Therefore, this new GSM version is utilized for cal-
culating the thermodynamic properties of the Cu-Al-Zn
and Cu-Mn-Ni ternary systems.
The basic equation of the general solution model for
a ternary system is:
ΔG x x L x x x
x x L
i
i
n
i
i
n
E = − + − +
+
=
=
∑1 2 12
0
1 2 12 3
2 3 23
1
0
2 1( ( ) )
∑
∑
− + − +
+ − + −
=
( ( ) )
( (
x x x
x x L x x
i
ik
k
n
2 3 23 1
3 1 31
0
3 1 31
2 1
2 1

 ) )x i2
(1)
Similarity coefficient  is defined as:
 
 
 
12 = +I I II/ ( ) (2)
 
 
 
23 = +II II III/ ( ) (3)
 
 
 
31 = +III III I/ ( ) (4)
and the deviation sum of squares can be calculated
using:

 I = + + +
− +
+
+ +
=
∑ 12 2 1 2 3 2 5
1
1
12
0
13
2
( )( )( )
( )
( )
i i i
L L
j k
i
l
n
i
( )( )
( )( )
j k j k
L L L L
k j
m
j
n
i i
+ + + +
− −
>=
∑∑ 3 50 12 13 12
1
13
1
(5)

 II = + + +
− +
+
+ +
=
∑ 12 2 1 2 3 2 5
1
1
21
0
23
2
( )( )( )
( )
(
i i i
L L
j k
i
l
n
i
)( )( )
( )( )
j k j k
L L L L
k j
m
j
n
i i k k
+ + + +
− −
>=
∑∑ 3 50 21 23 21 23
(6)

 III = + + +
− +
+
+ +
=
∑ 12 2 1 2 3 2 5
1
31
0
32
2
( )( )( )
( )
(
i i i
L L
j k
i
l
n
i
1 3 50
31 32 31 32)( )( )
( )( )
j k j k
L L L L
k j
m
j
n
i i k k
+ + + +
− −
>=
∑∑
(7)
The basic equation of the general solution model for
a quaternary system19 is:
L. GOMID@ELOVI] et al.: THERMODYNAMIC PROPERTIES AND MICROSTRUCTURES ...
48 Materiali in tehnologije / Materials and technology 50 (2016) 1, 47–53
Table 1: Composition, shape and production method of investigated samples
Tabela 1: Sestava, oblika in na~in izdelave preiskanih vzorcev
Sample Alloy
Composition (w/%)
Shape Production method
Al Cu Zn Mn Ni Fe
A1 NiCuFeMn / 32 / 1.5 65 1.5 rod, R 1.27 cm casting
A2 CuMnNi / 84 / 12 4 / wire, R 1 mm casting, extraction
A3 CuAlZn 4.54 68.14 27.31 / / / wire, R 3.5 mm casting
A4 CuAlZn 5.7 68.27 26.03 / / / rod, R 8 cm casting
ΔG x x L X x x L Xk
k
n
k k
k
n
E = − +
= =
∑ ∑1 2 12
0
1 12 1 3 13
0
1 132 1 2( ) (( ) ( )
( )
)
( ) (
− +
+ − +
= =
∑ ∑
1
2 1 21 4 14
0
1 14 2 3 23
0
2
k
k
k
n
k k
k
n
x x L X x x L X ( )
( )
)
( ) (
23
2 4 24
0
2 24 3 4 34
0
1
2 1
− +
+ − +
= =
∑ ∑
k
k
k
n
k k
k
n
x x L X x x L 2 13 34X
k
( ) )−
(8)
with
X x xi ij i i ij
k
k
k i j
k( ) ( )
,
= +
=
≠
∑ 
1
4
(9)
   i ij
k
(ij,ik) (ij,ik) (ji,jk)( ) / ( )= + (10)
and
 (ij,ik) ij
l
l
n
ik
l
l l l
L L=
+ + +
− +
+
=
∑ 12 2 1 2 3 2 5
1
0
2
( )( )( )
( )
(l m l m l m
L L L L
m l
m
l
n
ij
l
ik
l
ij
m
ik+ + + + + +
− −
>=
∑∑ 1 3 50 )( )( )
( )( m )
(11)
In Equation (11) the second part is different from
zero only if the sum of m and n is an even number, and it
applies to all the Lij parameters that Lkij = (-1)k Lkij. In all
the equations given, Lvij is the Redlich-Kister parameter
for the binary system ij, independent of the composition
and only dependent on the temperature; GE is the integ-
ral molar excess Gibbs energy for the ternary or quater-
nary system and xi is the mole fraction of component i.
Partial thermodynamic quantities are calculated
according to the equations:
G G x G x RTi i i i
E E E= + − =( )( / ) ln1 ∂ ∂  (12)
and:
a xi i i=  (13)
4 RESULTS AND DISCUSSION
The basic thermodynamic data on the constituent
binary subsystems, needed for the calculation of the
thermodynamic properties of the investigated systems,
were taken from the available literature data20–28 and
presented in the form of Redlich-Kister parameters in
Table 3.
The results for the integral molar excess Gibbs
energies of the investigated sections at the corresponding
temperatures, obtained with the general solution model,
are given analytically in polynomial forms (Table 4).
The general solution model and Equations (12) and
(13) were used for the calculation of the copper activities
in the selected sections of ternary systems Cu-Al-Zn and
Cu-Mn-Ni and for the calculation of the nickel activity in
the selected cross-section of quaternary system Ni-Cu-
Fe-Mn. The results of these calculations are presented in
a graphic form (Figures 1 to 3). The thermodynamic
L. GOMID@ELOVI] et al.: THERMODYNAMIC PROPERTIES AND MICROSTRUCTURES ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 47–53 49
Table 3: Redlich-Kister parameters for constitutive binary systems
Tabela 3: Redlich-Kister parametri za konstitutivne binarne sisteme
System ij Loij (T) L1ij (T) L2ij (T) L3ij (T)
Al-Cu20 –67094 + 8.555*T 32148 – 7.118*T 5915 – 5.889*T –8175 + 6.049*T
Cu-Zn21 –40695.54 + 12.65269*T 4402.72 – 6.55425*T 7818.1 – 3.25416*T 0
Al-Zn22 10465.55-3.39259T 0 0 0
Cu-Mn23 1118.55 – 5.6225T –10915.375 0 0
Cu-Ni24 11760 + 1.084T –1672 0 0
Mn-Ni25 –85853 + 22.715*T –1620 + 4.902*T 0 0
Fe-Ni26 –18782 + 3.7011*T 12308 – 2.7599*T 4457 – 4.1536*T 0
Cu-Fe27 +35625.8 – 2.19045*T –1529.8 + 1.15291*T +12714.4 – 5.18624*T +1177.1
Fe-Mn28 –3950 + 0.489*T +1145 0 0
Table 4: Polynomial form of integral molar excess Gibbs energies calculated using general solution model
Tabela 4: Oblika polinoma integralnih molskih odve~nih Gibbsovih energij, izra~unanih z uporabo splo{nega modela re{itev
System Cross-section T/K Gxs/J mol–1 R2
Ni-Cu-Fe-Mn Cu:Fe:Mn=20:1:1 1873 –2180.5*xNi3 – 6229*xNi2 + 7760.2*xNi + 643.38 1
Cu-Mn-Ni Mn:Ni=3:1 1773 13831*xCu3 – 24352*xCu2 + 18408*xCu – 7883.2 1
Cu-Al-Zn Al:Zn=1:2 1373 11751*xCu2 – 7231*xCu – 5241.8 0.9895
Figure 1: Dependence of nickel activity on the composition, for
cross-section Cu : Fe : Mn = 20 : 1 : 1 from quaternary Ni-Cu-Fe-Mn
system, calculated with GSM, at 1873 K
Slika 1: Odvisnost aktivnosti niklja od sestave, za presek Cu : Fe : Mn
= 20 : 1 : 1 v kvaternarnem sistemu Ni-Cu-Fe-Mn, izra~unana z upo-
rabo GSM, pri 1873 K
properties calculated with the general solution model are
related to the liquid phase of the system, so the tem-
perature, at which the calculation was carried out, was
selected according to that rule, taking into account the
melting points of all the metals in the investigated
system.
From Figure 1, it can be seen that the nickel activity
in section Cu : Fe : Mn = 20 : 1 : 1 and at T = 1873 K
shows a variable character of the deviation from Raoult’s
law, where up to xNi = 0.4 the deviation is positive, but
with a higher content of nickel in the alloy the deviation
becomes negative, indicating that a higher amount of
nickel in the alloy leads to a better miscibility of the
alloy components.
The copper activity in cross-section Mn : Ni = 3 : 1
and at T = 1773 K (Figure 2) shows a clear positive
deviation from Raoult’s law, which can even result in an
occurrence of layering.
The copper activity in cross-section Al : Zn = 1 : 2
and at T = 1373 K (Figure 3) exhibits an apparent
negative deviation from Raoult’s law, indicating that the
L. GOMID@ELOVI] et al.: THERMODYNAMIC PROPERTIES AND MICROSTRUCTURES ...
50 Materiali in tehnologije / Materials and technology 50 (2016) 1, 47–53
Figure 4: Microstructure of sample A1: a) LM (magnification of
500×), b) SEM (magnification of 4000×) and c) positions of EDX
analysis
Slika 4: Mikrostruktura vzorca A1: a) LM (pove~ava 500×), b) SEM
(pove~ava 4000×) in c) polo`aj EDX-analiz
Table 5: Results of EDX analysis of sample A1 in amount fractions,
(x/%)
Tabela 5: Rezultati EDX-analiz vzorca A1 v mno`inskih dele`ih,
(x/%)
Position
A1
Mn Fe Ni Cu
Spectrum 1 1.23 1.52 65.49 31.75
Spectrum 2 1.20 1.69 66.63 30.48
Spectrum 3 1.26 1.67 67.95 29.12
Spectrum 4 1.31 1.27 64.97 32.46
Figure 2: Dependence of copper activity on the composition, for
cross-section Mn : Ni = 3 : 1 from ternary Cu-Mn-Ni system, calcul
ated with GSM, at 1773 K
Slika 2: Odvisnost aktivnosti bakra od sestave, za presek Mn : Ni = 3 : 1
v ternarnem sistemu Cu-Mn-Ni, izra~unana z uporabo GSM, pri 1773
K
Figure 3: Dependence of copper activity on the composition, for
cross-section Al : Zn = 1 : 2 from ternary Cu-Al-Zn system, calculated
with GSM, at 1373 K
Slika 3: Odvisnost aktivnosti bakra od sestave, za presek Al : Zn = 1 : 2
v ternarnem sistemu Cu-Al-Zn, izra~unana z uporabo GSM, pri 1373 K
miscibility of the metals in the ternary Cu-Al-Zn system
is quite good.
The results of the microstructural analysis with light
optical microscopy and SEM-EDX for sample A1 are
given in Figure 4, with the chemical composition deter-
mined with EDX presented in Table 5.
The microphotograph obtained with LM (Figure 4a)
shows that the alloy structure consists of sharp-edged
polygonal grains.
The SEM image on Figure 4b reveals the structure of
sample A1 as a gray matrix with imbedded triangular
grains, but the EDX analysis shows that the grains and
the matrix have almost identical chemical compositions.
These findings are in agreement with the fact that copper
and nickel, two components that together account for
over 90 % of the alloy’s mass, form a continuous series
of solid solutions29.
L. GOMID@ELOVI] et al.: THERMODYNAMIC PROPERTIES AND MICROSTRUCTURES ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 47–53 51
Figure 5: Microstructure of sample A2: a) SEM (magnification of
2000×) and b) positions of EDX analysis
Slika 5: Mikrostruktura vzorca A2: a) SEM (pove~ava 2000×) in b)
polo`aj EDX-analiz
Table 6: Results of EDX analysis of sample A2 in amount fractions,
(x/%)
Tabela 6: Rezultati EDX-analiz vzorca A2 v mno`inskih dele`ih,
(x/%)
Position
A2
Mn Ni Cu
Spectrum 1 15.04 4.70 80.25
Spectrum 2 14.92 4.50 80.58
Spectrum 3 15.11 4.48 80.41
Spectrum 4 15.35 4.41 80.24
Figure 6: Microstructure of sample A3: a) LM (magnification of
500×), b) SEM-EDX (magnification of 1000×) and c) positions of
EDX analysis
Slika 6: Mikrostruktura vzorca A3: a) LM (pove~ava 500×), b)
SEM-EDX (pove~ava 1000×) in c) polo`aj EDX-analiz
Table 7: Results of EDX analysis of sample A3 in amount fractions,
(x/%)
Tabela 7: Rezultati EDX-analiz vzorca A3 v mno`inskih dele`ih,
(x/%)
Position
A3
Al Cu Zn
Spectrum 1 8.83 69.75 21.42
Spectrum 2 7.93 71.11 20.96
Spectrum 3 1.11 78.03 20.87
The results of the microstructural analysis with light
microscopy and SEM for sample A2 are given in Figure
5 and the chemical compositions determined with the
EDX analysis are presented in Table 6.
Technical difficulties like the fact that the maximal
magnification of the LM apparatus is just 500× and a
very small diameter (1 mm) of sample A2 prevented us
from getting a LM photograph.
The microstructure of sample A2 (Figure 5b) is
characterized by the grains irregular in the shape and
size, and the results of the EDX analysis presented in
Table 6 are consistent with the fact that copper forms
solid solutions with nickel and manganese29.
The results of the microstructural analysis with light
optical microscopy and SEM for sample A3 are given in
Figure 6, with the chemical compositions determined
with the EDX analysis presented in Table 7.
The microstructure of alloy A3, obtained with a LM
microphotograph (Figure 6a), consists of polygonal
grains with a significant variation in size.
The results of the microstructural analysis with light
optical microscopy and SEM-EDX for sample A4 are
given in Figure 7 and the chemical compositions deter-
mined with the EDX analysis are presented in Table 8.
The microstructure of sample A4 consists of polygonal
grains, which vary in size.
According to the phase diagram of the binary Cu-Zn
and Cu-Al systems29, the solid solubility of aluminum in
copper is approximately 18 % of amount fractions, and
for zinc it goes up to 30 % of amount fractions. Consi-
dering that the base material for samples A3 and A4 is
copper (w(Cu) = 68 %), it is reasonable to expect that
aluminum and zinc will dissolve in copper, creating solid
solutions. This was confirmed with the results of the
EDX analysis presented in Tables 7 and 8. In addition,
the EDX results indicate that the homogeneity of sample
A4 is quite good because there is no significant
difference in the chemical composition analyzed at
various measuring points.
5 CONCLUSION
Different shape-memory alloys belonging to ternary
systems Cu-Al-Zn and Cu-Mn-Ni and to quaternary sys-
tem Ni-Cu-Fe-Mn were investigated. The termodynamic
properties of these alloys were investigated analytically,
using the general solution model (GSM) and the known
Redlich-Kister parameters for the constitutive binary
systems. The thermodynamic analysis showed that the
alloys with high copper amounts from systems Cu-Al-Zn
and Ni-Cu-Fe-Mn display a good miscibility, while the
alloys from the Cu-Mn-Ni system tend to display
positive deviations from Raoult’s law, which can even
lead to layering.
The microstructures of the selected alloys were inve-
stigated experimentally by means of light optic micro-
scopy (LM) and scanning electron microscopy with
energy-dispersive X-ray spectrometry (SEM-EDX). The
microstructure analysis of the investigated alloy samples
L. GOMID@ELOVI] et al.: THERMODYNAMIC PROPERTIES AND MICROSTRUCTURES ...
52 Materiali in tehnologije / Materials and technology 50 (2016) 1, 47–53
Figure 7: Microstructure of sample A4: a) LM (magnification of
80×), b) SEM-EDX (magnification of 1000×) and c) positions of EDX
analysis
Slika 7: Mikrostruktura vzorca A4: a) LM (pove~ava 80×), b)
SEM-EDX (pove~ava 1000×) in c) polo`aj EDX-analiz
Table 8: Results of EDX analysis of sample A4 in amount fractions,
(x/%)
Tabela 8: Rezultati EDX-analiz vzorca A4 v mno`inskih dele`ih,
(x/%)
Position
A4
Al Cu Zn
Spectrum 1 12.14 63.45 24.41
Spectrum 2 11.95 64.22 23.83
Spectrum 3 12.03 63.60 24.37
Spectrum 4 12.09 63.66 24.25
Spectrum 5 12.10 63.15 24.76
revealed that the microstructure is built of polygonal
grains that can significantly vary in size. The EDX
analysis results provided the information about the alloy
chemical compositions and were, overall, in agreement
with the known facts about the investigated systems. The
results presented in this paper contribute to a better
understanding of the thermodynamic properties and mi-
crostructures of the investigated shape-memory alloys.
Acknowledgement
The authors are grateful to the Ministry of Education,
Science and Technological Development of the Republic
of Serbia for the financial support provided through
Projects 34005 "Development of ecological knowl-
edge-based advanced materials and technologies for
multifunctional application" and 172037 "Modern
multi-component metal systems and nanostructured
materials with different functional properties".
6 REFERENCES
1 W. M. Huang, Z. Ding, C. C. Wang, J. Wei, Y. Zhao, H. Purnawali,
Shape memory materials, Materials Today, 13 (2010) 7–8, 54–61,
doi:10.1016/S1369-7021(10)70128-0
2 D. Achitei, P. Vizureanu, N. Cimpoesu, D. Dana, Thermo-mechani-
cal fatigue of Cu-Al-Zn shape memory alloys, Proc. of 44th Interna-
tional October Conference on Mining and Metallurgy, Bor, 2012,
401–404
3 L. Janke, C. Czaderski, M. Motavalli, J. Ruth, Applications of shape
memory alloys in civil engineering structures - Overview, limits and
new ideas, Materials and Structures, 38 (2005), 578–592,
doi:10.1007/BF02479550
4 K. K. Jee, J. H. Han, W. Y. Jang, A method of pipe joining using
shape memory alloys, Materials Science and Engineering A,
438–440 (2006), 1110–1112, doi:10.1016/j.msea.2006.02.094
5 Y. Bellouard, Shape memory alloys for microsystems: A review from
a material research perspective, Materials Science and Engineering
A, 481–482 (2008), 582–589, doi:10.1016/j.msea.2007.02.166
6 D. Mantovani, Shape Memory Alloys: Properties and Biomedical
Applications, Journal of the Minerals Metals and Materials Society,
10 (2000), 36–44, doi:10.1007/s11837-000-0082-4
7 A. Melzer, D. Stöckel, Using shape-memory alloys, Medical Device
Technology, 6 (1995) 4, 16–23
8 Y. Luo, M. Higa, S. Amae, T. Yambe, T. Okuyama, T. Takagi, H.
Matsuki, The possibility of muscle tissue reconstruction using shape
memory alloys, Organogenesis, 2 (2005) 1, 2–5, doi:10.4161/org.
2.1.1757
9 M. Dolce, D. Cardone, R. Marnetto, Implementation and testing of
passive control devices based on shape memory alloys, Earthquake
Engineering and Structural Dynamics, 29 (2000), 945–968,
doi:10.1002/1096-9845(200007)29:7<945::AID-EQE958>3.0.CO;2-
#
10 S. Saadat, J. Salichs, M. Noori, Z. Hou, H. Davoodi, I. Bar-On, Y.
Suzuki, A. Masuda, An overview of vibration and seismic applica-
tions of NiTi shape memory alloy, Smart Materials and Structures,
11 (2002), 218–229, doi:10.1088/0964-1726/11/2/305
11 R. DesRoches, B. Smith, Shape memory alloys in seismic resistant
design and retrofit: a critical review of their potential and limitations,
Journal of Earthquake Engineering, 8 (2004) 3, 415–429,
doi:10.1080/13632460409350495
12 B. Kim, M. G. Lee, Y. P. Lee, Y. I. Kim, G. H. Lee, An earth-
worm-like micro robot using shape memory alloy actuator, Sensors
and Actuators A: Physical, 125 (2006), 429–437, doi:10.1016/
j.sna.2005.05.004
13 R. B. Gorbet, R. A. Russell, A novel differential shape memory alloy
actuator for position control, Robotica, 13 (1995), 423–430,
doi:10.1017/S0263574700018853
14 K. C. Chou, A general solution model for predicting ternary thermo-
dynamic properties, Calphad, 19 (1995), 315–325, doi:10.1016/
0364-5916(95)00029-E
15 K. C. Chou, S. K. Wei, A new generation solution model for pre-
dicting thermodynamic properties of a multicomponent system from
binaries, Metallurgical and Materials Transactions B, 28 (1997),
439–445, doi:10.1007/s11663-997-0110-7
16 Lj. Balanovi}, D. @ivkovi}, A. Mitovski, D. Manasijevi}, @. @ivko-
vi}, Calorimetric investigations and thermodynamic calculation of
Zn-Al-Ga system, Journal of Thermal Analysis and Calorimetry, 103
(2011) 3, 1055–1061, doi:10.1007/s10973-010-1070-8
17 L. Gomid`elovi}, I. Mihajlovi}, A. Kostov, D. @ivkovi}, Cu–Al–Zn
System: Calculation of thermodynamic properties in liquid phase,
Hemijska Industrija, 67 (2013) 1, 157–164, doi:10.2298/HEMIND
120306041G
18 G. H. Zhang, K. C. Chou, General formalism for new generation
geometrical model: application to the thermodynamics of liquid
mixtures, Journal of Solution Chemistry, 39 (2010), 1200–1212,
doi:10.1007/s10953-010-9570-5
19 G. H. Zhang, L. J. Wang, K. C. Chou, A comparison of different
geometrical models in calculating physicochemical properties of
quaternary systems, Calphad, 34 (2010) 4, 504–509, doi:10.1016/
j.calphad.2010.10.004
20 V. T. Witusiewicz, U. Hecht, S. G. Fries, S. Rex, The Ag–Al–Cu
system: Part I: Reassessment of the constituent binaries on the basis
of new experimental data, Journal of Alloys and Compounds, 385
(2004), 133–143, doi:10.1016/j.jallcom.2004.04.126
21 A. T. Dinsdale, A. Kroupa, J. Vizdal, J. Vrestal, A. Watson, A.
Zemanova, COST Action MP0602, Version 1.0, Thermodynamic
Database, Brno, Czech Republic, 2009 (http://www.cost.eu/COST_
Actions/mpns/Actions/MP0602)
22 S. Mey, Re-evaluation of the Al-Zn system, International Journal of
Materials Research, 84 (1993) 7, 451–455
23 C. He, Y. Du, H. L. Chen, S. Liu, H. Xu, Y. Ouyang, Z. K. Liu,
Thermodynamic modeling of the Cu–Mn system supported by key
experiments, Journal of Alloys and Compounds, 457 (2008),
233–238, doi:10.1016/j.jallcom.2007.03.041
24 J. Miettinen, Thermodynamic description of the Cu–Mn–Ni system
at the Cu–Ni side, Calphad, 27 (2003) 2, 147–152, doi:10.1016/
j.calphad.2003.08.003
25 J. Miettinen, Thermodynamic solution phase data for binary Mn-
based systems, Calphad, 25 (2001) 1, 43–58, doi:10.1016/S0364-
5916(01)00029-3
26 G. Cacciamani, A. Dinsdale, M. Palumbo, A. Pasturel, The Fe-Ni
system: Thermodynamic modelling assisted by atomistic calcula-
tions, Intermetallics, 18 (2010), 1148–1162, doi:10.1016/j.intermet.
2010.02.026
27 Q. Chen, Z. P. Jin, The Fe-Cu system: A thermodynamic evaluation,
Metallurgical and Materials Transactions A, 26 (1995) 2, 417–426,
doi:10.1007/BF02664678
28 W. Huang, An Assessment of the Fe–Mn system, Calphad, 13 (1989)
3, 243–252, doi:10.1016/0364-5916(89)90004-7
29 http://www.crct.polymtl.ca/fact/documentation/sgte/sgte_figs.htm
L. GOMID@ELOVI] et al.: THERMODYNAMIC PROPERTIES AND MICROSTRUCTURES ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 47–53 53

G. SU^IK et al.: THE RELATIONSHIP BETWEEN THERMAL TREATMENT OF SERPENTINE ...
55–58
THE RELATIONSHIP BETWEEN THERMAL TREATMENT OF
SERPENTINE AND ITS REACTIVITY
ODVISNOST MED TOPLOTNO OBDELAVO SERPENTINA IN
NJEGOVO AKTIVNOSTJO
Gabriel Su~ik1, Adriana Szabóová1, L’ubo{ Popovi~1, Damir Hr{ak2
1Technical University of Kosice, Faculty of Metallurgy, Department of Ceramics, Park Komenského 3, 04200 Kosice, Slovakia
2University of Zagreb, Faculty of Metallurgy, Aleja narodnih heroja, 44103 Sisak, Croatia
gabriel.sucik@tuke.sk
Prejem rokopisa – received: 2014-09-09; sprejem za objavo – accepted for publication: 2015-02-04
doi:10.17222/mit.2014.222
In this research the effect of the thermal treatment of chrysotile serpentine on the increase in the reactivity during the process of
its leaching in a diluted hydrochloric acid was investigated. Measurements were made on samples of 5 g taken from the heap by
selecting the fractions of 3–5 mm. The calcination in air of individual samples, required for the analysis, was carried out in an
electric muffle furnace at temperatures from 500 to 1100 °C at intervals of 50 °C. The specific surface areas of the calcined
samples were measured with the multipoint B.E.T. method and the relative density with a mercury high-pressure porosimeter.
The results were related with the yield of Mg2+ in an extract of 1 g of a ground serpentine fraction from 0 to 315 μm in 250 cm3
of 0.25 M HCl, taken after 5 min from a reactor stirred at 500 min–1 and at 20 °C. The strong relation between the temperature
of the serpentinite calcination and the rate of leaching was confirmed. The specific surface area of the examined serpentine rose
from 16.2 m2 g–1 at a calcination temperature of 600 °C to the maximum value of 45.2 m2 g–1 at a calcination temperature of 700
°C. At this temperature, the degree of dehydroxylation was 82 % and, at the same time, the maximum rate of dissolution of
Mg2+ was reached. Above this temperature, the specific surface area decreased and, at a temperature of 1100 °C, it fell to a value
of 2 m2 g–1, which also resulted in a reduction of the yield of Mg2+.
Keywords: serpentinite, calcination, specific surface area, apparent porosity, leaching rate, crystallinity
V tem ~lanku je bil preiskovan vpliv toplotne obdelave krizotil serpentina na pove~anje reaktivnosti pri procesu izlo~anja iz
raztopine solne kisline. Meritve so bile izvr{ene na 5 g vzorcih, vzetih na odlagali{~u, z lo~itvijo frakcije 3–5 mm. Kalcinacija
na zraku, posameznih vzorcev za analizo, je bila izvr{ena v elektri~ni retortni pe~i pri temperaturah od 500 °C do 1100 °C, v
intervalih po 50 °C. Specifi~na povr{ina kalciniranih vzorcev je bila izmerjena z ve~to~kovno B.E.T. metodo, relativna gostota
pa z `ivo-srebrnim visoko-tla~nim porozimetrom. Rezultati so bili primerjani z izkoristkom Mg2+ pri ekstrakciji iz 1 g osnovne
frakcije serpentina z zrnatostjo 0 do 315 μm v 250 cm3 0,25 M HCl, vzeti po 5 min iz reaktorja s hitrostjo me{anja 500 min–1 pri
20 °C. Potrjena je bila mo~na odvisnost med temperaturo kalcinacije serpentina in hitrostjo izlo~anja. Specifi~na povr{ina
preiskovanega serpentina je narasla iz 16,2 m2 g–1 pri temperaturi kalcinacije 600 °C na najve~jo vrednost 45,2 m2 g–1 pri
temperaturi kalcinacije 700 °C. Pri tej temperaturi je bila stopnja dehidroksilacije 82 % in isto~asno je bila dose`ena tudi
najve~ja hitrost raztapljanja Mg2+. Nad to temperaturo se je specifi~na povr{ina zmanj{ala in pri temperaturi 1100 °C padla na
2 m2 g–1, kar je vplivalo tudi na zmanj{anje izkoristka Mg2+.
Klju~ne besede: serpentinit, kalcinacija, specifi~na povr{ina, navidezna poroznost, hitrost izlo~anja, kristalini~nost
1 INTRODUCTION
The economic efficacy of chemical technologies is
closely related with the production speed. The raw ma-
terial, the subject of this paper and also of earlier papers,
is the waste micro-chrysotile material from the Dob{iná
area (Slovakia). The aim of the chemical treatment is the
production of chemically pure magnesium compounds1,
silica2–4 and ferric hydroxide5. In the previous papers6–8,
the following leaching agents were tested: hydrochloric
acid, acetic acid and ammonium chloride. The effect of
the leaching-agent concentration and the temperature on
the kinetics of the leaching of crude serpentine was
monitored.
The impacts of the thermal-treatment calcination,
where the key parameters were the degree of conversion
and the temperature of the dehydroxylation of serpen-
tine, were examined in previous papers9–14. According to
these papers, the calcination of serpentine of up to 80 %
resulted in an up to 30-time acceleration of the transfer
of Mg2+ in the solution compared to the leaching of
crude serpentine. The rapidity is attributed to the des-
truction of the layers of magnesium octahedron and the
release of internal links. In addition, there are also
changes in the specific surface area and bulk density,
relating to the porosity15. These parameters are directly
measurable and, in contrast to the qualitative parameter
change in the crystallinity, they may be related to the
kinetic parameters of the chemical reactions of the
solidus-liquidus type, where the control process is the
reaction at the phase interface.
2 EXPERIMENTAL WORK
The measurements of the specific-surface-area and
density criteria were implemented on fractions of 3–5
mm. The apparent/open porosity was measured on sam-
Materiali in tehnologije / Materials and technology 50 (2016) 1, 55–58 55
UDK 622.78:622.782:546 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)55(2016)
ples with a cube shape and a volume of 0.2–0.5 cm3. For
the chemical and thermal analysis, the samples were
ground in a Mn-steel spherical vibrating chamber. The
chemical composition of serpentine was analyzed with
ICP OES iCAP6300 and the results are summarized in
Table 1. By heating the crude serpentine, the thermal
dissociation to the origin of the so-called topotactic
structure16 of serpentine anhydride17 (Equation (1))
occurs in a temperature range of 540–660 °C, which is
characterized by a low chemical stability and is
accompanied by an increase in the specific surface area
and porosity:
Mg3Si2O5(OH)4 (s)
540 660° °⎯ →⎯⎯⎯⎯C C–
[3MgO + 2SiO2] (s) + 2H2O (g) (1)
chrysotile topotactic structure of oxides water
By heating them to above 700 °C, the MgO and SiO2
oxides react with each other on a stable forsterite and
amorphous silica17, stabilizing the structure and increas-
ing the resistance to acids. The chemical reaction is
expressed with Equation (2):
2[3MgO + 2SiO2] (s)
≥ °⎯ →⎯⎯700 C
3Mg2SiO4(s) + SiO2 (s) (2)
topotactic structure of oxides forsterite amorphous silica
Table 1: Chemical composition of SED serpentine
Tabela 1: Kemijska sestava SED serpentina
MgO SiO2 Al2O3 CaO Fe2O3 NiO L.O.I.
47.8 28.6 2.4 0.9 5.9 0.3 12.57
The thermochemical processes are identified with a
differential thermal analysis and a NETZSCH STA
449F3 Jupiter thermogravimetric instrument, and eva-
luated in the NETZSCH Proteus program. A graphic
recording with characteristic temperature points is shown
in Figure 1 and used for setting the experimental con-
ditions of the calcination.
The calcination in air of 5 g samples of coarse frac-
tions of 3–5 mm was carried out under controlled condi-
tions in an electric muffle furnace at 500–1100 °C, with
increments of 50 °C and a dwell time at a particular
temperature of 180 min. The samples were inserted into
the cold furnace and heated at a rate of 15 °C per minute.
With this procedure, 13 samples (SED500– SED1100)
were prepared. After the annealing procedure, the
residual loss due to annealing (the degree of conversion)
and dimensional changes were determined. In the first
step of the analysis, the weight and moisture content of
each sample were determined on a KERN MLB 50-3N
thermobalance, and the geometric volume of mercury
was determined on an Amsler 9/593 volume meter. The
pore size distribution and the open porosity of the
samples were measured with the method of high-
pressure mercury porosimetry using automatic Quanta-
chrome porosimeter Poremaster 33. The specific surface
areas of SA samples were measured with a surface-and-
pore analyzer through the sorption of nitrogen, with the
B.E.T. method. The results were compared with the
parameters of the reference sample of unannealed ser-
pentine (SEDraw).
3 RESULTS AND DISCUSSION
The measurement of open porosity a of calcined
samples SED500 to SED1100 did not confirm the
expected correlation between the chemical reactivity in
terms of the leaching rate of Mg2+ and a. It can be
claimed that the maximum a of almost 20 % was
measured on the samples exposed to the temperatures of
the forsterite formation with the maximum rate of around
900 °C, with the increase from the temperature of 700 °C
in accordance with the forsterite formation17. On the con-
trary, at the temperatures in the area of dehydroxylation,
a = 1.5 % at the level of raw serpentinite (SEDraw).
Intragranular pores had the major proportion, up to 2/3
a, as shown in Figure 2.
The only noticeable consistency between a and
specific surface area SA is the local minimum of 1.5 %
and 16 m2 g–1 at 600 °C, which could be explained with
the closing of the pores due to the impact of the oxi-
G. SU^IK et al.: THE RELATIONSHIP BETWEEN THERMAL TREATMENT OF SERPENTINE ...
56 Materiali in tehnologije / Materials and technology 50 (2016) 1, 55–58
Figure 2: Open porosity and pore character depending on the calcina-
tion temperature of SED
Slika 2: Odprta poroznost in zna~ilnost por v odvisnosti od tempe-
rature kalcinacije SED
Figure 1: Differential thermal and thermogravimetric analysis of
crude serpentine
Slika 1: Diferen~na termi~na in termogravimetri~na analiza surovega
serpentina
dizing reactions of Fe2+  Fe3+. Strong correlations
between the degree of conversion  and SA can be found
in Figure 3. The maximum SA of 45.2 m2 g–1 was
measured for sample SED700 with  = 82 %. Increasing
the temperature over the recrystallization temperature
causes sintering, which is shown as a decrease in a and
SA over the value of 2 m2 g–1.
The reactivity of ground calcinate was tested for
fractions of 0–315 μm. Figure 4 shows a graphical
representation of the function of the yield of Mg2+ Mg =
f(, T) in intervals of (10, 20 and 60) min. From this
dependence it follows that for the maximum efficiency
of leaching, it is necessary to calcine serpentine so that a
conversion of about 80 % is achieved and the calcination
temperature is in a range from 600 to 700 °C. This will
also guarantee the maximum surface area, which is an
important parameter affecting the kinetics of the
leaching. On the other hand, a porosity increase does not
indicate a leaching improvement because of the structure
stabilization due to the transformation to forsterite as
shown in Figure 5.
4 CONCLUSION
The chemical reactivity of calcined serpentine de-
pends primarily on the degree of structural crystallinity.
The leaching rate of the forsterite-crystal phase is similar
to the leaching rate of crude serpentine. It makes the
thermal activation meaningless.
The reactive serpentine formation at temperatures of
600–700 °C does not significantly depend on the dwell
time, unlike in the case of higher temperatures. At a
temperature above 700 °C, a partial forsterite formation
takes place.
At the temperature above 700 °C, the apparent poro-
sity increases, but the specific surface decreases. This
phenomenon is related to the recrystallization, the over-
all contraction of the volume and the sintering.
Acknowledgement
This work was supported by the Scientific Grant
Agency of the Ministry of Education of the Slovak
Republic and The Slovak Academy of Sciences – Grant
VEGA 1/0378/14.
G. SU^IK et al.: THE RELATIONSHIP BETWEEN THERMAL TREATMENT OF SERPENTINE ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 55–58 57
Figure 5: Comparison of the total open porosity and specific surface
area as the function of temperature
Slika 5: Primerjava skupne odprte poroznosti in specifi~ne povr{ine v
odvisnosti od temperature
Figure 3: Change in the specific surface of SED depending on the
calcination temperature
Slika 3: Spreminjanje specifi~ne povr{ine SED, v odvisnosti od
temperature kalcinacije
Figure 4: Dependence of the dissolution efficiency of magnesium
depending on the calcination condition for serpentinite (SED)
Slika 4: Odvisnost u~inkovitosti raztapljanja magnezija od pogojev
pri kalcinaciji serpentina (SED)
5 REFERENCES
1 L. Haurie, A. I. Fernández, J. I. Velasco, J. M. Chimenos, J. M. L.
Cuesta, F. Espinel, Polymer Degradation and Stability, 91 (2006) 9,
989–994, doi:10.1016/j.polymdegradstab.2005.08.009
2 V. V. Velinskii, G. M. Gusev, J. of Mining Science, 38 (2002) 4,
402–404, doi:10.1023/A:1023324206554
3 A. Pietriková, M. Búgel, M. Golja, Metalurgija, 43 (2004) 4,
299–304
4 A. Fedoro~ková, B. Ple{ingerová, G. Su~ik, P. Raschman, A.
Doráková, Int. J. of Mineral Processing, 130 (2014), 42–47,
doi:10.1016/j.minpro.2014.05.005
5 R. A. Silva, C. D. Castro, C. O. Petter, I. A. H. Schneider, Mine
Water – Managing the Challenges, IMWA 2011, Aachen, 2011,
469–473
6 A. Fedoro~ková, M. Hreus, P. Raschman, G. Su~ik, Minerals
Engineering, 32 (2012), 1–4, doi:10.1016/j.mineng.2012.03.006
7 A. Fedoro~ková, P. Raschman, Chemical Papers, 100 (2006) 5,
337–347
8 W. Gao, J. Wen, Z. Li, Industrial and Engineering Chemistry
Research, 53 (2014) 19, 7947–7955, doi:10.1021/ie4043533
9 J. N. Weber, R. T. Greer, American Mineralogist, 50 (1965), 450–464
10 A. F. Gualtieri, A. Tartaglia, Journal of the European Ceramic
Society, 20 (2000) 9, 1409–1418, doi:10.1016/S0955-2219(99)
00290-3
11 D. Hr{ak, J. Malina, A. B. Had`ipa{i}, Mater. Tehnol., 39 (2009) 6,
225–227
12 C. Viti, American Mineralogist, 95 (2010) 4, 631–638, doi:10.2138/
am.2010.3366
13 A. F. Gualtieri, C. Giacobbe, C. Viti, American Mineralogist, 97
(2012) 4, 666–680, doi:10.2138/am.2012.3952
14 P. Raschman, A. Fedoro~ková, G. Su~ik, Hydrometallurgy, 139
(2013), 149–153, doi:10.1016/j.hydromet.2013.08.010
15 B. Ple{ingerová, K. K. Tká~ová, ¼. Tur~ániová, Transactions of the
Technical University of Kosice, 4 (1994) 1, 79
16 J. R. Günter, H. R. Oswald, Bulletin of the Institute for Chemical
Research, Kyoto University, 53 (1975) 2, 249–255, [cited
2014-08-22], available from World Wide Web: http://hdl.handle.net/
2433/76601
17 G. W. Brindley, R. Hayami, Mineral Magazine, Pennsylvania, 35
(1965), 189–195
G. SU^IK et al.: THE RELATIONSHIP BETWEEN THERMAL TREATMENT OF SERPENTINE ...
58 Materiali in tehnologije / Materials and technology 50 (2016) 1, 55–58
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
59–67
DEFORMATIONS AND VELOCITIES DURING THE COLD
ROLLING OF ALUMINIUM ALLOYS
DEFORMACIJA IN HITROSTI PRI HLADNEM VALJANJU
ALUMINIJEVIH ZLITIN
Mitar Mi{ovi}1, Neboj{a Tadi}1, Milojica Ja}imovi}2, Mileta Janji}3
1University of Montenegro, Faculty of Metallurgy and Technology, D`ord`a Va{ingtona bb, 81000 Podgorica, Montenegro
2University of Montenegro, Faculty of Natural Sciences and Mathematics, D`ord`a Va{ingtona bb, 81000 Podgorica, Montenegro
3University of Montenegro, Faculty of Mechanical Engineering, D`ord`a Va{ingtona bb, 81000 Podgorica, Montenegro
mitarm@ac.me
Prejem rokopisa – received: 2014-10-02; sprejem za objavo – accepted for publication: 2015-01-20
doi:10.17222/mit.2014.250
This paper presents the analysis results of investigations on the deformations and velocities during the cold rolling of strips of
aluminium alloys AW2024 and AW5083 obtained with an application of the finite-element method (FEM) and the conventional
rolling theory. The results of the simulation obtained using the FEM software DEFORM-2D were analysed for the
displacements, effective strain and velocity of rolling. The diagrams of the changes in these values along the deformation zone
were presented together with the identified characteristics of the trajectories of the surface and axis as the boundary surfaces for
simulating the cold rolling of the strips. The finite effective strains were compared with the degree of reduction of the strip
thickness, and equations were derived for the dependence of their ratio on the starting thickness of a strip. The equations for the
dependence of the velocity of a strip and the roll ratio on the degree of reduction due to the control of forward and backward
slips were also derived. The reliability of the obtained results and derived equations was tested with experimental tests,
theoretical relationships for plane strain and the results for the forward slip published in the literature.
Keywords: cold rolling, FEM, displacements, effective strain, forward slip
^lanek predstavlja analizo rezultatov preiskav deformacije in hitrosti pri hladnem valjanju trakov aluminijevih zlitin AW2024 in
AW5083, dobljenih z uporabo metode kon~nih elementov (FEM) in konvencionalne teorije valjanja. Analizirani so bili rezultati
simulacije, dobljene s FEM programsko opremo DEFORM-2D, za raztezek, efektivno obremenitev in hitrost valjanja.
Predstavljeni so diagrami sprememb teh vrednosti vzdol` deformacijske cone, skupaj z identifikacijo zna~ilnosti trajektorij
povr{ine in osi, kot mejnih povr{in za simulacijo hladnega valjanja trakov. Kon~ne efektivne obremenitve so bile primerjane s
stopnjo redukcije debeline trakov in razvite so bile ena~be za odvisnost njihovega razmerja od za~etne debeline traku. Razvite
so tudi ena~be za odvisnost hitrosti traku in razmerja valjev na stopnjo redukcije zaradi kontrole prehitevanja in zaostajanja
traku. Zanesljivost dobljenih rezultatov in razvitih ena~b je bila preizku{ena s preizkusom, s teoreti~nimi odvisnostmi za
ravninsko napetost in za rezultate prehitevanja, objavljenega v literaturi.
Klju~ne besede: hladno valjanje, FEM, raztezek, efektivna obremenitev, prednji zdrs
1 INTRODUCTION
Cold rolling is the primary technological process for
the preparation of strips, thin strips and foils of alumi-
nium and its alloys. This process is applied for strips
with thicknesses of up to 4 mm, although strips with
larger thicknesses can be rolled as well.
The cold-rolling technology has been constantly
developed and improved with the intent to establish
reliable quality and high productivity1,2. Regardless of
the continuous development in all the areas, theoretical
models of the process are still the subjects of a com-
prehensive research3,4. The understanding and control of
deformation, kinematics and stress states are particularly
complex due to the conditions for a reliable planning and
control of the process5–7. The conventional theory of
rolling8,9, the experimental procedures for tracing the
measurement weights10–12 and the procedures for the
FEM simulation4,13,14 were used in the research.
This paper reports experimental results for cold strip
rolling and characteristic results for deformations and
velocities obtained via FEM. A connection and compa-
rison of the results obtained from different research
procedures are discussed as well.
Strips made of alloys AW2024 and AW5083 with
thicknesses from 4 to 0.6 mm were chosen for this re-
search. These two alloys are widely used in all areas of
metal material application and are also characterised by a
wide range of final mechanical properties.
The analysis of the cold rolling of strips, thin strips
and foils is based on the condition that it is a process
with plane strain. In this manner, the geometry of the
deformation zone and the change in the velocity of a
strip and roll can be described using the scheme shown
in Figure 1.
For the plane strain, the change in the width can be
neglected (b0 = b1), and thus the deformations and geo-
metric criteria for the dimensions of a strip and defor-
mation zone can be described with the relationships:
h = ln(h0/h1), l = ln(l1/l0), b = ln(b1/b0) = 0;
h = –l; b0/h0  10; (ld/hm)min = 2 – 3 (1)
Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67 59
UDK 669.715:621.771.23:519.61/.64 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)59(2016)
The length of the deformation zone (ld) for absolutely
solid rolls is calculated using the following equation:
l R h h R hd = − ≈Δ Δ Δ( / )2
2 (2)
However, a high value of the contact stress (the roll
pressure) for cold rolling causes an elastic deformation
of the roll with a radial compression (roll flattening).
Thus, the length of the deformation zone increases (Fig-
ure 1) and can be calculated using the following
equation:
l x x x R h xde = + = + +0 1 0 0
2Δ (3)
The value of x0 is determined using the Hertz
equation:
x
v
E
RP0
28 1
=
−
⋅
( )
π a
(4)
where v – Poisson’s ratio, E – the modulus of elasticity
of the roll material, and Pa – the average roll pressure.
The movements of a strip at the entrance of the roll
gap occur under the action of an active tangential friction
force between the rolls and the strip. The horizontal
velocity of the strip (Vsx) is determined using the hori-
zontal velocity of the roll (Vrx). These velocities are
altered along the deformation zone, thus creating a zone
of the backward slip (lb) in which the friction force is
active and the relationship of the velocities is described
as Vsx < Vrx, and a forward-slip zone (lf) in which the
friction force is reactive and the relationship of the velo-
cities is described as Vsx > Vrx (Figure 1). The relative
difference in the velocity (Vsx – Vrx)/Vrx in zone lb is
known as the backward slip, and in zone lf, it is known as
the forward slip. The forward and backward slips of a
strip represent significant and carefully researched
weights of the kinematics of the rolling process4,15–17.
A simulation of the rolling process for plane strain
can be accomplished using the 2D method of finite ele-
ments. The scheme of the process simulation is pre-
sented in Figure 2.
As cold rolling is symmetrical to the longitudinal
x-axis, it is sufficient to present the top half of a strip
thickness and the top roll for a 2D simulation. The lower
symmetrical portion of the strip and the lower roll can be
neglected by setting the boundary condition in such a
way that the displacement velocity in the y-axis direction
is equal to zero on the longitudinal x-axis. In addition,
when the friction force cannot accomplish the grasping
of a strip by the rolls, it is necessary to use pushers,
keeping in mind that the active friction force has a
limited value and the pushing force is also limited. The
movement velocity of a pusher is smaller than the velo-
city of the movement of a strip (0.1 Vsx) and, therefore
this force acts only while the rolls grasp the strip.
2 EXPERIMENTAL DETAILS AND THE FEM
SIMULATION
2.1 Rolling process
The laboratory rolling of this work was carried out to
assess the plane strain, the tolerance of dimensions, and
the strain hardening of the alloys. Alloys AW2024 and
AW5083 were examined. The final properties of the
alloys can be obtained via different processes of thermal
treatment (T conditions) and mechanical treatment (H con-
ditions). Alloy AW2024 can be thermally hardened so
that its final mechanical properties are obtained through
the thermal precipitation, whereas alloy AW5083 cannot
be thermally hardened and its final mechanical properties
are accomplished through strain hardening or annealing.
The initial 2024-alloy strips for the experimental
rolling were industrially rolled to a thickness of 5 ± 0.01
mm and, subsequently, soft annealed (the T0 condition)
and rolled. The modes of rolling and annealing for the
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
60 Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67
Figure 1: Deformation-zone geometry with and without an elastic
deformation of rolls: h0, b0, l0 / h1, b1, l1 – thickness, width and length
of the initial/rolled strip, respectively; ld, lde – length of deformation
zone (contact between strip and roll) with and without an elastic
deformation of rolls; lb, lf – length of backward and forward slip
zones; R – roll radius; Re – radius of an elastically deformed roll; h =
h0 – h1 – absolute strip thickness reduction; hm – mean strip thickness,
i.e., (h0 + h1)/2; Vsx, Vrx – horizontal velocities of strip and roll
Slika 1: Geometrija podro~ja deformacije z in brez elasti~ne defor-
macije valjev: h0, b0, l0 / h1, b1, l1 – debelina, {irina in dol`ina
za~etnega oziroma valjanega traku; ld, lde – dol`ina podro~ja defor-
macije (stik med trakom in valjem) z in brez elasti~ne deformacije
valjev; lb, lf – dol`ina podro~ja zaostajanja in prehitevanja; R – premer
valja; Re – premer elasti~no deformiranega valja; h = h0 – h1 – abso-
lutno zmanj{anje debeline traku; hm – povpre~na debelina traku, to je
(h0 + h1)/2; Vsx, Vrx – horizontalna hitrost traku in valja
Figure 2: Scheme of 2D-FEM simulation of the rolling process
Slika 2: Shema 2D-FEM simulacije postopka valjanja
preparation of strips with thicknesses of 4, 2 and 1.28
mm are shown in Figure 3. The initial thickness of the
5083-T0 alloy strips is 1.28 ± 0.01 mm. The choice of
dimensions and reductions was intended to ensure the
conditions corresponding to the plane strain. The highest
values of the reduction for all the strip thicknesses
correspond to the conditions of the maximum bite on the
laboratory rolling mill.
The rolling was completed on a laboratory duo
rolling mill having rolls with a diameter of 125 mm and
a peripheral velocity of Vr = 173.4 mm/s. The lubrication
of the strip and rolls was carried out with hydraulic oil.
All the testing processes were completed without the use
of tension on the strips.
The annealing was completed in a laboratory furnace
with a temperature regulation accuracy of ±2 °C.
2.2 FEM simulation of the rolling process
2.2.1 Software and mathematical formulation of tasks
The commercial DEFORM-2D software was chosen
for the simulation of the planned program of cold strip
rolling under the conditions of plane strain. The
DEFORM formulation of FEM for the mechanics of a
rigid-plastic body is based on the known principle of the
minimum virtual work expressed via the functional
balance between the external and internal forces18,19:
π = + −∫ ∫ ∫  d d dV V F v S
V
v
V
i i
S
(5)
where  – the equivalent or effective stress (equal to
yield stress Kf),  – the equivalent or effective strain rate
(determined on the basis of the components of the
velocity of deformation for the plane strain),  – the
Lagrange multiplier (equal to the mean stress),  v – the
volumetric strain rate, Fi – the surface traction (for the
processes with slip friction, this represents the friction
force at the strip/roll contact determined with the
normal rolling force and the friction coefficient) and vi –
the velocity (the slipping velocity of a strip on the roll
surface).
The first term in Equation (5) refers to the deforma-
tional work, the second term expresses the condition of
incompressibility and the third term refers to the impact
of the friction forces on the strip/roll contact surface.
The Newton-Raphson iterative method is used by
DEFORM to find solutions. This method is re-
commended for the majority of problems because it
usually converges, using fewer iterations than the other
methods19. All the simulations were completed with a
strip as a rigid-plastic, deformable body hardened by the
deformation, based on experimentally determined
equations and with the rolls elastically deformed to a
diameter determined with the Hitchcock formula.
2.2.2 Choice of the finite-element mesh
The finite-element mesh provides important baseline
data for the analysis of strains, kinematics, and stress
values. The choice of the shape and number of finite ele-
ments influences the calculation and the accuracy of the
solutions for the analysed values. Square isoparametric
finite elements represent a simple shape and are adequate
for a simulation of the plane-strain process18,20. For cold
rolling of strips, the geometry of the deformation zone
takes on a simple shape. If square elements are chosen in
a manner proportional to the strip thickness, the results
for the strips with different thicknesses and under diffe-
rent conditions of reductions can be easily compared.
The square isoparametric shape was chosen for these
reasons.
The choice of the number of elements, i.e., the mesh
density, should enable the required accuracy and conver-
gence of the solutions. Based on the recommendations
for the number of finite elements relative to the thick-
ness, the checks for a mesh of 4 to 20 elements on half of
a strip thickness were carried out for the simulation of
plane strain in thin elements19. The changes in the shape
of the displacement along the x and y axes were analysed
depending on the number of finite elements. The dis-
placements represent particularly important values
because the strains, velocities and stresses are calculated
on the basis of these values. For the sake of clarity, the
displacement diagram was considered in a simplified
form with two characteristic lines (trajectories) of nodes
corresponding to the boundary surfaces of a strip: the
line of contact between the metal and roll (the strip sur-
face) and the horizontal centreline (the longitudinal axis
of the strip). Since the characteristics of the displacement
diagrams change slightly if the number of finite elements
(FE) increases above 8, the complete further analysis
was carried out with a mesh of eight finite elements on a
half strip thickness (the shape of the displacement dia-
gram is analysed in additional detail in Section 3.3.1).
2.2.3 Choice of the coefficient of friction
The choice of the contact-friction condition holds
special importance for the cold-rolling process. A quanti-
tative indicator of the friction conditions in plastome-
chanics is the contact-friction coefficient (f). A func-
tional analysis of the friction conditions is thus
considered with respect to the change in the value of this
coefficient. The simulation and analysis of the process
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67 61
Figure 3: Modes of cold rolling/annealing of alloy-2024 strips
Slika 3: Na~ini hladnega valjanja/`arjenja trakov iz zlitine 2024
can be carried out on the basis of a constant value for the
friction coefficient. In this research, the simulation was
carried out with an adopted value for the friction coeffi-
cient based on Coulomb’s law. The coefficient of friction
was varied in a range of f = 0.06–0.2 for the purpose of
performing this check4,21. In the choice of the values of
the friction coefficient, an analysis of the displacement
diagrams was carried out in a manner similar to that of
the choice of the finite-element mesh. The changes in the
diagrams of displacements for different values of the
friction coefficient from the extracted interval are evident
and expected, but these values are not expressed to the
extent of a reliable quantification along the complete
deformation zone. Therefore, for further research on
deformations and velocities, the presented results are
given for the friction-coefficient value of f = 0.1, which
is often used for simulating similar rolling conditions22,23.
3 RESULTS AND DISCUSSION
3.1 Dimensions and deformation coefficients of rolled
strips
The dimensions of the strips, the thickness reduc-
tions, the thickness tolerances (±h) and the spread (b)
of the experimental rolling were determined. The thick-
ness tolerances of the rolled strips (±h) met the produc-
tion standards for the given thicknesses and reductions of
the tested alloys24.
The experimentally verified value of the spread (b)
was less than 2 %, which satisfied the condition for plane
strain25,26.
3.2 Strain hardening of the alloys
The strain hardening of the alloys was examined via
tensile testing of standard specimens. The values of the
yield stress (Kf), which depend on the degree of defor-
mation by elongation (l) and the coefficient (K) and the
index (n) of strain hardening were determined according
to Hollomon’s equation Kf = K ln. Hollomon’s equation
was chosen because it describes the area of plastic
deformation with a satisfactory accuracy. The beginning
of the interval corresponds to the conventional yield
strength (Rp0.2). The boundary values of the uniform
deformation intervals of the method of the tensile testing
by elongation are limited.
In the area of high deformation degree, the data can
be extrapolated but cannot be experimentally verified
using the tensile-testing method. For this reason, the
strain-hardening testing was carried out within the strip-
rolling process. The strips with the same initial dimen-
sions in the thermal state (T0) were rolled with different
thickness reductions. After the rolling, specimens for the
tensile testing were prepared from the strips, and the
yield strength Rp0.2 was determined. The degree of defor-
mation for the rolled strips was calculated on the basis of
the strip thickness change (h in Equation (1)). The
results for the standard deviation prove that Hollomon’s
equation Kf = f(h) fully applies to the strain hardening
in the area of high deformation degrees.
Thus, for the FEM simulation of rolling, the chosen
alloys are considered as the rigid-plastic materials that
can be hardened in the deformational sense according to
Hollomon's equation.
3.3 Deformations
3.3.1 Displacements
Displacement diagrams (x and y) for a 4-mm-
thick strip and the singled-out trajectories of the surface
and axes of the strip are presented in Figure 4. The
points at the entrance and exit of the deformation zone
were chosen according to the positions that are located
outside the boundary values of the displacements. The
displacements relative to the geometrical length of the
deformation zone determined by Equation (3) are also
traced in this analysis.
The change in the horizontal x-displacements
shows the presence of characteristic areas and different
mutual relationships of the boundary surfaces (Figure
4a). Area I is located at the entrance of the roll gap. The
deformation begins on the surface, before the geometric
entrance into the roll gap. The deformation is non-uni-
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
62 Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67
Figure 4: Displacements of mesh nodes in the directions of: a) x-axis
and b) y-axis for reduction  = 20 % (alloy 2024, h0 = 4 mm, 8 FE, f =
0.1)
Slika 4: Premiki vozli{~ v mre`i v smeri: a) x-osi in b) y-osi pri reduk-
ciji  = 20 % (zlitina 2024, h0 = 4 mm, 8 FE, f = 0,1)
form over the cross-section so that the differences bet-
ween the surface and axis are apparent. The other trajec-
tories between the surface and axis have the same shape
and are distributed inside this area at mutually uniform
distances. Area I ends near the vertical cross-section with
the same displacements. Area II is observed after Area I
inside the deformation zone. The same characteristics are
visible, but the positions of the curves for the surface and
axis are changed. Area III displays the same characte-
ristics of the trajectories as Area I, but the inhomoge-
neity is lower. Area IV covers the exit from the roll gap,
and the relationship of the values along the trajectories is
same as in Area II. At the exit from the roll gap, the
balance between the surface and axis is re-established.
The vertical displacements y on the surface show a
monotonic flow that follows the roll geometry in the roll
gap. Because of the rolling symmetry, the vertical dis-
placement on the axis is zero (Figure 4b). The other
trajectories have the same shape as the trajectories of the
surface and are regularly distributed (the same distances
on average).
The diagrams of the horizontal and vertical displace-
ments shown in Figure 4 retain the characteristic shapes
for all the tested conditions depending on the number of
finite elements and the friction coefficient.
3.3.2 Reduction of the strip thickness and indicators of
deformation
The reduction of thickness h (often referred to as the
deformation degree) in cold rolling is a particularly
important, influential process parameter. The production
processes for rolling aluminium and its alloys are pre-
dominantly carried out in an interval of h = 20–50 %.
The same interval was used for the experimental rolling
and simulation. The results obtained with the simulation
also show that the thickness reduction significantly influ-
ences the components of deformation. In accordance
with the goal of the work, the effective strain (ef) was
particularly singled out because the value and boundary
states (yield stress, stresses and forces) primarily depend
on ef. Figure 5 presents FEM diagrams of the ef change
along the deformation zone for a strip with the starting
thickness of 4 mm that was rolled with reductions of 20
and 50 %.
The first reduction (Figure 5a) exhibits an obvious
correlation with the diagrams of the x displacements.
The difference between the lengths of the trajectories on
the surface and axis, conditioned by the shape of the
deformation zone, causes large differences in the values
of ef on these trajectories. Indeed, the same areas as for
the x displacements can be extracted from the diagram
of ef. However, the finite values of ef differ between the
surface and axis, and thus the final strain over the cross-
section of a strip is non-uniform. The strain relationships
show that the deformation zone ends with a localisation
of the plastic deformation in the surface layer. The
localisation of the plastic deformation, i.e., the elastic
area around the axis and the plastic area in the surface
zone, causes phenomena that follow the inhomogeneous
deformation of the vertical cross-section. One of these
phenomena is the residual stress27.
The increase in the strip thickness reduction to 50 %
was followed by changes in ef, so that an inhomoge-
neous deformation (differences between the values for
the surface and axis) was formed at the entrance of the
roll gap and retained along the entire deformation zone
(Figure 5b).
3.3.3 Relationships of the strip thickness and the
effective strain
Final values h and ef for the trajectories of the sur-
face and the axis, and the average values on the vertical
cross-section of the strip with h0 = 4 mm were found (for
the strips with h0 = 2 mm and 1.28 mm analogous results
were obtained). Because h and ef are integral values,
the differences between their values are apparent. The
condition of plane strain obtained with the FEM simu-
lation can be checked by comparing these two indicators
of deformation. Specifically, for plane strain with
components d1, d2, and d3 and defined relationships
d2 = 0 and d1/d3 = –1, the following equation applies for
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67 63
Figure 5: Effective strain on the surface and axis of a strip rolled with
thickness reductions of: a) 20 % and b) 50 %. Parameters of the
simulation are as listed in Figure 4.
Slika 5: Efektivna napetost na povr{ini in v osi traku, valjanega z re-
dukcijo debeline: a) 20 % in b) 50 %. Parametri simulacije so nave-
deni na sliki 4.
the effective strain: def = (2/ 3)d1, i.e., ef = 1.155 1.28
According to Equation (1) 1 = h and, therefore, the
previous equation can be rewritten as follows:
ef = 1.155h, ef/h = 1.155 and
(ef2/h2)/(ef1/h1) = 1 (6)
The values for relationships h2/h1 and ef2/ef1
slightly differ for the same rolling program. Therefore,
the compatibility of the results obtained with the simu-
lation using Equation (6) can be found with a sufficient
accuracy. These results present an opportunity for a
simple correlation of the change in the dimensions and
the effective strain obtained with FEM.
The correlation of relationships ef/h and the change
in the dimensions were assessed. The change in ef/h
depending on the thickness reduction is similar for all
the starting thicknesses. The area of change is limited,
but all the values differ from 1.155. Therefore, the mean
values of ef/h for each initial thickness were calculated.
The results were determined with a 95 % confidence
interval. The confidence interval in all the cases entirely
covers the area of the changes in the values, indicating
that the mean values can be reliably used.
It is obvious that the relationship ef/h indicates a
regular change on the surface and axis and that it
depends on the starting thickness of the strip. Thus, the
results of ef/h = f(h0) were analytically processed, and
the equations used to approximate them were determined
with a satisfactory accuracy. According to the limited
number of data, the analytical processing of the results
was completed using the STATEGRAPH program for all
the equations checked with this software program pack-
age29. The linear equations were chosen.
The derived equations of dependence ef/h = f(h0)
were additionally checked for the cold rolling of the
strips with starting thicknesses of 6, 0.6 and 0.012 mm.
The thickness of 6 mm is the largest starting thickness
for cold rolling, 0.6 mm is the largest starting thickness
for rolling thin strips, and 0.012 mm is the thickness for
rolling the thinnest foils (aluminium foils with a thick-
ness of 6 μm are produced by doubling two 12-μm foils
prior to the rolling and separation after the rolling).
The lowest value of the ratio ef/h was obtained for
the axis of a 12-μm foil and it is 1.160. This value also
differs from the theoretical value of 1.155, although only
slightly. For the axis of the 6-mm strip, the difference is
approximately 2 %, which represents an allowed deflec-
tion from the condition of proportionality for the plane
strain. Therefore, the linear equations for the transition
of the thickness reduction into effective strains depend-
ing on the starting thicknesses of the strips are quite
reliable.
3.3.4 Deformation zone
The results in Figure 6 show the positions of the iso-
lines for ef in the deformation zone during the rolling of
a strip with a thickness of 4 mm. The DEFORM program
gives the possibility of presenting the deformation zone
as isolines that divide the zone into eight intervals with a
constant increment of ef/8. The shapes of isolines as
well as the areas between two consecutive isolines show
a non-uniform deformation of the cross-section, and thus
the theory of a homogeneous deformation over the
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
64 Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67
Figure 7: Isolines of the beginning of the plastic deformation zone
(ef = 0.002) for: a) alloy 2024, strip with h0 = 4 mm and b) alloy
5083, strip with h0 = 1.28 mm (y = h/2; simulation parameters as
listed in Figure 4)
Slika 7: Izolinije za~etka podro~ja plasti~ne deformacije (ef = 0,002)
za: a) zlitine 2024, trak h0 = 4 mm in b) zlitine 5083, trak debeline h0
= 1,28 mm (y = h/2; parametri simulacije so navedeni na sliki 4)
Figure 6: Isolines of ef for rolling a strip with a thickness of 4 mm
with reductions of: a) 20 % and b) 50 % (simulation parameters as
listed in Figure 4)
Slika 6: Izolinije ef pri valjanju 4 mm debelega traku, z redukcijo: a)
20 % in b) 50 %. Parametri simulacije so navedeni na sliki 4.
cross-section adopted for the conventional-rolling theory
cannot be applied.
The positions of the isolines on the boundary of
plasticity ef = 0.002 were chosen at the beginning. The
results for the chosen conditions are given in Figure 7.
The vertical line corresponding to the first point of the
contact between the strip and the roll is presented as the
"0" position at the start of the zone. The deformation on
the cross-section begins non-uniformly (similarly to
isoline B in Figure 6), first on the surface and prior to
the initial contact of the strip and the rolls. This "out-of-
contact" deformation is a consequence of the strain
gradient, which causes the pulling of the external region
of the strip and has been known and considered in nume-
rous works in the area of rolling21,25,30. It causes a signi-
ficant difference between different reductions of the strip
at the cross-section so that these cannot be reliably
quantified.
The end of the deformation zone on the vertical
cross-section is also non-homogeneous (isoline I in
Figure 6). However, these lines represent the boundary
of the plastic area of the complete vertical cross-section.
A detailed examination showed that the plastic defor-
mation continues (after isoline I) but it is localised on the
surface layer. Thus, the actual shape of the boundary of
the plastic deformation is elastic-plastic (an approximate
mirror image of the beginning of the plastic deforma-
tion). Such boundaries are completely in accordance
with the experimental results28 and schemes used for the
analysis of the process parameters (e.g., for use of "a
slab analysis")31.
3.4 Velocities
The kinematics of cold rolling are determined by the
velocities of the strip and roll in the deformation zone,
i.e., by the horizontal (Vsx, Vrx) and vertical (Vsy, Vry) velo-
cities. In the conventional theory of rolling, the horizon-
tal velocities are predominantly considered (Figure 1).
Because they can be directly (immediately) measured,
this situation offers an opportunity to include the control
of the process in the results of the analysis.
In reality, no adequate data are available for the
vertical velocity of strips for cold rolling in the conven-
tional theory. The FEM simulation ensures that these
data can also be obtained in the deformation zone.
3.4.1 Horizontal velocity
Figure 8a shows the relationship of the horizontal
velocity of a strip and the velocity of the rolls (Vsx/Vrx)
for surface and axis trajectories. The velocity for a con-
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67 65
Figure 9: Results for the backward slip depending on the reduction
degree obtained with the FEM simulation: a) length of the backward-
slip zone and b) backward slip for the simulation parameters as listed
in Figure 4
Slika 9: Rezultati za zaostajanje v odvisnosti od stopnje redukcije,
dobljene s FEM-simulacijo: a) dol`ina podro~ja zaostajanja in b)
zaostajanja za simulacijske parametre, navedene na sliki 4
Figure 8: Diagrams of relationship changes of: a) Vsx/Vrx and b) Vsy
on the surface and on the axis along the deformation zone. Alloy
2024, h0 = 4 mm, simulation parameters as listed in Figure 4.
Slika 8: Diagram sprememb odvisnosti: a) Vsx/Vrx in b) Vsy na povr{ini
in vzdol` osi podro~ja deformacije. Zlitina 2024, h0 = 4 mm,
parametri simulacije so navedeni na sliki 4.
stant time interval () is determined by the displace-
ment itself (Vsx = x/). As the roll velocity (Vrx) also
changes slightly, it is expected that the diagrams for
Vsx/Vrx in Figure 8a completely correspond to the
diagrams of the x displacement shown in Figure 4a.
Significant data for the backward slip are shown in
Figure 9. The zone of the backward slip (lb), i.e., its
length, covers an important portion of the deformation
zone and increases with an increase in the initial thick-
ness and the degree of reduction (Figure 9a). The lowest
lb/ld relationship is >2/3 of the total deformation zone.
The highest lb/ld relationship is >3/4 of the total defor-
mation zone, which means that for a neutral cross-sec-
tion, Vsx/Vrx = 1, it is necessary to obtain an adequate
increment of the displacement and its homogenous dis-
tribution on the cross-section of the strip. Both of these
factors require a certain length of the zone but also
depend on the initial strip thickness and the reduction
degree.
Diagrams of the backward slip depending on the
initial strip thickness and reduction degree are shown in
Figure 9b. A linear increase in the backward slip can be
observed with an increase in the reduction degree and a
slight difference in the tested strip thicknesses. As the
shape of the backward-slip diagram is notably close to a
straight line, the results are approximated using straight-
line equations. The scale for the backward slip expressed
in percentages is convenient for the approximation.
However, the results for relationship Vsx/Vrx are selected
for the equation because of a simple tracing of the values
relative to the roll velocity and the connections with the
forward slip. The derived equations are obtained and the
correlation coefficients for the equations have high
values. The other indicators also confirm their signifi-
cance.
The diagrams of the backward slip and the free term
in the equations of the Vsx/Vrx relationship show that the
backward slip is unavoidable (for h = 0 Vsx/Vrx > 0). In a
physical sense, the condition for the unavoidable back-
ward slip is completely justifiable, but it is not realistic
that it differs from zero if the reduction is equal to zero.
The equations should thus be accepted to show how the
backward slip increases from zero to certain values,
which are already in the area of small reduction degrees.
The forward slip changes in the interval of 2.4 to
12.5 %. According to the continuity equation, the for-
ward slip has a significantly lower value than the back-
ward slip (a smaller length of the zone causes a smaller
total displacement).
The check of the simulation results for the backward
and forward slips can be carried out on the basis of the
continuity equation. All the results of the simulations
satisfy this condition with a high level of accuracy.
Therefore, the forward slip can be determined on the
basis of the roll velocity, the backward slip and the
reduction level.
The forward and backward slips were thoroughly
tested and experimentally determined. Thus, the calcu-
lated values of the backward and forward slips provided
in15 amount to 19.2 and 4.1 %, respectively. The results
for the forward slip, which depends on the coefficient of
friction, given in4 are 4 to 6 %. The forward slip
depending on the velocity of the rolls and the reduction
given in16 reaches up to 4 %, and in4,17 this value is as
high as 10 %. All these results highly coincide with the
results presented in this work.
3.4.2 Vertical velocity
The diagrams in Figure 8b show the change in the
vertical velocity on the surface of a strip with a thickness
of 4 mm. These changes are characterised by the maxi-
mum values at the beginning of the deformation zone.
The maximum value of velocity Vsy is almost one order
of magnitude lower than the horizontal velocity. The
influence of this velocity on the total velocity/kinematics
of rolling is thereby symbolic.
4 CONCLUSIONS
This paper presents the results and analysis of the
deformations and kinematics during the cold rolling of
aluminium alloy strips. The relevant experimental rolling
was completed on a laboratory rolling stand using strips
with thicknesses ranging from 4 to 1.28 mm.
The FEM simulation of the process was completed
using the DEFORM-2D program with a mesh of eight
finite elements on a half thickness of a strip and a con-
stant friction coefficient of f = 0.1. The choice and
analysis of deformations and velocities were carried out
with the aim of comparison with the results from the
conventional rolling theory. The following results were
established:
The horizontal displacement diagram, presented with
the chosen trajectories, has four characteristic areas with
clear differences between the changes on the surface and
axis. The shape of the changes is also reflected in the
other strain components and velocities of cold rolling;
The relationship of the reduction degree and the
effective strain based on the derived equations ensures a
simple and reliable calculation of these indicators,
through which other dependencies can also be quantified
in the analysis of the rolling process;
With use of the FEM, the boundaries of the start and
end of the deformation zone can be reliably identified as
well as the out-of-contact deformation at the entrance
and exit of the deformation zone;
The values of the backward slip amount to 18–46 %
and linearly depend on the strip thickness reduction. The
forward slip lies in the interval of 2.4–12.5 % and can be
quantified on the basis of the dependence of these two
kinematic parameters of the cold-rolling process.
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
66 Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67
5 REFERENCES
1 P. Montmitonnet, P. Buessler, ISIJ International, 31 (1991), 525–538,
doi:10.2355/isijinternational.31.525
2 SIROLL ALU, Simens AG, Linz, 2012,1-28, available from World
Wide Web: www.industry.siemens.com/verticals/metals-industry/en/
metals/aluminum-rolling/Pages/home.aspx
3 I. Yarita, M. Katahama, K. Kenmochi, Kawasaki Steel Technical
Report, 41 (1999), 20–24
4 J. G. Lenard, Primer on Flat Rolling, 2nd ed., Elsevier, London 2014,
doi:10.1016/B978-0-08-099418-5.00005-6
5 X. Liu, J Iron Steel Res Int., 18 (2011) 1, 1–7, doi:10.1016/S1006-
706X(11)60001-0
6 A. Kroll, A. Vollmer, ABB Review, 4 (2004), 44–49
7 P. Hartley, C. E. N. Sturgess, C. Liu, G. W. Rowe, Int. Mater. Rev.,
34 (1989), 19–34, doi:10.1179/imr.1989.34.1.19
8 I. Y. A. Tarnovskii, A. A. Pozdeyev, V. B. Lyashkov, Deformation of
Metals During Rolling, Pergamon Press, Oxford 2013, 200–259,
doi:10.1016/B978-0-08-010223-8.50010-4
9 H. Changqing, D. Hua, C. Jie, H. U. Xinghua, Y. Shuangcheng, Proc.
Eng., 16 (2011), 745–754, doi:10.1016/j.proeng.2011.08.1150
10 F. E. Dolzhenkov, Metall. Min. Ind., 1 (2009) 1, 33–37
11 D. Pérez, F. J. Garcia-Fernandez, I. Diaz, A. A. Cuadrado, D. G.
Ordonez, A. B. Diez, M. Dominguez, Eng. Appl. Artif. Intel., 26
(2013) 8, 1865–1871, doi:10.1016/j.engappai.2013.05.009
12 S. Zhang, B. Song, X. Wang, D. Zhao, Appl Math Model, 38 (2014),
3485–3494, doi:10.1016/j.apm.2013.11.061
13 R. Mei, L. Changsheng, X. Liu, Finite Elem Anal Des, 61 (2012),
44–49, doi:10.1016/j.finel.2012.06.006
14 S. H. Zhang, G. L. Zhang, J. S. Liu, C. S. Li, R. B. Mei, Finite Elem
Anal Des, 46 (2010), 1146–1154, doi:10.1016/j.finel.2010.08.005
15 E. B. Li, A. K. Tieu, W. Y. D. Yuen, Opt Laser Eng, 39 (2003),
479–488, doi:10.1016/S0143-8166(02)00030-1
16 E. B. Li, A. K. Tieu, W. Y. D. Yuen, J Mater Process Tech, 133
(2003), 348–352, doi:10.1016/S0924-0136(02)01049-X
17 A. K. Tieu, Y. J. Liu, Tribol. Int., 37 (2004), 77–183, doi:10.1016/
S0301-679X(03)00048-3
18 P. Hartley, I. Pillinger, C. Sturgerss, Numerical Modelling of Mate-
rial Deformation Processes: Research, Development and Application,
Springer-Verlag, London 1992, doi:10.1007/978-1-4471-1745-2
19 J. Fluhrer, DEFORMTM 2D-User’s Manual, Scientific Forming
Technologies Corp., Ohio, 2004
20 R. S. Prakash, P. M. Dixit, G. K. Lal, J Mater Process Tech, 52
(1995), 338–358, doi:10.1016/0924-0136(94)01728-J
21 Y. M. Hwang, H. H. Hsu, J Mater Process Tech, 88 (1999), 97–104,
doi:10.1016/S0924-0136(98)00390-2
22 J. G. Lenard, S. Zhang, J Mater Process Tech, 72 (1997), 293–301,
doi:10.1016/S0924-0136(97)00183-0
23 Y. J. Liu, A. K. Tieu, D. D. Wang, W. Y. D. Yuen, J Mater Process
Tech, 111 (2001), 142–145, doi:10.1016/S0924-0136(01)00541-6
24 Aluminium Alloy-EN Standards for Rolled Aluminium, Aalco
Metals Ltd, 2014, available from World Wide Web: www.aalco.
co.uk/datasheets/Aluminium-Alloy-EN-Standards-for-Rolled-Alumi
nium_51.ashx
25 P. F. Thomson, J. H. Brown, Int J Mech Sci, 24 (1982), 559–576,
doi:10.1016/0020-7403(82)90048-0
26 A. B. Richelsen, V. Tvergaard, Int J Mech Sci, 46 (2004), 653–671,
doi:10.1016/j.ijmecsci.2004.05.013
27 N. Tadi}, PhD Thesis, University of Montenegro, Podgorica, 2011
28 W. A. Backofen, Deformation processing, Addison-Wesley Pub. Co.,
1975
29 STATGRAPHICS Centurion XVI – trial version, 2014, available
from World Wide Web: www.statgraphics.com
30 H. R. Le, M. P. F. Sutcliffe, Int J Mech Sci, 43 (2001), 1405–1419,
doi:10.1016/S0020-7403(00)00092-8
31 W. F. Hosford, R. M. Cadell, Metal Forming-Mechanics and
Metallurgy, 4th ed., Cambridge University Press, Cambridge 2014
M. MI[OVI] et al.: DEFORMATIONS AND VELOCITIES DURING THE COLD ROLLING OF ALUMINIUM ALLOYS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 59–67 67

M. KOVA^I^, D. NOVAK: PREDICTION OF THE CHEMICAL NON-HOMOGENEITY OF 30MnVS6 BILLETS ...
69–74
PREDICTION OF THE CHEMICAL NON-HOMOGENEITY OF
30MnVS6 BILLETS WITH GENETIC PROGRAMMING
NAPOVEDOVANJE NEHOMOGENOSTI KEMIJSKE SESTAVE PRI
GREDICAH 30MnVS6 S POMO^JO GENETSKEGA
PROGRAMIRANJA
Miha Kova~i~, Damir Novak
[tore Steel d.o.o., @elezarska cesta 3, 3220 [tore, Slovenia
miha.kovacic@store-steel.si
Prejem rokopisa – received: 2014-11-13; sprejem za objavo – accepted for publication: 2015-02-18
doi:10.17222/mit.2014.280
[tore Steel Ltd. is a small and flexible steel plant. The plant also produces the 30MnVS6 steel grade, which is used for crack
connection rods in the automotive industry. The chemical elements are not uniformly distributed over the billet cross-sections,
consequently influencing the final product properties. The chemical distribution depends mainly on the casting parameters. The
article presents an attempt at predicting the chemical non-homogeneity of 30MnVS6 billets. With respect to the
chemical-element distribution (% C, % Si, % Mn, % V, % S) over the billet cross-sections and the casting parameters (casting
speed, casting temperature, meniscus level), several models for the chemical non-homogeneity prediction were developed by
means of the genetic-programming method. The results show that the most influential parameter is the casting speed. The results
of modeling can be practically implemented in order to reduce the chemical non-homogeneity of the billets.
Keywords: steel, casting, billets, chemical composition, non-homogeneity, modelling, genetic programming
[tore Steel je majhna, a prilagodljiva jeklarna. Proizvajajo tudi jeklo 30MnVS6, ki se uporablja za ojnice, ki se izdelujejo z
lomljenjem, za avtomobilsko industrijo. Kemijski elementi niso enakomerno porazdeljeni po prerezu gredice, kar posledi~no
vpliva na lastnosti kon~nega izdelka. Razporeditev kemijskih elementov je najbolj odvisna od parametrov vlivanja. V ~lanku je
predstavljen poskus napovedovanja kemijske nehomogenosti gredic jekla 30MnVS6. Glede na razporeditve kemijskih
elementov (% C, % Si, % Mn, % V, % S) po prerezu gredice in parametre vlivanja (hitrost vlivanja, temperatura vlivanja, nivo
taline), se je izdelalo ve~ modelov, za napovedovanje kemijske nehomogenosti gredic, s pomo~jo metode genetskega
programiranja. Rezultati ka`ejo, da je najvplivnej{i parameter hitrost vlivanja. Rezultati modeliranja se lahko uporabijo v praksi
z namenom zmanj{anja kemijske nehomogenosti gredic.
Klju~ne besede: jeklo, litje, gredice, kemijska sestava, nehomogenost, modeliranje, genetsko programiranje
1 INTRODUCTION
Due to a gradual solidification during the continuous
casting of steel, chemical-composition variations occur,
which influence the cast and, consequently, the pro-
cessed-material properties; therefore, their optimization
is essential.1
In the previous research2, the chemical composition
of the cross-section of a high-grade pipeline slab was
measured point by point. The results indicated that a ne-
gative segregation inside the central line is more severe
than that outside the central line, and that the highest
positive segregation of the elements appears close to the
inner sides of the negative segregation strips. In addition,
the segregation of the elements in the central area is
higher than that in the outer and inner arc areas.
Article3 discusses the manufacturing of bearing steels
of low distortion potential. The 100Cr6 steel billets were
spray formed to achieve metallurgical homogeneity. The
microstructures and properties of the billets produced
under different thermal conditions were studied and
evaluated. A heat-transfer model for a growing billet was
established in order to investigate the thermal profiles of
the billets during spray forming. An apparent correlation
between the cooling and solidification conditions of the
deposit and its metallurgical properties was revealed by
means of a numerical simulation and an experiment.
Gheorghies et al.4 developed a theoretical model that
was adapted for studying the steel continuous-casting
technology. The model is based on the system theory,
considering input/output, command and control para-
meters. It can be used to describe the physicochemical
processes, thermal processes, chemical processes and the
characteristics of the cast material on the basis of the
above-mentioned stages.
In the research described in5 LIBS scanning measure-
ments were performed on samples displaying segrega-
tion. The resulting quantified elemental maps correlated
very well with the data obtained with the conventional
methods.
In research6 an artificial-intelligence analyzer of the
mechanical properties of rolled steel bars was proposed
using neural networks. The complex correlation among
the steel bar properties, the billet compositions and the
control parameters of manufacturing was developed. The
developed analyzer could be used in practice in order to
improve the steel quality.
Materiali in tehnologije / Materials and technology 50 (2016) 1, 69–74 69
UDK 669.18:004.89:621.74 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)69(2016)
Hwang et al.7 tried to minimize the center segregation
with the help of a developed coupled temperature/dis-
placement finite-element model. The center segregation,
center porosity, homogeneity of elements and equiaxed
crystal zone were improved.
This paper discusses the use of the genetic-pro-
gramming method for predicting the chemical non-
homogeneity of the 30MnVS6 billets, used in rolled
conditions, for forging crack connection rods in the
automotive industry. Genetic programming is an evolu-
tionary computation-based methodology of artificial
intelligence (AI); it is similar to the genetic algorithm
(GA).8 Genetic programming (GP) is capable of solving
many different problems in industry; however, it uses
different natural phenomena in comparison to many
other AI-based approaches, such as artificial neural net-
works (ANN), swarm intelligence (SI) and gravitational
search algorithm (GSA). For a comparison of the
practical uses of the above-mentioned methods in indu-
strial applications see reference literature, for
example.9–12
The problem is stated in section 2. In the subsequent
section the experimental background and, afterwards, the
essence of the chemical non-homogeneity prediction are
presented. The analysis of the modelling results are
presented in section 4 and, finally, the main contributions
of the research and the guidelines for further research are
given in the last section.
2 EXPERIMENTAL BACKGROUND
Steelmaking begins with scrap melting in an elec-
tric-arc furnace. The melting bath, which is heated up to
the tapping temperature required for the further treat-
ment procedure, is discharged into a casting ladle.
After achieving the proper melt temperature in the
melting bath, the billets are continuously cast. The melt
flows through a sliding-gate system and ladle shroud
towards a tundish.
After filling up the tundish with the help of the
mold-filling system with tundish stoppers and sub-
merged pouring tubes, the casting is established. The
billets, with a square section of 180 mm, are cast. After
reaching a certain melting-bath level, the potentiometer
starts the flattening system, which drags the billet out of
the mold. In this way, the continuous casting is estab-
lished. Each billet goes through the cooling zone toward
the gas cutters, where it is cut and laid off onto the
cooling bed.
The data for the analysis was collected on the basis of
30 consecutively cast batches of the 30MnVS6 steel in
[tore Steel Ltd. (Table 1) from May to September 2011.
The data was taken from the technological documen-
tation of the cast batches and from the chemical archive.
The goal was to get as wide a range of influential para-
meters as possible, namely:
– the contents of C, Si, Mn, S and V in the tundish
(w/%)
– the average melt temperature in the tundish (°C)
– the average meniscus level (mm)
– the average casting speed (m/min)
– the average strand temperature in the cooling zone
(°C).
From each of the selected 30 batches, a billet was
taken from the middle of the casting and a slice was cut
out. For the chemical analysis, optical emission spec-
troscopy was used (instrument SPECTRO LAVMC12A).
Five spark spots were used for determining the chemical
non-homogeneity (Figure 1).
For example, the carbon content obtained from five
spark spots on the sample from batch number 1 is pre-
sented in Figure 2.
The carbon non-homogeneity Cn–h can be easily cal-
culated:
C
C
C
i
i
n h−
==
∑
1
5
(1)
where i is the individual spot size and C is the average
of all five carbon-content values for each spot:
C
C i
i= =
∑
1
5
5
(2)
Similarly, the non-homogeneity for each individual
chemical element can be calculated.
The experimental data and the non-homogeneities for
individual chemical elements are presented in Table 1.
M. KOVA^I^, D. NOVAK: PREDICTION OF THE CHEMICAL NON-HOMOGENEITY OF 30MnVS6 BILLETS ...
70 Materiali in tehnologije / Materials and technology 50 (2016) 1, 69–74
Figure 2: Carbon content at each spark spot
Slika 2: Vsebnost ogljika na posami~nem mestu ob`iga
Figure 1: Sample from the billet slice with the spark spots
Slika 1: Vzorec iz rezine, odrezane iz gredice, s to~kami ob`iga
3 MODELLING OF CHEMICAL NON-HOMO-
GENEITY WITH GENETIC PROGRAMMING
Genetic programming is probably the most general
evolutionary optimization method8 and it has already
been found useful for several different applications in
[tore Steel Ltd.13–17. The organisms that undergo an
adaptation are in fact mathematical expressions (models)
for chemical non-homogeneity, consisting of the avail-
able function genes (i.e., basic arithmetical functions)
and terminal genes (i.e., independent input parameters
and random floating-point constants). In our case, the
models consist of the function genes of addition (+),
subtraction (–), multiplication (*) and division (/), while
terminal genes include:
– the contents of C (C), Si (SI), Mn (MN), S (S) and V
(V) in the tundish
– the average melt temperature in the tundish (TM)
– the average meniscus level (ML)
– the average casting speed (SPEED) and
– the average strand temperature in the cooling zone
(TC).
Random computer programs of various forms and
lengths are generated by means of selected genes at the
beginning of the simulated evolution. Afterwards, the
varying of the computer programs during several itera-
tions, known as generations, is performed by means of
genetic operations. For the progress of the population,
only the reproduction and crossover are sufficient. A new
generation is obtained after the completion of various
M. KOVA^I^, D. NOVAK: PREDICTION OF THE CHEMICAL NON-HOMOGENEITY OF 30MnVS6 BILLETS ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 69–74 71
Table 1: Experimental data
Tabela 1: Eksperimentalni podatki
B
at
ch
C (w
/%
)
S
i
(w
/%
)
M
n
(w
/%
)
S (w
/%
)
V (w
/%
)
A
ve
ra
ge
m
el
t
te
m
pe
ra
tu
re
in
tu
nd
is
h
(°
C
)
A
ve
ra
ge
m
en
is
cu
s
le
ve
l
(m
m
)
A
ve
ra
ge
ca
st
in
g
sp
ee
d
(m
/m
in
)
A
ve
ra
ge
st
ra
nd
te
m
pe
ra
tu
re
in
th
e
co
ol
in
g
zo
ne
(°
C
)
C
no
n-
ho
m
og
en
ei
ty
(w
/%
)
S
i
no
n-
ho
m
og
en
ei
ty
(w
/%
)
M
n
no
n-
ho
m
og
en
ei
ty
(w
/%
)
S
no
n-
ho
m
og
en
ei
ty
(w
/%
)
V
no
n-
ho
m
og
en
ei
ty
(w
/%
)
S
um
of
C
,
S
i,
M
n,
S
an
d
V
ch
em
ic
al
no
n-
ho
m
og
en
ei
ty
(w
/%
)
1 0.29 0.55 1.43 0.05 0.09 1543 74.53453 1.135516 1092.7 35.28 34.30 34.17 31.52 33.57 168.83
2 0.29 0.58 1.44 0.058 0.1 1548 76.11422 1.128306 1091.669 5.61 1.77 2.19 10.81 2.41 22.79
3 0.28 0.59 1.43 0.05 0.09 1543 74.34058 1.114705 1093.121 10.32 6.98 7.49 19.51 6.71 51.01
4 0.3 0.6 1.43 0.051 0.1 1542 74.38523 1.148343 1110.401 22.01 23.24 24.73 22.12 24.72 116.82
5 0.29 0.59 1.46 0.052 0.1 1545 74.55143 1.143362 1100.045 40.36 39.00 38.98 27.54 36.84 182.72
6 0.29 0.57 1.45 0.051 0.11 1544 74.62807 1.144693 1100.519 33.73 34.43 35.10 38.21 34.41 175.89
7 0.29 0.58 1.47 0.05 0.1 1551 74.61523 1.115993 1092.786 33.68 38.06 38.73 36.24 40.04 186.76
8 0.3 0.57 1.43 0.058 0.1 1530 74.5125 1.136773 1123.988 32.08 35.08 33.65 26.26 33.60 160.68
9 0.29 0.58 1.45 0.05 0.1 1548 74.49432 1.136511 1123.195 22.17 22.17 20.65 23.32 20.46 108.76
10 0.3 0.57 1.45 0.059 0.1 1536 74.46816 1.139951 1121.601 5.64 5.86 6.76 22.05 8.12 48.43
11 0.3 0.63 1.44 0.059 0.11 1538 74.54645 1.142406 1121.38 33.94 33.88 31.97 27.54 33.80 161.13
12 0.29 0.58 1.43 0.06 0.1 1553 74.59237 1.116969 1116.805 40.13 40.46 39.14 36.08 38.92 194.72
13 0.29 0.58 1.43 0.055 0.1 1549 74.52381 1.117879 1111.379 24.40 28.32 26.53 42.90 29.29 151.43
14 0.3 0.55 1.43 0.045 0.09 1544 74.4881 1.133849 1105.889 25.87 26.01 25.05 23.44 23.03 123.41
15 0.29 0.61 1.45 0.052 0.1 1549 74.32708 1.126732 1108.002 19.93 23.92 23.15 23.20 24.09 114.29
16 0.28 0.6 1.42 0.053 0.1 1541 74.39798 1.150479 1118.254 32.04 33.77 32.41 26.27 33.59 158.09
17 0.28 0.57 1.46 0.06 0.1 1548 74.36111 1.133132 1120.187 21.63 22.42 20.75 24.63 20.04 109.46
18 0.28 0.58 1.44 0.058 0.1 1543 74.37007 1.165916 1119.613 32.89 33.99 33.22 30.33 33.84 164.27
19 0.28 0.59 1.45 0.05 0.11 1547 74.39048 1.130368 1123.507 35.47 39.57 38.79 39.55 39.49 192.87
20 0.29 0.6 1.45 0.058 0.1 1546 74.49107 1.134441 1115.035 12.77 7.91 6.69 12.62 6.95 46.94
21 0.29 0.63 1.48 0.06 0.1 1545 74.50678 1.139199 1119.22 23.28 23.68 23.88 14.81 23.63 109.29
22 0.29 0.6 1.45 0.059 0.1 1542 74.49815 1.144132 1127.137 39.29 40.28 39.51 38.89 38.91 196.88
23 0.29 0.6 1.42 0.052 0.1 1547 74.3956 1.110585 1126.187 7.69 6.32 6.86 15.46 5.16 41.48
24 0.29 0.58 1.42 0.056 0.1 1558 74.46951 1.092817 1124.571 9.03 8.00 7.44 21.41 7.21 53.10
25 0.29 0.61 1.43 0.06 0.1 1555 74.32818 1.082388 1120.363 23.68 20.73 21.52 25.79 22.54 114.26
26 0.28 0.61 1.46 0.052 0.11 1543 74.47059 1.125766 1132.468 3.85 0.88 1.50 6.82 0.92 13.97
27 0.29 0.59 1.44 0.052 0.1 1557 74.41155 1.103096 1116.793 21.61 23.94 23.45 17.61 22.57 109.19
28 0.28 0.58 1.44 0.055 0.1 1546 74.31526 1.132103 1133.436 23.16 21.12 21.20 26.43 20.55 112.46
29 0.29 0.58 1.44 0.055 0.1 1556 74.43371 1.101385 1128.988 32.11 37.68 39.59 42.06 38.09 189.53
30 0.29 0.62 1.45 0.052 0.1 1548 74.42917 1.09335 1126.615 21.81 23.84 23.54 17.37 22.90 109.47
computer programs and this generation is then evaluated
and compared with the experimental data.
The process of changing and evaluating organisms is
repeated until the termination criterion of the process is
fulfilled. This was the prescribed maximum number of
generations.
For the process of simulated evolutions, the following
evolutionary parameters were selected: the size of the
population of organisms – 500; the greatest number in a
generation – 100; the reproduction probability – 0.4; the
crossover probability – 0.6; the greatest permissible
depth of population (6); the greatest permissible depth,
after the operation, of the crossover of two organisms –
10; and the smallest permissible depth of organisms
when generating new organisms – 2. Genetic operations
of reproduction and crossover were used. For the
selection of the organisms, the tournament method with
tournament size 7 was used. For the model, the fitness
average relative deviation from the monitored data was
selected. It is defined as:
Δ =
−
=
∑
E P
E
n
i i
ii
n
1
(3)
where n is the size of the monitored data, Ei and Pi are
the actual (experimental) and predicted sums of the C,
Si, Mn, S and V chemical non-homogeneity, respec-
tively.
We have developed 100 independent civilizations of
mathematical models for the chemical non-homogeneity
prediction. Each civilization had its most successful
organism – mathematical model. The most successful
organism of all the civilizations is presented here:
(4)
with an average relative deviation of 27.82 %. It is
obvious that only C (the content of C in the tundish), S
(the content of S in the tundish), V (the content of V in
the tundish) and SPEED (the average casting speed) are
included in the model.
4 ANALYSIS OF THE RESULTS
A randomly driven process builds the fittest and the
most complex models from generation to generation and
uses the ingredients that are the most suitable for an
experimental environment adaptation. For curiosity’s
sake, the analysis of the genes (parameters) excluded
from the models is presented in the next figure (Figure
3).
Based on the number of the genes excluded from 100
mathematical models, we may make assumptions about
the influence of the parameters on the chemical non-
homogeneity. It is clear from the figure that out of 100
genetically obtained mathematical models only 21
models do not include the parameter of the average
casting speed (SPEED) and fewer than 30 out of 100
models do not have the parameter of the contents of C
(C), Si (SI), and V (V) included in the tundish. We can,
therefore, speculate that they are probably the most
important parameters influencing the chemical
non-homogeneity.
Figure 4 shows the calculated influences of indivi-
dual parameters on the chemical non-homogeneity using
M. KOVA^I^, D. NOVAK: PREDICTION OF THE CHEMICAL NON-HOMOGENEITY OF 30MnVS6 BILLETS ...
72 Materiali in tehnologije / Materials and technology 50 (2016) 1, 69–74
Figure 3: Frequency of genes excluded from the best 100 mathema-
tical models for chemical non-homogeneity prediction
Slika 3: Frekvenca izlo~enih genov na podlagi najbolj{ih 100-ih mate-
mati~nih modelov za napovedovanje kemijske nehomogenosti
Figure 4: Calculated influences of individual parameters on chemical
non-homogeneity while separately changing them within the range
from Table 1
Slika 4: Izra~unani vplivi posami~nih parametrov na kemijsko neho-
mogenost pri le-njihovem spreminjanju znotraj obmo~ja, navedenega
v tabeli 1
the developed model (Equation (4)) while separately
changing the individual parameters within the range
from Table 1. The dashes at individual total decarburi-
zation ranges represent the calculated average chemical
non-homogeneity of all 30 collected samples, which is
98.88 %. The average chemical non-homogeneity value
of the collected data is 122.96 %.
Figure 5 shows the calculated correlation between
the casting speed and the chemical non-homogeneity
while separately changing the average casting speed.
This can be calculated and, consequently, used for
individual influential parameters as a tool for selecting
the optimal casting speed. It should be noted that the
average value of the average casting speed for all 30
cases is 1.127 m/min (a standard deviation of 0.0191
m/min).
5 CONCLUSION
The purpose of this research was to predict the
chemical non-homogeneity of 30MnVS6 steel grade
billets. The data for the analysis was collected on the
basis of 30 consecutively cast batches. The distribution
of the chemical elements (% C, % Si, % Mn, % V, % S)
over the billet cross-sections and the casting parameters
(casting speed, casting temperature, meniscus level)
were gathered. On the basis of the gathered data, several
models for predicting the chemical non-homogeneity
were developed by means of the genetic-programming
method. There were 100 different models developed and
only the best one was used for the chemical non-homo-
geneity prediction. The relative average deviation
between the actual and the predicted scrap was 27.82 %.
In addition, the frequencies of the genes excluded from
the best 100 mathematical models for the chemical non-
homogeneity prediction were analyzed. The results show
that the parameters influencing the chemical non-homo-
geneity the most are the casting speed and the contents
of C, Si and V in the melt. Also, the influences of
individual parameters on the chemical non-homogeneity
were calculated while separately changing individual
parameters, using the best genetically developed model.
The calculation shows that the variation in the speed
changes the chemical non-homogeneity from 39 to 324
%, while the calculated average chemical non-homoge-
neity of all 30 collected samples is 98.88 %. Finally, the
correlation between the casting speed and the chemical
non-homogeneity was calculated while separately chang-
ing the average casting speed. The results of the research
can be used as a tool for selecting the optimum casting
speed. In the future, a larger sample size will be used and
a methodology for the other steel grades will be imple-
mented.
6 REFERENCES
1 W. R. Irving, Continuous casting of steel, Institute of Materials,
London 1993
2 J. Liu, Y. Bao, X. Dong, T. Li, Y. Ren, S. Zhang, Distribution and
segregation of dissolved elements in pipeline slab, Journal of
University of Science and Technology Beijing, Mineral, Metallurgy,
Material, 14 (2007) 3, 212–218, doi:10.1016/S1005-8850(07)
60041-3
3 C. Cui, U. Fritsching, A. Schulz, R. Tinscher, K. Bauckhage, P.
Mayr, Spray forming of homogeneous 100Cr6 bearing steel billets,
Journal of Materials Processing Technology, 186 (2005) 3, 496–504,
doi:10.1016/j.jmatprotec.2005.02.250
4 C. Gheorghies, I. Crudu, C. Teletin, C. Spanu, Theoretical Model of
Steel Continuous Casting Technology, Journal of Iron and Steel
Research, 16 (2009) 1, 12–16, doi:10.1016/S1006-706X(09)60003-0
5 F. Boué-Bigne, Laser-induced breakdown spectroscopy applications
in the steel industry: Rapid analysis of segregation and decarbu-
rization, Spectrochimica Acta Part B: Atomic Spectroscopy, 63
(2008) 10, 1122–1129, doi:10.1016/j.sab.2008.08.014
6 W. Wang, X. Hu, L. Ning, R. Bülte, W. Bleck, Improvement of
center segregation in high-carbon steel billets using soft reduction,
Journal of University of Science and Technology Beijing, Mineral,
Metallurgy, Material, 13 (2006) 6, 490–496, doi:10.1016/S1005-
8850(06)60100-X
7 R. C. Hwang, Y. J. Chen, H. C. Huang, Artificial intelligent analyzer
for mechanical properties of rolled steel bar by using neural net-
works, Expert Systems with Applications, 37 (2010) 4, 3136–3139,
doi:10.1016/j.eswa.2009.09.069
8 J. R. Koza, Genetic programming III, Morgan Kaufmann, San Fran-
cisco 1999
9 M. Hrelja, S. Klancnik, T. Irgolic, M. Paulic, Z. Jurkovic, J. Balic,
M. Brezocnik, Particle swarm optimization approach for modelling a
turning process, Advances in Production Engineering & Mana-
gement, 9 (2014) 1, 21–30, doi:10.14743/apem2014.1.173
10 M. Hrelja, S. Klancnik, J. Balic, M. Brezocnik, Modelling of a turn-
ing process using the gravitational search algorithm, International
Journal of Simulation Modelling, 13 (2014) 1, 30–41,
doi:10.2507/IJSIMM13(1)3.248
11 M. Chandrasekaran, D. Devarasiddappa, Artificial neural network
modeling for surface roughness prediction in cylindrical grinding of
Al-SiCp metal matrix composites and ANOVA analysis, Advances in
Production Engineering & Management, 9 (2014) 2, 59–70,
doi:10.14743/apem2014.2.176
12 N. Senthilkumar, T. Tamizharasan, V. Anandakrishnan, An ANN
approach for predicting the cutting inserts performances of different
geometries in hard turning, Advances in Production Engineering &
Management, 8 (2013) 4, 231–241, doi:10.14743/apem2013.4.170
13 M. Kova~i~, B. [arler, Genetic programming prediction of the natu-
ral gas consumption in a steel plant, Energy, 66 (2014) 1, 273–284,
doi:10.1016/j.energy.2014.02.001
14 M. Kova~i~, B. Jurjovec, L. Krajnc, Ladle nozzle opening and gene-
tic programming, Mater. Tehnol., 48 (2014) 1, 23–26
M. KOVA^I^, D. NOVAK: PREDICTION OF THE CHEMICAL NON-HOMOGENEITY OF 30MnVS6 BILLETS ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 69–74 73
Figure 5: Calculated correlation between the casting speed and che-
mical non-homogeneity while separately changing the average casting
speed
Slika 5: Izra~unana korelacija med livno hitrostjo in kemijsko neho-
mogenostjo pri spreminjanju povpre~ne livne hitrosti
15 M. Kova~i~, S. Sen~i~, Modeling of PM10 emission with genetic
programming, Mater. Tehnol., 46 (2012) 5, 453–457
16 M. Kova~i~, B. [arler, Application of the genetic programming for
increasing the soft annealing productivity in steel industry, Materials
and Manufacturing Processes, 24 (2009) 3, 369–374, doi:10.1080/
10426910802679634
17 M. Kova~i~, Genetic programming and Jominy test modeling,
Materials and Manufacturing Processes, 24 (2009) 7, 806–808,
doi:10.1080/10426910902841050
M. KOVA^I^, D. NOVAK: PREDICTION OF THE CHEMICAL NON-HOMOGENEITY OF 30MnVS6 BILLETS ...
74 Materiali in tehnologije / Materials and technology 50 (2016) 1, 69–74
F. KARACA, B. AKSAKAL: EFFECT OF THE TiBN COATING ON A HSS DRILL WHEN DRILLING THE MA8M Mg ALLOY
75–79
EFFECT OF THE TiBN COATING ON A HSS DRILL WHEN
DRILLING THE MA8M Mg ALLOY
VPLIV TiBN PREVLEKE NA HSS SVEDRU PRI VRTANJU MA8M
Mg ZLITINE
Faruk Karaca1, Bünyamin Aksakal2
1Firat University, Technology Faculty, Dept. Mech. Eng., Elazig, Turkey
2Yildiz Technical University, Faculty of Chemical and Metallurgy, Dept. Metallurgy and Mater. Eng., Istanbul, Turkey
fkaraca@firat.edu.tr
Prejem rokopisa – received: 2014-11-30; sprejem za objavo – accepted for publication: 2015-01-28
doi:10.17222/mit.2014.290
Mg alloy MA8M is used in the aerospace and food industries and in biomedical applications due to its lightness and
biocompatibility. This paper presents the drilling performance of the standard HSS and TiBN-coated drill bits during the
machining of the MA8M Mg alloy at various drill rotational speeds and feed rates. After the experiments, the surface roughness,
topography and chip formation were analyzed. Atomic-force microscopy (AFM) and surface profilometer were used for this
purpose. It was observed that higher drilling feed rates and drill rotational speeds lead to lower surface-roughness values.
TiBN-coated drill bits exhibited undesired surface qualities.
Keywords: MA8M Mg alloy, TiBN coating, twist drills, surface quality, multiple regression analysis, chip formation
Mg zlitina MA8M se uporablja v letalstvu, prehrambeni industriji in v biomedicini, ker je lahka in biokompatibilna. ^lanek
predstavlja obna{anje pri vrtanju z normalnimi HSS in s svedri s TiBN prevleko na konici, pri obdelavi Mg zlitine MA8M pri
razli~nih vrtljajih svedra in hitrostih podajanja. Po eksperimentih je bila analizirana hrapavost povr{ine, topografija in nastanek
ostru`ka. V ta namen sta bila uporabljena mikroskop na atomsko silo (AFM) in profilometer povr{ine. Ugotovljeno je, da ve~ja
hitrost podajanja in vrtenja svedra povzro~i manj{o hrapavost povr{ine. S TiBN prevle~eni svedri so povzro~ali neustrezno
kvaliteto povr{ine.
Klju~ne besede: MA8M Mg zlitina, TiBN prevleka, vija~ni svedri, kvaliteta povr{ine, multipla regresijska analiza, nastanek
ostru`ka
1 INTRODUCTION
Nowadays, reducing the energy consumption and
providing a better surface quality in several manufac-
turing industries are vital for economic production
cycles. Many manufacturing industries substituted the
materials, e.g., steel was replaced with light metals or
plastics to decrease the energy consumption and/or
increase the strength/weight ratio. Although light metals
such as aluminum or magnesium are easier to machine,
the magnesium alloys have a higher specific strength and
stiffness than aluminum alloys.1–3 Metallic implants
made of stainless steel, titanium or cobalt-chromium
alloys are used for stress shielding and revision sur-
geries, improving the quality of life and the healthcare
system. On the other hand, due to their low density and
compatibility, magnesium alloys are also very promising
as orthopedic biomaterials compared to the other me-
tallic alloys such as stainless steel and titanium alloys.4
However, the unsatisfactory corrosion resistance of mag-
nesium alloys limit their application to a great extent.5
To overcome these undesired problems, in some
studies, the microstructures and mechanical properties of
magnesium alloys were processed with cyclic closed-die
forging. Using this method under various processing
conditions resulted in the desired grain size, microstruc-
tural parameters and growth mechanical properties.3 The
surface integrity of a machined magnesium alloy used
for biomedical implants could have a critical impact on
its corrosion resistance. The influence of the cutting edge
radius and the cooling method on the surface integrity
was investigated. Cryogenic machining using a large
edge-radius tool led to a thicker grain-refinement layer
that remarkably enhanced the corrosion performance of
the magnesium alloy.5 During another surface-integrity
treatment, synergistic dry cutting/finish burnishing of
magnesium-calcium implants resulted in a good surface
finish, high compressive hook-shaped, low-residual
stress profile and extended strain hardening of the sub-
surface with little change in the grain size.4
The high-speed dry-machining process investigated
with a finite-element model predicts that the most hazar-
dous outcome, the chip ignition during machining mag-
nesium alloys, does not occur during high-speed dry
cutting with sharp PCD (polycrystalline diamond) tools.6
The effect of coated drills on the minimum-quantity-
lubrication drilling of magnesium alloys was experi-
mentally investigated using a carbon-coated HSS drill
and the AZ91 magnesium alloy. Such a coating and the
minimum-quantity-lubrication condition limited the
temperature to below the hazardous level and, hence,
Materiali in tehnologije / Materials and technology 50 (2016) 1, 75–79 75
UDK 669.721.5:621.95:620.179.11 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)75(2016)
both the drill wear and the magnesium adhesion were
successively reduced.7 On the other hand, the machining
of the AZ91 magnesium alloy with both the TiN-coated
and PCD tools was conducted to find the influence of
tool coatings in the machining of magnesium. It showed
an excessive tool wear of the TiN carbides even at low
cutting speeds, while the PCD coatings showed better
results at low film thicknesses.1
A limited number of the investigations were per-
formed in the field of machinability of magnesium
alloys. In this study, experimental drilling of the MA8M
magnesium-alloy sheet was conducted. The effects of the
drilling parameters such as the drill speed, the diameter,
the feed rate and the TiBN coating on the hole-surface
quality were analyzed with respect to the surface rough-
ness, the burr and chip formation and the hole accuracy.
After these evaluations, the optimum surface quality was
determined. I was also found that an increased drill feed
rate increased the roughness, an increased drill speed de-
creased the roughness and the TiBN coating increased
the surface roughness. The present study could be refe-
renced to similar studies.
2 EXPERIMENTAL WORK
In this study, MA8M Mg-alloy sheets having dimen-
sions of 100 × 100 × 8 mm3 were used. Two different
kinds of the cutting-tool material, namely HSS and
TiBN-coated HSS twist drills were used to compare their
effects on the surface roughness of the machined holes
and the chip formation. Some pilot experiments were
performed. Since the best results were obtained with a
6-mm drill diameter of the HSS tools, the TiBN coating
was considered only with this diameter and the others
were ignored. The tool diameter, feed rate and rotational
speed of the cutting tool were changed to explore their
effects. A full factorial experimentation was applied
using the parameters. In order to statistically identify the
correlation between the applied parameters and the sur-
face roughness, the multiple-regression-analysis method
with a confidence interval of p = 0.05 was used. The
drill-bit coating parameter variation of the regression
was analyzed separately. The surface-roughness
measurements were performed with a Mitutoyo SJ 210
profilometer with a 0.5 mm/s measuring speed and 0.25
× 4 mm length in line with the ISO 1997 standard. The
average surface roughness Ra (μm) was used to evaluate
the hole-surface roughness after the measurements were
repeated three times. Furthermore, an atomic force
microscope (AFM) was used to determine the 3D surface
topography with a scanning area of 40 × 40 μm2 and a
rate of 0.15 Hz (Park System XE-100).
3 RESULTS AND DISCUSSION
The surface roughness, surface topography, chip
formations and drill-bit coating were under consideration
when evaluating the machinability characteristics of the
MA8M Mg alloy. The measured values are discussed in
the following sections.
3.1 Average surface roughness (Ra)
The surface integrity is an important parameter in
machinability and is directly related to the surface
roughness.4,5 The effects of all the experimental para-
meters including the drill diameter, the feed rate, the drill
speed and the drill coating on the hole-surface rough-
ness, analyzed in a body with contour graphics, are given
in Figure 1. In order to plot these contour graphics, the
weighted least square method was used. As seen in
Figure 1, the surface roughness of the machined holes
has a tendency to decrease with an increase in the feed
rate for all drill types. When three drill diameters were
evaluated, the maximum surface roughness was obtained
at a 4-mm drill diameter, while 6-mm and 8-mm drills
presented similar values. The roughness reached the
highest values at 1710 min–1 when the 4-mm-diameter
drill bit was used and then became reduced at 2730
min–1, as shown in Figure 1.
On the other hand, the TiBN-coated HSS drill bit
enhanced the surface roughness to the highest rates
(Figure 1) compared to the other three drill bits. There
was a significant difference in the roughness between the
holes drilled with the TiBN-coated and uncoated 6-mm
drill bits. On the other hand, the roughness was reduced
with the increased drill diameter when the contour
graphics were analyzed. However, a drill speed of 1080
min–1 caused the lowest roughness as the TiBN coating
had an inverse effect on the surface roughness during the
drilling of the MA8M alloy. Although the increased drill
speed caused a decrease in the surface roughness at all
three feed rates and with the 6-mm and 8-mm drills, the
surface roughness increased with the TiBN coating as
shown in Figure 1.
F. KARACA, B. AKSAKAL: EFFECT OF THE TiBN COATING ON A HSS DRILL WHEN DRILLING THE MA8M Mg ALLOY
76 Materiali in tehnologije / Materials and technology 50 (2016) 1, 75–79
Figure 1: Variation in surface roughness with feed rate, drill speed,
drill diameter, in comparison with TiBN-coated drill bits
Slika 1: Spreminjanje hrapavosti povr{ine s podajanjem, hitrostjo
vrtenja svedra, premerom svedra, v primerjavi s svedrom s TiBN
prevleko
The drill speed of 1710 min–1 led to the peak rough-
ness from among the three drill speeds when using the
TiBN-coated drill. This speed was suggested as a switch
point aspect for the surface roughness (Figure 1). Espe-
cially the rates of 80 mm/min and 40 mm/min exhibited
similar roughnesses with all three drill speeds. The peak
value of the surface roughness (2.84 μm) occurred at the
speed of 1080 min–1, the feed rate of 20 mm/min and the
diameter of 4 mm, as shown in Figure 1. The drill speed,
feed rate and diameter were also presented as rather
parallel results when statistically analyzed. The drill
speed had the maximum effect on the surface roughness
when partial correlation values of the three parameters
were analyzed. The partial correlation values identifying
the effectiveness of the drill speed, the feed rate and the
diameter were Rpart(speed) = –0.62, Rpart(force) = –0.45
and Rpart(dia.) = –0.29, respectively. Statistical results
confirmed the above contour plots of the experiment.
The most effective parameter was the drill speed, and the
least effective parameter was the diameter of the drill bit,
also resulting from multiple regressions.
It was established with the analysis that the surface
roughness increased with a decreased drill diameter.
When Mg alloys are machined in a dry drilling con-
dition, a flank build-up occurs on the cutting tool and the
workpiece due to the adhesion.1 The flank build-up
formation on the Mg alloys causes sufficient temperature
at the tool-workpiece contact and the adhesion of the
workpiece material on the cutting tool leads to an in-
creased cutting-edge radius and to an increase in the
machining forces.1,8 The flank build-up formation, in the
current study, increased the cutting forces and, hence, the
performance of the cutting tool was reduced. This
suggests that the smallest drill diameter was not enough
to sufficiently overcome the cutting forces and a drilling
failure occurred. On the other hand, the largest drill
diameter was found to be more successful at removing
the chips, resulting in a lower surface roughness. In order
to achieve smooth surfaces when drilling MA8M, both
higher drill speed and feed rate were required.
3.2 TiBN Coating
In the present study, the surface roughness was in-
creased to a specific value at a drill speed of 1710 min–1,
and then it was diminished with the increasing drill
speed up to 2730 min–1. The cutting energy exceeded the
plastic-deformation force of the chip and a more effec-
tive drilling can be the reason for this fact. Moreover, it
was reported that higher cutting temperatures obtained
with higher cutting speeds cause material softening on
the shear plane, easier cuts and a smoother machined
surface.4,9,10 Hence, the maximum drill speed of 2730
min–1 caused a decrease in the surface roughness after
the peak value reached at 1710 min–1. On the contrary,
for the TiBN coating, the critical level of the roughness
did not distinctly differ from the uncoated drill bits.
According to the general trend, a roughness decrease was
observed via the increased drill speed. On the other hand,
the surface roughness was increased with a decreased
feed rate under all experimental conditions.
Larger cutting forces act more effectively in closing
surface cracks and pores. However, there is a limit to the
positive effect of the rolling force and beyond certain
levels, such a force acts as the initiating source of cracks
and cold welds, deteriorating the surface.4 However, the
TiBN coating had a minor effect on the surface-rough-
ness increment that was also found with the statistical
analysis (the partial correlation of 0.2745). On the other
hand, the feed rate had an extremely strong effect (the
partial correlation of -0.66) on the surface-roughness
reduction when compared with the drill speed and the
coating. In the present study, the TiBN coating of HSS
did not have the desired effect on the surface quality
because of a lower wear resistance. As the Mg alloys
caused the tool wear of the TiN-coated cutting tools, as
reported before,1,11 this approach had similar effects on
the MA8M drilling operation.
3.3 Chip formation
The achievement of a successful and better drilling
operation is also indicated by removed chip formations.
Moreover, the chip formations can show some variations
due to the workpiece material and operation parameters
such as cutting speed, feed rate or depth of cut.12 For in-
stance, when a regular broken chip or an irregular broken
chip is formed on a workpiece with elastic properties, no
chip breakers should be provided. Similarly, a workpiece
material with elastoplastic properties produces a conti-
nuous fragmentary chip or a continuous chip with a
wedge-shaped texture, and if the workpiece has plastic
properties then the result is a continuous type of chip.12
Long chips are usually not desirable because they can
tangle along the drill body and have to be removed ma-
nually.13 Instead, well broken chips are associated with a
smooth drilling process.
The diameter slightly influenced the chip formation,
and it seemed that a larger diameter provided for a better
chip formation. The feed rate seemed a more effective
parameter than the drill diameter. Most of the continuous
and a few irregular small-particle chips were produced
when the TiBN-coated drill bits were used. The analysis
suggests that the TiBN coating spoiled the drilling-pro-
cess performance and the hole quality, which was also
confirmed by the chip formation. The chip-formation
impairment was derived from an unstable material
adhesion on the drill, i.e., the surface roughness was
increased by the TiBN coating. Higher drill speeds
caused longer chip lengths and higher radii of helical
spirals caused flutes that made it hard to drill and hard to
push the chips away. Small-string chips are presented in
Figure 2a obtained with drilling at the speed of 1080
min–1, the 4-mm diameter and the 40-mm/min feed rate.
By comparing Figure 2b with Figure 2a and by
keeping the other parameters constant, it is seen that the
F. KARACA, B. AKSAKAL: EFFECT OF THE TiBN COATING ON A HSS DRILL WHEN DRILLING THE MA8M Mg ALLOY
Materiali in tehnologije / Materials and technology 50 (2016) 1, 75–79 77
smaller and irregular chips occurred as the drill diameter
changed from 4 mm to 6 mm. Since the drill diameter
was changed exclusively from 4 mm to 8 mm, a slight
difference was observed in the chip formation (Figures
2a and 2b). An insignificant difference in the chip for-
mation was observed when Figures 2b and 2c were com-
pared, being almost identical with Figure 2a. Obviously,
no significant difference in the chip formation was
observed between Figures 2a, 2b and 2c. Therefore, the
chip formation was not influenced by the drill diameter
and the feed rate. Figure 2d shows rather different chip
formations formed due to the TiBN coating including a
few long helical spiral chips that mostly had short
strings. An increased drill speed increased the amount of
the long helical spiral chips as seen from Figures 2d, 2e
and 2f. Long-string chip formations were obtained espe-
cially at 2730 min–1 (Figure 2f) though small-diameter
helical chips were observed for the holes drilled at 1710
min–1 (Figure 2e). Irregular, small chip formations
presented in Figure 2f are also seen in Figure 2e. In this
case, the drill speed and the TiBN coating were observed
to be more effective than the feed rate in the chip for-
mation.
3.4 Atomic-force-microscopy (AFM) observation
The maximum surface height of 500 nm was reached
(Figure 3). The longitudinal grooves were unclear and
shallow when Figures 3 to 5 were compared. The
drill-speed effect obtained at 2730 min–1, the 6-mm drill
diameter and the 40-mm/min feed rate was clearly ob-
served in both Figures 3 and 4. The maximum height of
F. KARACA, B. AKSAKAL: EFFECT OF THE TiBN COATING ON A HSS DRILL WHEN DRILLING THE MA8M Mg ALLOY
78 Materiali in tehnologije / Materials and technology 50 (2016) 1, 75–79
Figure 4: AFM topography of a specimen drilled at 2730 min–1, 6 mm
and 40 mm/min
Slika 4: AFM-topografija vzorca, vrtanega pri 2730 min–1, 6 mm in
40 mm/min
Figure 2: Chip formations for drilling conditions of: a) 1080 min–1,
4 mm dia and 40 mm/min feed rate, b) 1080 min–1, 6 mm dia and
40 mm/min feed rate, c) 1080 min–1, 6 mm dia and 80 mm/min feed
rate, d) 1080 min–1, 6 mm dia, 80 mm/min feed rate by TiBN-coated
drill bits, e) 1710 min–1, 6 mm dia, 80 mm/min feed rate by
TiBN-coated drill bits, f) 2730 min–1, 6 mm dia, 80 mm/min feed rate
by TiBN-coated drill bits
Slika 2: Nastanek ostru`kov pri pogojih vrtanja: a) 1080 min–1, 4 mm
premera in hitrostjo podajanja 40 mm/min, b) 1080 min–1, premer
6 mm in hitrost podajanja 40 mm/min, c) 1080 min–1, premer 6 mm in
hitrost podajanja 80 mm/min, d) 1080 min–1, premer 6 mm in hitrost
podajanja 80 mm/min pri svedru s TiBN prevleko, e) 1710 min–1,
premer 6 mm in hitrost podajanja 80 mm/min pri svedru s TiBn
prevleko, f) 2730 min–1, premer 6 mm in hitrost podajanja 80 mm/min
pri svedru s TiBN prevleko
Figure 5: AFM topography of a specimen drilled at 1710 min–1 and
40 mm/min with a TiBN-coated drill
Slika 5: AFM-topografija vzorca, vrtanega pri 1710 min–1, s TiBN
prevleko in 40 mm/min
Figure 3: AFM topography of a specimen drilled at 1710 min–1, 6 mm
and 40 mm/min
Slika 3: AFM-topografija vzorca, vrtanega pri 1710 min–1, 6 mm in
40 mm/min
the surface was elevated at about 1000 nm, and the
longitudinal grooves are clearly seen in Figure 4. The
TiBN-coating effect on the surface topography was
observed at about 1750 nm, being measured as the maxi-
mum height and rare longitudinal grooves of the surface
(Figure 5). The drill speed and feed rate were 1080
min–1 and 20 mm/min, respectively, and the lowest level
of the present experimental study led to the surface
topography of Figure 5. When Figures 3 to 5 are
compared it is found that the drill speed and TiBN
coating had an excessive negative effect on the surface
topography during the drilling of the MA8M Mg alloy.
On the AFM graphs, the maximum undulation was
observed for the TiBN-coated drill.
4 CONCLUSION
The experimental work and the analysis showed that
the MA8M alloy has a lower machinability capacity.
Although the drill-speed increasing was performed rather
well, the results for the roughness and chip formation
were not obtained with the AFM graphs. The drill speed
has a direct relationship with the cutting speed and an
increase in the cutting speed resulted in smoother sur-
faces compared with lower cutting speeds.13 But the in-
crement of the cutting speeds led to higher cutting tem-
peratures, and the temperature increase also decreased
both the cutting performance of a drill bit and the surface
integrity. For this reason, the drill speed should not be
performed at extremely high levels. On the other hand,
the feed rate can be increased by increasing the drill
speed if smooth surfaces are required. However, the
feed-rate increment has a positive effect on the surface
roughness, and it should be used carefully with smaller
diameters of drill bits. According to the results of the
experimental work, the TiBN coating was not appro-
priate for the MA8M drilling operation when compared
with the HSS drill bit. However, the wear resistance of
the TiBN-coated drilling tools should be investigated.
5 REFERENCES
1 H. K. Tönshoff, J. Winkler, The influence of tool coatings in ma-
chining of magnesium, Surface and Coatings Technology, 94 (1997),
610–616, doi:10.1016/S0257-8972(97)00505-7
2 H. Y. Wu, C. C. Hsu, J. B. Won, P. H. Sun, J. Y. Wang, S. Lee, C. H.
Chiu, S. Torng, Effect of heat treatment on microstructure and me-
chanical properties of the consolidated Mg alloy AZ91D machined
chips, J. of Materials Processing Technology, 209 (2009),
4194–4200, doi:10.1016/j.jmatprotec.2008.11.001
3 W. Guo, Q. Wang, B. Ye, H. Zhou, Microstructure and mechanical
properties of AZ31 magnesium alloy processed by cyclic closed-die
forging, J. of Alloys and Compounds, 558 (2013), 164–171,
doi:10.1016/j.jallcom.2013.01.035
4 M. Salahshoor, Y. B. Guo, Surface integrity of magnesium-calcium
implants processed by synergistic dry cutting-finish burnishing,
Procedia Engineering, 19 (2011), 288–293, doi:10.1016/j.proeng.
2011.11.114
5 Z. Pu, J. C. Outeiro, A. C. Batista, O. W. Dillon, D. A. Jr. Puleo, I. S.
Jawahir, Surface integrity in dry and cryogenic machining of AZ31B
Mg alloy with varying cutting edge radius tools, Procedia Engineer-
ing, 19 (2011), 282–287, doi:10.1016/j.proeng.2011.11.113
6 M. Salahshoor, Y. B. Guo, Cutting mechanics in high speed dry ma-
chining of bio medical magnesium-calcium alloy using internal state
variable plasticity model, Int. J. of Machine Tools and Manufacture,
51 (2011), 579–590, doi:10.1016/j.ijmachtools.2011.04.004
7 S. Bhowmick, A. T. Alpas, The role of diamond like carbon coated
drills on minimum quantity lubrication drilling of magnesium alloys,
Surface and Coatings Technology, 205 (2011), 5302–5311,
doi:10.1016/j.surfcoat.2011.05.037
8 Y. H. Celik, Investigating the effects of cutting parameters on the
hole quality in drilling the Ti-6Al-4V alloy, Mater. Tehnol., 48
(2014) 5, 653–659
9 A. Mavi, I. Korkut, Machinability of a Ti-6Al-4V alloy with cryo-
genically treated cemented carbide tools, Mater. Tehnol., 48 (2014)
4, 577–580
10 A. Altin, The effect of the cutting speed on the cutting forces and
surface finish when milling chromium 210 Cr12 steel hardfacings
with uncoated cutting tools, Mater. Tehnol., 48 (2014) 3, 373–378
11 H. Çaliskan, A. Erdogan, P. Panjan, M. S. Gök, A. C. Karaoglanli,
Micro-abrasion wear testing of multilayer nanocomposite TiAlSiN/
TiSiN/TiAlN hard coatings deposited on the AISI H11 steel, Mater.
Tehnol., 47 (2013) 5, 563–568
12 V. P. Astakhov, S. V. Shvets, M. O. M. Osman, Chip structure classi-
fication based on mechanics of its formation, J. of Materials Pro-
cessing Technology, 71 (1997), 247–257, doi:10.1016/S0924-0136
(97)00081-2
13 K. Feng, J. Ni, D. A. Stephenson, Continuous chip formation in drill-
ing, Int. J. of Machine Tools & Manufacture, 45 (2005), 1652–1658,
doi:10.1016/j.ijmachtools.2005.03.011
F. KARACA, B. AKSAKAL: EFFECT OF THE TiBN COATING ON A HSS DRILL WHEN DRILLING THE MA8M Mg ALLOY
Materiali in tehnologije / Materials and technology 50 (2016) 1, 75–79 79

C. OZAY et al.: APPLICATION OF THE TAGUCHI METHOD TO SELECT THE OPTIMUM CUTTING ...
81–87
APPLICATION OF THE TAGUCHI METHOD TO SELECT THE
OPTIMUM CUTTING PARAMETERS FOR TANGENTIAL
CYLINDRICAL GRINDING OF AISI D3 TOOL STEEL
UPORABA TAGUCHI METODE ZA IZBIRO OPTIMALNIH
PARAMETROV ODREZAVANJA PRI TANGENCIALNEM
CILINDRI^NEM BRU[ENJU ORODNEGA JEKLA AISI D3
Cetin Ozay1, Hasan Ballikaya2, Vedat Savas1
1Department of Mechanical Engineering, Faculty of Technology, University of Firat, 23119 Elazig, Turkey
2Ortakoy Vocational High School, University of Aksaray, Aksaray, Turkey
cozay@firat.edu.tr, hballikaya@aksaray.edu.tr, vsavas@firat.edu.tr
Prejem rokopisa – received: 2014-12-06; sprejem za objavo – accepted for publication: 2015-02-10
doi:10.17222/mit.2014.293
The purpose of this research was an analysis of the optimum cutting conditions for the lowest surface roughness in tangential
cylindrical grinding of the AISI D3 tool steel using the Taguchi method. In this experimental investigation, the surface
roughness with various wheel speeds, workpiece speeds, feed rates and depths was observed. The surface roughness was
investigated employing the Taguchi design of experiments and an analysis of variance (ANOVA). Significant machining
parameters were identified using the signal-to-noise ratio. The results of the experiments indicate that the wheel speed and feed
rate have dominating effects on the surface roughness during the cutting. The developed new grinding process can be used in the
machining industries in order to determine the optimum cutting parameters for the minimum surface roughness.
Keywords: tangential cylindrical grinding, Taguchi method, surface roughness, ANOVA
^lanek predstavlja analizo optimalnih pogojev rezanja s Taguchi metodo, za doseganje najmanj{e hrapavosti pri tangencialnem
cilindri~nem bru{enju orodnega jekla AISI D3. Pri teh eksperimentih je bila opazovana hrapavost povr{ine pri razli~nih hitrostih
kolesa, razli~nih hitrostih obdelovanca ter razli~nih hitrostih in globinah podajanja. Povr{inska hrapavost je bila preiskovana z
uporabo Taguchi-jeve postavitve preizkusov in analize variance (ANOVA). Pomembni parametri obdelave so bili identificirani z
uporabo razmerja signal – hrup. Rezultati eksperimentov ka`ejo, da imata hitrost kolesa in hitrost podajanja prevladujo~o vlogo
pri hrapavosti povr{ine in parametrih rezanja. Razvoj novega postopka bru{enja se lahko uporabi v strojni industriji za dolo~anje
optimalnih pogojev rezanja pri minimalni hrapavosti povr{ine.
Klju~ne besede: tangencialno cilindri~no bru{enje, Taguchi metoda, hrapavost povr{ine, ANOVA
1 INTRODUCTION
Recently, increased production quality, a reduction in
the process time, an improvement in dimensional accu-
racy and improved workplace-safety conditions have
been studied with respect to the development of produc-
tion methods. The Taguchi method and ANOVA analysis
used for determining the optimum values reduce the
number of experimental studies while providing more
accurate results.
1.1 Grinding
Grinding is a finishing process used to improve the
surface finish and the abrasive materials and tighten the
tolerance on flat and cylindrical surfaces by removing a
small amount of the material. Grinding is an essential
process for the final machining of the components re-
quiring smooth surfaces and precise tolerances.
Grinding methods vary depending on the cutting tool
and the shape, the location and the movement of the
workpiece. The cylindrical grinding method is one of
these methods. The cylindrical grinding method is a final
machining method, often used for processing interior and
exterior surfaces of cylindrical workpieces. This method
increases the surface quality of the workpieces and pro-
vides the required measurement and tolerance. More-
over, grinding is a method, which has a significant effect
on the corrosion rate, the fracture strength, the abrasion
and the magnetic features of a workpiece.
Large-diameter grinding wheels are used as the
cutting tools for the cylindrical grinding method. They
are composed of a grinding wheel, abrasive particles and
sealants that join them together. Some problems occur
while fixing these cutting tools to the bench and during
their operation because of the errors that occurred during
the manufacturing of these tools and because they are of
large dimensions. The wheel is not balanced since hard
particles were not homogeneously dispersed during the
manufacturing of the wheel. An unbalanced wheel in-
creases the centrifugal force while turning and cannot
reach high speeds; the vibration increases, the surface
quality deteriorates and the explosion risk of the wheel
increases since the contact of the wheel with the work-
piece is imbalanced. Such problems affect the manufac-
turing and manufacturer adversely.
Materiali in tehnologije / Materials and technology 50 (2016) 1, 81–87 81
UDK 621.923:519.233.4:620.179.11 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)81(2016)
The grinding process has an important place in
highly delicate procedures performed during the manu-
facturing of the molds used in the molding sector. Some
studies of the cylindrical grinding method have been
conducted recently.
Cylindrical grinding is used to grind the external or
internal diameter of a rigidly supported and rotating
workpiece. Although the term cylindrical grinding may
also be applied to centerless grinding, it generally refers
to a workpiece that is ground in a chuck or between the
supporting centers. Cylindrical grinders can be used to
grind all types of hard or soft workpieces to a high
degree of accuracy and very fine surface finishes1.
Over the years, researchers focused on improving the
performance of machining operations with the aim of
minimizing the costs and improving the quality of manu-
factured products2. Hassui and Diniz3 studied the effects
of grinding parameters on the surface roughness and
vibration and investigated the relation between the
vibration and the surface roughness during the grinding
of the AISI 52100 steel. Kwak4 stated that the geometric
errors occurring during the grinding process were due to
the rigidity of the grinding system and thermal effects.
He also indicated that an accurate determination of the
grinding parameters is of great importance for the reduc-
tion of such errors. During the cylindrical grinding of
Al/SiC-metal-matrix composites, Thiagarajan et al.5 stu-
died the effects of the grinding parameters on the grind-
ing force, the surface roughness and the heat that
emerges during the grinding. Fan and Miller6 developed
a force model for grinding with segmental wheels. Both
experimental and analytical results showed the average
grinding force of the wheels in comparison with the con-
ventional wheels. Larger spaces between the segments
further reduce the average force and increase the surface
roughness and peak force.
Hecker and Liang7 theoretically examined the rela-
tionship between the thickness of a chip that was unde-
formed in the cylindrical grinding process and the
arithmetic surface roughness. They tried to verify this
relation experimentally using the cylindrical grinding
method. Gavas et al.8 machined four different materials
using the helical scan-grinding (HSG) method. They
investigated the effects of the cutting parameters on the
surface roughness and roundness. They stated that the
helical scan-grinding method decreased the surface
roughness in comparison with the conventional cylindri-
cal grinding method. Go³¹bczak and Koziarski9 investi-
gated the cutting capabilities of CBN grinding wheels in
the grinding process. They studied the components of
grinding forces, the surface roughness and the stresses
on the surface layer.
A cylindrical grinder is a type of a grinding machine
used to shape the outside of an object. The cylindrical
grinder can work on a variety of shapes. However, the
object must have the central axis of the rotation. This in-
cludes but is not limited to the shapes such as a cylinder,
an ellipse, a cam or a crankshaft10. Agarwal and Ven-
kateswara Rao11 investigated the relation between the
chip thickness and the surface roughness in the grinding
of ceramic materials. Nguyen and Zhang12 investigated
the performance of a new, segmented, grinding-wheel
system using the surface integrity of ground components
as a criterion. The experimental results showed that the
segmented grinding wheel had some obvious advantages
in comparison with the standard wheel.
Rodrigo et al.13 studied the effects of tangential
cutting forces on the surface roughness in the grinding
process using cutting fluids flowing at different flow
rates. Koshy et al.14 used the centerless-grinding-process-
ing method by positioning the workpiece tangentially to
the grinding wheel. They stated that this method creates
a better surface finish than the conventional methods.
Upon the literature review, it is observed that the
studies are mostly focused on the effects of the cutting
parameters on the surface quality. It is determined that
there is a limited number of the studies conducted on the
selection of cutting tools or the location and movements
of cutting tools and workpieces with respect to each
other. A new approach to the cylindrical grinding method
is introduced in this study. Within this new method,
called the tangential cylindrical grinding, the outer sur-
face of a cylindrical workpiece is in a tangential contact
with the cutting tool and the axes of the cutting tool and
the workpiece are made to be tangent to each other
(Figure 1). This method allows us to use small grinding
wheels instead of the grinding wheels with large diame-
ters used with the conventional grinding. Moreover, the
C. OZAY et al.: APPLICATION OF THE TAGUCHI METHOD TO SELECT THE OPTIMUM CUTTING ...
82 Materiali in tehnologije / Materials and technology 50 (2016) 1, 81–87
Figure 1: Position of the contact between the grinding wheel and
workpiece
Slika 1: Polo`aj stika med brusnim kolesom in obdelovancem
negative effects that may occur during the grinding with
the wheels with large diameters are eliminated.
1.2 Taguchi method and ANOVA
The Taguchi experimental-design method minimizes
the number of experiments, enabling experimental stu-
dies to be conducted in a shorter and easier way. This
method was first introduced by Genichi Taguchi, a Japa-
nese engineer. This method reduces the number of expe-
riments that would otherwise last for a long time and
have high costs15.
Analysis of variance is the predominant statistical
method used to interpret experimental data and make the
necessary decisions on whether this method is the most
objective one. The column effect is used by Taguchi as a
simplified ANOVA to subjectively identify the columns
that have large influences on the response15. The aim of
the analysis of variance is to evaluate the significance of
the cutting parameters on the surface roughness for this
paper. It gives a clear picture of how much the cutting
parameters affect the response and the level of signifi-
cance of the factor considered. Statistically, there is a
tool called an F test allowing us to see which design
parameter has a significant effect on the quality charac-
teristic. Usually, when F > 4, it means that the change in
the design parameter has a significant effect on the
quality characteristic16.
Chang and Kuo17 machined aluminium-oxide cera-
mics with laser machining using the Taguchi experimen-
tal-design method. They investigated the effects of the
cutting parameters on the surface roughness and machin-
ing ratio in the experimental studies they conducted.
Prabhu and Vinayagam18 used SAE20W40 nanocarbon-
reinforced oil in the machining of the AISI D3 tool steel
with the grinding method. They used the Taguchi experi-
mental-design method in planning and evaluating their
experimental studies. Kwak and Kim19 investigated the
effects of machining parameters on the geometric errors
that appear on the surface, during the surface grinding,
using the Taguchi and surface-response methods. They
also developed a secondary surface-response model for a
prior detection of any geometric error. Nalbant et al.20
used the Taguchi method to find the optimum cutting
parameters for the surface roughness in turning. They
provided experimental results to illustrate the effective-
ness of this approach.
Gür21 alloyed the surface of medium-carbon steel
using the PTA method and he investigated the wear
resistance of this alloyed coating layer via the Taguchi
method. He used the Taguchi design according to the L18
orthogonal array and experimentally evaluated the
factors affecting the wear of the coating layer. The
preparation of experimental study plans and an easy
interpretation of the conducted experimental studies are
the primary advantages of the Taguchi experimental-
design method. An experimental study plan is formed
following the determination of the relevant parameters to
be used and their related levels and the selection of the
orthogonal arrays appropriate for their levels of freedom.
It is then converted into the performance characteristic
called the S/N (signal/noise) ratio for the interpretation of
such experimental studies. The most commonly used
performance characteristics are the smallest the best, the
biggest the best and the nominal the best characteristics.
The largest S/N ratio represents the value of a parameter
for the optimum level15.
2 EXPERIMENTAL WORK
2.1 Experimental set-up and measurements
A VMC-850 Johnford vertical, processing, centered
workbench was used for designing the experimental
set-up of the tangential cylindrical grinding as a new
method. The flowchart for the optimization of the cutting
parameters in the tangential cylindrical grinding process
is shown in Figure 2.
The set-up mechanism shown in Figure 3 was in-
stalled for the workpiece to rotate around its own axis at
desired rotation rates in the experimental set-up. A
Micromaster 400 inverter was used for rotational adjust-
ments. Furthermore, a grinding wheel used as the cutting
tool could be easily installed on or demounted from a
milling machine, acting as a milling tool. For the work-
piece’s rotation with no backlash, the workpiece was
supported with the center on the other side.
Rotational adjustments were calibrated with an Ex-
tech Instruments 461880 tachometer and a vibration-
C. OZAY et al.: APPLICATION OF THE TAGUCHI METHOD TO SELECT THE OPTIMUM CUTTING ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 81–87 83
Figure 2: Flowchart for optimizing the cutting parameters in tangen-
tial cylindrical grinding process
Slika 2: Potek optimizacije parametrov rezanja pri postopku tangen-
cialnega cilindri~nega bru{enja
measuring device. The parallelism adjustment of the
workpiece was checked by means of a comparator.
Ø20 × 60 mm AISI D3 tool steel was cut using a saw
to conduct tangential cylindrical grinding experiments.
The length of the workpiece was determined as per L =
2D ratio, where L is the length and D is the diameter of
the workpiece. A center hole was opened on one side in
order to enable the workpiece to be supported by the
center. The machining was conducted by making the
grinding wheel turn around its own axis, while being
positioned tangentially to the workpiece’s axis and with
the feed motion providing the depth of cut.
The AISI D3 tool steel is used for manufacturing
mold plates, powder-metallurgy tools, cold extrusion,
punch dies, ceramic forming dies and cold punches. A
45s SIOUX 2145GP grinding wheel with a grain size of
100 μm and a diameter of R = 44 mm was used to
conduct the tangential cylindrical grinding experiments.
The average surface-roughness values of the parts
machined with the tangential cylindrical grinding me-
thod were measured using Mitutoyo Surftest S-J210. The
surface-roughness measurements were taken along the
cylindrical workpiece’s axis and in the direction of the
feed rate. The measurements were taken from four
different points of the machined surfaces and the average
of the measurements was calculated.
2.2 Experimental study
L18 orthogonal arrays were used in the machining of
the AISI D3 tool steel with the tangential cylindrical
grinding method in this experimental study. The number
of experiments and usage levels of the parameters were
specified on the Taguchi toolbar of the Minitab 15 soft-
ware program. Because of conducting these experiments,
the surface-roughness values obtained from the surface
of the relevant workpiece were converted into the S/N
ratio in the Minitab 15 software program. The following
criteria were taken into consideration when determining
the levels of the parameters used in the experimental stu-
dies:
• The levels of the depth of cut are generally deter-
mined as the depth of cut left for the grinding process
and the values above this value.
• Relevant values were obtained from the tables
according to the number of rotations of the grinding
wheel, the number of rotations of the workpiece, the
material features of the materials to be machined and
the cutting tools to be used and the features of the
workbench.
• The smallest feed rate of the workbench that was
used for determining the parameters of the axial feed
speed and higher feed rates were taken into conside-
ration. Table 1 illustrates the parameters and the
related levels to be used in the experiments.
Table 1: Machining parameters
Tabela 1: Parametri obdelave
Cutting parameters Unit Symbol
Levels
Level 1 Level 2 Level 3
Depth of cut mm A 0.005 0.01 -
Wheel speed min–1 B 1000 1500 2000
Workpiece speed min–1 C 220 320 420
Table feed rate mm/min D 3.2 7.9 12.6
Considering the freedom degrees of these parameters,
it was determined that the use of the L18 orthogonal array
was appropriate. Table 2 illustrates the L18 orthogonal
array used in the experimental studies.
Table 2: L18 orthogonal array for the experiments
Tabela 2: L18 ortogonalna namestitev preizkusov
Trial
No
Levels of parameters
Depth of
cut (mm)
Wheel speed
(min–1)
Workpiece
speed (min–1)
Table feed rate
(mm/min)
1 0.005 1000 220 3.2
2 0.005 1000 320 7.9
3 0.005 1000 420 12.6
4 0.005 1500 220 3.2
5 0.005 1500 320 7.9
6 0.005 1500 420 12.6
7 0.005 2000 220 7.9
8 0.005 2000 320 12.6
9 0.005 2000 420 3.2
10 0.01 1000 220 12.6
11 0.01 1000 320 3.2
12 0.01 1000 420 7.9
13 0.01 1500 220 7.9
14 0.01 1500 320 12.6
15 0.01 1500 420 3.2
16 0.01 2000 220 12.6
17 0.01 2000 320 3.2
18 0.01 2000 420 7.9
3 RESULTS
Table 3 illustrates the results obtained from the pro-
cessing of the AISI D3 tool steel, performed with the
Taguchi test-design method, and the corresponding S/N
C. OZAY et al.: APPLICATION OF THE TAGUCHI METHOD TO SELECT THE OPTIMUM CUTTING ...
84 Materiali in tehnologije / Materials and technology 50 (2016) 1, 81–87
Figure 3: Tangential cylindrical grinding set-up mechanism
Slika 3: Postavitev mehanizma tangencialnega cilindri~nega bru{enja
ratio values. The average surface roughness and the
cutting parameters are present in Figures 4 to 6.
Examining Table 3 and Figures 4 and 6, it is ob-
served that the surface roughness increases with the
increase in the workpiece speed. During the grinding
process, the workpiece speed is selected according to the
grinding wheel and the characteristics of the workpiece
material. Besides, there must be a proper ratio between
the workpiece and the grinding-wheel speeds. In this
study, it is found that as the workpiece speed increases,
this ratio diverges from its proper value, the dynamic
hardness of the grinding wheel decreases and the surface
quality is reduced due to irregular abrasion. Examining
Table 3 and Figures 4 and 5, it is observed that the
surface roughness decreases while the grinding-wheel
speed increases. Since the amount of the chips that each
cutting wheel removes decreases with the increase in the
grinding-wheel speed, it can be asserted that the ma-
chining force per wheel decreases and the vibration is
reduced correspondingly. In consequence, it is obvious
from the results that the surface quality is good.
Table 3: Experimental results obtained from the studies of the average
surface roughness and S/N ratio
Tabela 3: Rezultati, dobljeni iz {tudija povpre~ne hrapavosti povr{ine
in razmerja S/N
Trial
No
Levels of parameters
Surface
roughness
(Ra)
(μm)
S/N
ratio
Depth
of cut
(mm)
Wheel
speed
(min–1)
Work-
piece
speed
(min–1)
Table feed
rate
(mm/min)
1 0.005 1000 220 3.2 0.37 8.635
2 0.005 1000 320 7.9 0.46 6.744
3 0.005 1000 420 12.6 0.59 4.582
4 0.005 1500 220 3.2 0.34 9.370
5 0.005 1500 320 7.9 0.42 7.535
6 0.005 1500 420 12.6 0.55 5.192
7 0.005 2000 220 7.9 0.35 9.118
8 0.005 2000 320 12.6 0.43 7.330
9 0.005 2000 420 3.2 0.36 8.873
10 0.01 1000 220 12.6 0.56 5.036
11 0.01 1000 320 3.2 0.46 6.744
12 0.01 1000 420 7.9 0.52 5.679
13 0.01 1500 220 7.9 0.41 7.744
14 0.01 1500 320 12.6 0.54 5.352
15 0.01 1500 420 3.2 0.41 7.744
16 0.01 2000 220 12.6 0.48 6.375
17 0.01 2000 320 3.2 0.39 8.178
18 0.01 2000 420 7.9 0.44 7.130
When Table 3 and Figures 4 to 6 are examined, it is
seen that the surface roughness increases with the
increase in the depth of cut and the feed rate.
The surface roughness generally increases as the feed
rate increases. It can be concluded that with the increase
in the depth of cut and feed rate, the cutting forces and,
accordingly, the vibration increase during the processing
of the AISI D3 tool steel with the tangential cylindrical
grinding method. In addition, with the increasing vibra-
tion, the surface roughness increases during the machin-
ing process. In a study conducted by Demir it is stated
C. OZAY et al.: APPLICATION OF THE TAGUCHI METHOD TO SELECT THE OPTIMUM CUTTING ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 81–87 85
Figure 5: Effect of parameters "grinding-wheel speed" and "table feed
rate" on the average surface roughness in machining of AISI D3 tool
steel
Slika 5: Vpliv parametrov "hitrost brusilnega kolesa" in "hitrost poda-
janja mize" na povpre~no hrapavost in obdelavnost orodnega jekla
AISI D3
Figure 4: Effect of processing parameters on the average surface
roughness of AISI D3 tool steel
Slika 4: Vpliv procesnih parametrov na povpre~no hrapavost povr{ine
orodnega jekla AISI D3
Figure 6: Effect of parameters "workpiece speed" and "depth of cut"
on the average surface roughness in machining of AISI D3 tool steel
Slika 6: Vpliv parametrov "hitrost obdelovanca" in "globina reza" na
povpre~no hrapavost povr{ine pri obdelavi orodnega jekla AISI D3
that the cutting section and the grinding force increase
because of the increase in the depth of cut, and thus, the
abrasion rate of the wheel and the average surface rough-
ness increase22,23.
Table 4 illustrates the S/N ratios, which correspond
with the surface-roughness rates obtained according to
the parameter levels during the processing of the AISI
D3 tool-steel material. In this table, the optimum levels
of the processing parameters are shown with (a).
Table 4: Effect of the factors at each level on the surface roughness in
the machining of AISI D3 tool steel (S/N ratio)
Tabela 4: Vpliv faktorjev na vsakem nivoju na hrapavost povr{ine pri
obdelavi orodnega jekla AISI D3 (S/N razmerje)
Cutting Parameters Symbol
Average S/N rate (dB)
Level 1 Level 2 Level 3
Depth of cut A 7.49 a 6.67 –
Wheel speed B 6.24 7.16 7.83a
Workpiece speed C 7.71a 6.98 6.53
Table feed rate D 8.26 a 7.33 5.64
When the results were examined, it was determined
that the A1, B3, C1 and D1 levels indicated the optimum
parameters in the evaluation of the effects of the cutting
parameters on the average surface roughness.
In conclusion, it is evidently seen in the experimental
studies conducted that the cutting parameters lead to
results that are similar to those for the external surface
obtained in the cylindrical grinding process with the
tangential cylindrical grinding method. The study allows
the grinding wheels with a smaller diameter to be used in
conventional cylindrical grinding instead of the grinding
wheels with a large diameter. In addition, the problems
that may occur when fixing and using the grinding
wheels with a large diameter are minimized. Moreover, it
is possible to perform the grinding process on a turning/
milling machine.
In this paper, the aim of the analysis of variance is to
evaluate the effects of the cutting parameters on the
average surface roughness. The analysis clearly shows
how much the cutting parameters affect the response and
the level of significance of the factor considered. Table 5
shows the results of ANOVA for the average surface
roughness. It can be seen from this table that the table
feed rate and the wheel speed are the most significant
cutting parameters affecting the surface roughness.
Therefore, based on the S/N and ANOVA analyses, the
optimum cutting parameters are the depth of cut at level
1, the wheel speed at level 3, the workpiece speed at
level 1 and the feed rate at level 1. However, the most
significant cutting parameter contributing the most to the
quality characteristic, i.e., the average surface roughness
is the table feed rate (56.27 %). The error rate of 4.56 %
was found with the ANOVA.
Table 5: ANOVA analysis of the average surface roughness obtained
during the processing of AISI D3 tool steel
Tabela 5: Analiza ANOVA povpre~ne hrapavosti povr{ine, dobljene
pri obdelavi orodnega jekla AISI D3
Cutting
parameters
Degree
of
freedom
Sum of
square Variance F ratio
Contri-
bution
(% )
Depth of cut
(mm) 1 3.041 3.041 30.586 7.940
Wheel speed
(min–1) 2 7.711 3.855 38.780 20.279
Workpiece
speed (min–1) 2 4.253 2.126 21.393 10.946
Table feed rate
(mm/min) 2 21.044 10.522 105.830 56.270
Error 10 0.994 0.099 – 4.562
Total 37.044 – – –
4 CONCLUSIONS
In this study, the AISI D3 tool steel was machined
with tangential cylindrical grinding introduced as a new
method. Using the variance analysis (ANOVA), the
effects of different parameter levels on the average
surface roughness were analyzed. The experimental stu-
dies were evaluated using the Taguchi experimental
design method and ANOVA; the following results were
obtained:
• It was determined that the average surface roughness
increases with the increase in the workpiece speed
during the processing of the AISI D3 tool-steel
material with the tangential grinding method. It was
observed that the best value of the average surface
roughness was obtained at the first level.
• It was observed that the average surface-roughness
rate increases with the increase in the depth of cut
and the axial-feed rate during the processing of the
AISI D3 tool-steel material. The most appropriate
levels were the first levels.
• It was observed that the average surface roughness
decreases with the increase in the grinding-wheel
speed during the processing of the AISI D3 tool-steel
material.
• The ANOVA results for all the cutting parameters
affecting the average surface roughness showed that
the configuration analysis has a certain effect.
• Using the Taguchi method and the S/N rate, para-
meter levels A1, B3, C1, D1 were used to obtain the
optimum average surface roughness.
• The tangential cylindrical grinding process as a new
method enabled us to use the grinding wheels with
smaller diameters compared to the diameter of the
grinding wheels used in conventional cylindrical
grinding and the obtained surface quality was similar
to the quality of fine grinding.
C. OZAY et al.: APPLICATION OF THE TAGUCHI METHOD TO SELECT THE OPTIMUM CUTTING ...
86 Materiali in tehnologije / Materials and technology 50 (2016) 1, 81–87
Acknowledgement
The authors would like to acknowledge the Firat
University, Turkey, for the financial support (Project No.
FUBAP-TEF.11.06).
5 REFERENCES
1 H. A. Youssef, H. El-Hoppy, Machining Technology, Machine Tools
and Operations, CRC Press, 2008
2 N. H. Rafai, M. N. Islam, An investigation into dimensional accuracy
and surface finish achievable in dry turning, Machining Science and
Technology, 13 (2009) 4, 571–589, doi:10.1080/
10910340903451456
3 A. Hassui, A. E. Diniz, Correlating surface roughness and vibration
on plunge cylindrical grinding of steel, International Journal of
Machine Tools & Manufacture, 43 (2003), 855–862, doi:10.1016/
S0890-6955(03)00049-X
4 J. S. Kwak, Application of Taguchi and response surface methodolo-
gies for geometric error in surface grinding process, International
Journal of Machine Tools and Manufacture, 45 (2005) 3, 327–334,
doi:10.1016/j.ijmachtools.2004.08.007
5 C. Thiagarajan, R. Sivaramakrishnan, S. Somasundaram, Experi-
mental evaluation of grinding forces and surface finish in cylindrical
grinding of Al/SiC metal matrix composites, Proceedings of the
Institution of Mechanical Engineers, Part B: Journal of Engineering
Manufacture, 225 (2011) 9, 1606–1614, doi:10.1177/
0954405411398761
6 X. Fan, M. H. Miller, Force analysis for grinding with segmental
wheels, Machining Science and Technology: An International
Journal, 10 (2006) 4, 435–455, doi:10.1080/10910340600996142
7 R. L. Hecker, S. Y. Liang, Predictive modelling of surface roughness
in grinding, International Journal of Machine Tools and Manufacture,
43 (2003) 8, 755–761, doi:10.1016/S0890-6955(03)00055-5
8 M. Gavas, Ý. Karacan, E. Kaya, A novel method to improve surface
quality in cylindrical grinding, Experimental Techniques, 35 (2011)
1, 26–32, doi:10.1111/j.1747-1567.2009.00575.x
9 A. Go³¹bczak, T. Koziarski, Assessment method of cutting ability of
CBN grinding wheels, International Journal of Machine Tools and
Manufacture, 4 (2005) 11, 1256–1260, doi:10.1016/j.ijmachtools.
2005.01.008
10 C. N. de Souza, R. E. Catai, P. R. de Aguiar, M. H. Salgado, E. C.
Bianchi, Analysis of diametrical wear of grinding wheel and round-
ness errors in the machining of steel, J. of the Braz. Soc. of Mech.
Sci. & Eng. by ABCM, XXVI (2004) 2, 209, doi:10.1590/S1678-
58782004000200013
11 S. Agarwal, P. Venkateswara Rao, A probabilistic approach to predict
surface roughness in ceramic grinding, International Journal of
Machine Tools and Manufacture, 45 (2005) 6, 609–616, doi:10.1016/
j.ijmachtools.2004.10.005
12 T. Nguyen, L. C. Zhang, Performance of a new segmented grinding
wheel system, International Journal of Machine Tools and Manufac-
ture, 49 (2009) 3–4, 291–296, doi:10.1016/j.ijmachtools.2008.10.015
13 D. M. Rodrigo, C. B. Eduardo, E. C. Rodrigo, R. A. Paulo, Analysis
of the different forms of application and types of cutting fluid used in
plunge cylindrical grinding using conventional and super abrasive
CBN grinding wheels, International Journal of Machine Tools and
Manufacture, 46 (2006) 2, 122–131, doi:10.1016/j.ijmachtools.2005.
05.009
14 P. Koshy, Y. Zhou, C. Guo, R. Chand, Novel kinematics for cylin-
drical grinding of brittle materials, Annals of CIRP, 54 (2005),
289–292, doi:10.1016/S0007-8506(07)60105-X
15 P. J. Ross, Taguchi Techniques for Quality Engineering, McGraw-
Hill, 1995
16 W. H. Wang, Y. S. Tarng, Design optimization of cutting parameters
for turning operations based on the Taguchi method, Journal of Ma-
terials Processing Technology, 84 (1998), 122–129, doi:10.1016/
S0924-0136(98)00079-X
17 C. W. Chang, C. P. Kuo, Evaluation of surface roughness in
laser-assisted machining of aluminum oxide ceramics with Taguchi
method, International Journal of Machine Tools and Manufacture, 47
(2007), 141–147, doi:10.1016/j.mach tools.2006.02.009
18 S. Prabhu, B. K. Vinayagam, AFM investigation in grinding process
with nano fluids using Taguchi analysis, Int. J. Adv. Manuf.
Technol., 60 (2012), 149–160, doi:10.1007/s00170-011-3599-5
19 J. S. Kwak, I. K. Kim, Parameter optimization of surface grinding
process based on Taguchi and response surface methods, Key Engi-
neering Materials, 306–308 (2006), 709–714, doi:10.4028/www.
scientific.net/KEM.306-308.709
20 M. Nalbant, H. Gokkaya, G. Sur, Application of Taguchi method in
the optimization of cutting parameters for surface roughness in turn-
ing, Materials and Design, 28 (2007) 4, 1379–1385, doi:10.1016/
j.matdes.2006.01.008
21 A. K. Gür, Investigating the wear behaviour of FeCrC/B4C powder
alloys coating produced by plasma transferred arc weld surfacing
using the Taguchi method, MP Materials Testing, 55 (2013) 6,
462–467, doi:10.3139/120.110463
22 H. Demir, A. Güllü, Investigation the effects of processing parame-
ters and wheel hardness on the surface roughness and grinding
forces, Gazi University Journal of the Faculty of Engineering and
Architecture, 3 (2008), 577–584, http://www.mmfdergi.gazi.edu.tr/
article/view/1061000403
23 C. Ozay, H. Ballikaya, V. Savas, Investigation on surface roughness
of D3 tool steel using tangential cylindrical grinding method, Proc.
of the EuroTecS 2013, European Conference of Technology and
Society, Sarajevo, Bosnia and Herzegovina, 2013, 377–384
C. OZAY et al.: APPLICATION OF THE TAGUCHI METHOD TO SELECT THE OPTIMUM CUTTING ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 81–87 87

V. D. MILA[INOVI] et al.: EFFECTS OF FRICTION-WELDING PARAMETERS ON THE MORPHOLOGICAL PROPERTIES ...
89–94
EFFECTS OF FRICTION-WELDING PARAMETERS ON THE
MORPHOLOGICAL PROPERTIES OF AN Al/Cu BIMETALLIC
JOINT
VPLIV PARAMETROV TORNEGA VARJENJA NA MORFOLO[KE
LASTNOSTI Al/Cu BIMETALNEGA SPOJA
Veljko D. Mila{inovi}1, Radovan V. Radovanovi}2, Mijat D. Mila{inovi}1,
Bojan R. Gligorijevi}3
1University of Belgrade, Faculty of Mechanical Engineering, Kraljice Marije 16, 11120 Belgrade, Serbia
2Academy of Criminalistic and Police Studies, Cara Du{ana 196, 11080 Belgrade, Serbia
3University of Belgrade, Innovation Center of Faculty of Technology and Metallurgy, Karnegijeva 4, 11120 Belgrade, Serbia
v.milasinovic@gmail.com
Prejem rokopisa – received: 2014-12-14; sprejem za objavo – accepted for publication: 2015-01-20
doi:10.17222/mit.2014.304
The objective of this research is to consider the effects of certain parameters of the friction-welding process on the morphology
of an aluminum/copper joint. The effect of the following parameters was monitored: the operating time, the operating pressure,
the forging time and the forging pressure. The speed was constant during the binding process and reached 1500 min–1. The
preparation of the welding materials was performed in accordance with the industrial production conditions. With the SEM-EDS
analysis, it was found that the morphology of the Al/Cu interface slightly changes when we change the distance from the
rotation axis, irrespective of the combination of the friction-welding parameters. Apart from this, the joined effects of the
operating pressure of 48 MPa and the forging pressure of 160 MPa caused a morphological change of the Al/Cu interface, while
the forging time at the moment of the combined pressurizing effect significantly influenced the modification of the Al/Cu
interface shape within a very narrow time interval of only a few seconds.
Keywords: friction welding, bimetallic joint, interface, aluminum, copper, SEM-EDS
Cilj te raziskave je obravnava vpliva nekaterih parametrov procesa tornega varjenja na morfologijo spoja aluminij/baker.
Pregledan je bil vpliv naslednjih parametrov: ~as delovanja, tlak pri obratovanju, ~as kovanja in tlak pri kovanju. Hitrost 1500
min–1 je bila med spajanjem konstantna. Priprava materialov za varjenje je bila izvr{ena skladno s pogoji industrijske
proizvodnje. S pomo~jo SEM-EDS analiz je bilo ugotovljeno, da se morfologija spoja Al/Cu rahlo spreminja s spreminjanjem
razdalje od rotirajo~e osi, ne glede na kombinacijo parametrov procesa tornega varjenja. Poleg tega je skupni u~inek delovnega
tlaka 48 MPa in tlaka pri kovanju 160 MPa povzro~il morfolo{ke spremembe spoja Al/Cu, medtem ko ~as kovanja, v trenutku
kombiniranega stiskanja mo~no vpliva na spremembo oblike Al/Cu spoja v zelo ozkem temperaturnem intervalu samo nekaj
sekund.
Klju~ne besede: torno varjenje, bimetalni spoj, stik, aluminij, baker, SEM-EDS
1 INTRODUCTION
In energetics, bonding elements, originally used for
bonding copper and aluminum cables, were made of
copper (Figures 1a to 1c) and a joint between aluminum
(Al) and copper (Cu) (Al/Cu joint) was most frequently
made by creating a mechanical contact between the two
metals, e.i., by crimping them (Figure 1a).1 The bonding
of Al cables onto the Cu busbars in electrical substations
was conducted using Cu lugs where the connection
between an Al cable and a Cu lug was made by crimping
(Figure 1b)2 or by tightening a screw (Figure 1c)3 of the
inserted Al cable. As a result of the high copper price,
new and more cost-effective solutions were found, such
as Al lugs with an inserted Cu ring (Figure 1d).4 How-
ever, the use of such Al/Cu joints proved to be unreliable
in the long run. The reasons for this were the following:
1) the weakening of the joint during its use at higher
temperatures (due to different linear Al and Cu expan-
sion coefficients) and 2) a large transient electrical resis-
tance at the interface, which is one of the deficiencies of
a joint based on the mechanical type of connection.
Further research, with the objective to find ways for
achieving a tighter bond between Al and Cu, led to the
discovery of Al/Cu bonding elements where the connec-
tion between Al and Cu is achieved with an inter-
diffusion of the solid-state atoms. The procedure, with
which such a connection is formed, is called friction
welding.
Materiali in tehnologije / Materials and technology 50 (2016) 1, 89–94 89
UDK 621.791/.792:669.71:669.3 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)89(2016)
Figure 1: Cu bonding elements: a) Cu butt connector, b) Cu crimping
lug, c) Cu lug with a screw, d) Al lug with the inserted Cu ring1–4
Slika 1: Cu priklju~ni elementi: a) Cu valjasti priklju~ek, b) Cu pri-
klju~ek z zanko, c) Cu priklju~ek z vijakom, d) Al priklju~ek z vlo-
`enim Cu obro~kom1–4
Friction welding represents a process of bonding
similar or dissimilar materials that occurs at the tempe-
ratures below the melting temperature of the base mate-
rial that is the easiest to melt and are high enough to
allow the optimum interdiffusion distance between the
atoms of the metals, with the aim of achieving the
required tensile strength.5 In practice, there are two
methods of friction welding, different only in the method
of the energy supply required for the process of weld-
ing/bonding metals. These are the continuous and inertia
friction welding. In terms of continuous friction welding,
the energy required for welding is supplied from a steady
power source all the way to the forging phase, while the
energy required for inertia friction welding is obtained
from accumulated flywheel energy.6–8
In addition to the parameters of the welding process
(time, pressure and speed), a significant impact on the
quality of the achieved connection between Al and Cu is
also made by the contact surfaces.9,10 The preparation of
the bonding surfaces is of special importance since each
type of surface impurities may disturb the required
quality level of the joint.11 Due to the tendencies of Al
and Cu to oxidize under ambient conditions,12–14 it is
recommended that the cleaning of the contact surfaces of
these metals should be performed immediately before
the welding procedure. In addition to the superficial
impurities, the quality of an Al/Cu joint may be affected
by the impurities present in the volumes of Al and Cu,
which, apart from affecting the conductivity of the basic
materials, might also induce a production of various
compounds at the very joint during the welding proce-
dure, subsequently increasing the local contact resistance
upon applying the Al/Cu bonding elements.
The greatest advantages of the friction-welding pro-
cedure over the other procedures are a low power con-
sumption and a short duration of the procedure.15 On the
other hand, its greatest shortcoming is the fact that the
bonding is achieved only within a narrow scope of the
welding parameters. Therefore, all of these parameters
may easily be exaggerated, which might lead to the
occurrence of intermetallic phases affecting the joint
strength and its electrical conductivity.9,16 The bonding
elements produced during the friction-welding procedure
are divided into three product families: bimetallic Al-Cu
butt connectors (Figure 2a)1, bimetallic Al-Cu cable lugs
(Figure 2b)1 and bimetallic Al-Cu bolt connectors (Fig-
ure 2c).4 These product families are different from one
another in terms of the construction intended for parti-
cular types of use. The lugs are used in substations in the
process of connecting Al cables to Cu busbars, the Al/Cu
bolt connectors are used for cable endings, and the
connectors are used for the continuation of an Al to a Cu
cable and vice versa. Each of the three types of products
was manufactured in several standard sizes depending on
the diameter of the cable and voltage.
The objective of this paper is to evaluate the effects
of the friction-welding parameters (time, pressure and
speed) on the morphological properties of an Al/Cu
joint, i.e., to define the parameters, at which a flat sur-
face of the Al/Cu joint is produced. The joint is made by
continuous friction welding, of the samples shaped as
cylindrical bars. The samples were prepared on the basis
of the standard procedure, under industrial conditions
and within a serial production.
2 EXPERIMENTAL PART
The materials whose bonding was performed by
friction welding were Cu and Al cylindrical bars. The Cu
bars, with a 99.99 % purity and dimensions of Ø22 × 60
mm, were produced by cutting pieces of the Cu cathode
and subsequently shaping them, with forging, on an
eccentric press. After the forging, Cu bars were ther-
mally treated for 30 min at 300 °C in the air atmosphere;
then they were taken out of the furnace and cooled in the
ambient air under standard conditions and at room tem-
perature. The Al bars, with a 99.5 % purity and dimen-
sions of Ø25 × 90 mm, were produced by cutting pieces
from the bars of 6 m in length. These dimensions match
the standard dimensions of the samples for bimetallic
connectors for medium voltage (1–35 kV).
Friction-welding contact surfaces, i.e., Al/Cu
cylinder bases were prepared on a lathe with the machine
treatment, with the aim of eliminating the present oxides
and other impurities that might have significantly im-
peded the quality of the bimetallic joint.5 Such a method
of preparing friction-welding surfaces was chosen since
it is most frequently applied under real industrial condi-
tions.
Friction welding was performed on a machine, pro-
duced in the Cable Factory (FKS) in Jagodina, in the
production facility producing cable accessories. This
machine has two sample carriers, facing each other and
controlling the process of welding. One carrier is
enabled to rotate around its axis, while the other has the
possibility of translation in the axial direction. As a
result, during the process of friction welding, there is a
possibility of a simple control of the rotational speed
through the first carrier, and also the control of the ope-
rating pressure through the other carrier. In this probe,
the Al rod is positioned on the rotating carrier, while the
Cu rod is on the carrier allowing axial movements. The
rods are brought into contact and adjusted across the
V. D. MILA[INOVI] et al.: EFFECTS OF FRICTION-WELDING PARAMETERS ON THE MORPHOLOGICAL PROPERTIES ...
90 Materiali in tehnologije / Materials and technology 50 (2016) 1, 89–94
Figure 2: Examples of the Al-Cu bonding elements produced with
friction welding: a) bimetal Al-Cu butt connector, b) bimetal Al-Cu
cable lug, c) bimetal Al-Cu bolt connector1
Slika 2: Vzorci spojenih Al-Cu elementov, izdelanih s postopkom
tornega varjenja: a) bimetalni Al-Cu valjasti priklju~ek, b) bimetalni
Al-Cu kabelski priklju~ek, c) bimetalni Al-Cu sorni{ki priklju~ek1
cylinder axes to ensure the maximum contact of the work
surfaces and achieve approximately the same quality of
the joint across the entire work surface.
The procedure of friction welding was performed in
two short consecutive phases. The first phase was
achieved by reaching an operating pressure of P1 =
32–48 MPa (depending on the sample) on the work
surface of the rotating Al rod by translating the Cu rod in
the axial direction. The Al rod rotated with the initial
rotation speed of 1500 min–1. After the first phase, during
which the heat was generated due to the contact-area
friction, in the second phase, an additional Cu injection
to the rotating Al rod was conducted under a pressure of
P2 = 0–160 MPa. The first phase was executed within a
period of t1 = 1.5 s, whereas the other took an interval
of t2 = 4 s. The temperature in the zone of the material
bonding was measured during the welding using a FLIR
thermal imaging camera with a shooting range within a
temperature interval from 0–350 °C. Due to the reprodu-
cibility of the tests of the microstructural and morpholo-
gical properties of the Al/Cu bimetallic joints, acquired
in the described regimes of friction welding, at least five
joints were produced with each of the four chosen regi-
mes (Figure 3a).
The diameter of the sample in Figure 3a was reduced
to Ø20 mm along the entire sample length. The subse-
quent slow cutting of the basic materials (Cu and Al)
was performed in the transverse direction using a water-
cooled saw at a distance of approximately 10 mm from
an Al/Cu bimetallic joint. The samples obtained this way
were cut (halved) along the axis of the cylinder using the
same procedure, with which the relevant surfaces for
examining the morphology of the Al/Cu joints were
obtained. For the needs of the microscopic examination,
the further preparation of the relevant surfaces also in-
cluded the hot mounting process, grinding and polishing
(Go{a Institute ltd.). One half of each sample was
mounted in bakelite with a graphite filling using the hot
mounting procedure, with the face area of the relevant
surface facing upwards. The hot mounting procedure
lasted for 20–30 min per sample and it was performed at
a temperature of 100–120 °C and under a pressure of 5–6
bar. Grinding was performed using waterproof SiC
abrasive papers P-120, P-240, P-400, P-800, P-1000,
P-1500 and P-2000 with abundant amounts of water.
Polishing was performed using Al2O3 suspensions with
the average particle diameter of 1 μm.
The microstructural and morphological properties of
the Al/Cu bimetallic joints were examined with a
scanning electron microscope (SEM). The SEM analysis
was performed at the Faculty of Mining and Geology at
the University of Belgrade using a JEOL JSM–6610LV
microscope connected to an INCA350 energy-disper-
sive-spectroscopy unit for the X-ray analysis (EDS). The
electron-acceleration voltage applied during the exami-
nation was 20 kV, while the electron source was a fila-
ment made of tungsten. For observing a possible pre-
sence of porosity, secondary electrons were used, while
the differences in the chemical content were observed
using back-scattered electrons and EDS analyzers. Prior
to the SEM-EDS analysis, all of the samples were
cleaned in ethanol and acetone using an ultrasonic bath,
with the aim of eliminating the residual impurities from
the previous phases of the sample preparation.
3 RESULTS AND DISCUSSION
In the friction-welding procedure, the bond between
Al and Cu is made due to the interdiffusion of the atoms
belonging to these metals through the Al/Cu border
surface. In this paper, the interdiffusion was detected
with the EDS analysis. It was noticed that Al and Cu do
not diffuse each other to the same extent; at the same
distance from the interface, the Cu concentration in the
Al basis was always larger than the Al concentration in
the Cu basis (Figures 4a and 4b). The most probable
reason for the larger concentration of Cu is the fact that
the activation enthalpy of the Cu bulk diffusion in Al
(Q = 136 kJ/mol)17 is smaller than the one required for
the Al diffusion in Cu (Q = 165 kJ/mol).18
3.1 Effects of the speed
With an increase in the distance from the rotation
axis, the speed also increases. The points at large dist-
ances from the rotation axis cross longer distances
because they have longer routes than the points closer to
the axis (s = r·/2·) and thus a greater wear of the con-
V. D. MILA[INOVI] et al.: EFFECTS OF FRICTION-WELDING PARAMETERS ON THE MORPHOLOGICAL PROPERTIES ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 89–94 91
Figure 4: EDS analysis along the axis of an Al/Cu sample
Slika 4: EDS-analiza vzdol` aksialne osi vzorca Al/Cu
Figure 3: Representative appearance of an Al/Cu bimetallic joint
immediately after: a) the friction-welding process and b) after the
preparations for metallographic observations
Slika 3: Zna~ilen izgled bimetalnega spoja Al/Cu, takoj po: a) postop-
ku tornega varjenja in b) po pripravi za metalografijo
tact surfaces occurs at the large distances from the rota-
tion axis. Figures 5a to 5c show the appearance of the
Al/Cu interface of sample 1.
By comparing Figure 5a, where the appearance of
the interface close to the rotation axis is shown, with
Figure 5b, where the appearance of the interface at the
mid-distance between the rotation axis and the sample
edge is shown, and Figure 5c, where a joint close to the
sample edge is shown, the trend of an increased wear of
the surface roughness at the copper part, caused by the
lathe preparation of the contact surfaces, can be clearly
seen. In Figures 5d to 5f the appearance of the interface
at the mid-distance between the axis and the sample edge
and close to the edge of sample 3, produced under a
different regime, is shown. For this sample, the trend of a
decreasing surface roughness at the interface, from the
rotation axis towards the sample edge, was noticed,
proving that such a phenomenon does not depend on the
regimes of the applied friction-welding parameters, but
instead, it is a typical occurrence in the friction-welding
process.
3.2 Effects of the operating pressure
The aim of the operating pressure is to provide fric-
tion for producing the heat necessary for the inter-
diffusion of the Al and Cu atoms as well as for realizing
the Al/Cu bimetallic connection.5 In Figures 6a and 6b,
representative microstructural presentations of the Al/Cu
bimetallic-joint morphologies are given for the middle
and peripheral parts of sample 1, produced by the effects
of the operating pressure of PR=32 MPa in the period of
tR = 1.5 s, without the application of the forging pressure
(PU = 0, tU = 0). Figures 6c and 6d show the mor-
phologies of the joint at the middle and peripheral parts
of sample 2, respectively, produced with the operating
V. D. MILA[INOVI] et al.: EFFECTS OF FRICTION-WELDING PARAMETERS ON THE MORPHOLOGICAL PROPERTIES ...
92 Materiali in tehnologije / Materials and technology 50 (2016) 1, 89–94
Figure 6: Appearance of the longitudinal cross-section of an Al/Cu
bimetallic joint produced by friction welding: a) appearance of the
joint of sample 1 close to the rotation axis, b) appearance of sample 1
very close to the sample edge, c) appearance of the sample 2 joint
close to the rotation axis, d) appearance of sample 2 very close to the
sample edge, e) appearance of sample 3 close to the rotation axis, f)
appearance of sample 3 very close to the sample edge, g) appearance
of sample 4 close to the rotation axis, h) appearance of sample 4 very
close to the sample edge
Slika 6: Izgled vzdol`nega prereza bimetalnega spoja Al/Cu, izdela-
nega s tornim varjenjem: a) izgled spoja vzorca 1, blizu rotacijske osi,
b) izgled spoja vzorca 1, blizu roba vzorca, c) izgled spoja vzorca 2,
blizu rotacijske osi, d) izgled spoja vzorca 2, blizu roba vzorca, e)
izgled vzorca 3, blizu rotacijske osi, f) izgled vzorca 3, blizu roba
vzorca, g) izgled vzorca 4, blizu rotacijske osi, h) izgled vzorca 4,
blizu roba vzorca
Figure 5: Appearance of the longitudinal cross-section of an Al/Cu
bimetallic joint produced by friction welding: a) appearance of the
joint of sample 1 close to the rotation axis, b) appearance of sample 1
at the mid-distance between the rotation axis and the sample edge, c)
appearance of the sample 1 joint very close to the sample edge, d)
appearance of sample 3 close to the rotation axis, e) appearance of
sample 3 at the mid-distance between the rotation axis and the sample
edge, f) appearance of sample 3 very close to the sample edge
Slika 5: Izgled vzdol`nega prereza bimetalnega spoja Al/Cu, izde-
lanega s tornim varjenjem: a) izgled spoja vzorca 1, blizu osi rotacije,
b) izgled vzorca 1 na sredini med rotacijsko osjo in robom vzorca, c)
izgled spoja v vzorcu 1, blizu roba vzorca, d) izgled vzorca 3, blizu
rotacijske osi, e) izgled vzorca 3 na sredini razdalje med osjo rotacije
in robom vzorca, f) izgled vzorca 3, blizu roba vzorca
pressure, increased by 50 % when compared to the
operating pressure for sample 1.
By comparing the morphologies of the joint pre-
sented in Figures 6a and 6b with the morphologies
presented in Figures 6c and 6d, it is noticeable that the
roughness of the interface, resulting from the lathe
preparation of the contact surfaces of sample 2, is
slightly lower than the roughness for sample 1. This
indicates that the increase in the operating pressure of up
to 50 % during the rotation without forging did not
significantly affect the change of the original morpho-
logy of the joint. The reason for this is the fact that the
applied operating pressure was most probably lower than
the pressure required for causing the deformation and
change of the Al/Cu interface shape in the friction-
welding process.
It should also be mentioned that the operating
pressure, increased by 50 % reached 48 MPa, while the
yield strength required for the deformation of pure Cu
within an interval of 300–400 °C was of a similar order
of magnitude (Figure 7).19 This means that the applied
operating pressure could, in theory, deform the pure
copper at the temperatures reached during the friction-
welding process. However, the purity of the Al/Cu joint
probably exceeds the mentioned strength of the pure Cu
since the material in the vicinity of the joint most
probably becomes stronger due to the dissolution in the
process of friction welding.20 The strengthening caused
by the dissolution occurs due to the interdiffusion of the
Al and Cu atoms, as proven to occur in the process (Fig-
ure 4).
It is known that the substitutional solid solutions with
a face-centered cubic lattice, such as the solid Cu solu-
tion in Al and Al in Cu, show a prominent dependency
on the strengthening by dissolution, even at increased
temperatures, since the dissolved atoms affect the ther-
mal component of the Peierls-Nabarro stress.20 Aside
from this, it should be mentioned that the effect of the
strengthening by dissolution also significantly depends
on the atomic size, the relative size of the modulus of
elasticity, as well as on the relative valence.20 In Figure
4, it can be noticed that the amount of dissolved atoms
increases towards the interface, meaning that the effect
of the strengthening by dissolution was the largest pre-
cisely on the interface of the Al/Cu joint.
To define the friction-welding parameters, due to
which the interface is deformed and a joint without the
lathe-induced roughness is obtained, in addition to the
operating pressure (PR), further examinations also
included the additional effect of the forging pressure
(PU). The additional introduction of the forging pressure
PU that is significantly larger than the pure-Cu yield
strength was necessary for causing the change in the
Al/Cu interface within a short interval, which is longer
than in the case when friction welding is performed
under the operating-pressure effects.
3.3 Effects of the forging pressure
The forging pressure during the continuous friction
welding starts to apply at the moment of the termination
of the rotation.10 Its role is to squeeze out all the
impurities from the bonding area and to create a strong
bond.
Figures 6e and 6f show the formations of the bond of
sample 3 in the center and very close to the sample edge.
The production regime of sample 3 is different from the
sample 2 regime: after the termination of the operating-
pressure effects PR, the forging-pressure effects of PU =
160 MPa start to be effective in a duration of tU = 2 s. By
comparing Figure 6e with 6c and Figure 6f with 6d, a
significant decrease in the roughness of the Al/Cu inter-
face is noticed. Even though the operating pressure did
not significantly affect the morphology of the joint, it is
evident that it provides a sufficient amount of heat by
friction, causing a noticeable change in the shape of the
Al/Cu bimetallic joint forged under the pressure of 160
MPa.
Figures 6g and 6h show the morphology of the
sample 4 joint formed in the middle and peripheral parts
relative to the rotation axis. Sample 4 was produced in a
time twice as long as the time for the forging-pressure
effects of tU = 4 s spent for obtaining sample 3. When
comparing Figures 6g and 6h to Figures 6e and 6f, an
almost complete absence of the roughness of the Al/Cu
interface is noticeable in Figures 6g and 6h. This con-
firms the importance of the forging time as in the dura-
tion of tU = 4 s a much larger deformation of the Al/Cu
interface is formed than under the forging-pressure PU
effects in the duration of tU = 2 s.
V. D. MILA[INOVI] et al.: EFFECTS OF FRICTION-WELDING PARAMETERS ON THE MORPHOLOGICAL PROPERTIES ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 89–94 93
Figure 7: Changes in the yield-strength values of copper and its alloys
depending on the temperature19
Slika 7: Sprememba vrednosti meje te~enja pri bakru in njegovih zliti-
nah, odvisno od temperature19
4 CONCLUSIONS
By increasing the distance from the rotation axis, the
wear of the contact surfaces increases as well. Such an
occurrence is noticed in various combinations of fric-
tion-welding parameters, thus proven to be a property of
the process.
The effect of the operating pressure of 48 MPa does
not significantly affect the shape of the Al/Cu interface
during the first phase of bonding, but it provides a suffi-
cient amount of heat for the interdiffusion of the atoms
of the Al and Cu metals, taking place in short-term
intervals of the friction-welding process.
The combined effect of the operating pressure of 48
MPa and the forging pressure of 160 MPa changes the
shape of the Al/Cu interface.
The forging time for the combined effect of the pres-
sures significantly affects the morphology of an Al/Cu
joint within a very narrow time interval. After 2 s of
forging, the superficial roughness caused by the lathe
preparation is still obvious, but immediately after 4 s of
forging, the Al/Cu interface becomes almost completely
flat, i.e., without the presence of the roughness caused
during the lathe preparation of the surfaces.
5 REFERENCES
1 http://www.mms-jagodina.com
2 http://www.feman.net
3 http://www.elliottelectric.com
4 http://www.marel.rs
5 N. Ratkovi}, A. Sedmak, M. Jovanovi}, V. Lazi}, R. Nikoli}, B.
Krsti}, Quality Analysis of Al-Cu Joint Achieved by Friction Weld-
ing, Technical Gazette, 16 (2009) 3, 3–7
6 W. Kinley, Inertia Welding: Simple in Principle and Application,
Welding and Metal Fabrication, (1979), 585–589
7 N. I. Fomichev, The Friction Welding of New High Speed Tool
Steels to Structural Steels, Welding Production, (1980), 35–38
8 K. G. K. Murti, S. Sundaresan, Parameter Optimisation in Friction
Welding Dissimilar Materials, Metal Construction, 15 (1983) 6,
331–335
9 M. Sahin, Friction Welding of Different Materials, UNITECH –
International Scientific Conference, Bulgaria, Gabrovo, 2010,
131–134
10 M. Sahin, C. Misirli, Mechanical and Metallurgical Properties of
Friction Welded Aluminium Joints, Aluminium Alloys - New Trends
in Fabrication and Applications, InTech, Rijeka 2012, 277–300,
doi:10.5772/51130
11 R. N. Shubhavardhan, S. Surendran, Friction Welding to Join
Dissimilar Metals, International Journal of Emerging Technology
and Advanced Engineering, 2 (2012) 7, 200–210
12 M. J. Freiría Gándara, Aluminium: The Metal of Choice, Mater.
Tehnol., 47 (2013) 3, 261–265
13 G. Papadimitropoulos, N. Vourdas, V. Em Vamvakas, D. Dava-
zoglou, Deposition and Characterization of Copper Oxide Thin
Films, Journal of Physics: Conference Series, 10 (2005), 182–185,
doi:10.1088/1742-6596/10/1/045
14 P. Keil, D. Lützenkirchen-Hecht, R. Frahm, Investigation of Room
Temperature Oxidation of Cu in Air by Yoneda-XAFS, AIP Con-
ference Proceedings, 882 (2007), 490–492, doi:10.1063/1.2644569
15 V. I. Vill, Friction Welding of Metals, AWS, New York 1962
16 M. Sahin, Joining of Aluminium and Copper Materials with Friction
Welding, International Journal of Advanced Manufacturing Tech-
nology, 49 (2010) 5–8, 527–534, doi:10.1007/s00170-009-2443-7
17 E. A. Brandes, G. B. Brook, Smithells Metals Reference Book, 7th
ed., Butterworth-Heinemann, Oxford 1992, 13.16
18 D. R. Askeland, P. P. Fulay, The Science and Engineering of Mate-
rials, 5th ed., Thomson, New York 2006, 152
19 M. Li, S. J. Zinkle, Physical and Mechanical Properties of Copper
and Copper Alloys, Comprehensive Nuclear Materials, 4 (2012),
667–690, doi:10.1016/B978-0-08-056033-5.00122-1
20 \. Drobnjak, Fizi~ka Metalurgija – Fizika ^vrsto}e i Plasti~nosti 1,
TMF, Beograd 1990, 207–215
V. D. MILA[INOVI] et al.: EFFECTS OF FRICTION-WELDING PARAMETERS ON THE MORPHOLOGICAL PROPERTIES ...
94 Materiali in tehnologije / Materials and technology 50 (2016) 1, 89–94
O. LYUBIMOVA et al.: CHARACTERISATION OF THE MECHANICAL AND CORROSIVE PROPERTIES ...
95–100
CHARACTERISATION OF THE MECHANICAL AND CORROSIVE
PROPERTIES OF NEWLY DEVELOPED GLASS-STEEL
COMPOSITES
KARAKTERIZACIJA MEHANSKIH IN KOROZIJSKIH LASTNOSTI
NOVO RAZVITIH KOMPOZITOV STEKLO-JEKLO
Olga Lyubimova1, Ekaterina Gridasova2, Alexander Gridasov2, Gerrit Frieling3,
Martin Klein3, Frank Walther3
1Department of Mechanics and Mathematical Modeling, Far Eastern Federal University (FEFU), Vladivostok, Russia
2Department of Welding Production, Far Eastern Federal University (FEFU), Vladivostok, Russia
3Department of Materials Test Engineering (WPT), TU Dortmund University, Dortmund, Germany
gerrit.frieling@tu-dortmund.de
Prejem rokopisa – received: 2014-12-17; sprejem za objavo – accepted for publication: 2015-02-17
doi:10.17222/mit.2014.305
This paper presents a preliminary research about a newly developed glass-steel composite created with diffusion welding of
glass (C49-1) and carbon steel (St3sp). The main conclusions on the process of forming a diffusion zone during the welding of
glass and steel are made. The results of quasi-static and cyclic mechanical tests and corrosion investigations are presented and
interpreted on the basis of the microstructure developed during diffusion welding and described in this article.
Keywords: hardening of glass, diffusion welding, glass-steel composite material, tensile tests, cyclic tests, fatigue, corrosion,
microstructure
^lanek predstavlja uvodne raziskave novo razvitega kompozita steklo-jeklo, izdelanega z difuzijskim varjenjem stekla (C49-1)
in ogljikovega jekla (St3sp). Postavljeni so glavni zaklju~ki o postopku nastajanja difuzijske cone med varjenjem stekla in jekla.
Na osnovi razvoja mikrostrukture pri difuzijskem varjenju so predstavljeni in razlo`eni rezultati kvazi stati~nih, cikli~nih
mehanskih in korozijskih preizkusov.
Klju~ne besede: utrjevanje stekla, difuzijsko varjenje, steklo-jeklo kompozitni material, natezni preizkusi, cikli~ni preizkusi,
utrujenost, korozija, mikrostruktura
1 INTRODUCTION
Glass shows a diverging behavior for different kinds
of deformation. On the one hand, it has a relatively high
compressive strength; however, on the other hand, it has
a relatively low tensile strength. The values for the
compressive strength of glass are in the range of
500–1250 MPa. So, when working in the compressive
mode, glass can compete with the structural-grade
carbon steel. At the same time, the tensile strength of
glass is much lower: 30–50 MPa.1,2 This is due to the fact
that the strength of glass is largely dependent on the state
of its surface. The most well-known methods for the
hardening of glass allow the creation of compressive
stresses in the surface layers of glass (glass hardening,
ion exchange, surface crystallization) and surface hard-
ening (mechanical polishing, removal of the defective
surface layer by etching the glass, application of protec-
tive coatings). It is proposed to study the methods, which
allow the hardening of glass by eliminating not only the
surface micro-defects but also the internal micro-defects,
as well as providing insulation from the effects of the
environment.3,4 As a result, the composite material would
be larger and stronger. Diffusion welding5 is used widely
for joining the elements of simple configuration. It finds
a wide application in electronics, aviation and other
industries. However, despite the extensive study of the
mechanism interaction of various materials during the
diffusion welding, a general theory about the relationship
between glass and metal has not yet been developed.
The aim of this study is to investigate the mechanical
and corrosive properties of this newly developed glass-
steel composite material. Therefore, quasi-static, cyclic
and potentiodynamic polarization tests are carried out.
Furthermore, light microscopic studies are performed to
build the basis for microstructure-oriented assessments
of the mechanical and chemical results.
2 EXPERIMENTAL METHODS
2.1 Processing of glass-metal composites
The rod was made as illustrated in Figure 1a. The
mobile cover (3) provides holes for gas outlets and the
possibility for a forward movement along the longitudi-
nal axis of the shell (2). The movement of the cover (3)
is dependent on the shrinkage of the rod (1) during the
heating process.
The technological regime (pressure during diffusion
bonding p = 0.25 MPa) is divided into six stages (Figure
Materiali in tehnologije / Materials and technology 50 (2016) 1, 95–100 95
UDK 669.018.9:620.17:620.193 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)95(2016)
1b). In the first step, which corresponds to the interval
(0, t1), the composite rod is heated from T0 = 20 °C to the
temperature of welding Tw (around 800 °C). Given the
limited deformation ability of glass to produce a lasting
connection, Tw = Tg, where Tw = (0.4 … 0.6) Tm, with Tg
as the temperature of glass softening and Tm as the tem-
perature of the melting metal from which the shell (2)
was made. In the interval (t1, t2), Tw is maintained,
providing a reliable connection between glass and steel.
It is then cooled down to the temperature of annealing
Trel. To reduce the initial stress, the annealing is carried
out in the interval (t3, t4). This is a necessity at such tem-
peratures, as the viscosity is still quite low, for example,

"= 1012 Pa s; the residual stresses are removed after
15 min and, with 
" = 1013.5 Pa s, after 4 h. In the interval
(t4, t5), the required cooling is carried out up to T1, with a
rate of 3–5 K/min. After 
’  1015 Pa s, the residual
stresses in the glass do not disappear so that at the last
stage – the sixth stage – a more rapid cooling at 15–25
K/min is possible.
2.2 Microscopy
The microstructure of the hybrid material was
investigated with light microscopy. Therefore, slices of
the specimen were cut off using a cutting wheel. Before
using a microscope, these slices were cold mounted,
ground and polished.
2.3 Tensile and fatigue testing
The quasi-static properties with respect to the com-
pression, tension and torsion of the glass-metal compo-
site, consisting of chemical-laboratory glass brand C49-1
(3C-5Na) and carbon steel St3sp, were determined using
a tensile testing system (Shimadzu, 1000 kN, UH-I)
(Figure 2). Glass C49-1 has the following chemical
composition: 67.5 % SiO2; 20.3 % B2O3; 3.5 % Al2O;
8.7 % Na2O.
Carbon steel St3sp comprises the following quantities
of the alloying elements: 0.14–0.22 C; 0.15–0.30 Si;
0.4–0.65 Mn; 0.05 S; 0.04 P; 0.3 Cr; 0.3 Ni;
0.3 Cu; 0.08 As; 0.01 N.
The geometries of the specimens and their dimen-
sions are shown in Figure 3 and Table 1, respectively.
Table 1: Geometrical characteristics of the specimens, in mm
Tabela 1: Geometrijske zna~ilnosti vzorcev, v mm
d l0 l D h r lg dg hg
10.0 100.0 105.0 16.0 10.0 3.0 105.0 7.5 30.0
O. LYUBIMOVA et al.: CHARACTERISATION OF THE MECHANICAL AND CORROSIVE PROPERTIES ...
96 Materiali in tehnologije / Materials and technology 50 (2016) 1, 95–100
Figure 2: Tensile testing system Shimadzu 1000 kN UH-I
Slika 2: Natezni preizku{evalni sistem Shimadzu 1000 kN UH-I
Figure 1: a) Schematic view of a composite rod: 1 – rod of glass
C49-1 (3C5Na), 2 – outer shell of steel St3sp, 3 – mobile cover, b)
temperature welding: T – temperature welding, lg 
 – logarithm of the
dynamic viscosity of glass
Slika 1: a) Shematski prikaz kompozitne palice: 1 – palica iz stekla
C49-1 (3C5Na), 2 – zunanji ovoj iz jekla St3sp, 3 – premi~ni pokrov,
b) temperaturno varjenje: T – temperature varjenja, lg 
 – logaritem
dinami~ne viskoznosti stekla
Figure 3: Geometries of the specimens for: a) compression and b) ten-
sile tests
Slika 3: Geometrija vzorcev za: a) tla~ni preizkus in b) natezni preiz-
kus
For the investigation of the fatigue behavior, load-
increase tests and constant-amplitude tests were
performed.6,7 The specimens were subjected to a stress
ratio of R = –1 and a frequency of f = 10 Hz using sinu-
soidal load-time functions at room temperature. For the
dynamic tests, the specimen geometry shown in Figure
3b was used. In order to measure the total strain and to
determine the plastic-strain amplitude, an extensometer
was attached to the specimen.
Figure 4 shows a specimen mounted onto the servo-
hydraulic testing system (Shimadzu, 20 kN, EHF-LV20)
with the attached extensometer. The testing system has
the maximum load capacity of 20 kN and the extenso-
meter has a range of ± 4.0 % of the total strain. The
fatigue strength of the glass-metal composite rod for the
ultimate number of cycles Nlimit = 2·106 was estimated
with a stepwise load-increase test according to the
measured material response and afterwards validated
with constant-amplitude tests.
2.4 Corrosion testing
The corrosion behavior of the St3sp steel and of the
hybrid interface was investigated with potentiodynamic
polarization measurements in a 0.1 mol L–1 NaCl so-
lution at pH7; the pH values were adjusted using a 0.1
mol L–1 KOH solution. A standard three-electrode sys-
tem, consisting of a saturated calomel electrode (SCE) as
the reference electrode and a graphite electrode as the
counter electrode, was used for the measurements8. As
the working electrodes, the St3sp steel and the cross-sec-
tion of a round specimen (Figure 3b) for fatigue testing
were used. The specimens were abraded with emery
paper of 18–5 μm. Then, a copper wire was attached to
the specimens and the whole specimens, except for their
front sides, were embedded in the epoxy resin. Before
each polarization measurement, the electrolyte was
purged for 30 min with argon and afterwards the open-
circuit potential was measured for 30 min. The measure-
ments were conducted using a potentiostat (Gamry,
PCI4300) at a scanning rate of 0.1 mV/s. The experi-
mental set-up of the potentiodynamic-polarization
measurements is shown in Figure 5.
3 RESULTS AND DISCUSSION
3.1 Microstructure
The St3sp steel was found to have a ferritic structure
with around 8 % of pearlite. After polishing and analyz-
ing the border-welding zones of the samples, there were
still no visible cracks or voids. The analysis of micro-
sections showed the presence of a full contact without
cracks or spills in the welding zone (Figure 6). The
microscopic examination of the welding zone clearly
shows a development of the interaction at the boundary
of the contact. A bright zone in the glass confirms this
development. The results of the spectral analysis indicate
a penetration of cations into the glass to a depth of 30
microns and more. This is in accordance with the results
O. LYUBIMOVA et al.: CHARACTERISATION OF THE MECHANICAL AND CORROSIVE PROPERTIES ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 95–100 97
Figure 4: Fatigue testing system: a) overview and b) detail
Slika 4: Sistem za preizku{anje utrujanja: a) pregled in b) detajl
Figure 6: a) Micrograph of the welded joint between metal and glass,
b) micrograph of the cross-section
Slika 6: a) Posnetek podro~ja zvarjenega spoja med kovino in ste-
klom, b) posnetek preseka
Figure 5: Potentiodynamic-polarization measurement set-up
Slika 5: Sistem za merjenje potenciodinami~ne polarizacije
of the corrosion tests, which also indicated the formation
of the glass-metal phase.
Figure 7 shows the grains in the transition area
between steel and glass. It is apparent that the grain size
of the ferritic structure in close vicinity of the glass is
smaller than at bigger distances. The explanation for this
behavior has not been found yet, but it will be a topic of
further investigations of this glass-steel composite
material.
3.2 Deformation behavior
By performing the compression, tensile and torsional
tests, the quasi-static properties of the selected glass, the
steel and the glass-steel composite were determined. The
stress-strain curves for compression and tension of the
composite material are depicted in Figure 8. The results
for different materials and tests are shown in Tables 2
and 3.
The maximum compressive stress of the composite
material is more than 13 times higher than the maximum
compressive stress of the C49-1 glass and it is twice as
high as for the St3sp carbon steel. The maximum tensile
stress of the composite sample is twice as much as that
of glass and half of that of carbon steel. The Young’s
modulus and shear modulus of the composite have
intermediate values of the modules of the base materials,
glass and steel.
Table 2: Mechanical characteristics of base materials and glass-steel
composite rod
Tabela 2: Mehanske zna~ilnosti osnovnih materialov in kompozitne
palice steklo-jeklo
Fmax/kN max/MPà
Glass (C49-1) compressivestrength 2.55 55.00
Glass (C49-1) tensile strength 3.40 77.00
Steel (St3sp) compressivestrength 30.05 389.20
Steel (St3sp) tensile strength 29.00 368.96
Composite
glass-steel
compressive
strength 58.50 745.00
Composite
glass-steel tensile strength 12.70 162.00
Table 3: Mechanical characteristics of base materials and glass-steel
composite rod
Tabela 3: Mehanske zna~ilnosti osnovnih materialov in kompozitne
palice steklo-jeklo
Glass (C49-1) Steel (St3sp) Compositeglass-steel
Young’s
modulus, (MPà) 0.73·10
5 2.10·105 1.40·105
Shear modulus,
(MPà) 0.30·10
5 0.80·105 0.56·105
3.3 Fatigue beahvior
For an evaluation of the cyclic-deformation behavior
and for an estimation of the fatigue strength, a stepwise
load-increase test was performed with a glass-metal
composite specimen as depicted in Figure 9. During this
test, the plastic-strain amplitude was measured as the
parameter of the material reaction, which characterizes
the proceeding fatigue damage as the basis for the esti-
mation of the fatigue strength.9 The estimation of the
fatigue strength was validated with constant-amplitude
tests, whose lifetimes (the number of cycles to failure)
are shown with the S-N curve.
Starting at a,start = 6.4 MPa, the stress amplitude was
increased by 6.4 MPa each 104 cycles until failure. The
plastic-strain amplitude a,p increases slightly within the
first nine steps, but remains almost constant within each
step. In the tenth step, there is a visible increase com-
pared to the ninth step, but the plastic-strain amplitude
still does not increase within the step. This is different
for the eleventh step at the stress amplitude of 70.4 MPa.
In the twelfth step, there is a considerable increase of the
O. LYUBIMOVA et al.: CHARACTERISATION OF THE MECHANICAL AND CORROSIVE PROPERTIES ...
98 Materiali in tehnologije / Materials and technology 50 (2016) 1, 95–100
Figure 7: Micrograph of the grains in the transition area between steel
and glass
Slika 7: Posnetek zrn v jeklu v prehodnem podro~ju v steklo
Figure 8: Stress-strain curve of the glass-metal composite for: a) com-
pression and b) tension
Slika 8: Krivulja napetost – raztezek kompozita steklo-kovina pri: a)
tla~nem in b) nateznem preizkusu
plastic-strain amplitude and, finally, there is an exponen-
tial increase in the thirteenth step.
As the fatigue strength is linked to the first increase
in the material-reaction quantity within one step,9 the
estimate from this load-increase test for the endurance
limit is 70 MPa. This estimation was validated with the
constant-amplitude tests with Nlimit = 2·106 as the
ultimate number of cycles. Ten tests with different stress
amplitudes were conducted and the results are shown as
an S-N curve in Figure 10. The highest stress amplitude
used is 127 MPa, which is close to the tensile strength of
162 MPa (Table 3) for the glass-metal composite. The
specimen with a load amplitude of 64 MPa reached the
ultimate number of cycles without fracture.
For this set of specimens, the modified Basquin
equation was found to be a = 245 MPa (Nf)–0.09.
If the estimation of the fatigue strength from the
load-increase test (70 MPa) is compared to the highest
stress amplitude without failure (64 MPa), it becomes
apparent that the estimation on the basis of the
plastic-strain amplitude is a very suitable tool for this
composite material, with only one specimen. The stress
amplitude of the highest run-out specimen is about 40 %
of the tensile strength of the composite material.
3.4 Corrosion behaviour
Figure 11 shows the Tafel plots for the steel jacket
and the hybrid interface between glass and steel. The
steel jacket shows a marginally less noble corrosion
potential Ecorr than the hybrid interface. However, the
corrosion current density icorr (i.e., the corrosion rate) of
the glass-metal joint is approximately four times higher
than that of the metal jacket. Consequently, the welding
of glass and metal influences the corrosion resistance,
suggesting that a glass-metal phase is formed through the
welding process.
4 CONCLUSIONS
The formation of a glass-metal phase was determined
with micrographs and corrosion tests. Furthermore, a
dependency of the steel grains as a function of the dist-
ance to the transition zone was found. The newly deve-
loped glass-metal composite material has quasi-static
properties, comparable to those of the St3sp steel. Using
a stepwise load-increase test and constant-amplitude
tests, the endurance limit of the composite was evaluated
as 70 MPa. This value is about 40 % of its tensile
strength. The corrosion rate of the composite material
was ascertained to be four times higher than the corro-
sion rate of the St3sp carbon steel.
This glass-metal material has potential applications,
especially for compressive loads, for example, in civil
engineering. The analysis of the glass-metal phase, the
examination of the grain-size distribution in the carbon
steel near the transition zone as well as the mechanical
investigations with a superimposed corrosive impact
during the cyclic testing, i.e., corrosion-fatigue investi-
gations, are planned as the further investigation steps.
O. LYUBIMOVA et al.: CHARACTERISATION OF THE MECHANICAL AND CORROSIVE PROPERTIES ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 95–100 99
Figure 11: Tafel plots of the metal jacket and glass-steel composite
Slika 11: Taflov diagram kovinskega ovoja in kompozita steklo-jeklo
Figure 9: Stepwise load-increase test with a glass-metal composite
Slika 9: Preizkus stopni~astega nara{~anja obremenitve kompozita
steklo-jeklo
Figure 10: S-N curve for glass-steel composite
Slika 10: S-N-krivulja za kompozit steklo-jeklo
Acknowledgements
The study was supported by the Federal Program
"Nanosystems" Activity 1.2, contract ¹ 14.575.21.0009,
unique identifier ASR RFMEFI57514X0009.
5 REFERENCES
1 V. P. Pukh, L. G. Baikova, M. F. Kireenko, L. V. Tikhonova, T. P.
Kazannikova, A. B. Sinani, Atomic structure and strength of
inorganic glasses, Physics of the Solid State, 47 (2005) 5, 876–881,
doi:10.1134/1.1924848
2 E. A. Gridasova, Increase of strength properties of glass as a result of
metallization by diffusion welding, Ph. D. Dissertation, Vladivostok,
2013, 134
3 E. A. Gridasova, O. N. Lyubimova, K. N. Pestov, G. L. Kayak,
Patent ¹2428389 RF, MPK C03C 27/02, Method of making
steklometallokompozita – ¹2009149794, Zayav. 31.12.2009, Publ.
10.09.2011, Bul. ¹ 25, 6
4 O. N. Lyubimova, E. A. Gridasova, Method of hardening during
diffusion bonding of glass to the metal, Welding and diagnostic
materials, 2010, ¹ 6, 31–45
5 V. A. Bachin, Theory, technology and diffusion welding equipment –
M: Mechanical Engineering, 1991, 352
6 F. Walther, Microstructure-oriented fatigue assessment of construc-
tion materials and joints using short-time load increase procedure,
MP Materials Testing, 56 (2014) 7–8, 519–527, doi:10.3139/120.
110592
7 F. Walther, D. Eifler, Cyclic deformation behavior of steels and
light-metal alloys, Materials Science and Engineering A, 468–470
(2007), 259–266, doi:10.1016/j.msea.2006.06.146
8 P. Wittke, M. Klein, F. Walther, Corrosion Fatigue Behaviour of
Creep-Resistant Magnesium Alloy Mg-4Al-2Ba-2Ca, Procedia
Engineering, 74 (2014), 78–83, doi:10.1016/j.proeng.2014.06.228
9 P. Starke, F. Walther, D. Eifler, "PHYBAL": a short-time procedure
for a reliable fatigue-life calculation, Advanced Engineering Mate-
rials, 12 (2010) 4, 276–282, doi:10.1002/adem.200900344
O. LYUBIMOVA et al.: CHARACTERISATION OF THE MECHANICAL AND CORROSIVE PROPERTIES ...
100 Materiali in tehnologije / Materials and technology 50 (2016) 1, 95–100
M. PRIJANOVI^ TONKOVI^, J. LAMUT: PHASE ANALYSIS OF THE SLAG AFTER SUBMERGED-ARC WELDING
101–107
PHASE ANALYSIS OF THE SLAG AFTER SUBMERGED-ARC
WELDING
ANALIZA FAZ V @LINDRI PRI OBLO^NEM VARJENJU POD
PRA[KOM
Marica Prijanovi~ Tonkovi~1, Jakob Lamut2
1High Mechanical Engineering School, [egova 112, 8000 Novo Mesto, Slovenia
2University of Ljubljana, Faculty of Natural Sciences and Engineering, A{ker~eva cesta 12, 1000 Ljubljana, Slovenia
marica.prijanovic-tonkovic@guest.arnes.si
Prejem rokopisa – received: 2015-01-16; sprejem za objavo – accepted for publication: 2015-03-04
doi:10.17222/mit.2015.014
The quality of a weld depends, to a large extent, on the filler material and type of welding. Welds and surfacing welds were
produced with submerged-arc welding. The welding current was varied. The flux with a fineness of 0.2–1.8 mm was used.
During the welding process, the welding flux melted and the liquid slag was formed. After the input of heat was stopped, the
solidification of the slag began and different mineral phases started to precipitate. We found out that besides the basic
constituents listed by the manufacturer of the welding flux, the alloying elements and deoxidizers from the flux and from the
melt are also present in the slag. Based on these results, it can be concluded that in the case of reusing the welding slag for the
production of welding flux, it is important to consider the composition of the welding slag.
Keywords: submerged-arc welding, welding flux, welding slag
Na kakovost zvara ima velik vpliv dodajni material ter postopek varjenja. Zvari in navari so bili izdelani po postopku varjenja
pod pra{kom. Tok varjenja se je spreminjal. Uporabljen je bil varilni pra{ek, zrnatosti od 0,2 do 1,8 mm. Med varjenjem se je
varilni pra{ek stalil in nastala je teko~a `lindra. Po kon~anem dovajanju toplote se je za~elo strjevanje `lindre in v `lindri so se
izlo~ale razli~ne mineralne faze. Ugotovili smo, da na mineralno sestavo `lindre vplivajo poleg osnovnih sestavin, ki jih navaja
proizvajalec varilnega pra{ka, tudi legirni elementi in dezoksidanti iz pra{ka in iz taline. Na osnovi rezultatov preiskav
sklepamo, da je pri ponovni uporabi varilne `lindre za izdelavo varilnega pra{ka pomembno, da se upo{teva tudi sestavo varilne
`lindre.
Klju~ne besede: varjenje pod pra{kom, varilni pra{ek, `lindra po varjenju
1 INTRODUCTION
The quality of welds and surfacing welds depends on
the chemical composition of the filler and base material
as well as on the method of welding, determining the
heat input which influences the development in the heat-
affected zone and in the melt during the welding process.
The slag, formed during the process of flux melting,
plays an important role.1 The slag accompanies the metal
during the wire fusion and it covers the melt and protects
it until it solidifies. The welding slag can be used again
for the flux preparation.2
The important welding parameters are the welding
current, the voltage and the speed of welding.3 If the
current is too high, degassing in the weld is weak and
cracks occur. Raising the current increases the depth of
remelting the base material.4 Surfacing with a low
welding current (450 A) is recommended in the case of
using the alloyed agglomerated fluxes.5
Slag is formed during the melting of the fluxes that
lead to the ionisation of the arc atmospere and deoxi-
dation of the melt, enable the passage of the alloying
elements into the melting bath and prevent oxidation of
carbon, manganese and silicon. Many types of welding
fluxes are in use and they differ by the main mineral
content and the presence of metal additives.6–10 Recycled
SAW slag can be used as a welding flux.8
For the investigations, the agglomerated flux labelled
as FB 12.2,11 suitable for reaching a high toughness in
multipass welds, was used. The width of the heat-
affected zone (HAZ) of steel OCR12 VM12,13 welded
with the submerged-arc welding method depends on the
welding current.14 Figure 1a shows the microstructure at
the HAZ/weld border at a current of 470 A, and Figure
1b shows the effect of a current of 610 A. The mineral
composition of the slag, formed during the welding
process, was determined.
2 EXPERIMENTS
Four different welds were produced, two welds and
two surfacing welds. Figure 2a presents a steel sample,
used for welding. Figure 2b presents a surfacing weld.
The welding was carried out on a device for submerged-
arc welding made by Iskra. The chemical compositions
of the welding pieces (steel OCR12 VM) and the filler
material are presented in Table 1.
The applied basic welding flux11 is used for automa-
tic welding and surfacing of construction steels.
Materiali in tehnologije / Materials and technology 50 (2016) 1, 101–107 101
UDK 621.791.793:66.046.58 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)101(2016)
The first joining of the welding pieces was performed
with the MAG procedure. Table 2 shows the welding
parameters. The current and the voltage were varied
during the welding process. The time of welding and the
mass of the used welding flux were measured.
Table 2: Welding parameters
Tabela 2: Parametri varjenja
Specimen
Wire
diameter
(mm)
Voltage
(V)
Current
(A)
Specimen 1 − weld 3.2 28 470
Specimen 2 − weld 3.2 27 610
Specimen 3 −
surfacing weld 3.2 29 450
Specimen 4 −
surfacing weld 3.2 28 628
The samples of the flux and slags were examinated
with metallography, scanning electron microscopy
(SEM) of type JEOL JSM 5610 and analysed with elec-
tron dispersive spectroscopy (EDS) of type JEOL JSM
5610 and X-ray diffraction (XRD) of type PANalitical
X’pert PRO.
3 WELDING-FLUX COMPOSITION
The agglomerated commercial basic welding flux of
type FB 12.2 was used for the welding. The chemical
composition of the welding flux (Table 3) and the
basicity of the flux according to the Boniszewski index
of 1.70 were taken from the catalogue.11
For the microscopic investigation of the welding flux
and welding slag, metallographic samples were prepared.
After grinding and polishing, the samples were ready for
the metallographic analysis. Figure 3 shows a SEM
image of a welding flux composed of non-metallic and
metallic materials.
M. PRIJANOVI^ TONKOVI^, J. LAMUT: PHASE ANALYSIS OF THE SLAG AFTER SUBMERGED-ARC WELDING
102 Materiali in tehnologije / Materials and technology 50 (2016) 1, 101–107
Figure 1: Microstructure of the weld on the HAZ/weld border with a
current of: a) 470 A and b) 610 A
Slika 1: Mikrostruktura zvara na meji CTV/zvar pri toku varjenja: a)
470 A in b) 610 A
Table 1: Chemical composition of base steel and filler material
Tabela 1: Kemijska sestava osnovnega jekla in dodajnega materiala
Type of material SIST EN 10027-2
Chemical composition (w/%)
C Si Mn P S Cr Mo V
Welding piece: OCR12 VM (1.2379) 1.50 0.4 0.4 0.03 0.03 11.5 0.8 0.85
Filler material 0.08 0.35 1.4 / 0.03 5.0 0.85 /
Figure 2: Macro-shot of the welding piece for: a) weld and b) sur-
facing weld
Slika 2: Makroposnetek varjenca za: a) zvar in b) navar
Table 3: Chemical composition of welding flux11
Tabela 3: Kemijska sestava varilnega pra{ka11
Welding
flux
Chemical composition (w/%)
SiO2 +
TiO2
CaO +
MgO
Al2O3 +
MnO CaF2
FB 12.2 20 30 25 20
The distribution of the elements in the welding flux
found with EDS is presented in Figure 4. The area
where, on the EDS chart, aluminium overlaps with tita-
nium aluminium oxide, which contains mass fractions of
1.5 % of titanium oxide, is presented.
The region where only calcium can be found indi-
cates the presence of calcium fluorite CaF2. Calcium
overlapping with silicon is typical for calcium silicon
(CaSi) and wollastonite (CaO·SiO2). As expected, three
elements – aluminium, silicon and potassium – overlap
with the presence of alumosilicate containing potassium. The presence of silico manganese (SiMn) in the
welding powder indicates a distribution of manganese
and silicon.
Figure 5 depicts the grains of calcium silicon (CaSi)
that contain mass fractions of 46 % of calcium and mass
fractions of 54 % of silicon and an inclusion of ferro-
silicon with minimum amounts of w(Si) = 45 %, w(Al)
= 2.7 % and about w(Fe) = 52 %. The calcium silicon
grains have a slag inclusion from the production of
calcium silicon (CaSi) (Figure 6).
Figure 7 presents an XRD diagram of the welding
flux that shows the presence of aluminium and magne-
sium oxides (periclase) and calcium fluorite.
M. PRIJANOVI^ TONKOVI^, J. LAMUT: PHASE ANALYSIS OF THE SLAG AFTER SUBMERGED-ARC WELDING
Materiali in tehnologije / Materials and technology 50 (2016) 1, 101–107 103
Figure 3: Microstructure of SAW flux
Slika 3: Mikrostruktura EPP pra{ka
Figure 6: Microstructure of calcium silicon with included slag
Slika 6: Mikrostruktura kalcij silicija z vklju~kom `lindre
Figure 4: EDS distribution of elements; SEM; EDS
Slika 4: EDS-prikaz porazdelitve elementov; SEM; EDS
Figure 5: Microstructure of calcium silicon with a ferro-silicon
inclusion
Slika 5: Mikrostruktura kalcij silicija z vklju~kom fero silicija
Figure 7: XRD of the welding flux FB 12.2
Slika 7: XRD-diagram varilnega pra{ka FB 12.2
4 RESULTS AND DISCUSSION
4.1 Influence of the welding current on the welding-
flux usage
During the process of welding, the welding flux
melts and a liquid slag is formed. The arc is hidden in
the liquid slag, which means that the welding procedure
is welder friendly. The welding current influences the
time of welding and the consumption of the welding
flux. Different currents were applied during the welding
procedures; the surfacing weld was welded at currents of
450 A and 628 A, while the weld was welded at currents
of 470 A and 610 A. Figure 8 represents the used flux in
correlation with the welding current. Increasing the
current power for the surfacing welds leads to a larger
welding-flux consumption, while the time of the welding
is decreased to a small degree.
4.2 Slag of the weld at the current of 470 A
The welding of specimen 1 was carried out at the
current of 470 A. The slag was formed on the weld toe in
the shape of a circular arch in the center, 2–3 cm wide
and 3–5 mm thick. The weld/slag border is smooth, but
some unmelted grains of the welding flux remained on
its top.
Figure 9 shows the slag microstructure on the weld
of steel OCR12 VM after the application of agglome-
rated welding flux FB 12.2. During the welding process,
a liquid slag was formed, which solidified during the
cooling. Aluminium and manganese oxides included in
the welding flux reacted and formed a solid-solution
spinel structure. The welding flux contained manganese
oxide and, after the melting, it reacted with magnesium
oxide which was also included in the flux.
During the welding, chromium from the base mate-
rial and the welding wire is oxidized. Chromium oxide
together with aluminium, magnesium and manganese
oxides forms a spinel structure with a solid-solution
composition (MgO,MnO)·(Al2O3,Cr2O3). After the weld-
ing, spinel contains mass fractions of 2.8 % MnO and
mass fractions of 3.3 % Cr2O3.
On the left side of Figure 9, there is magnesium
oxide, encircled with spinel. Solid magnesium oxide
reacts with the liquid slag and thus spinel (MgO, MnO)·
(Al2O3,Cr2O3) is formed on its surface. The resulting
spinel prohibits a dessolution of magnesium oxide in the
liquid slag.
The solidified welding slag has the following compo-
sition: magnesium oxide, spinel, a lamelar phase
(w(MgO) = 21.1 %, w(Al2O3) = 18.8 %, w(SiO2) = 39.8
% and w(K2O) = 10.2 %) and the matrix (w(CaO) = 46.2
%, w(MgO) = 11.8 %, w(SiO2) = 28.9 %, w(Al2O3) = 2.8
%, w(MnO) = 4.2 % and w(F) = 4.5 %).
Figure 10 shows an XRD diagram of the slag formed
at the welding current of 470 A. The diagram shows the
peaks for aluminium oxide, spinel and calcium fluorite.
In the sample, the undissolved flux from the surface of
the circular arch of the slag is also observed.
4.3 Slag of the weld at the current of 610 A
Figure 11 shows a SEM image of the slag formed
during the welding at the current of 610 A. During the
welding with a higher current, the unreacted leftovers of
magnesium oxide in the welding slag are surrounded
with spinel. The spinel generated during the welding
with the current of 610 A contains mass fractions of 3.9
% MnO, mass fractions of 24.3 % MgO, mass fractions
of 5.2 % Cr2O3 and mass fractions of 66.6 % Al2O3. In
the slag, tiny drops of metal (white particles) are located
at the border between the spinel and magnesium oxide
and in the lamelar phase.
M. PRIJANOVI^ TONKOVI^, J. LAMUT: PHASE ANALYSIS OF THE SLAG AFTER SUBMERGED-ARC WELDING
104 Materiali in tehnologije / Materials and technology 50 (2016) 1, 101–107
Figure 10: XRD diagram of the slag formed at the welding current of
470 A
Slika 10: Rentgenogram `lindre, nastale pri varjenju s tokom 470 A
Figure 8: Welding current and flux consumption
Slika 8: Varilni tok in poraba varilnega pra{ka
Figure 9: Microstructure of the welding slag of specimen 1
Slika 9: Mikrostruktura varilne `lindre preizku{anca 1
Figure 12 shows a SEM image of the slag, having a
metalic particle with a size of 50 μm and a composition
of mass fractions of 27.6 % Si, mass fractions of 4.8 %
Ti, mass fractions of 4.6 % Cr, mass fractions of 33.8 %
Mn (label 4), while the rest is iron. The particle was
formed during the metling of silico manganese and the
welding wire. Its composition is not similar to the com-
positions of the filler or the base material. In the slag ma-
trix, a lamellar phase and small spinel crystals containing
manganese and chromium oxides are embedded.
Welding with a higher current increases the amount
of MnO by mass fractions of 1.1 % and Cr2O3 by mass
fractions of 1.9 % in the spinel, compared to the welding
with the current of 470 A.
Figure 13 presents an XRD diagram of the slag
formed at the welding with current of 610 A. The
diagram shows that the slag is composed of aluminium
oxide, spinel, calcium fluorite and periclase (MgO).
4.4 Slag of the surfacing weld at the current of 628 A
The specimens listed under numbers 3 and 4 are the
slags of the surfacing welds. Figure 14 presents the slag
formed during the surfacing with the current of 628 A. In
the base (matrix) of the slag, there are spinel, a lamelar
phase and the leftovers of a free magnesium oxide. Com-
pared to the surfacing with 450 A, during the surfacing
with 628 A, the slag has more aluminium oxide and less
calcium oxide. The matrix of the slag (label 2) is com-
posed of mass fractions of 32.7 % CaO, 12.3 % MgO,
25.6 % SiO2, 16.5 % Al2O3, 5.7 % MnO, 2.7 % F and 3.2
% of K2O. In the slag, there are drops of metal and most
of them are located close to the lamelar phase.
4.5 Composition of the spinel and matrix of the slag
Figure 15 gives a graphical presentation of the con-
tents of the oxides in the slag base (the matrix), in which
the phases with higher melting points are precipitated.
From the diagram it is evident that specimens 1 (470 A)
and 3 (450 A) were welded with lower currents and have
the same contents of silicon oxide and different contents
of calcium oxide and alumina. Specimens 2 and 4,
welded with higer currents (610 A, 628 A), also have
different contents of calcium oxide. The content of
calcium oxide in a slag decreases with the increasing
current, while the content of aluminium oxide increases
(Figure 15). At lower welding currents, the slag includes
phases with lower melting points, while at higher
welding currents more liquid slag is formed due to the
melting of the oxides with higher melting points such as
aluminium and magnesium oxides as well as manganese
and chromium oxides.
M. PRIJANOVI^ TONKOVI^, J. LAMUT: PHASE ANALYSIS OF THE SLAG AFTER SUBMERGED-ARC WELDING
Materiali in tehnologije / Materials and technology 50 (2016) 1, 101–107 105
Figure 14: Micrograph of the phases in slag of specimen 4
Slika 14: Mikroposnetek faz v `lindri preizku{anca 4
Figure 12: SEM image of welding slag; place of analysis 2
Slika 12: SEM-posnetek varilne `lindre; mesto analize 2
Figure 13: XRD diagram of slag formed during the welding at the
current of 610 A
Slika 13: Rentgenogram `lindre, nastale pri varjenju s tokom 610 A
Figure 11: SEM image of welding slag; place of analysis 1
Slika 11: SEM-posnetek varilne `lindre; mesto analize 1
The content of the formed spinel depends on the
welding current. With an increase in the welding current
the content of manganese oxide increases from 2.8 to 3.9
% of mass fractions and chromium oxide also increases
from 3.3 to 7.3 % of mass fractions. Figure 16 presents
the change in the composition of the spinel in relation to
the welding current.
We calculated the basicity15 of the main components
in the slag with Equation (1):
B =
+ + + + + + ⋅ +
+ ⋅
CaO MgO BaO CaF Na O K O (MnO FeO)
SiO
2 2 2
2
0 5
0 5
.
. (Al O TiO ZrO2 3 2 2+ + )
(1)
The content of the base (matrix) of a slag and its
basicity also vary with respect to the welding current.
The basicity is calculated with Equation (1) and it
changes with the welding current, as presented in Table
4.
Lower welding currents also diminish the loss of the
alloying elements. The calculated values of the basicity
show that the basicity of the matrix of a slag is higher at
lower welding currents (Figure 17) because of the
content of Al2O3 in the slag melt.
5 CONCLUSIONS
During the submerged-arc welding of steel OCR 12
VM, slag solidifies above the weld in the form of a cir-
cular arch. The contact area with the slag has a smooth
glassy shine. On the contrary, at the contact with the slag
surface, the leftovers of undissolved welding flux are
observed.
The agglomerated welding flux is in the form of
pellets with a size of 0.2–1.8 mm. During the welding,
the slag is liquid. The solidified weld slag contains alu-
minum and magnesium oxides, spinel, a lamellar phase
and the matrix. The composition of the matrix is similar
to that of mineral cuspidin16 (3CaO·2SiO2·CaF2) but it
also contains aluminium, potassium and magnesium
oxides. The magnesium oxide in the welding slag is
surrounded by spinel.
Spinel16 with a complex composition (MgO,MnO)·
(Al2O3,Cr2O3) is formed from magnesium and aluminium
oxides, with smaller amounts of manganese and chro-
mium oxides.
The composition of the welding-slag matrix depends
on the welding current. At higher welding currents more
aluminium oxide is formed.
The welding slag generated during the submerged
welding can be a valuable raw material for the produc-
tion of a new welding flux.
6 REFERENCES
1 H. Grajon, Bases métallurgiques du soudage, Publications de la
Soudure Autogène, 41, Institute de Soudure, Paris, 1989
2 K. Singh, V. Sahni, S. Pandey, Slag Recycling in Submerged Arc
Welding and its Influence on Chemistry of Weld Metal, Asian
Journal of Chemistry, 21 (2009) 10, 047–051
M. PRIJANOVI^ TONKOVI^, J. LAMUT: PHASE ANALYSIS OF THE SLAG AFTER SUBMERGED-ARC WELDING
106 Materiali in tehnologije / Materials and technology 50 (2016) 1, 101–107
Figure 17: Basicity of the base (matrix) of the welding slag
Slika 17: Bazi~nost osnove (matice) varilne `lindre
Figure 15: Relative contents of oxides and fluorine in the bases
(matrices) of welding slags
Slika 15: Relativna vsebnost oksidov in fluora v osnovi (matici)
varilnih `linder
Table 4: Basicity of the base (matrix) of a slag
Tabela 4: Bazi~nost osnove (matice) `lindre
Welding flux Current (A) Matrixbasicity
Specimen 1 − weld 470 2.14
Specimen 2 − weld 610 1.68
Specimen 3 − surfacing weld 450 1.94
Specimen 4 − surfacing weld 628 1.50
Figure 16: Composition of the spinel in relation to the welding
current
Slika 16: Sestava spinela v odvisnosti od toka pri varjenju
3 Submarged arc welding, Copyright 1982, Muller Electric Mfg. Co.,
(Rev. 11/85), (http://www.millerwelds.com/pdf/spec_sheets/Sub
merged.pdf)
4 S. Duri}, B. Sabo, M. Perovi}, P. Da{i}, Matemati~ni model odvis-
nosti oblike in dimenzij zvara od parametrov navarajanja pri
postopku EPP – II. del, Varilna tehnika, 59 (2010) 3, 20–40
5 R. Kej`ar, Platiranje konstrukcijskih jekel z navarjanjem, Kovine
zlitine tehnologije, 28 (1994) 2, 95–100
6 R. Kej`ar, B. Kej`ar, Dodajni materiali na osnovi izbranih sinteti~nih
repromaterialov z dodatkom alkalijskih oksidov, Kovine zlitine
tehnologije, 28 (1994) 3
7 A. M. Paniagua-Mecado, V. M. Lopez-Hirata, Chemical and physical
propertis of flux for SAW low-carbon steels, Instituto Politechnico
National Mexico, 2011 (www.intechnopen.com)
8 J. Singh, K. Singh, J. Garg, Reuse of Slag as Flux in Submerged Arc
Welding & its Effect on Chemical Composition, Bead Geometry &
Microstructure of the Weld Metal, International Journal of Surface
Engineering & Materials Technology, 1 (2011) 1, 24–27
9 T. Lau, G. S. Weatherly, A. McLean, Gas/Metal/Slag Reactions in
Submerged Arc Welding Using CaO-Al2O3 Based Fluxes, Welding
Journal, 65 (1986) 2, 343–347
10 D. Mahto, A. Kimar, Novel Methos of Productivity Improvement and
Waste Reduction Through Recycling of Submerged Arc Welding
Slag, Jordan Journal of Mechanical and Industrial Engineering, 4
(2010) 4, 451–466
11 http://www.elektrode.si/html/slo/katalog/index_katalog.html
12 B. Kosec, G. Kosec, M. Sokovi}, Case of temperature field and
failure analysis of die-casting die, Journal of Achievements in Mate-
rials and Manufacturing Engineering, 20 (2007) 1/2, 471–474
13 F. Legat, Orodna jekla v praksi, samozal. F. Legat, Medium, @irov-
nica 2013
14 M. P. Tonkovi~, A. Nagode, L. Kosec, Mehanizem nastanka sekun-
darnega ledeburita med varjenjem orodnega jekla, IRT 3000, 5
(2012), 17–21
15 I. Polajnar, Varjenje pod pra{kom I. del: Varilni procesi in oprema,
In{titut za varilstvo, Specializacija IWE/IWT, Ljubljana 2013/2014
16 F. Trojer, Die oxydische Kristallphasen der anorganischen Indu-
strieprodukte, E. Schweizerbartsche Verlagsbuchhandlung, Stuttgart
1963
M. PRIJANOVI^ TONKOVI^, J. LAMUT: PHASE ANALYSIS OF THE SLAG AFTER SUBMERGED-ARC WELDING
Materiali in tehnologije / Materials and technology 50 (2016) 1, 101–107 107

M. SAHIN: OPTIMIZING THE PARAMETERS FOR FRICTION WELDING STAINLESS STEEL TO COPPER PARTS
109–115
OPTIMIZING THE PARAMETERS FOR FRICTION WELDING
STAINLESS STEEL TO COPPER PARTS
OPTIMIRANJE PARAMETROV PRI TORNEM VARJENJU
NERJAVNEGA JEKLA NA BAKRENE DELE
Mumin Sahin
Department of Mechanical Engineering, Trakya University, 22180 Edirne, Turkey
mumins@trakya.edu.tr
Prejem rokopisa – received: 2015-01-25; sprejem za objavo – accepted for publication: 2015-02-13
doi:10.17222/mit.2015.023
St-Cu (stainless steel and copper) parts were friction welded with the aim to optimize the process parameters in the present
study. The joints obtained with various process-parameter combinations were subjected to a tensile test. Empirical relationships
were developed to predict the strength of the joints using RSM (the response-surface methodology) and the coherency of the
model was tested. The tensile properties, microhardness variations, SEM, the EDS analysis and X-ray diffraction (XRD)
analysis of the welded specimens were evaluated. It was found, with an ANOVA analysis, that the friction pressure/friction time
relation has the largest influence on the tensile strength of the joints followed by the rotational speed. However, it was also
found that the formation of intermetallics at the interface is responsible for a higher hardness and lower tensile strength of the
friction-welded stainless steel-copper joints.
Keywords: friction welding, metallurgy, response-surface methodology, tensile strength
V predstavljenem delu so bili deli St-Cu (nerjavno jeklo in baker) torno varjeni z namenom optimizacije procesnih parametrov.
Spoji, dobljeni z razli~nimi procesnimi parametri, so bili preizku{eni z nateznim preizkusom. Razvite so bile empiri~ne
odvisnosti za napovedovanje trdnosti spojev s pomo~jo RSM (Metodologija odgovora povr{ine) in izvr{ena je bila koherenca
modela. Ocenjene so bile natezne lastnosti, spreminjanje mikrotrdote, SEM, EDS analiza in rentgenska difrakcija (XRD)
zvarjenih vzorcev. Iz ANOVA analize je bilo ugotovljeno, da ima torni tlak/~as trenja najve~ji vpliv na natezno trdnost spojev,
sledi pa mu hitrost vrtenja. Ugotovljeno je bilo, da je ve~ja trdota in manj{a natezna trdnost torno varjenih spojev posledica
nastanka intermetalne zlitine na stiku nerjavno jeklo-baker.
Klju~ne besede: torno varjenje, metalurgija, metodologija odgovora povr{ine, natezna trdnost
1 INTRODUCTION
Parts made of different materials are known to be
cost-effective. The life cycle of the materials, especially
in corrosive media, is prolonged. Many ferrous and non-
ferrous alloys can be friction welded. Friction welding
can be used to join metals of widely different thermal
and mechanical properties. The combinations that can be
friction welded cannot be joined with other welding
techniques because of the formation of brittle phases that
make the joint poor with respect to mechanical pro-
perties. Friction welding prevents distortion of the mate-
rials, as heat is not applied.
The welding technology is widely used in manufac-
turing. The development of new welding methods gained
importance along with the developing technology.1–4
Welding of different metals and their alloys is a common
application in engineering solutions. Fusion welding is
almost impossible in such cases due to incompatible
physical characteristics and chemical compositions of
different metals and alloys. As a result, friction welding
was developed. Several researches worked on the heat in
friction welding.5–7 In friction welding, heat is generated
at the interface of the workpieces since mechanical ener-
gy is dissipated as heat during the rotation under
pressure. Friction welding is a solid-state welding pro-
cess, using the heat generated through the mechanical
friction with a moving workpiece, with an addition of an
upsetting pressure to plastically displace the material.
Friction welding is generally used to join the parts that
are axially symmetrical and have circular cross-sections.
However, it can be easily used to join parts without cir-
cular cross-sections, with the aid of automation devices
and computerized control facilities.8 It is an energy saver
since heat is not applied.
The friction time and pressure, the upset time and
pressure and the speed of rotation are the principal
variables in friction welding.9–12 There are two types of
friction-welding techniques: continuous-drive friction
welding and inertia friction welding. Different metals
have different hardness values and different melting
points. Interface activity during friction welding forms
brittle intermetallic phases or eutectics with low melting
points. Clean welding surfaces are also of prime import-
ance.13–15 In welded St-Cu joints, the joint strength in-
creases with the increasing upset pressure up to the
critical value. An increase in the friction time causes a
lower strength of a St-Cu joint compared to the Cu base
metal.16 A deformation of the material during friction
welding is generally due to the diffusion involving a
Materiali in tehnologije / Materials and technology 50 (2016) 1, 109–115 109
UDK 621.791.1:669.14.018.8:669.3:519.61/.64 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)109(2016)
migration of lattice defects, which can be influenced by
an external electric field.17 Sintered powder metallurgical
preforms have a low mass, high stiffness and, therefore,
their natural frequency is high. Having inherent porosity,
they can also be good dampeners besides possessing the
latent lubricant.18 Maalekian19 found that the formation
of hard interlayers, such as intermetallic phases, when
joining dissimilar materials may cause a joint to become
brittle. Further, Sahin et al.20,21 showed that the interme-
tallic phases formed in the interface cause a decrease in
the strength of the joints.
However, based on the literature review, Murti and
Sundaresan22 carried out a study about a parameter opti-
mization using a statistical approach based on factorial-
experiment-design friction welding of dissimilar
materials. The response-surface methodology (RSM) is a
collection of mathematical and statistical techniques that
are useful for designing a set of experiments, developing
a mathematical model, analyzing the optimum combina-
tion of the input parameters and graphically expressing
the values.23 To obtain the maximum strength, it is essen-
tial to have complete control over the relevant process
parameters as demonstrated.24
Therefore, in this work, an attempt was made to opti-
mize the process parameters of continuous-drive friction
welding to achieve the maximum tensile strength of
stainless steel-copper parts using the response-surface
methodology. Tensile tests were performed on the
welded test parts. A microstructure analysis, EDS anal-
ysis, XRD analysis and microhardness variations were
also carried out on the test parts.
2 EXPERIMENTAL WORK
In the experiments, AISI 304 austenitic stainless-steel
and copper parts having a diameter of 10 mm were made
using the continuous-drive friction-welding process
parameters. The chemical composition and mechanical
properties of the stainless-steel and copper parts are
presented in Tables 1 and 2, respectively, as given in25.
Different combinations of the process parameters
were used to carry out the trial runs. Process parameters
were tested by varying one of the factors while keeping
the rest of them at constant values. The working range of
each process parameter was determined for a smooth
appearance without any observable defects. The selected
levels of the process parameters and design matrix with
their units and notations are presented in Tables 3 and 4.
However, in order to examine the intermetallic phases
formed at the interface of the joints, SEM (scanning
electron microscopy) and EDS (energy-dispersive X-ray
spectroscopy) were applied to the joints. Examinations
were carried out with an SEM-JEOL JSM 5410 LV
microscope and in the field of 200 kV. In addition, the
weld zones of the joints were analyzed in this work since
an XRD analysis of the phase constituents in the weld
zone is of a great importance.
Then, the strength of the joints was related to the
hardness variation within the HAZ. The hardness vari-
ations across the welding regions of the joints were
measured using a 0.3 kg load Vickers microhardness test.
3 RESULTS AND DISCUSSION
3.1 Empirical relationships and the optimization
The responses, the tensile-strength (TS) values of
friction-welded joints, are the functions of the friction-
welding parameters such as the friction pressure per
second (F), the forging pressure per second (D) and the
rotational speed per second (N) and they can be ex-
pressed as:
M. SAHIN: OPTIMIZING THE PARAMETERS FOR FRICTION WELDING STAINLESS STEEL TO COPPER PARTS
110 Materiali in tehnologije / Materials and technology 50 (2016) 1, 109–115
Table 1: Chemical composition of austenitic stainless steel used in the experiment25
Tabela 1: Kemijska sestava avstenitnega nerjavnega jekla, uporabljenega pri preizkusu25
Material % C % P % S % Mn % Si % Cr % Ni Tensile strength(MPa)
AISI 304
(X5CrNi1810) < 0.07 < 0.045 < 0.030 < 2.0 < 1.0 17–19 8.5–10.5 825
Table 2: Chemical composition of copper used in the experiment
Tabela 2: Kemijska sestava bakra, uporabljenega pri preizkusu
Copper % Sn % Pb % Zn % P % Mn % Fe % Ni % Si % Mg % Al % Bi % S % Sb % Cu Tensile strength(MPa)
0.00222 <0.0020 <0.0010 0.00137 <0.0005 0.0381 <0.0010 0.00745 0.00376 0.00500 <0.0005 0.00251 <0.0020 99.93 300
Table 3: Feasible working limits of friction-welding parameters
Tabela 3: Obmo~je delovnih parametrov pri tornem varjenju
Parameter Notation Unit
Level
–1.68 (–1) (0) (+1) +1.68
Friction pressure/Friction time F MPa/s 3.78 5.82 8.82 11.82 13.86
Upset pressure/Upset time D MPa/s 2.96 5 8 11 13.04
Rotational speed/sec N s–1 18.46 20.5 23.5 26.5 28.54
TS = f {F, D, N} (1)
The second-order polynomial (regression) equation
used to represent the response surface Y (TS) is as
follows:
Y = b0 + bixi + bii xi
2+ bij xi xj 2)
and for three factors, the selected polynomial could be
expressed as:
TS = b0 + b1(F) + b2(D) + b3(N) + b12(FD) + b13(FN) +
+ b23(DN) + b11(F
2) + b22(D
2) + b33(N
2) (3)
Regression coefficients are b1, b2, b3, … b44 where b0
is the average of the responses and they depend on the
respective linear, interaction and squared terms of the
factors as shown in26,27. The significance of each
coefficient was determined with a t-test and p-values,
listed in Table 5.
The value of a coefficient was calculated using the
Design-Expert software. The values of the probability>F
of less than 0.05 indicate that the model terms are
significant. In this case, F, D, N, FD, FN, DN, F2, D2 and
N2 are significant model terms. The values greater than
0.1 point out that the model terms are not significant.
The results of multiple linear regression coefficients for
the second-order response surface model are given in
Table 6.
The final empirical relationship was obtained using
only these coefficients, and the developing final empiri-
cal relationship for the tensile strength is given below:
TS = [221.36 + 18.16F + 10.90D + 13.05N – 6.62FD –
9.12FN + 2.88DN – 14.74F2 – 12.80D2 – 6.08N2] (4)
The ANOVA (analysis of variance) technique was
used to check the adequacy of the developing empirical
relationship. In this investigation, the desired level of
confidence was taken to be 95 %. The relationship is
considered adequate if the calculated F-value of the
model developed does not go over the standard tabulated
F-value and the calculated R-value of the developed
relationship exceeds the standard tabulated R-value for a
desired level of confidence. It was found that the above
model is adequate. In the same way, interactions FD, FN,
DN had significant effects. A lack of fit was not signi-
ficant though it was desired. The normal probability plot
of the residuals for the tensile strength is shown in
Figure 1. It reveals that the residuals are on a straight
line, which means that the errors are distributed nor-
mally. Each predicted value matches well its experimen-
tal value, as shown in Figure 2. The response-surface
methodology (RSM) was used to optimize the friction-
welding parameters in this study. The response contours
can assist in the prediction of the response for any zone
in the experimental field as observed in24,25.
M. SAHIN: OPTIMIZING THE PARAMETERS FOR FRICTION WELDING STAINLESS STEEL TO COPPER PARTS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 109–115 111
Table 4: Design matrix and the corresponding output response
Tabela 4: Postavljena matrika in ustrezni dobljeni odgovori
Standard
order
Run
order
Original value Tensile
strength
(MPa)F D N
16 1 8.82 8 23.5 223
18 2 8.82 8 23.5 222
4 3 11.82 11 20.5 190
9 4 3.78 8 23.5 150
20 5 8.82 8 23.5 218
6 6 11.82 5 26.5 210
15 7 8.82 8 23.5 223
1 8 5.82 5 20.5 130
7 9 5.82 11 26.5 213
5 10 5.82 5 26.5 180
8 11 11.82 11 26.5 215
12 12 8.82 13.04 23.5 217
11 13 8.82 2.96 23.5 160
10 14 13.86 8 23.5 216
3 15 5.82 11 20.5 150
19 16 8.82 8 23.5 222
17 17 8.82 8 23.5 219
13 18 8.82 8 18.46 200
14 19 8.82 8 28.54 215
2 20 11.82 5 20.5 195
Table 5: ANOVA test results for the response of the tensile strength
Tabela 5: Rezultati ANOVA preizkusa za odgovore natezne trdnosti
Sum of
squares
Mean
square F-value
P-value
si
gn
if
ic
an
t
Source df prob > F
Model 14726.69 9 1636.30 13.39 0.0002
A-F 4503.47 1 4503.47 36.85 0.0001
B-D 1622.62 1 1622.62 13.28 0.0045
C-N 2325.93 1 2325.93 19.03 0.0014
AB 351.12 1 351.12 2.87 0.1209
AC 666.12 1 666.12 5.45 0.0417
BC 66.13 1 66.13 0.54 0.4789
A^2 3132.15 1 3132.15 25.63 0.0005
B^2 2360.38 1 2360.38 19.31 0.0013
C^2 532.80 1 532.80 4.36 0.0633
Residual 1222.11 10 122.21
Lack of fit 1199.28 5 239.86 52.52 0.0003
Pure error 22.83 5 4.57
Cor. total 15948.80 19
Std. dev. 11.05 R-squared 0.9234
Mean 198.40 Adj. R-squared 0.8544
Table 6: Estimated regression coefficients
Tabela 6: Ocenjeni regresijski koeficienti
Factor
Estimated regression
coefficients
Tensile strength (MPa)
Intercept 221.36
F–friction force/friction time 18.16
D–upset force/upset time 10.90
N–rotational speed 13.05
FD –6.62
FN –9.12
DN 2.88
F2 –14.74
D2 –12.80
N2 –6.08
The end of the response plot shows the maximum
achievable tensile strength. Figures 3 and 4 show that
the tensile strength increases with the increasing friction
pressure/time relation and rotational speed and then it
decreases.
The maximum tensile strength of the friction-welded
joints was attained under the following welding con-
ditions: a friction pressure/time relation of 8.82 MPa/s (a
friction pressure of 75 MPa and a friction time of 8.5 s),
an upset pressure/time relation of 8 MPa/s (an upset
pressure of 160 MPa and an upset time of 20 s) and a
rotational speed of and 23.5 s–1, showing the accuracy of
the model.
During the welding processes, the strength of the
welds obtained with dissimilar materials strongly de-
pends on the temperature attained by each substrate.
Differences in the mechanical and thermophysical
properties and behaviour of the substrates at the interface
influence the quality of the joints during the welding as
reported in20,21.
3.2 Metallurgical analysis
The macrophotography of the joints is given in Fig-
ure 5. There is no evidence of cracking or other defects
in the joints. Due to the variations in the strength of the
materials, an appreciable variation in the width of the
HAZ (heat-affected zone) region is evident from the
joints. However, the microstructure of stainless steel is
characterized by equiaxed grains, in the austenitic-grain
structure being the natural structure of this type of steel
at room temperature (Figure 6).
M. SAHIN: OPTIMIZING THE PARAMETERS FOR FRICTION WELDING STAINLESS STEEL TO COPPER PARTS
112 Materiali in tehnologije / Materials and technology 50 (2016) 1, 109–115
Figure 4: Contour plots of the process parameters for the tensile
strength
Slika 4: Prikaz obrisov procesnih parametrov na natezno trdnost
Figure 2: Correlation graph of the response
Slika 2: Prikaz korelacije odziva
Figure 3: Response plots of the process parameters for the tensile
strength
Slika 3: Prikaz odziva procesnih parametrov na natezno trdnost
Figure 1: Normal probability plot of residuals
Slika 1: Normalna verjetnost izrisa ostankov
However, copper is formed of eutectic particles,
having dark points indicating that it is a mixture of pure
copper and cuprous oxide, dispersed in the ground
copper (Figure 7). The effect of melting was minimal at
the interface because the heat-affected zone (HAZ) was
small (Figure 8).
It is also observed that the joints have larger deforma-
tions on the Cu side compared to the steel side (Figure
5). Welding flashes occur on the copper side of the inter-
face because the melting temperature of copper is lower
than the melting temperature of steel. However, stainless
steel does not undergo an extensive deformation while
copper undergoes an extensive melting because of the
high generated and concentrated frictional heat.
Since copper has a higher thermal conductivity than
steel, the heat-affected zone on the copper side is wider
M. SAHIN: OPTIMIZING THE PARAMETERS FOR FRICTION WELDING STAINLESS STEEL TO COPPER PARTS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 109–115 113
Figure 9: EDS analysis of the intermetallic-phase zone of a joint: a)
SEM, b) EDS spectrum
Slika 9: EDS-analiza intermetalne faze na stiku: a) SEM, b) EDS
spekter
Figure 7: Microstructure of copper
Slika 7: Mikrostruktura bakra
Figure 8: Image of the interface of a joint
Slika 8: Posnetek stika v spoju
Figure 5: Macrophotography of a joint
Slika 5: Makroposnetek spoja
Figure 6: Microstructure of stainless steel
Slika 6: Mikrostruktura nerjavnega jekla
than that of the steel side. There is no change in the grain
size on the steel side. The presence of small particles on
the copper side reveals hardening on this side. There are
equiaxed  grains and Cu2O particles on the copper side.
The interface elements of both materials diffused along
the interface and some intermetallic phases were formed
at the interface as reported in20,21.
The EDS analysis performed at a defined zone of the
interface showed that the interface was formed of 2.70 %
C, 0.98 % Cr, 2.96 % Fe and 93.36 % Cu (Table 7).
Thus, the presence of intermetallic phases at the inter-
face is obvious. Copper-oxide films were broken into
pieces due to an excessive deformation at the interface
caused by the rotation (Figure 9). According to Figure
10, the X-ray diffraction results for friction-welded
stainless steel-copper joints indicated that FeCu4 and
Cu2NiZn intermetallics were formed in the welding
zone. The thickness of the layer containing the inter-
metallic phases varied between 8.72 μm and 17.53 μm
(Figure 11).
3.3 Microhardness measurement
The microhardness of a joint was measured across
the weld region and the values were plotted as shown in
Figure 12. The microhardness is maximum at the inter-
face; this may be due to the formation of brittle inter-
metallics, and it is one of the reasons for a lower tensile
strength of dissimilar joints.
4 CONCLUSIONS
Stainless-steel and copper parts were successfully
friction joined in this work. The following important
conclusions were obtained from this investigation:
• Empirical relationships were developed to predict the
tensile strength of the friction-welded stainless-steel
and copper parts incorporating process parameters at
a 95 % confidence level. The friction-welding para-
meters were optimized with the response-surface
methodology to attain the maximum tensile strength.
• The maximum tensile strength of 223 MPa was
attained in the friction-welded joints under the
following welding conditions: a friction pressure/
time relation of 8.82 MPa/s, an upset pressure/time
relation of 8 MPa/s and a rotational speed of 23.5 s–1.
• The friction pressure/friction time relation was found
to have the greatest influence on the tensile strength
of the joints, followed by the rotational speed.
• Various intermetallic phases such as FeCu4 and
Cu2NiZn occurred at the interface. The formation of
intermetallics at the interface is responsible for the
M. SAHIN: OPTIMIZING THE PARAMETERS FOR FRICTION WELDING STAINLESS STEEL TO COPPER PARTS
114 Materiali in tehnologije / Materials and technology 50 (2016) 1, 109–115
Figure 12: Microhardness variation across the joint
Slika 12: Spreminjanje mikrotrdote preko spoja
Figure 10: XRD results for the welding zone of a joint
Slika 10: Rezultati rentgenske analize (XRD) v zvaru stika
Table 7: EDS analysis of the defined zone at the interface
Tabela 7: EDS-analiza ozna~enega podro~ja na stiku
Element (w/%) (x/%)
C 2.70 12.72
Cr 0.98 1.07
Fe 2.96 3.00
Cu 93.36 83.21
Total 100.00
Figure 11: Thicknesses of the intermetallic phases at the interface
Slika 11: Debelina intermetalnih faz na stiku
higher hardness and lower tensile strength of the
friction-welded stainless steel-copper joints.
• The intermetallic phases at the interface are also
expected to play a role in the hardness variations.
Acknowledgement
The author would like to thank Trakya Univer-
sity/Edirne–Turkey, Hema Industry/Çerkezköy–Turkey
and TUBITAK MRC/Gebze–Turkey for the help in the
experimental part of the study.
5 REFERENCES
1 V. I. Vill, Friction Welding of Metals, AWS, New York 1962
2 W. Kinley, Inertia welding: simple in principle and application,
Welding and Metal Fabrication, (1979), 585–589
3 N. I. Fomichev, The friction welding of new high speed tool steels to
structural steels, Weld. Prod., (1980), 35–38
4 C. R. G. Ellis, Friction Welding; some recent applications of friction
welding, Welding and Metal Fabrication, (1977), 207–213
5 K. P. Imshennik, Heating in friction welding, Weld. Prod., (1973),
76–79
6 T. Rich, R. Roberts, Thermal analysis for basic friction welding,
Met. Const. and British Weld. J., (1971), 93–98
7 A. Sluzalec, International Journal of Mechanical Sciences, 32 (1990)
6, 467–478, doi:10.1016/0020-7403(90)90153-A
8 M. Sahin, H. E. Akata, Journal of Materials Processing Technology,
142 (2003) 1, 239–246, doi:10.1016/S0924-0136(03)00589-2
9 M. Sahin, H. E. Akata, Industrial Lubrication & Tribology, 56 (2004)
2, 122–129, doi:10.1108/00368790410524074
10 M. Sahin, Assembly Automation, 25 (2005) 2, 140–145,
doi:10.1108/01445150510590505
11 M. Sahin, Materials and Design, 28 (2007) 7, 2244–2250,
doi:10.1016/j.matdes.2006.05.031
12 A. W. E. Nentwig, Friction welding of cross section of different
sizes, Schweissen und Schneiden/Welding & Cutting, 48 (1996) 12,
236–237
13 B. S. Yýlbaº, A. Z. ªahin, N. Kahraman, A. Z. Al-Garni, Journal of
Materials Processing Technology, 49 (1995) 3–4, 431–443,
doi:10.1016/0924-0136(94)01349-6
14 A. Z. ªahin, B. S. Yýlbaº, M. Ahmed, J. Nickel, Journal of Materials
Processing Technology, 82 (1998) 1–3, 127–136,
doi:10.1016/S0924-0136(98)00032-6
15 W. B. Lee, S. B. Jung, Zeitschrift für Metallkunde, 94 (1993) 12,
1300–1306, doi:10.3139/146.031300
16 S. A. Fabritsiev, A. S. Pokrovsky, M. Nakamichi et al., Journal of
Nuclear Materials, 258 (1998) 2, 2030–2035, doi:10.1016/S0022-
3115(98)00126-3
17 L. Fu, S. G. Du, Journal of Materials Science, 41 (2006) 13,
4137–4142, doi:10.1007/s10853-006-6224-5
18 K. Jayabharath, M. Ashfaq, P. Venugopal et al., Materials Science
and Engineering A: Structural Materials: Properties, Microstructure
and Processing, 454–455 (2007), 114–123, doi:10.1016/j.msea.2006.
11.026
19 M. Maalekian, Science and Technology of Welding & Joining, 12
(2007) 8, 738–759, doi:10.1179/174329307X249333
20 M. Sahin, Industrial Lubrication and Tribology, 61 (2009) 6,
319–324, doi:10.1108/00368790910988435
21 M. Sahin, E. Çil, C. Misirli, Journal of Materials Engineering &
Performance, 22 (2013) 3, 840–847, doi:10.1007/s11665-012-0310-4
22 K. G. K. Murti, S. Sundaresan, Parameter optimisation in friction
welding dissimilar materials, Metal Construction, 15 (1983) 6,
331–335
23 R. Karthikeyan, V. Balasubramanian, International Journal of
Advanced Manufacturing Technology, 51 (2010) 1–4, 173–183,
doi:10.1007/s00170-010-2618-2
24 S. T. Selvamani, K. Palanikumar, Measurement: Journal of the
International Measurement Confederation, 53 (2014), 10–21,
doi:10.1016/j.measurement.2014.03.008
25 C. W. Wegst, Stahlschlüssel, Verlag Stahlschlüssel Wegst GmbH,
Marbach 1995
26 G. E. P. Box, W. H. Hunter, J. S. Hunter, Statistics for experiment,
John Wiley Publications, New York 1978
27 A. I. Khuri, J. Cornell, Response Surfaces, Design and Analysis,
Marcel Dekker, New York 1996
M. SAHIN: OPTIMIZING THE PARAMETERS FOR FRICTION WELDING STAINLESS STEEL TO COPPER PARTS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 109–115 115

U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
117–125
WEDM CUTTING OF INCONEL 718 NICKEL-BASED
SUPERALLOY: EFFECTS OF CUTTING PARAMETERS ON THE
CUTTING QUALITY
WEDM REZANJE NIKLJEVE SUPERZLITINE INCONEL 718:
VPLIV PARAMETROV REZANJA NA KVALITETO REZANJA
Ulaº Çaydaº, Mustafa Ay
University of Firat, Technology Faculty, Department of Mechanical Engineering, 23119 Elazið, Turkey
ucaydas@firat.edu.tr, ucaydas@gmail.com
Prejem rokopisa – received: 2015-01-26; sprejem za objavo – accepted for publication: 2015-02-19
doi:10.17222/mit.2015.026
Investigations of the effects of machining parameters on the cutting quality of wire-electrical-discharge-machining (WEDM)
cutting of an annealed Inconel 718 nickel-based superalloy are described in this paper. The cutting-quality characteristics
considered are the kerf width, the recast-layer thickness and the surface roughness of the cut specimens. The essential process
input parameters were identified as the pulse-on time, the pulse-peak current, and the injection pressure. The analysis of
variance (ANOVA) technique was used to find the parameters affecting the cut quality. The regression analysis was used for the
development of empirical models able to describe the effects of the process parameters on the quality of the WEDM cutting.
The ANOVA results show that the pulse-on time and the pulse-peak current are significant variables affecting the surface
roughness of wire-EDMed Inconel 718. The surface roughness, the kerf width and the recast layer of the test specimens
increased as these two variables increased. The measured and modelled results were in good agreement with respect to the
correlation coefficients Ra, RLT and Kü.
Keywords: wire-electrical-discharge machining (WEDM), Inconel 718, recast layer, surface morphology, analysis of variance
V ~lanku je opisana raziskava vplivov parametrov obdelave na kvaliteto reza, z `i~no elektroerozijo (WEDM) rezane `arjene
nikljeve superzlitine Inconel 718. Upo{tevane so bile karakteristike reza, kot je {irina reza, debelina nataljenega sloja in
hrapavost povr{ine odrezanega vzorca. Ugotovljeno je, da so bistveni vhodni parametri procesa impulzi v ~asu, tok vrha impulza
in tlak vbrizgavanja. Za iskanje parametrov, ki vplivajo na kvaliteto reza je bila uporabljena tehnika analiza variance (ANOVA).
Regresijska analiza je bila uporabljena za razvoj empiri~nega modela, ki lahko opi{e vpliv procesnih parametrov na kvaliteto
WEDM rezanja. Rezultati iz ANOVE ka`ejo, da sta impulz v ~asu in tok vrha impulza pomembni spremenljivki pri rezanju
Inconela 718 z `i~no elektroerozijo. Hrapavost povr{ine, {irina zareze in debelina pretaljene plasti so nara{~ali pri preizkusnih
vzorcih, ~e sta nara{~ali ti dve spremenljivki. Izmerjeni in modelirani rezultati so se dobro ujemali s korelacijskimi koeficienti
Ra, RLT in Kü.
Klju~ne besede: `i~na elektroerozija (WEDM), Inconel 718, pretaljena plast, morfologija povr{ine, analiza variance
1 INTRODUCTION
Among nickel-based alloys, the superalloy Inconel
718 is one of the most important ones. Due to its high
corrosion and high temperature resistance, it is
commonly used in the space industry, in particular for
the hot parts of plane, marine and industrial gas-turbine
engines, rocket engines, nuclear reactors, submarines,
pressure tanks, steam-turbine generators and other
high-temperature applications.1–5 Despite this widespread
use, it is classified as a difficult-to-machine material due
to its unique characteristics such as low thermal
conductivity, hardness, work hardening and the presence
of abrasive carbide particles in its microstructure.3,6
Due to its thermal machining ability, wire-electri-
cal-discharge machining (WEDM) has been an important
manufacturing method as it provides effective solutions
for machining difficult-to-machine materials, such as
zirconium, nimonic, titanium, nickel, etc., that cannot be
machined with traditional methods.7 Nonetheless, there
are problems that remain to be solved such as the
stresses that occur on the top layer where the material
solidifies and in the formation of the heat-affected zone,
micro-cracks, porosity, local hardness/softness, grain
growth and small alloys that get transferred via dielectric
fluids or the tool.8,9 For a complete and efficient WEDM,
the machining parameters should be selected in
accordance with the nature of the material.
Effects of the EDM-process parameters have been
investigated. Kanlayasiri and Boonmung10 studied the
effects of the pulse-on time, pulse-off time, pulse-peak
current and wire tension (a wire electrode made of Cu-35 %
of mass fractions of Zn; the wire was KH Sodick with
0.25 mm in diameter, tolerating a tension of up to 900
MPa) on the surface roughness of the DC53 die steel.
Kuriakose and Shunmugam11 investigated the characte-
ristics of a WEDMed Ti6Al4V surface. Gökler and
Ozanözgü12 experimentally investigated the effects of the
cutting parameters on the surface roughness in the
WEDM process for the 1040, 2379 and 2738 steels.
Miller et al.13 investigated the effects of the EDM process
parameters, particularly the spark cycle time and
Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125 117
UDK 669.245:621.9.048:519.61/.64 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(1)117(2016)
spark-on time on thin cross-section cuts of the Nd–Fe–B
magnetic material. Ramakrishnan and Karunamoorty14
developed a neural-network model in order to predict
process outputs and to establish the optimum process
parameters. Kang and Kim15 investigated the EDM
characteristics of nickel-based heat resistant alloys;
Hastelloy-X. Bai et al.16 developed electrical discharge
surface alloying of superalloy Haynes 230 with Al and
Mo. Aspinwall et al.17 developed 3D topographic maps
of workpiece surfaces, microstructural and microhard-
ness depth-profile data of Inconel 718.
So far, the most common workpiece materials used in
the research of WEDM have been tool or die steels. Very
little research was done to investigate the machinability
of Inconel 718 with respect to WEDM. As a result, the
effects of the machining parameters on the formation of
the recast layer and the surface integrity during the
WEDM of Inconel 718 remain to be elucidated in detail.
2 EXPERIMENTAL WORK
In this study, an annealed Inconel 718 nickel alloy
with the AMD 5596 standard number was used. The
chemical composition of this material is given in Table
1. WEDM cutting experiments were performed using an
Accutex CNC WEDM machine. In the conducted experi-
ments, a CuZn37 wire with a tensile strength of 900 N/m2
and a diameter of 0.25 mm was used. In the experimental
studies, the pulse current (A), pulse duration (μs) and
dielectric flushing circulation pressure (MPa) were cho-
sen as variable parameters. With these cutting parameters
and their levels, the Taguchi L9 experimental-design
method was chosen as the basis and a total of 9 experi-
ments were performed. The experimental parameters and
their levels are provided in Table 2 and the sequence of
the experiments is provided in Table 3.
Table 3: Experimental layout using an L9 orthogonal array
Tabela 3: Eksperimentalna postavitev z uporabo L9 ortogonalne
matrike
Exp. No Pulsed current(A)
Pulse-on
duration (ìs)
Flushing
pressure (MPa)
1 8 5 1.27
2 8 7 1.47
3 8 9 1.76
4 10 5 1.47
5 10 7 1.76
6 10 9 1.27
7 12 5 1.76
8 12 7 1.27
9 12 9 1.47
At the end of the experiments, the width of each cut
was measured at five points along the cutting channel in
order to determine the kerf widths. The measurements
were performed using a profile-measuring microscope
with a sensitivity of 0.002 mm (Model 98-0001,
SCHERR-TUMICO, USA). In order to obtain the sur-
face images, determining the microstructures of the
surfaces and the heat-affected zones, the surface that was
perpendicular and adjacent to the surface that was being
machined was chosen and these surfaces were polished
using 200–1200 mesh abrasive paper and a 3-μm
diamond paste. The polished surfaces were etched with
the electrolytic-etching method in a 50 % HCl and 50 %
methanol solution and at a 4.5 V voltage.1 The micro-
structural analysis of the specimens was done with a
scanning electron microscope (SEM) and a three-dimen-
sional atomic-force microscope (AFM). In order to
determine the elements and phases on the machined
surfaces, EDS (energy dispersive spectrograph) and
XRD (X-ray diffraction) analyses were performed.
Microhardness tests were conducted with a Leica
Q500/L hardness tester using the Vickers scale. The
surface roughness was obtained using a Mitutoyo
Surftest SJ-201 portable device.
3 RESULTS AND DISCUSSION
3.1 Effect of the cutting parameters on the surface
structure
In Figure 1, an image obtained after the cutting of
the Inconel 718 nickel alloy with WEDM is shown. The
surface is composed of spherical grains detached from
the material that cannot be removed from the plain space
with the liquid pressure, debris that melt and stick to the
U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
118 Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125
Table 1: Chemical composition of Inconel 718
Tabela 1: Kemijska sestava Inconel 718
Element Composition (w/%)
Ni 53.60
Cr 18.20
Nb 5.06
Mo 3.04
Ti 0.97
Al 0.44
C 0.052
B 0.003
Si 0.10
S <0.002
P <0.005
Fe Balance
Table 2: Machining parameters of WEDM in this study
Tabela 2: Parametri obdelave z WEDM v tej {tudiji
Cutting conditions Settings
Level 1 Level 2 Level 3
Pulsed current (A) 8 10 12
Gap voltage (V) 39 – –
Pulse-on duration (μs) 5 7 9
Pulse-off duration (μs) 9 – –
Wire tension (N) 10 – –
Wire speed (mm/s) 15 – –
Flushing-circulation pressure (MPa) 1.27 1.47 1.76
Table-feed ratio (mm/min) 10
surface in drops, cracks, residues and randomly dis-
persed craters of various sizes. These craters are charac-
terized as cavities formed by the spherical chips
detached from the surface due to the effect of the sparks
between the wire and the workpiece during the ma-
chining. In terms of the surface appearance, it was
determined that the densities of the machined surfaces
are similar, whereas the densities of the craters and hills
on the surface are different, depending on the variations
in the current and pulse duration.
Surface images of the specimens cut under different
currents, pulse durations and fluid-circulation pressures
are seen on Figure 2. As can be seen, when the strength
of the current is constant, the surface gets rougher as the
pulse duration increases. The pulse duration is the time
duration of the discharge applied to the wire electrode. In
other words, it is the time of the spark discharge that
forms between the wire and the workpiece. Therefore, an
increase in the pulse duration leads to an increased trans-
fer of heat onto the surface of the workpiece and a for-
mation of larger craters on the surface of the workpiece.
Similarly, the surface was observed to deteriorate also
with an increased current.
An increase in the current intensity leads to a more
intense energy discharge to the workpiece surface; thus,
more chips detach from the surface. The increase in the
amount of detached chips from the surface, at the same
time, deteriorates the surface roughness.18 The results of
the experiments show that the liquid-circulation pressure
had little effect on the surface structure. Similarly, pre-
vious studies indicate that while the surface roughness
increases with the increasing current and pulse duration,
and wire fractures are prevented using an appropriate
liquid-circulation pressure, the effect of the liquid-circu-
lation pressure on the surface roughness, the thickness of
the recast layer and the microstructure is very low.9,19
3.2 Effect of the cutting parameters on the recast-layer
thicknesses
The basic metallurgical effects that occur during the
cutting with WEDM are the formation and width of the
recast layer and/or the heat-affected region. Figure 3
shows a SEM photo obtained from the surface adjacent
to a machined surface. When the image is examined, it is
seen that three different regions formed after the cutting
operation. At the very top, there is the recast layer. This
layer developed because of the rapid cooling and
re-solidification of the molten material with an auxiliary
liquid-circulation pressure, without being removed from
the cutting channel during the cutting with WEDM. The
heat-affected region is the region where no melting
occurs during the cutting, but the base metal undergoes
microstructural changes under the influence of the result-
ing heat. At the very bottom, there is the base material.
U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125 119
Figure 1: SEM micrograph of a WEDM-cut surface of Inconel 718
Slika 1: SEM-posnetek povr{ine reza Inconel 718 po rezanju z
WEDM
Figure 2: SEM micrographs of a surface WEDM-processed under
different machining conditions
Slika 2: SEM-posnetki z WEDM obdelane povr{ine pri razli~nih po-
gojih obdelave
Figure 3: SEM of a kerf cross-section showing the RLT and HAZ
Slika 3: SEM-prikaz pre~nega preseka zareze, ki ka`e RLT in HAZ
EDS and XRD results for the recast layer are pro-
vided in Figures 4 and 5. When the analyses are eva-
luated, the results show that the recast layer is enriched
with oxygen and carbon. This phenomenon was
explained by T. R. Newton et al.20 to be the result of the
electrolysis taking place in the dielectric liquid. In addi-
tion, the presence of copper and zinc compounds in the
recast layer is evident. While Inconel 718 contains
0.08 % of mass fractions of Cu, the EDS results for the
recast layer indicate that the density of Cu increases.
Even though there is no Zn in the structure, the presence
of this element in the recast layer indicates that these
elements diffuse from the wire electrode into the
workpiece, as reported in the previous studies.21
Table 4: Microhardness distribution through the cut surface section
Tabela 4: Razporeditev mikrotrdote skozi prerezano podro~je na
povr{ini
Measurement
region
Experiment number
1 2 3 4 5 6 7 8 9
Base material 418 410 408 416 420 416 413 406 410
Recast layer 387 392 400 375 402 402 396 400 402
Heat-affected
zone 415 474 422 424 426 412 418 410 418
Table 4 gives the average microhardness values for
the three regions defined. While the hardness of the
heat-affected region shows similarity to the base mate-
rial, the microhardness of the recast layer shows a slight
decrease when compared to the base material. The
reason for this is assumed to be a decrease in the hard-
ness of the recast layer, being a result of the dissolution
of the 
’ (N3Al,Ti) secondary phases and carbides back
into the main matrix, after the heat treatment of the
specimens. In their study, T. R. Newton et al.20 indicated
that the hardness of the recast layer reduces due to the
wire-electrical-discharge machining of the age-hardened
Inconel 718 alloy. In their study, a reduced chromium
density in the recast layer and the existence of Cu and Zn
in this region cause the properties of the region to
change.
Since the outermost recast layer is in direct contact
with the environment, the determination of the thickness
of this layer plays an important role when assessing the
finish cut. Figure 6 shows the effects of the cutting para-
meters on the recast-layer thickness (RLT). In addition,
in Figure 7, SEM images show a variation in the RLT
depending on the experimental parameters. When these
figures are analyzed, it is seen that while the effect of the
circulation pressure on the recast layer is small, the
effects of the current and the pulse duration are large.
Under constant current conditions, the average RLT in-
creases with the increasing pulse duration. Since the
amount of heat transferred onto the surface of the work-
piece increases with the increasing pulse duration, the
amount of the material to be melted from the surface
increases as well.
U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
120 Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125
Figure 5: X-ray analysis of the recast layer
Slika 5: Rentgenska analiza pretaljene plasti
Figure 6: Main-effect plots of WEDM cutting factors for the RLT
Slika 6: Diagrami glavnih vplivnih faktorjev pri WEDM rezanju na
RLT
Figure 4: EDS analysis of the recast layer
Slika 4: EDS-analiza pretaljene plasti
As the pulse duration increases, the isothermal curves
of the material spread more into the sublayers and the
amount of the heat spreading under the surface covers a
large area. This, in turn, causes a large region to be in-
fluenced by the heat and the RLT to increase.22 More-
over, the average RLT increases considerably with
increased current values under constant pulse-duration
conditions. As the current increases, the discharge
energy gets more intense. High discharge energy causes
more material to melt. In parallel with the thickness of
the melted material, the RLT increases as well. In their
study, Kanlayasiri and Boonmung10 mentioned that in the
cases where the current and pulse duration were high, the
machined surface was rougher and thermal damages
were deeper since the discharge energy acting on the
workpiece surface was more intense. For this reason,
they indicated that the current and pulse duration should
be kept at low levels, but also that in the cases where
these two parameters are kept at low levels, the duration
and cost of the machining process increase.
3.3 Effect of the cutting parameters on the surface
roughness
During wire-electrical-discharge machining, the dis-
charge energy released with each discharge produces a
very high amount of heat at the point where the spark
hits the workpiece. This heat causes small pieces on the
surface to melt and detach via vaporization. Spherical
chips removed from the surface with each discharge
change the machined surface into a crater structure. The
surface morphology depends on the sizes of the craters
that form due to the contact between the sparks and the
surface. Therefore, the surface profile is a function of the
current and pulse duration, determining the intensity of
the applied sparks.23
U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125 121
Figure 7: Cross-sectional SEM micrographs of a surface WEDM-processed under different machining conditions
Slika 7: SEM-posnetki presekov z WEDM obdelane povr{ine pri razli~nih pogojih obdelave
Figure 8: Main-effect plots of WEDM cutting factors for the surface
roughness
Slika 8: Diagrami glavnih vplivnih faktorjev pri WEDM rezanju na
hrapavost povr{ine
Figure 8 demonstrates the effects of the variables
such as the current intensity, the pulse duration and the
dielectric flushing pressure on the surface roughness, Ra.
The fact that the current intensity and pulse duration
have large effects on the surface roughness is clearly
seen on the graphs. If the current is low, the intensity of
the sparks that hit the surface of the piece with each
discharge is also lower. This leads to a smoother surface
due to a better erosion effect. Additionally, the amount of
heat transferred to the surface of a piece decreases with a
shortening pulse duration; therefore, less metal is melted.
Thus, with the craters being more superficial, the surface
roughness decreases.23,24
The lowest surface roughness of 2.08 μm was
measured at the end of experiment no. 1, while the
highest surface roughness of 3.01 μm was measured at
the end of experiment no. 9. In Figure 9, SEM and 3D
AFM images of specimens 1 and 9 are shown. On the
AFM image, it is possible to see the maximum and mini-
mum heights of the surface profile and hillocks and
valleys. A higher roughness can be assessed on the basis
of the values for the Z depth of the AFM profile. When
both figures are taken into account, it is observed that the
surface-roughness values of the specimens show a ten-
dency to decrease while all the cutting parameters
decrease.
3.4 Effect of the WEDM parameters on the kerf width
Figure 10 shows a schematic representation of a kerf
formed during wire-electrical-discharging. During
wire-electrical-discharge machining, pieces are removed
from the workpiece via melting as a result of the high
temperature caused by spark discharges, occurring bet-
ween the wire and the workpiece and the kerf occurs
when these pieces are removed from the intermediate
region with the liquid-circulation pressure. The kerf
width varies depending on the parameters used during
the machining. Alias et al.25 proved that as the feed ratio
increases, the kerf width decreases and, as a result of the
experiments, low feed ratios increased the kerf width.
Somashekhar and Ramachandran26 did not recommend
machining at high feed ratios as high feed rates increase
the distortions in the kerf width. Tosun et al.27 demon-
U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
122 Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125
Figure 9: SEM and three-dimensional AFM images of the first and
ninth experiments
Slika 9: SEM- in tridimenzionalni AFM-sliki prvega in devetega
preizkusa
Figure 11: Main-effect plots of WEDM cutting factors for the kerf
width
Slika 11: Diagrami glavnih vplivnih faktorjev pri WEDM rezanju na
{irino zareze
Figure 10: Schematic representation of the kerf width
Slika 10: Shemati~en prikaz {irine zareze
strated experimentally and mathematically that the
open-circuit voltage and the pulse duration have a high
influence on the kerf width and the material-lift ratio.
They indicated that, for the kerf-width control, the volt-
age is three times more effective than the pulse duration.
Graphs showing the effects of the experimental para-
meters on the kerf width are given in Figure 11. As it
can be seen, the average top kerf width varies dramati-
cally, depending on the variations in the current intensity
and pulse duration. An increase in the pulse duration
causes a longer-term spark shock and, accordingly, the
removal of more chips from the surface. As a result,
depending on the amount of the chips removed from the
cutting zone, the kerf width increases. Similarly, an
increase in the kerf width was observed with the
increasing current intensity. If the current is low, the
intensity of the sparks that hit the surface of the piece
with each discharge is also low. High-intensity sparks
lead to the widening of the kerf by creating a deeper
corrosion effect on the surface.
Figure 12 gives images of specimens 1 and 9 where
the lowest and highest kerf widths were obtained, respec-
tively. Under the conditions of experiment 1 where the
current intensity (8 A), pulse duration and liquid-circu-
lation pressure (5 μs, 1.27 MPa) were at their lowest
values, the kerf width was measured as 0.335 mm. Under
the conditions of experiment 9 where the current inten-
sity, pulse duration and liquid-circulation pressure (12 A,
9 μs, 1.47 MPa) were at their highest values, the kerf
width was measured as 0.375 mm.
4 STATISTICAL ANALYSIS OF THE
EXPERIMENTAL RESULTS
In this study, an analysis of the WEDM experimental
results was performed using the analysis of variance
(ANOVA) method. This method is used to statistically
examine the impacts of the machining parameters on the
operational performance. The experiment numbers and
the measurements obtained at the end of the experiments
are collectively presented in Table 5. The ANOVA
results for the RLT, surface roughness and kerf width in
the case of WEDM can be seen in Tables 6 to 8, respec-
tively. According to the results in Table 6, the current
intensity has a major effect of as much as 84 % on the
U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125 123
Figure 12: Light micrographs of the kerf width obtained at different
cutting conditions: a) pulsed current = 8 A, pulse-on duration = 5 μs,
flushing pressure (kg/cm2) = 13, b) pulsed current = 12 A, pulse-on
duration = 9 μs, flushing pressure (kg/cm2) = 15
Slika 12: Svetlobna posnetka {irine zareze, dobljena pri razli~nih
pogojih rezanja: a) tok pulza = 8 A, trajanje pulza = 5 μs, tlak
splakovanja (kg/cm2) = 13, b) tok pulza = 12 A, trajanje pulza = 9 μs,
tlak splakovanja (kg/cm2) = 15
Table 7: Results of ANOVA for the surface roughness
Tabela 7: Rezultati ANOVA za hrapavost povr{ine
WEDM cutting
parameter
Degree of
freedom
(df)
Sum of
square
(SSA)
Mean
square F P (%)
Pulsed current
(A) 2 8.154 4.077 13.83 82.17
Pulse-on
duration (μs) 2 1.21 0.60 0.42 12.18
Flushing
pressure (MPa) 2 0.16 0.08 0.05 1.57
Error – – – – 4.08
Total 6 9.524 – – 100
Table 8: Results of ANOVA for the kerf width
Tabela 8: Rezultati ANOVA za {irino reza
WEDM cutting
parameter
Degree of
freedom
(df)
Sum of
square
(SSA)
Mean
square F P (%)
Pulsed current
(A) 2 0.000636 0.000318 3.65 54.87
Pulse-on
duration (μs) 2 0.000508 0.000254 2.34 43.83
Flushing
pressure (MPa) 2 0.000014 0.0000074 0.04 1.17
Error – – – – 0.13
Total 6 0.05939 – – 100
Table 5: Experimental results
Tabela 5: Rezultati eksperimentov
Exp. No RLT (μm)
Surface
roughness
(μm)
Kerf width
(mm)
1 2.61 2.08 0.335
2 2.65 2.31 0.348
3 2.69 2.51 0.351
4 2.73 2.61 0.349
5 2.76 2.69 0.359
6 2.80 2.89 0.365
7 2.78 2.95 0.353
8 2.84 3.00 0.367
9 2.87 3.01 0.375
Table 6: Results of ANOVA for the recast-layer thickness
Tabela 6: Rezultati ANOVA za debelino pretaljene plasti
WEDM cutting
parameter
Degree of
freedom
(df)
Sum of
square
(SSA)
Mean
square F P (%)
Pulsed current
(A) 2 0.04969 0.02484 15.11 83.65
Pulse-on
duration (μs) 2 0.00962 0.00481 0.58 16.16
Flushing
pressure (MPa) 2 0.00008 0.00004 0.00 0.15
Error – – – – 0.04
Total 6 0.05939 – – 100
RLT. The effect of the pulse duration on the RLT is 16 %,
whereas the liquid-circulation pressure has no effect on
the RLT. Similarly, Tables 7 and 8 clearly show that the
current intensity has a much larger effect on the surface
roughness and kerf width than the pulse duration.
5 MATHEMATICAL MODELING OF THE
EXPERIMENTAL RESULTS
In this study, mathematical modeling of the RLT,
surface roughness and kerf width were performed by
employing the multiple-linear-regression method on the
results of the cutting experiments with WEDM. In
addition, the appropriateness of the models obtained with
the method of regression was investigated with ANOVA.
Regression models with dependent parameters of the
current intensity (Ip), pulse duration (Ton) and liquid-cir-
culation pressure (Ps), and control parameters of the
recast-layer thickness (RLT), surface roughness (Ra) and
kerf width (Kw) are expressed as follows:
Recast-layer thickness
RLT = a0 + a1 × IP + a2 × Ton – a3 × Ps (1)
Surface roughness
Ra = a0 + a1 ×IP + a2 × Ton + a3 × Ps (2)
Kerf width
Kw = a0 + a1 × IP + a2 × Ton – a3 × Ps (3)
The regression and correlation coefficients obtained
are presented collectively in Table 9. When these coeffi-
cients are substituted into the equations, the mathema-
tical models of the recast-layer thickness (RLT), surface
roughness (Ra) and kerf width (Kw) are obtained as:
RLT = 2.18 + 0.0450 × IP + 0.0200 × Ton –
– 0.00140 × Ps (4)
Ra = 0.307 + 0.172 × IP + 0.0642 × Ton +
+ 0.0130 × Ps (5)
Kw = 0.278 + 0.00508 × IP + 0.00450 × Ton –
– 0.000325 × Ps (6)
The correlation coefficients of the models from Table
9 show that the models that are quite appropriate.
Experimental values of the recast-layer thickness (RLT),
surface roughness (Ra) and kerf width (Kw) are given in
Table 10 along with their predicted values calculated
with the obtained mathematical models.
Table 9: Regression and correlation coefficients
Tabela 9: Regresijski in korelacijski koeficienti
Regression
coefficients Correlation coefficients (r)
a0 a1 a2 a3
Recast-layer
thickness 2.18 0.0450 0.0200 0.00140 98.1 %
Surface
roughness 0.307 0.172 0.0642 0.0130 97.8 %
Kerf width 0.278 0.00508 0.00450 0.000325 95.9 %
Figures 13 to 15 show graphs of these values in their
respective order. As it is seen on the figures, the expe-
rimental and numerical values show distributions quite
close to the regression lines.
U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
124 Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125
Figure 15: Comparison of experimental and predicted kerf-width
values
Slika 15: Primerjava eksperimentalno dolo~ene in napovedane {irine
zareze
Figure 13: Comparison of experimental and predicted recast-layer
thickness values
Slika 13: Primerjava eksperimentalno dolo~ene in napovedane debe-
line pretaljenega sloja
Figure 14: Comparison of experimental and predicted surface-rough-
ness values
Slika 14: Primerjava eksperimentalno dolo~ene in napovedane hrapa-
vosti povr{ine
6 CONCLUSIONS
Based on the conducted research and investigations,
the following conclusions can be drawn:
1. In the WEDM cutting process, the liquid-circulation
pressure has little effect on the recast-layer thickness,
surface roughness and kerf width, whereas the
current and pulse duration are the most important
parameters determining the quality of a cut. The
highest quality cut was obtained with experiment 1
where the current and pulse duration were at their
lowest values.
2. Considering the results of ANOVA, the intensity of
the current was statistically found to have a larger
effect on the surface roughness, kerf width and RLT
than the pulse duration.
3. Taking the correlation coefficients into account, the
obtained linear-regression models yield estimates
with appropriate error rates. These results, in turn,
show that the obtained models are appropriate.
7 REFERENCES
1 A. Sharman, R. C. Dewes, D. K. Aspinwall, Journal of Materials
Processing Technology, 118 (2001), 29–35, doi:10.1016/S0924-0136
(01)00855-X
2 A. Hasçalik, M. Ay, Optics & Laser Technology, 48 (2013),
554–564, doi:10.1016/j.optlastec.2012.11.003
3 M. Ay, U. Çaydaº, A. Hasçalik, Materials and Manufacturing
Processes, 25 (2010), 1–6, doi:10.1080/10426914.2010.502953
4 E. O. Ezugwu, International Journal of Machine Tools & Manu-
facture, 45 (2005), 1353–1367, doi:10.1016/j.ijmachtools.2005.
02.003
5 D. Dudzinski, A. Devillez, A. Moufki, D. Larrouque`re, V. Zerrouki,
J. Vigneau, International Journal of Machine Tools & Manufacture,
44 (2004), 439–456, doi:10.1016/S0890-6955(03)00159-7
6 M. Ay, U. Çaydaº, A. Hasçalik, International Journal of Advanced
Manufacturing Technology, 44 (2012), 439–456, doi:10.1007/
s00170-012-4385-8
7 D. Scot, S. Boyina, K. P. Rajurkar, Int. J. Prod. Res., 29 (1991),
2189–2207, doi:10.1080/00207549108948078
8 P. Bleys, J. P. Kruth, B. Lauwers, B. Schacht, V. Balasubramanian, L.
Froyen, J. Van Humbeeck, Advanced Engineering Materials, 8
(2006), 15–25, doi:10.1002/adem.200500211
9 A. Hasçalik, U. Çaydaº, Journal of Materials Processing Technology,
148 (2004), 362–367, doi:10.1016/j.jmatprotec.2004.02.048
10 K. Kanlayasiri, S. Boonmung, Journal of Materials Processing
Technology, 187–188 (2007), 26–29, doi:10.1016/j.jmatprotec.2006.
11.220
11 S. Kuriakose, M. S. Shunmugam, Materials Letters, 58 (2004),
2231–2237, doi:10.1016/j.matlet.2004.01.037
12 M. Ý. Gökler, A. M. Ozanözgü, International Journal of Machine
Tools & Manufacture, 40 (2000), 1831–1848, doi:10.1016/S0890-
6955(00)00035-3
13 S. F. Miller, C. C. Kao, A. J. Shih, J. Qu, International Journal of
Machine Tools & Manufacture, 45 (2005), 1717–1725, doi:10.1016/
j.ijmachtools.2005.03.003
14 R. Ramakrishnan, L. Karunamoorty, Journal of Materials Processing
Technology, 207 (2008), 343–349, doi:10.1016/j.jmatprotec.2008.
06.040
15 S. H. Kang, D. E. Kim, KSME International Journal, 17 (2003),
1475–1484, doi:10.1007/BF02982327
16 C. Y. Bai, C. H. Koo, C. C. Wang, Materials Transactions, 45 (2004),
2878–2885, doi:10.2320/matertrans.45.2878
17 D. K. Aspinwall, S. L. Soo, A. E. Berrisford, G. Walder, CIRP
Annals, 57 (2008) 1, 187–190, doi:10.1016/j.cirp.2008.03.054
18 M. S. Hewidy, T. A. El-Taweel, M. F. El-Safty, Journal of Materials
Processing Technology, 169 (2005), 328–336, doi:10.1016/
j.jmatprotec.2005.04.078
19 Y. S. Liao, J. T. Huang, H. C. Su, Journal of Materials Processing
Technology, 71 (1997), 487–493, doi:10.1016/S0924-0136(97)
00117-9
20 T. R. Newton, S. N. Melkote, T. R. Watkins, R. M. Trejo, L. Reister,
Materials Science and Engineering A, 513–514 (2009), 208–215,
doi:10.1016/j.msea.2009.01.061
21 J. T. Huang, Y. S. Liao, W. J. Hsue, Journal of Materials Processing
Technology, 87 (1999), 69–81, doi:10.1016/S0924-0136(98)00334-3
22 U. Çaydaº, Investigation of the Machinability of Ti6Al4V alloy by
electrical discharge and electrochemical machining processes, PhD
Thesis, Firat University Graduate School of Natural and Applied
Sciences, 2008, 95–98
23 Y. H. Guu, K. Hou, Materials Science and Engineering A, 466
(2007), 61–67, doi:10.1016/j.msea.2007.02.035
24 H. Ramasawmy, L. Blunt, K. P. Rajurkar, Precision Engineering, 29
(2005), 479–490, doi:10.1016/j.precisioneng.2005.02.001
25 A. Alias, B. Abdullah, N. M. Abbas, Procedia Engineering, 41
(2012), 1806–1811, doi:10.1016/j.proeng.2012.07.387
26 K. P. Somashekhar, N. Ramachandran, J. Mathew, Int. J. Adv.
Manuf. Technol., 51 (2010), 611–626, doi:10.1007/s00170-010-
2645-z
27 N. Tosun, C. Cogun, A. Gnan, Machining Science and Technology, 7
(2003), 209–219, doi:10.1081/MST-120022778
U. ÇAYDAª, M. AY: WEDM CUTTING OF INCONEL 718 NICKEL-BASED SUPERALLOY: EFFECTS ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 117–125 125
Table 10: Estimated values and calculations obtained with mathematical models
Tabela 10: Dolo~ene in z dobljenim matemati~nim modelom izra~unane vrednosti
Exp. No
Recast-layer
thickness Surface roughness Kerf width
Experimental
RLT
Estimated
RLT
Experimental
Ra
Estimated
Ra
Experimental
Kt
Estimated
Kt
1 2.61 2.61111 2.08 2.15889 0.335 0.334444
2 2.65 2.65444 2.31 2.26556 0.348 0.348444
3 2.69 2.68444 2.51 2.47556 0.351 0.351111
4 2.73 2.72444 2.61 2.57556 0.349 0.349111
5 2.76 2.76111 2.69 2.76889 0.359 0.358444
6 2.80 2.80444 2.89 2.84556 0.365 0.365444
7 2.78 2.78444 2.95 2.90556 0.353 0.353444
8 2.84 2.83444 3.00 2.96556 0.367 0.367111
9 2.87 2.87111 3.01 3.08889 0.375 0.374444

N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
127–135
INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE
BEHAVIOR OF SELF-COMPACTING CONCRETE
VPLIV IZKOPANIH SEDIMENTOV NA KR^ENJE
SAMOZGO[^EVALNEGA BETONA
Nasr-Eddine Bouhamou, Fouzia Mostefa, Abdelkader Mebrouki, Karim Bendani,
Nadia Belas
LCTPE Laboratory, Civil Engineering Department, Mostaganem University, Route de Belahcel BP 227, 27000 Mostaganem, Algeria
bendanik@yahoo.fr
Prejem rokopisa – received: 2013-10-16; sprejem za objavo – accepted for publication: 2015-02-10
doi:10.17222/mit.2013.252
Every year, millions of cubic meters of dams and restraints are dredged as part of the management and prevention procedures all
over the world. These dredged sediments are considered as natural waste leading to environmental, ecological and even
economic problems associated with their processing and depositing. Nevertheless, in the context of a sustainable development
policy, a way of their management is open aiming at assessing the sediments as a building material and, particularly, as a new
binder that can be industrially exploited and that can improve the physical, chemical and mechanical characteristics of the
concrete. This study is a part of the research made at the Civil Engineering Department at the University of Mostaganem
(Algeria) on the impact of the mud dredged from the Fergoug Dam on the behavior of self-consolidating concrete (SCC) in the
fresh and hardened states, such as the mechanical performance and its impact on different deformations (shrinkage). The work
aims to assess this mud in the SCC and to show possible interactions between the constituents. The obtained results provide the
details needed for producing the SCC based on calcined mud.
Keywords: sediment, calcined mud, self-consolidating concrete, fresh state, hard state, shrinkage
Vsako leto po vsem svetu kot varovalni ukrep izkopljejo milijone kubi~nih metrov sedimentov iz jezov in zadr`evalnikov. Ti
izkopani sedimenti so obravnavani kot naravni odpadek, ki povzro~a okoljske, ekolo{ke in celo ekonomske te`ave pri njihovi
predelavi ali odlaganju. Vseeno, v kontekstu politike trajnostnega razvoja je postavljena pot za njihovo obdelavo, za oceno teh
sedimentov kot gradbenega materiala in {e posebej kot novega veziva, ki ga je mogo~e industrijsko izkoristiti in ki lahko
izbolj{a fizikalne, kemijske in mehanske lastnosti betona. Ta {tudija je del raziskovalnega dela, opravljenega na oddelku za
gradbeni{tvo Univerze v Mostaganemu (Al`irija), o vplivu izkopanega mulja iz jezu Fergoug na vedenje samozgo{~evalnega
betona (SCC) v sve`em in v strjenem stanju, na mehanske lastnosti in na razli~ne deformacije (kr~enje). Namen je oceniti to
blato v SCC in pokazati morebitne interakcije med sestavinami. Prikazani rezultati so dobra mo`nost za izdelavo SCC na osnovi
kalciniranega blata.
Klju~ne besede: sediment, kalcinirano blato, samozgo{~evalni beton, sve`e stanje, trdo stanje, kr~enje
1 INTRODUCTION
This work addresses general problems associated
with durable development. The search for new building
materials indicates that the research is focused on the
possibilities of reusing industrial and natural waste as an
alternative to the currently used materials that will
become scarce in the near future.
River sediments in dams can be seen as an environ-
mental and economic threat. The big quantities generated
by the silting phenomena make the authorities perplexed
about their depositions.
An assessment of these sediments made in the area of
civil engineering proved that they could be used as build-
ing materials (bricks, aggregates and cementitious mate-
rials).
As these studies are in their first step, the reuse of
sediments in concrete is poorly documented, particularly
for self-compacting concrete.
It is known that SCC contains a very high amount of
a paste that requires a large quantity of cement; so, its
partial substitution with sediments can be a solution, re-
ducing the use of cement.
2 IDENTIFICATION OF FERGOUG DAM
Regarded as one of the most silted dams in Algeria,
the Dam of Fergoug (Figure 1) gave rise to the interest
of several Algerian researchers who contributed various
studies towards highlighting the causes, making it a
disastrous dam.
These studies were primarily focused on explaining
the phenomenon of silting, trying to find solutions for it
and assessing the possibility of using the dredged sedi-
ments.1,2
During its existence, the dam underwent several ope-
rations of dredging; the first one was carried out in 1984
and 1986 when more than 10 million m3 of mud were
removed. The second operation of dredging was carried
out in 1992 when 6.5 million m3 of mud was evacuated.
The last operation was launched in 2005, at a cost of 800
million DA, which is 130 DA/m3, and carried out by the
National Agency for Dams.
Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135 127
UDK 691.004.8:658.567.1 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 50(1)127(2016)
In spite of these various attempts, the rate of the
current silting is estimated to be 97.77 % according to
the bulletin published by the Agency for the Chergui
Hydro Channel [ABHC]3.
3 MATERIALS AND METHODS
3.1 Used materials
3.1.1 Cement
The cement used was CEM I 42.5, made at the
Zahana Factory (west of Algeria). The average physical
and chemical characteristics of this cement are given in
Table 1.
Table 1: Physical properties and chemical analysis of the cement
Tabela 1: Fizikalne lastnosti in kemijska analiza cementa
Physical properties of cement
Bulk density (g/cm3) 1.18
Specific gravity (g/cm3) 3.13
Fineness (Blaine) (cm2/g) 3180
Chemical analysis of cement, w/%
SiO2 20.90
CaO 63.93
MgO 1.45
Fe2O3 5.93
Al2O3 5.10
SO3 0.86
Na2O 0.17
K2O 1.34
Loss on ignition 0.86
Carbonates –
CO2 –
H2O 0.6
3.1.2 Mud
The mud was taken downstream of the dam (Figure
2a) and activated thermally by burning it in a slow oven
at a temperature of (750 ± 5) °C at a rate of 5 °C/min for
5 h, followed by steaming, crushing and sieving to 80 μm
(Figure 2b). The calcined mud (Figure 2c) was obtained
and stored away from the air and any moisture. The
N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
128 Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135
Figure 3: Fergoug-mud grain-size distribution (SIBELCO Laboratory,
France, 2011)
Slika 3: Razporeditev velikosti zrn iz blata Fergoug (SIBELCO Labo-
ratory, France, 2011)
Figure 1: Fergoug Dam and its effluents
Slika 1: Jez Fergoug s pritoki
Figure 2: Mud-preparation stages
Slika 2: Stopnje priprave blata
Table 2: Physical characteristics of the calcined mud
Tabela 2: Fizikalne zna~ilnosti kalciniranega blata
Test Values
Bulk density (g/cm3) 0.53
Specific gravity (g/cm3) 2.62
Fineness (Blaine) (cm2/g) 7964.00
Table 3: Chemical composition of the calcined mud
Tabela 3: Kemijska sestava kalciniranega blata
Component Content, w/%
SiO2 54.69
CaO 14.25
MgO 3.08
AL2O3 15.49
Fe2O3 7.50
SO4 –
Loss on ignition 1.87
physical and chemical characteristics of the calcined
mud are presented in Tables 2 and 3.
The grain-size analysis of the mud was carried out at
the SIBELCO Laboratory (France). The results of the
analysis of this mud are shown with a grading curve in
Figure 3.
3.1.3 Aggregates
The aggregates used in this work consist of broken
particles of limestone with a well graded distribution,
obtained from the quarry of Kristel (the Oran area) and
the sand from the siliceous sea, from the quarry of Sidi
Lakhdar (the area of Mostaganem). Table 4 gives the
characteristics of the aggregates for the whole of the con-
crete compositions.
Table 4: Physical characteristics of different aggregates
Tabela 4: Fizikalne zna~ilnosti razli~nih agregatov
Sea sand
(Ss)
Quarry
sand (Sc) Gravel (G)
Class 0/2 0/3 3/8 8/15
Nature Silicious Limestone Limestone Limestone
Specific gravity
(g/cm3) 2.56 2.68 2.66 2.66
Absorption (%) – – 0.86 0.40
Fineness
modulus 1.64 2.63 – –
Sand equivalent
(ES) 83.18 88.96
3.1.4 Admixture
The admixture employed is a superplasticizer pro-
vided by the GRANITEX Society, a high water reducer,
containing synthesized polymers combined according to
the NF EN 934-2 standard. The MEDAFLOW113 allows
obtaining concretes and mortars of a very high quality,
maintaining the workability and avoiding the segrega-
tion. Its density is 1.12 g/cm3 and its proportioning can
vary from 0.8 % to 2.5 % of the binder mass.
3.2 Concrete mixtures
Four self-consolidating concretes were made to study
the substitution effect of the cement with the calcined
mud on the behavior in the fresh and hard states of the
SCC. The concretes were made by adopting the method
of the volume of the paste. The admixture proportioning
is calculated in order to limit the segregation and
bleeding and to obtain a distribution ranging between 60
cm and 75 cm.
The aggregate proportioning (G/S), the water/binder
ratio (W/B) and the volume of the paste were kept con-
stant for all the compositions of the SCCs. Tests were
carried out on the concretes containing various percen-
tages of the substitution mud with respect to the volume
of cement, i.e., (10, 15 and 20) %. The compositions of
various mixtures are presented in Table 5.
Table 5: Mix proportions of concretes
Tabela 5: Razmerje me{anic v betonih
Mix proportion
(kg/m3)
Concrete mixes
SCCR SCCM10 SCCM15 SCCM20
Cement 450 420 408 395
Calcined mud – 35 52 66
Water 225 218 216 213
Admixture 5.7 7.5 8.3 9.8
W/B 0.5 0.5 0.5 0.5
Ss 560 560 560 560
Sc 251 251 251 251
Gravel (3/8) 333 333 333 333
Gravel (8/15) 499 499 499 499
SCCR: Reference Concrete (0 % calcined mud)
SCCM10: Concrete with 10 % of calcined mud
SCCM15: Concrete with 15 % of calcined mud
SCCM20: Concrete with 20 % of calcined mud
3.2.1 Tests of the fresh concrete
The characterization made in the fresh state of the
concretes was limited to the tests recommended by
AFGC4 including the slump flow, L-box, sieve stability
and bleeding.
3.2.2 Tests of the hardened concrete
3.2.2.1 Mechanical strength
The mechanical compressive strength is an essential
characteristic of a concrete material and a fundamental
parameter of our study. Consequently, its evolution was
measured for all the formulations of concrete studied
within this work.
The samples used to determine the mechanical com-
pressive strength of various studied concretes are cylin-
drical test tubes with a diameter of 11 cm and a height of
22 cm. Once removed from the mold, they are preserved
in water for a certain period (1, 7, 28 and 90) d. To mea-
sure the average tensile strength, three samples (7 cm ×
7 cm × 28 cm) were broken at various ages by means of
three-point-bending tensile tests.
3.2.2.2 Shrinkage deformation
It is interesting here to observe the deformations of
the hardened material (after 24 h). For this purpose, two
series of samples were produced and preserved in two
different environments with and without a hydrous ex-
change with the external medium. The tests were per-
formed to measure the total and endogenous shrinkage
deformations. The shrinkage deformations were mea-
sured with a refractometer used on the prismatic test
tubes with the size of 7 cm × 7 cm × 28 cm, placed in an
air-conditioned room with a relative humidity of (20 ± 1)
°C or (50 ± 5) %, according to the following two con-
ditions:
• A hydrous exchange of the material with the environ-
ment: the total shrinkage is obtained and it represents
the sum of the endogenous and drying shrinkages.
• No hydrous exchange with the environment because
the test tubes are covered with one or two sheets of
N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135 129
self-adhesive aluminum paper: the endogenous
shrinkage is identified.
The drying-shrinkage deformation is obtained from
the difference between the total and endogenous shrink-
age deformations.
Once removed from the mold, six test tubes relative
to each concrete (three for the total shrinkage and three
for the endogenous shrinkage) were tested over a very
short period: first at its beginning, later the periodicity of
the measurement increased with the time.
4 RESULTS AND DISCUSSIONS
4.1 Index and the activity coefficient of the calcined
mud
The index of the activity noted, Ip, is defined as the
ratio between compressive strengths fp(t) and f0(t) (Equa-
tion (1)) with respect to the strength of the standardized
mortar containing 25 % of the calcined mud as the ce-
ment substitution p and the reference-mortar strength
(with cement only) (Table 6):
I p
f t
f t
( )
( )
( )
=
p
0
(1)
Table 6: Results of the compressive-crushing tests
Tabela 6: Rezultati tla~nih poru{nih preizkusov
Age f0(t)/MPa fp(t)/MPa I(p)
28 d 47.4 39.20 0.83
90 d 52.30 45.08 0.86
4.2 Fresh states
The characterization results carried out on the con-
cretes are presented in Table 7.
Table 7: Workability test results
Tabela 7: Rezultati preizkusov obdelavnosti
Concrete SCCR SCCM10 SCCM15 SCCM20
Slump flow R/cm 66.5 65.4 63.6 63.3
T50 cm/s, slump flow 3.5 3.3 3.1 3.2
(H2/H1)/% (L-Box) 0.85 0.83 0.82 0.80
T40/s (L-Box) 3.4 3.5 3.7 3.6
(H2/H1)/% (L-Box) 8.47 7.55 6.90 4.55
Bleeding, ‰ 1.25 1.18 1.12 1.15
4.3 Workability (slump-flow test)
It can be noted that all of the SCCs comply with the
criterion of the flow spread where specified values lie
between 63.3 cm and 66.5 cm (Figure 4), causing a
lower viscosity. Although no limit is given for the times
of the flow spread, the time needed to reach a 50 cm dia-
meter (T50) is close to the values usually noted, i.e., 3 s.
4.4 Flowability (L-box test)
The L-box test is used to assess the filling and pass-
ing ability of SCC. This is a widely used test suitable for
a laboratory as well as site use. A concrete can be
accepted if the fill ratio (H2/H1) of the L-box is higher
than 0.8 4; the flowing times can be measured in order to
assess the viscosity. The obtained results clearly show
that the concrete present satisfies the ratio of 0.80–0.85.
4.5 Sieve-stability test
For this test, the results in Table 7 show that all the
SCCs have a segregation rate lower than 15 %, indicating
a good stability.4 When determined as 0    5 %, the
resistance to the segregation is maintained to be "too
significant" which is true in the case of SCCM20 where
the paste is too viscous to run out through the sieve.5
4.6 Bleeding test
The results in Table 7 indicate that all the concretes
meet the recommended value. The values obtained vary
between 1.12 ‰ and 1.25 ‰.
4.7 Hardened state
4.7.1 Evolution of the mechanical compressive strength
According to the obtained results and curves repre-
sented in Figure 5, it is clear that the SCCR displays a
good compressive performance at various testing stages
compared to the concretes with the substitute material
(SCCM). It is observed that over short periods the con-
cretes have similar amplitudes; on the other hand, over
long periods, their amplitudes start to move away from
that of the SCC, except for the SCCM10, which displays
a trend close enough to the SCCR observed between the
28th and 90th day.
The evolution of the compressive strength with res-
pect to time shows that during a short-term period the
N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
130 Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135
Figure 4: Diagram of the diameter of the slump flow
Slika 4: Diagram premera pri preizkusu razleza
evolution rate of the strength of the SCCMs is deve-
loping more slowly than that of the SSCR; nevertheless,
it can be noted that the variation between the measured
values is slightly different, especially for the SCCM10.
The measured values indicate that the SCCR reached
(48, 60 and 82) % of its compressive strength after 28 d
for the testing ages of (3, 7 and 14) d, respectively. How-
ever, the SCCM10, SCCM15 and SCCM20 reached bet-
ween 48–51 % after 3 d, 64–65 % after 7 d and 89–92 %
after 14 d, which can only be explained with the densifi-
cation effect. The calcined mud acted as the filler, filling
in the pores and increasing the compactness of the
cementitious matrix.
On the other hand, between the 28th and 90th day, it
is the SCCM10 that evolves distinctly among the
SCCMs. For these testing ages, the effect of the pozzo-
lanic activity on the concretes can be distinguished. The
evolution of this compressive strength can also be clearly
noticed on the histogram presented in Figure 6, which
also shows the effect of the variation of the M/C ratio
(the proportioning of the mud to cement) on the strength
evolution.
However, it is interesting to note that even with the
20 % substitution, the compressive strength remains
within the reasonable limit of 30 MPa recommended by
various construction specifications for building concre-
tes.
4.7.2 Evolution of the mechanical tensile strength
It is known that the factors influencing the evolution
of the compressive strength also influence the evolution
of the tensile strength of concrete. The obtained results
show that SCCMs developed a low tensile strength com-
pared to the reference SCCR.
On Figure 7, we can see that the development of the
tensile strengths follows the trend of the compressive
strength and the evolutions of the strengths in the early
period are identical, confirming the hypothesis that the
mud acts as the filler.
It can also be noticed on Figure 7 that the tensile
strength of the SCCM10 displays a convergence towards
that of the SCCR as of the 28th day with respect to the
two other concretes, indicating a sufficiently attenuated
amplitude.
Figure 8 clearly shows that the reference concrete
exhibits the best performances followed by the SCCM10.
In general, the loss of the resistance to compression or
traction does not exceed the mean value of 25 % com-
pared to the reference concretes irrespective of the age
and the rate of substitution.
N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135 131
Figure 8: Histogram of the evolution of the tensile strengths with
respect to the mud proportion
Slika 8: Histogram razvoja natezne trdnosti glede na dele` blata
Figure 6: Histogram of evolution of the compressive strength with
respect to the mud proportion
Slika 6: Histogram razvoja tla~ne trdnosti glede na dele` blata
Figure 5: Compressive strength of concretes versus time
Slika 5: Tla~na trdnost betona v odvisnosti od ~asa
Figure 7: Evolution of the tensile strength at different ages
Slika 7: Razvoj natezne trdnosti pri razli~ni starosti
4.8 Different deformations
It is known that the factor with the highest influence
on the shrinkage is the quantity of the water used. For
this reason and in order to better determine the problem,
the W/B ratio is kept equal to 0.5. The compositions of
the tested SCCs only differ in the proportion of the cal-
cined mud substituting the cement, so all the differences
in the behavior of the concretes are related only to this
parameter.
4.8.1 Endogenous shrinkage
The endogenous deformations measured from the
first day onwards are displayed on Figure 9. The types
of the kinetics of the endogenous-shrinkage
deformations of the concretes are rather similar; at the
beginning they take similar and almost identical forms
and start to differ with time.
Since this shrinkage is a consequence of the hydra-
tion phase6,7, it gives evidence of its kinetics and the
quantity of formed hydrates. It can be observed that
during the period from the 28th to 90th day, the orders of
the magnitudes of the SCCMs approach that of the
SCCR in comparison with the values displayed at the
early age. At the 28th day, the SCCMs display reduced
values, varying between 5.5 % and 15.5 % compared to
the SCCR; on the other hand, between the 60 and 90th
day, they are reduced by 4.6 % to 12.45 % compared to
the SCCR, which shows that the deformations evolve in
the same spindle, whereas the coming together of the
values is a proof of the pozzolanic activity of the mud,
which begins, a priori, after the 28th day.
The reduction in the endogenous shrinkage in the
presence of the calcined mud can be explained with the
fact that some hydrates (calcium aluminate) obtained via
the pozzolanic reaction between the silica and the port-
landite, are slightly expansive8, compensating for the
dimensional variations due to the shrinkage.9 The shrink-
ages show the same development according to the me-
chanical compressive strength10, the auto-desiccation
known as the principal phenomenon, which governs the
endogenous shrinkage growing under the influence of a
high strength.
4.8.2 Drying shrinkage
This shrinkage develops from the surfaces exposed to
the external environment; it is determined by calculating
the difference between the total shrinkage and the endo-
genous shrinkage and it is presented with respect to time
and mass loss. Figure 10 shows that the curves obtained
for the SCC appear in a spindle, indicating that, due to
drying, the component is not modified by the M/C ratio.
During the first phase, the values of the shrinkage of the
SCCMs decrease compared to the SCCR, by the percen-
tages varying between 21 % and 52 % on the 7th and
28th day. On the other hand, during the second phase, the
evolution starts to slow down until becoming almost
constant up to the 90 day. Thus, it can be noted that the
difference between the measured values decreases com-
pared to the first phase, as it varies between 16 % and
30 %, which proves that the amplitude of the shrinkage
starts to stabilize.
Nevertheless, whatever the age, the SCCR always
shows the highest values, contrary to the SCCMs as their
values decrease with the increase in the rate of the
calcined-mud substitution.
4.8.3 Mass loss
In order to try to improve the drying shrinkage, we
also carried out follow-up tests of the test-tube mass
shrinkage in order to quantify the hydrous exchanges.
The curves from Figure 11 show that the mass losses
are in agreement with the evolution of the drying
shrinkage; the values of the mass loss are very high and
almost identical as of the first day for all the concretes;
all the curves start with a linear part whose slope seems
to decrease slightly with the time. In the intermediate
zone, the curves are strongly nonlinear, which explains
the reduction to low values due to the evaporation. After
60 d, the mass losses start to decrease very slightly; they
N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
132 Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135
Figure 10: Evolution of the drying shrinkage versus time
Slika 10: Razvoj kr~enja zaradi su{enja v odvisnosti od ~asa
Figure 9: Evolution of the endogenous shrinkage versus time
Slika 9: Razvoj notranjega kr~enja v odvisnosti od ~asa
are weaker for the concrete with the mud-cement com-
bination than for the reference concrete. Finally, the
effect of the M/C ratio undoubtedly changes the hyd-
rous-pressure development of the curves with respect to
the degree of saturation.11
Figure 12 also shows that various drying-shrinkage
evolution curves for the mass loss are presented in two
phases; the first schematizes the first water departure
without any consequence for the shrinkage, while the se-
cond one schematizes the evolution of the shrinkage with
the water loss. The explanation is based the type of water
and two families of pores: the water contained in the
large pores leaves the material without causing a shrink-
age, whereas the water contained in the small pores
generates contractions of the material.12
4.8.4 Total shrinkage
It evolves very quickly for all the types of the test
tubes kept in the air because of their sizes that make the
desiccation more favorable. At the early stage, the
shrinkage is almost independent of the composition of
the concrete. The values of the shrinkage are concen-
trated in the same spindle and the effect of the addition
takes place only after the first week, with a slight
superiority for the SCCR. After a long period, the pre-
sence of the mud decreases the final shrinkage with
respect to the proportion of the substitute material.
Figure 13 shows a similar evolution of deformation
for all the SCCs. This is the reason why at the very early
stage we find it difficult to distinguish between the repre-
sentative graphs for each concrete. The order of the
magnitude on the 7th day presents reductions, which
vary from 14 % to 36 % for the SCCMs compared to the
SCCR. After the 7th day, the shrinkage of the reference
SCC evolves much more quickly and it is distinguished
from the others up to the 90th day, having a similar
amplitude.
It is also observed that for this shrinkage the refe-
rence SCC shows the highest values at all stages.
On Figure 14, the total shrinkage is plotted with res-
pect to the lg (t) scale, rather than the drying shrinkage,
N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135 133
Figure 13: Evolution of the total shrinkage versus time
Slika 13: Razvoj skupnega skr~ka v odvisnosti od ~asa
Figure 11: Evolution of the loss of mass versus time
Slika 11: Razvoj izgube mase v odvisnosti od ~asa
Figure 14: Total shrinkage according to the logarithmic scale
Slika 14: Skupen skr~ek prikazan v logaritemski skali
Figure 12: Drying shrinkage versus the loss of mass
Slika 12: Kr~enje pri su{enju v odvisnosti od izgube mase
which is only a low estimate. Three phases appear; dur-
ing the first one, the shrinkage gradually increases, dur-
ing the second one, it evolves linearly with the logarithm
of time and during the third one, the curve of the shrink-
age inflects towards an asymptotic value.
5 CONCLUSIONS
Our study made it possible to confirm the possibility
of using the mud from the Fergoug Dam as a partial sub-
stitute material for cement.
The principal conclusions obtained are as follows:
1. The study of the behavior of the SCC in the fresh
state, with respect to the proportioning of the cal-
cined mud, allows us to make the following obser-
vations:
– The results indicate that the SCCs containing cal-
cined mud are more viscous and less workable
compared to the reference concrete because, beyond
the critical proportioning, the viscosity of the
concrete increases with the increased amount of the
substitute material.
– Only the SCCM20 presented too large a segregation
according to the sieve test.
– In addition to the bleeding test, the values obtained
with the tests of the simple flow, the L-box and the
sieve stability decrease when the proportion of the
calcined mud increases.
– A densification of the microstructure had a favorable
effect by decreasing the bleeding, thus providing us
with an idea on the improvement of the paste/aggre-
gate interface.
2. In the hardened state, the mechanical tests of the
compressive and tensile strengths carried out on the
test tubes made of self-consolidating concrete with
various amounts of the mud, in comparison with the
reference concrete made of cement alone, gave the
following results:
– The evolution of the strength is influenced by the
M/C ratio (the rate of the substitution of cement with
the calcined mud); the results indicate that the
compressive and tensile strengths decrease with the
increase in the calcined-mud content; these values
are very tolerable for the concretes employed for
buildings constructions.
– The best values of the SCCM compressive and
tensile strengths are obtained with the SCCM10, but
the SCCR always exhibits the highest values.
However, the use of a mineral addition involves the
formation of a new CSH, which fills in the pores of
the hardened cement paste and densifies the
structure of the paste, leading to a reduction in the
porosity. These effects lead to an improvement of the
mechanical strength.
3. With the tests of different deformations, carried out
on the same concretes, we followed the evolution of
free deformations in endogenous and drying condi-
tions. The obtained results show that the calcined
mud tends to slightly decrease the variety of defor-
mations and it can be concluded that their kinetics is
associated with the physicochemical mechanism, so:
– The endogenous shrinkage decreases with the in-
crease in the proportioning of the substitute calcined
mud.
– The compactness of the microstructure and the
refinement of the pores lead to a fall in the perme-
ability and prevent the diffusivity of water, conse-
quently, decreasing the drying shrinkage and the
mass loss.
– The total shrinkage follows the same principle, being
influenced by the endogenous shrinkage more than
by the drying shrinkage; it also decreases in accord-
ance with the M/C ratio, exhibiting a trend whereby
the total shrinkage is an intrinsic phenomenon of the
concrete.
– The analysis of the phenomenon of shrinkage in the
presence of calcined mud indicates that this addition
contributes to a decrease in the shrinkage amplitudes
compared to the reference concrete. The 20 % sub-
stitution seems to be the best option with respect to
its contribution to the improvement of the micro-
structure and, finally, to the decrease in the
shrinkage effects.
4. Finally, it is deduced that the calcined mud, due to its
reactivity and fineness, influences the mechanical and
other properties of the concretes:
– The pozzolanic reaction starts to be perceptible over
a long period by improving the compressive and
tensile strengths.
– The 10 % substitution of cement is the optimum
content to give the best mechanical performances,
followed by the 15 % and 20 % substitutions.
– The 20 % substitution is the optimum solution for a
reduction in the shrinkage and mass loss due to an
improvement in the microstructure, making the
matrix more compact and minimizing the diffusivity,
followed by the 15 % and 10 % substitutions.
– In conclusion, it can be deduced that the substitution
of 15 % of cement by the mud is the most interesting
option that proves to be optimal since it is the
average proportion that satisfies the two criteria: the
strength improvement and the shrinkage reduction.
6 REFERENCES
1 B. Remini, Qualification du transport solide dans le bassin versant de
l’oued Isser, Application à l’envasement du barrage de Béni Amrane,
2eme CMEE Alger, 2002
2 A. Semcha, Valorisation des sédiments de dragage: Applications
dans le BTP, cas du barrage de Fergoug, Doctoral Thesis, University
of Reims, Champagne-Ardenne, France, 2006
3 ABHC, Agence du bassin hydrographique Chott-Chergui de l‘ouest,
2007
4 AFGC, Bétons Auto-Plaçants Recommandations provisoires,
Association française de Génie Civil, 2008
N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
134 Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135
5 F. Cussigh, M. Sonebi, G. Schutter, Project testing SCC-segregation
test method, Proceedings of the Third international RILEM
conference on self-compacting concrete, 2003, 311–322,
http://www.rilem.org/gene/main.php?base=500218&id_publi-
cation=38
6 Andra, Référentiel géologique du site de Meuse/Haute-Marne sur le
stockage géologique des déchets radioactifs à haute activité et à vie
longue, Tome 2: matériaux cimentaires, 2005
7 I. Yurtdas, Couplage comportement mécanique et dessiccation des
matériaux à cimentaire: étude expérimentale sur mortiers, Doctoral
Thesis, Université des sciences et technologie de Lille, 2003
8 L. Courard, A. Darimont, M. Sschouterden, F. Ferauche, X. Willem,
R. Degeimbre, Durability of mortars modified with metakaolin,
Cement and Concrete Research, 33 (2003) 9, 1473–1479,
doi:10.1016/S0008-8846(03)00090-5
9 J. J. Brook, M. A. Megat Johar, Effect of metakaolin on creep and
shrinkage of concrete, Cement and Concrete Research Composite, 23
(2001), 495–502, doi:10.1016/S0958-9465(00)00095-0
10 B. Persson, Self-desiccation and its importance in concrete tech-
nology, Materials and Structures, 30 (1997) 5, 293–305, doi:10.1007/
BF02486354
11 P. Turcry, Retrait et fissuration des bétons autoplaçants: influence de
la formulation, Doctoral Thesis, Ecole Centrale de Nantes, France,
2004
12 A. Neville, Properties of Concrete, 4th Edition, John Wiley and Sons,
1996
N. E. BOUHAMOU et al.: INFLUENCE OF DREDGED SEDIMENT ON THE SHRINKAGE BEHAVIOR ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 127–135 135

J. ZACH et al.: STUDY OF THE PROPERTIES AND HYGROTHERMAL BEHAVIOUR ...
137–140
STUDY OF THE PROPERTIES AND HYGROTHERMAL
BEHAVIOUR OF ALTERNATIVE INSULATION MATERIALS
BASED ON NATURAL FIBRES
[TUDIJ LASTNOSTI IN HIGROTERMALNO OBNA[ANJE
ALTERNATIVNIH IZOLACIJSKIH MATERIALOV NA OSNOVI
NARAVNIH VLAKEN
Jiøí Zach, Martina Reif, Jitka Hroudová
Brno University of Technology, Faculty of Civil Engineering, Veveøí 331/95, 602 00 Brno, Czech Republic
zach.j@fce.vutbr.cz, reif.m@fce.vutbr.cz, hroudova.j@fce.vutbr.cz
Prejem rokopisa – received: 2014-08-01; sprejem za objavo – accepted for publication: 2015-03-02
doi:10.17222/mit.2014.169
The paper describes the results of a research focused on the development of natural thermal and acoustic insulation materials.
They are mainly materials based on locally available agricultural waste (mainly in the third-world countries). In particular, this
waste includes stems of rice and other plants. The paper describes the behaviour of these materials under different humidity
conditions and the possibility to influence their properties by varying the ratio between organic and inorganic binders. Using the
results, general conditions for the suitability of the application of these materials in building constructions were defined.
Keywords: natural fibres, moisture content, thermal conductivity, rice stems, cotton stems, thermal insulation
^lanek opisuje rezultate raziskav usmerjenih v razvoj naravnih materialov za toplotno in akusti~no izolacijo. To so prete`no
materiali, ki predstavljajo lokalno dosegljive kmetijske odpadke (ve~inoma v tretjem svetu). To so zlasti stebla ri`a in drugih
rastlin. ^lanek opisuje obna{anje teh materialov pri razli~nih pogojih vla`nosti in mo`nosti za vplivanje na njihove lastnosti pri
razli~nih razmerjih organskih in anorganskih veziv. Na podlagi rezultatov so bili dolo~eni splo{ni pogoji za primernost in
uporabnost teh materialov v gradbeni{tvu.
Klju~ne besede: naravna vlakna, vsebnost vlage, toplotna prevodnost, stebla ri`a, stebla bomba`a, toplotna izolacija
1 INTRODUCTION
Considering the ever-increasing demand for insu-
lation materials, the general need for a sustainable
development and an effort to limit the exploitation of
raw-material sources, fibrous insulation materials based
on organic fibres from agriculture (hemp, flax, waste
textile fibres, rice and cotton-plant stems, sheep wool,
etc.) have been developed at Brno University of Technol-
ogy, Faculty of Civil Engineering, for many years. The
development of these materials was also carried out with
regard to the findings of foreign experts recorded in the
literature in specialized databases (Thomson Reuters,
Scopus). We considered the research of the scientific
teams from Europe and Asia.1–8
These materials represent highly progressive building
materials with a low carbon footprint and a low pri-
mary-energy input. They are locally available and easily
renewable raw materials that can substitute non-renew-
able materials used in the production of insulation
materials (e.g., foam plastic materials). In addition, the
production of these materials has lower energy costs
compared to the production of many modern synthetic
insulators (e.g., mineral wool). The experiments per-
formed during the previous research revealed that these
materials show properties comparable with the synthetic
insulation materials available on the market.9 However,
in terms of the thermal-insulation properties, these natu-
ral fibre-based materials exhibit a different hygrothermal
behaviour due to different structures of the insulations
and also a low thermal conductivity of the natural fibres
compared to the glass or mineral ones.
2 INPUT RAW MATERIALS AND TEST
MIXTURES
For the production of specimens, alkali-activated
blast-furnace slag containing SiO2 (42.24 %), CaO
(44.87 %), Al2O3 (2.73 %) and MgO (5 %) was used
with Fe2O3, MnO, TiO2, P2O5, Na2O, K2O and sodium
water glass. Rice and cotton-plant stems were used as the
organic binders. The specific quantities of the raw mate-
rials are listed in Table 1.
Table 1: Overview of the specimen compositions
Tabela 1: Pregled sestave vzorcev
Mixture
No.
Rice stems
(kg m–3)
Cotton
stems
(kg m–3)
Slag
(kg m–3)
Water glass
(kg m–3)
1 100 – 300 200
2 120 – 250 190
3 – 100 300 200
Materiali in tehnologije / Materials and technology 50 (2016) 1, 137–140 137
UDK 677.1.004.8:691.1:699.8 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 50(1)137(2016)
3 METHODOLOGY
After removing the formwork, the specimens were
stored under laboratory conditions at 23±2 °C and a rela-
tive humidity of 50±5 %. After 28 d, the specimens were
dried and their bulk density was determined according to
EN 1602.10 Subsequently, hygroscopicity was deter-
mined for the specimens stored at +23 °C and different
relative-humidity conditions. The equal-sorption humi-
dity as well as the thermal conductivity depending on the
humidity were determined. The thermal conductivity was
determined at the stable state in accordance with EN
1266711 and ISO 8301.12 Acoustic properties were also
determined for the specimens. Specifically, the dynamic
stiffness of the material was determined in accordance
with ISO 9052-113 and the sound-absorption coefficient
was determined in accordance with EN ISO 1165414 and
ISO 10534-1.15
4 RESULTS AND DISCUSSION
The bulk density was determined for the samples in
the dried state in accordance with EN 160210 and the
specific values are shown in Table 2.
Table 2: Summary of individual bulk densities v of the specimens
Tabela 2: Pregled posameznih gostot v vzorcev
Mixture No. 1 2 3
v (kg m–3) 440 430 430
The average bulk density ranged from 430 kg m–3 to
440 kg m–3 and its values for different fillers in mixtures
1 and 3 were also comparable. Next, the moisture con-
tent under laboratory conditions was determined. The
samples were stored in the laboratory, at the temperature
of 23 °C and the relative humidity of 50 % and were sub-
sequently dried at a temperature of 105 °C. Afterwards,
the natural-moisture content prior to drying was
measured using the gravimetric method. The resulting
values are shown in Figure 1.
The natural-water content of a material depends on
the amount of the pores present in the material, their
size, openness and the degree to which they are inter-
connected. The pore system depends on the compaction
during the production, the amount of inorganic binders
and the choice of organic fillers. It is clear from the
graph that the cotton-stem sample has a higher natural
moisture than the specimen with the same amount of rice
stems. Increasing their amounts also increases the natural
moisture. In addition, hygroscopicity of the specimens
was determined. The specimens were stored in the envi-
ronments with the following parameters:
• in laboratory conditions – a relative humidity of 50 %
and a temperature of 23 °C
• in humid conditions – a relative humidity of 95 %
and a temperature of 23 °C
The amounts of the moisture absorbed by the speci-
mens into their structures are listed in Table 3.
Table 3: Summary of measured moistures wm at varied relative humi-
dity  and temperature of +23 °C
Tabela 3: Povzetek izmerjenih vla`nosti wm pri spreminjanju relativne
vla`nosti  in temperaturi +23 °C
/%
Mixture No. 1 Mixture No. 2 Mixture No. 3
wm/%
0 0 0 0
50 8.10 12.57 12.23
95 33.04 47.00 42.11
A sorption curve was constructed from the obtained
data. The determination of the sorption isotherm was
carried out at 23 °C. The moisture content of the spe-
cimens was measured at the relative-humidity levels of
50 and 95 %. The results of the measurement are shown
in Figure 2.
The amount of the absorbed moisture depends on the
structure of the given material, the temperature and the
relative humidity of the environment, to which the
material is exposed. With its growing value, the moisture
content in the material rises nonlinearly.
In the next stage, the thermal conductivity in
dependence on the specimen moisture was determined in
J. ZACH et al.: STUDY OF THE PROPERTIES AND HYGROTHERMAL BEHAVIOUR ...
138 Materiali in tehnologije / Materials and technology 50 (2016) 1, 137–140
Figure 2: Moisture content wm for the samples depending on relative
humidity  at the temperature of 23 °C
Slika 2: Vsebnost vlage wm v vzorcih, v odvisnosti od relativne
vla`nosti  pri temperaturi 23 °C
Figure 1: Natural-moisture content (wm/%)
Slika 1: Vsebnost naravne vlage (wm/%)
accordance with EN 1266711 and ISO 8301.12 The
measurement was carried out at the natural weight
moisture content state, at dry and at moist state. The
results are listed in Table 4 and Figure 3.
Table 4: Thermal conductivity  in dependence on relative humidity 
Tabela 4: Toplotna prevodnost  v odvisnosti od relativne vla`nosti 
Mixture
No. /% 0 50 95
1
wm/% 0 8.10 33.04
 (W m–1 K–1) 0.087 0.106 0.160
2
wm/% 0 12.57 47.00
 (W m–1 K–1) 0.085 0.106 0.177
3
wm/% 0 12.23 42.11
 (W m–1 K–1) 0.099 0.122 0.161
As Figure 3 indicates, the thermal conductivity
depends on the relative humidity; as its value increases,
the value of the thermal conductivity increases as well.
This is most apparent for the specimens containing the
organic cotton-plant filler. Among the specimens
containing rice stems, the thermal conductivity in the
damp condition significantly increased for specimen 2
with a lower content of binder and a higher porosity (it is
possible to see it in the area where the relative humidity
was above 50 %). Next, the dynamic stiffness of the
material was tested using the resonance method in
accordance with ISO 9052-1.13 The measured values are
listed in Table 5, including the category pad and
acoustic-material group.16
Table 5: Summary of dynamic stiffness s’
Tabela 5: Povzetek dinami~ne togosti s’
Mixture
No.
s’
(MPa m–1)
Category
pad*
Acoustic-insulation
properties*
1 10.91 Category I Dynamically soft
2 13.15 Category I Dynamically soft
3 12.47 Category I Dynamically soft
* according to CSN 73053216
It can be assumed from the obtained values that all
the specimens have good acoustic-insulation properties
in terms of the impact sound and can be classified as the
group of dynamically soft acoustic-insulation materials
for which s’  30 MPa m–1 applies. Considering the
dynamic-stiffness values, the materials can be used as
impact-sound insulators. Finally, the sound-absorption
coefficient was determined. The measurement was
carried out using an acoustic resonator in accordance
with ISO 10534-115 and from the obtained values the
weighted sound-absorption coefficient w was calculated
in line with EN ISO 11654.14 The resulting values are in
Table 6, listing different sound-absorption categories.
Table 6: Weighted sound-absorption coefficient w for the frequency
of 500 Hz
Tabela 6: Izmerjeni koeficient vpijanja zvoka w pri frekvenci 500 Hz
Mixture No. w (–) Class of soundabsorption
1 0.75 C
2 0.80 B
3 0.90 A
It is clear from Table 6 that the highest value of the
weighted sound absorption was found for specimen 3,
with the cotton-plant stems. Among the samples contain-
ing the rice-plant stems, specimen 2, with a higher
amount of organic fillers and a reduced binder amount,
has higher values.
5 CONCLUSION
The research into the possibility of using agricultural
waste from the rice and cotton production brought
interesting findings for the production of thermal and
acoustic-insulation materials. The structure of the speci-
mens was very porous as the coarse fractions of cotton
and rice stems form an open porous system. The
developed materials reached good values of the thermal
conductivity, which can be influenced by increasing or
reducing the amounts of the fillers or the binders. The
amount of water glass must be chosen according to the
need for activating the slag used. The thermal-conduc-
tivity value is also influenced by the sensitivity of the
material to the relative humidity. In terms of the acoustic
properties, the materials acted as good sound absorbers.
Especially the porous structure created by cotton stems
exhibited high values of the sound-absorption coeffi-
cient. To find the optimum formula for the lowest possi-
ble values of the thermal conductivity and good acoustic
properties while maintaining a sufficient material
consistency is the task for further research. The speci-
mens also exhibited very good properties from the point
of view of dynamic stiffness. All the samples can be
classified as dynamic soft insulation materials.
With respect to the hygrothermal transport, it was
found that these materials are sensitive to humidity;
under normal conditions, they exhibit sorptive moisture
comparable with wood. With a higher moisture content,
the thermal insulating properties of these materials
J. ZACH et al.: STUDY OF THE PROPERTIES AND HYGROTHERMAL BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 137–140 139
Figure 3: Dependence of thermal conductivity  on relative humidity 
Slika 3: Odvisnost toplotne prevodnosti  od relativne vla`nosti 
degrade; however, a more significant degradation occurs
in the environments with a higher humidity. It can, there-
fore, be stated that as long as these insulators (e.g., in the
form of ETICS) are not exposed to a very high humidity
or direct weather actions, they are able to function within
the structure and can be successfully applied in many
developing countries where they can provide protection
not only from the negative effects of cold but also from
the high temperatures during the summer (Central Asia,
for instance). The research confirmed that binding by
means of a slag-based alkali-activated binder could result
in a very good ratio between the thermal insulation and
mechanical properties. On the contrary, for instance,
alkali-bound insulators obtained simply by being pressed
or bound by means of bicomponent fibres, can be used in
the structures with a lower mechanical load, putting
relatively low demands on the production technology
and for this reason, they can find use also in industrially
less developed countries.2,9
Acknowledgements
This paper was elaborated with the financial support
of the projects GA 13-21791S and project No. LO1408
"AdMaS UP – Advanced Materials, Structures and Tech-
nologies", supported by Ministry of Education, Youth
and Sports under the "National Sustainability Pro-
gramme I".
6 REFERENCES
1 K. W. Corscadden, J. N. Biggs, D. K. Stiles, Sheep’s wool insulation:
A sustainable alternative use for a renewable resource?, Resources,
Conservation and Recycling, 86 (2014), 9–15, doi:10.1016/
j.resconrec.2014.01.004
2 K. Wei, C. Lv, M. Chen, X. Zhou, Z. Dai, D. Shen, Development and
performance evaluation of a new thermal insulation material from
rice straw using high frequency hot-pressing, Energy and Buildings,
87 (2015) 1, 116–122, doi:10.1016/j.enbuild.2014.11.026
3 D. P. L. Murphy, H. Behring, Arable crop materials for insulation in
buildings, Biomass for Energy and Industry, 10th European Confe-
rence and Technology Exhibition on Biomass for Energy and Indu-
stry, Würzburg, 1998, 176–179
4 W. D. Brouwer, Natural fibre composites: Where can flax compete
with glass?, Sample Journal, 36 (2000) 6, 18–23
5 S. A. Ibraheem, A. Ali, A. Khalina, Development of Green Insulation
Boards from Kenaf Fibres, Part 2: Characterizations of Thermal and
Water Absorption, Key Engineering Materials, 462–463 (2011),
1331–1336, doi:10.4028/www.scientific.net/KEM.462-463. 1331
6 S. A. Ibraheem, A. Ali, A. Khalina, Development of Green Insulation
Boards from Kenaf Fibres, Part 1: Development and Characteriza-
tions of Mechanical Properties, Key Engineering Materials, 462–463
(2011), 1343–1348, doi:10.4028/www.scientific.net/KEM.462-463.
1343
7 S. Mukhopadhyay, D. Annamalai, R. Srikanta, Coir Fiber for Heat
Insulation, Journal of Natural Fibers, 8 (2011) 1, 48–58, doi:10.1080/
15440478.2010.551001
8 S. A. Ibraheem, A. Ali, A. Khalina, Development of Green Insulation
Boards from Kenaf Fibres and Polyurethane, Polymer-Plastics
Technology and Engineering, 50 (2011) 6, 613–621, doi:10.1080/
03602559.2010.551379
9 A. Korjenic, V. Petránek, J. Zach, J. Hroudová, Development and
performance evaluation of natural thermal-insulation materials
composed of renewable resources, Energy and Buildings, 43 (2011)
9, 2518–2523, doi:10.1016/j.enbuild.2011.06.012
10 EN 1602 Thermal insulating products for building applications –
Determination of the apparent density, 2013
11 EN 12667 Thermal performance of building materials and products –
Determination of thermal resistance by means of guarded hot plate
and heat flow meter methods – Products of high and medium thermal
resistance, 2001
12 ISO 8301 Thermal insulation – Determination of steady-state ther-
mal resistance and related properties – Heat flow meter apparatus,
2010
13 ISO 9052-1 Acoustics – Determination of dynamic stiffness – Part 1:
Materials used under floating floors in dwellings, 1989, reviewed in
2011
14 EN ISO 11654 Acoustics – Sound absorbers for use in buildings –
Rating of sound absorption, 1997
15 ISO 10534-1 Acoustics – Determination of sound absorption coeffi-
cient and impedance in impedance tubes – Part 1: Method using
standing wave ratio, 1996, reviewed in 2011
16 ^SN 730532 Acoustics – Protection against noise in buildings and
evaluation of acoustic properties of building elements – Require-
ments, 2010
J. ZACH et al.: STUDY OF THE PROPERTIES AND HYGROTHERMAL BEHAVIOUR ...
140 Materiali in tehnologije / Materials and technology 50 (2016) 1, 137–140
U. ÖZMEN, B. OKUTAN BABA: PREDICTION OF THE ELASTIC MODULUS OF CHICKEN-FEATHER-REINFORCED PLA ...
141–146
PREDICTION OF THE ELASTIC MODULI OF
CHICKEN-FEATHER-REINFORCED PLA AND A COMPARISON
WITH EXPERIMENTAL RESULTS
NAPOVEDOVANJE MODULOV ELASTI^NOSTI PLA, OJA^ANEGA
S PI[^AN^JIM PERJEM IN PRIMERJAVA Z
EKSPERIMENTALNIMI REZULTATI
Uður Özmen, Buket Okutan Baba
Celal Bayar University, Engineering Faculty, Mechanical Engineering Department, 45140 Muradiye/Manisa, Turkey
u.ozmen@hotmail.com
Prejem rokopisa – received: 2014-08-07; sprejem za objavo – accepted for publication: 2015-03-04
doi:10.17222/mit.2014.182
The purpose of this study is to obtain the elastic moduli, the key material property, of random discontinuous fiber composites
with experiments and micromechanical models and to compare them. The proposed study makes it possible to assess the elastic
moduli of chicken-feather fiber (CFF)/PLA green composites with different CFF mass fractions and to determine the feasibility
of the micromechanical models for the CFF/PLA composites. For this purpose, initially, CFF/PLA composites including 2, 5 or
10 % chicken-feather mass fractions were extruded and standard tensile specimens for ISO 527 were formed with the
injection-molding method. Tensile tests were carried out in accordance with the standards and the elastic moduli were calculated
using the stress-strain curve. Then, using six different micromechanical models, the elastic moduli of the CFF/PLA composites
with different mass fractions were calculated and compared with the experimental results. The results of the experiments and the
models indicated that the presence of chicken feather increased the elastic moduli of all the composites in comparison with the
pure PLA. According to the experimental data, the maximum increase in the elastic moduli of the composites with the presence
of CFF was found to be 5.4 %. The maximum error in the prediction is about 16.8 % for the composite with a chicken-feather
rate of 10 % when Manera’s model is used. Among the micromechanical models, the ones that gave more converging results for
the prediction of the elastic moduli of the CFF/PLA composites are Pan’s 2-D, IROM (the inverse rule of mixtures),
Nielsen-Chen and Halpin-Tsai models. A comparison of the results of these six models shows that the maximum deviation (the
percentage error in prediction) is the smallest (1.4 %) for the Nielsen-Chen model. Therefore, the Nielsen-Chen model is the
most appropriate model for the prediction of the elastic moduli of the CFF/PLA composites.
Keywords: chicken feather, PLA, green composite, micromechanical models
Namen te {tudije je dobiti module elasti~nosti kot glavne lastnosti materiala, kompozitov z naklju~nimi vlakni, z eksperimenti in
z mikromehanskimi modeli ter njihova primerjava. Predlagana {tudija bo omogo~ila dolo~itev modula elasti~nosti kompozita iz
vlaken pi{~an~jega perja (CFF)/PLA, z razli~nim masnim dele`em CFF in ugotoviti izvedljivost mikromehanskih modelov za
CFF/PLA kompozite. V ta namen so bili najprej kompoziti CFF/PLA z 2 %, 5 % in 10 % pi{~an~jih peres, ekstrudirani in
izdelani so bili standardni natezni preizku{anci po ISO 527, z metodo brizganja v formo. Natezni preizkusi so bili izvr{eni v
skladu s standardom in modul elasti~nosti je bil izra~unan iz krivulje napetost-raztezek. Nato so bili izra~unani moduli
elasti~nosti kompozita CFF/PLA, z uporabo {estih razli~nih mikromehanskih modelov in primerjani z rezultati preizkusov.
Rezultati preizkusov in modelov so pokazali, da prisotnost pi{~an~jih peres pove~a modul elasti~nosti vseh kompozitov, v
primerjavi s ~istim PLA. Eksperimentalni podatki so pokazali, da je najve~je pove~anje modula elasti~nosti kompozita s CFF za
5,4 %. Pri uporabi Manera modela je bila najve~ja napovedana napaka okrog 16,8 % pri kompozitu z 10 % dele`em pi{~an~jega
perja. Med mikromehanskimi modeli, ki imajo najve~ji raztros rezultatov pri napovedovanju modula elasti~nosti CFF/PLA
kompozitov so Pan’s 2-D, IROM (Inverse Rule of Mixtures), Nielsen-Chen in Halpin-Tsai model. Primerjava rezultatov teh
{estih modelov ka`e, da je najve~je odstopanje (odstotek pri napovedi) najmanj{e (1,4 %) pri Nielsen-Chen modelu. Torej je
Nielsen-Chen model najbolj primeren za napovedovanje modula elasti~nosti CFF/PLA kompozitov.
Klju~ne besede: pi{~an~je perje, PLA, zelen kompozit, mikromehanski modeli
1 INTRODUCTION
Since the straw-reinforced cement in the antique
ages, random discontinuous short-fiber-reinforced com-
posites have always existed.1 A lot of experimental
studies have been done on this subject. Many are about
natural-fiber-reinforced ones because of the structure of
natural fibers that cannot be converted into long fibers.
Waste-based,2 synthetic-based,3 plant-based4–7 and ani-
mal-based8–12 fibers have been studied by many
researchers. Experimental studies are substantially
important to get the highest efficiency from the random
discontinuous short-fiber-reinforced composites. Experi-
mental studies are still hard to perform, as
time-consuming and costly tests need to be done to
characterize the composite materials.13 Therefore,
researchers developed certain micromechanical models
for calculating the properties of the composites to avoid
a great number of tests. Modelling of the random
discontinuous short-fiber-reinforced composites is quite
hard compared to the long and oriented ones. For the
situations where the well-known method named the rule
of mixtures for predicting the elastic moduli of com-
posite materials could not provide proper predictions,
Materiali in tehnologije / Materials and technology 50 (2016) 1, 141–146 141
UDK 66.017:620.168 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 50(1)141(2016)
Christensen and Waals,14 Pan,15 Manera,16 Nielsen-
Chen17 and Halpin-Tsai18,19 proposed new models based
on the rule of mixtures. While for some models the range
of validity is limited to specific mass fractions, other
models are different. Although each of these models was
developed for all the random discontinuous short-fiber-
reinforced composites, they may give unexpected results
for some of the composites.
The purpose of this study is to predict the elastic mo-
duli of the composites with different chicken-feather
mass content via the models developed for random dis-
continuous short-fiber-reinforced composites. The volu-
me fractions, with which these models express their sen-
sitive results, were investigated and the volume-fraction
range for these composite materials was determined.
2 EXPERIMENTAL WORK
Two materials were used for the composites, poly
lactic acid (PLA) as the matrix and chicken-feather
fibers (CFFs) as the reinforcement. The CFFs were
procured from a local company in Manisa/Turkey. The
CFFs (or barbs) were separated from the rachis (Figure
1a). They were washed with hot water for sterilization
and kept in water for 24 h. Then, the CFFs were kept in
an oven at 60 °C for 6 h in order to dehumidify them.
The lengths of the CFFs varied from 10 mm to 30 mm.
Their diameters were measured to be approximately
20–40 μm using a light microscope (Figure 1b). The
PLA was purchased from Resinex BMY AS in Istan-
bul/Turkey. The type of the PLA was NatureWorks
3052D with a density of 1.24 g/cm3.
The CFF/PLA composites with CFF mass fractions
of (2, 5 and 10) % were manufactured by extrusion. Ini-
tially, the PLA granules and the CFFs were mixed with a
mechanical mixer. The mixtures were extruded using a
U. ÖZMEN, B. OKUTAN BABA: PREDICTION OF THE ELASTIC MODULUS OF CHICKEN-FEATHER-REINFORCED PLA ...
142 Materiali in tehnologije / Materials and technology 50 (2016) 1, 141–146
Figure 1: a) Parts of a chicken feather, b) light-microscope image of
the barb
Slika 1: a) Deli pi{~an~jega peresa, b) svetlobni posnetek str`ena
peresa in perja
Table 1: Tensile-test specimens of composites with different chicken-feather mass fractions
Tabela 1: Natezni preizku{anci kompozitov z razli~nim masnim dele`em pi{~an~jega perja
Pure PLA
Tensile specimens containing 2 % of chicken feather
Tensile specimens containing 5 % of chicken feather
Tensile specimens containing 10 % of chicken feather
Figure 2: Specimen during the tensile test
Slika 2: Vzorec med nateznim preizkusom
ThermoFisher Scientific EuroLab 16 XL twin-screw
extruder. The barrel temperature-zone profile and the
screw speed were 165/175/185/195/205 °C and 150
min–1, respectively. Following the extrusion process, the
composites were chopped, with a pelletizer, into pieces
with a length of 0.3 cm. Then, the granules were manu-
factured as the tensile specimens in line with the ISO
527 standards, with a PERMAK injection-molding
machine. The barrel temperatures and pressure were
154/160/154 °C and 147 bar, respectively. The tensile
specimens are shown in Table 1.
The length of the tensile specimens was 88 mm, the
width of the narrow and wide sections was 5 and 10 mm,
respectively. The thickness of the specimens was 4 mm
and the gage length was 40 mm. The tensile test was
conducted to determine the elastic moduli of the
composites. Each specimen was tested at a speed of
1 mm/min, using a 100 kN Shimadzu Autograph testing
machine (Figure 2). At least five identical specimens
were tested per specimen type.
3 MICROMECHANICAL MODELS
The micromechanical models that can predict the
elastic moduli of random discontinuous fiber composites
are presented below.
The fiber volume fraction (Vf) used in the equations
is calculated as follows:13
V
M
M M Mf
f
f
f
f
c f
m
=
+
−

 
(1)
where Mf and f indicate the mass and the density of the
fiber material. Mc indicates the mass of the composite
material and m indicates the density of the matrix.
3.1 Christensen-Waals model
Christensen and Waals14 investigated a composite-
material system whose random fiber orientation has three
dimensional directions. They considered both the fiber
orientation and the fiber/matrix interaction. They cal-
culated the elastic moduli of the composite materials
with low fiber fractions for the plane stress using the
formula below:
E
c
E c E= + +
3
1f m( ) c < 0.2 (2)
where E indicates the elastic modulus of the composite
material, Ef indicates the elastic modulus of the fiber
material, Em indicates the elastic modulus of the matrix
material and c indicates the fiber volume fraction.
3.2 Pan’s model
Pan15 put forward a new approach for the prediction
of the elastic moduli of the randomly oriented fiber com-
posites. First, he used the rule of mixtures for the parts
where the fibers were not unidirectional; he used the
relationship between the fiber volume fraction and the
fiber-area ratio. He obtained the equations below for 2-
and 3-dimensional cases:
E E V E V2
1
1
1D
f f m f= + −
⎛
⎝
⎜
⎞
⎠
⎟π π
(3)
E E V E V3
1
2
1
1D
f f m f= + −
⎛
⎝
⎜
⎞
⎠
⎟π 2π
(4)
where, E indicates the elastic modulus of the composite
material, Ef indicates the elastic modulus of the fiber
material, Em indicates the elastic modulus of the matrix
material and Vf indicates the fiber volume fraction.
3.3 Inverse rule of mixtures (IROM)
The rule-of-mixtures model is frequently used to pre-
dict the elastic moduli of the composite materials with
continuous and unidirectional fibers. Since the inverse
rule of mixtures indicates that the loading is perpendi-
cular to the fibers, it is used to predict the elastic moduli
of the random discontinuous short-fiber-reinforced
composites in this study.13 The inverse rule of mixtures is
expressed as:
E
V
E
V
E
= +
−⎛
⎝
⎜⎜
⎞
⎠
⎟⎟
−
f
f
f
m
1
1
(5)
where E indicates the elastic modulus of the composite
material, Ef indicates the elastic modulus of the fiber
material, Em indicates the elastic modulus of the matrix
material and Vf indicates the fiber volume fraction.
3.4 Manera’s model
Manera16 developed a new equation by making some
assumptions on Puck’s micromechanical model and
simplifying it. These assumptions include a high fiber
orientation ratio, two-dimensional random-fiber range,
and he also considered the randomly oriented discon-
tinuous fibers as the laminate of an unlimited number of
layers. Manera’s equation is as follows:
E V E E E= +⎛
⎝
⎜
⎞
⎠
⎟ +f f m m
16
45
2
8
9
(6)
where E indicates the elastic modulus of the composite
material, Ef indicates the elastic modulus of the fiber
material, Em indicates the elastic modulus of the matrix
material and Vf indicates the fiber volume fraction.
3.5 Nielsen-Chen model
Nielsen and Chen built a new model for the predic-
tion of the elastic moduli of the composite materials
using the rule of mixtures and the inverse rule of
mixtures:17
U. ÖZMEN, B. OKUTAN BABA: PREDICTION OF THE ELASTIC MODULUS OF CHICKEN-FEATHER-REINFORCED PLA ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 141–146 143
E E E= +∏ ⊥
3
8
5
8
(7)
where;
E E V E V∏ = + −f f m f( )1 (8)
E
E E
E V V E⊥
=
− +
f m
f f f m( )1
(9)
and E indicates the elastic modulus of the composite
material, Ef indicates the elastic modulus of the fiber
material, Em indicates the elastic modulus of the matrix
material and Vf indicates the fiber volume fraction.
3.6 Halpin-Tsai model
The Halpin-Tsai model is quite complicated in
comparison with the other micromechanical models.18,19
Its consideration of the aspect ratio is the most important
advantage of this model. Therefore, the Halpin-Tsai
model converges better to the experimental results. The
Halpin-Tsai model is as follows:
E E
v
v
=
+
−m
f
f
1
1


(10)
where  and  are:


=
−
E
E
E
E
f
m
f
m
1
(11)
 =
2L
D
(12)
and E indicates the elastic modulus of the composite
material, Ef indicates the elastic modulus of the fiber
material, Em indicates the elastic modulus of the matrix
material and Vf indicates the fiber volume fraction. L
and D indicate the fiber length and diameter, respec-
tively.
4 RESULTS AND DISCUSSION
The elastic moduli of PLA and CFF/PLA composites
were obtained with tensile testing and the results are
shown in Table 2. The elastic modulus of PLA was
experimentally measured as 3004 MPa. The chicken-
feather addition increased the elastic modulus by
approximately 2.6 % for the mass fraction of 2 %, while
increasing the elastic modulus by 2.4 % for the mass
U. ÖZMEN, B. OKUTAN BABA: PREDICTION OF THE ELASTIC MODULUS OF CHICKEN-FEATHER-REINFORCED PLA ...
144 Materiali in tehnologije / Materials and technology 50 (2016) 1, 141–146
Figure 3: Comparison of the elastic moduli of the composite materials
obtained with different micromechanical models
Slika 3: Primerjava modulov elasti~nosti kompozitnih materialov,
dobljenih iz razli~nih mikromehanskih modelov
Table 3: Elastic moduli of composite materials obtained with micro-
mechanical models
Tabela 3: Moduli elasti~nosti kompozitnih materialov iz mikro-
mehanskih modelov
C
hi
ck
en
-f
ea
th
er
m
as
s
fr
ac
ti
on
s
(w
/%
)
C
hr
is
te
ns
en
-W
aa
ls
M
od
el
IR
O
M
Pa
n’
s
M
od
el
(2
-D
)
Pa
n’
s
M
od
el
(3
-D
)
M
an
er
a’
s
M
od
el
N
ie
ls
en
-C
he
n
M
od
el
H
al
pi
n-
T
sa
i
M
od
el
2 3130(1.7)
3034
(1.6)
3019
(2.1)
3317
(7.6)
2882
(6.3)
3040
(1.4)
3034
(1.6)
5 3315(7.8)
3079
(0.1)
3040
(1.2)
3319
(7.9)
3194
(3.8)
3092
(0.5)
3079
(0.1)
10 3614(14.1)
3153
(0.4)
3074
(2.9)
3322
(4.9)
3697
(16.8)
3179
(0.4)
3153
(0.4)
*The values given in the parentheses are percentage differences
between the elastic moduli obtained with the micromechanical models
and experiments.
Table 2: Experimental data for the elastic moduli of CFF/PLA com-
posites
Tabela 2: Eksperimentalni podatki za module elasti~nosti CFF/PLA
kompozita
Chicken-feather
mass fractions,
(w/%)
Chicken-feather
volume fractions,
(Vf /%)
Elastic modulus
(E/MPa)
0 0 3004 (255)
2 2.76 3083 (83)
5 6.83 3076 (309)
10 13.4 3166 (132)
*The values given in the parentheses are standard deviations.
fraction of 5 %. Similarly, its increase for the composite
materials with a chicken-feather mass fraction of 10 %
was 5.4 %. The increasing CFF content raised the elastic
modulus as expected. This increase in the elastic modu-
lus validated the predictions made with the micro-
mechanical models.
A comparison of the elastic moduli obtained with the
models and tests for the composite materials including
the chicken-feather mass fractions of (2, 5 and 10) % is
shown in Table 3. To show the differences between the
results of the models and the tests, the elastic moduli
from Table 3 are plotted as shown in Figure 3. Accord-
ing to the literature, the Christensen-Waals model gives
accurate results for some composite materials containing
a mass fraction of up to 20 %. In this study, the
CFF/PLA composites having a mass fraction of 2 %
show a deviation of 1.7 %, which is a close to the experi-
mental results, as mentioned above. The materials
including mass fractions of 5 and 10 % show deviations
of 7.8 % and 14.1 %, respectively, which means that the
results diverge from the experimental data when the
mass fraction of the material increases. Consequently,
the Christensen-Waals model does not predict the elastic
modulus very well for a fiber mass fraction of more than
2 %.
IROM shows a deviation of 1.6 % from the predic-
tion of the elastic modulus of the composites including a
2 % chicken-feather content when compared with the
experimental results. IROM produces accurate values
converging by more than 99 % to the prediction of the
elastic moduli of the composites having mass fractions
of 5 and 10 %. Pan’s 2-D model shows the results close
to the experimental elastic-modulus values of the com-
posites including (2, 5 and 10) % of CFF. Their devi-
ations are (2.1, 1.2 and 2.9) %, respectively. On the other
hand, the elastic modulus found with Pan’s 3-D model is
higher than the test data even at low fiber mass fractions.
Manera’s model fails to give good predictions of the
elastic modulus for high fiber mass fractions. Therefore,
it is not possible to use Manera’s model for the predic-
tion of the elastic moduli of the composites including
high fiber contents. However, the model gives reasonable
predictions for the fiber mass fractions of about 5 %.
Manera’s model exhibits a deviation of 3.8 % for the
composite including a mass fraction of 5 %.
Another elastic-modulus-prediction model is the
Nielsen-Chen model. The predictions of this model are
close to the test data. It exhibits deviations of (1.4, 0.5
and 0.4) % for the elastic moduli of the composites
including (2, 5 and 10) % of the CFF content, respec-
tively, when compared with the experimental results.
Hence, the Nielsen-Chen model is the best model
producing the closest and most reliable results among the
models investigated in this study. As seen in Figure 2,
the Halpin-Tsai model exhibits the results similar to
IROM and its deviations are very close to the ones found
with IROM. The Halpin-Tsai model is also one of the
most appropriate models to predict the elastic moduli of
the CFF/PLA composites.
5 CONCLUSION
In this study, the elastic moduli of the composites
including mass fractions of (2, 5 and 10) % of chicken-
feather fibers were predicted using six different micro-
mechanical models and the results were compared with
the experimental data. The results are given below:
Although the Christensen-Waals model gives good
predictions for low fiber mass fractions, it does not fit
well the test data for high fiber mass fractions. Pan’s 3-D
model and Manera’s model do not produce reliable
results when used to predict the elastic moduli of the
CFF/PLA composites.
The comparison with the experimental results shows
that Pan’s 2-D model, the inverse rule of mixtures and
the Halpin-Tsai model can be used to predict the elastic
moduli of the CFF/PLA composites.
From the comparison, it can be seen that the Niel-
sen-Chen model gives the best predictions of the elastic
modulus of the CFF/PLA composites. Consequently, the
most appropriate model to predict the elastic moduli of
the CFF/PLA composites is the Nielsen-Chen model.
Acknowledgements
This project was supported by the Celal Bayar
University Research Funds 2013/35.
6 REFERENCES
1 F. P. La Mantia, M. Morreale, Green composites: A brief review,
Composites: Part A, 42 (2011), 579–588, doi:10.1016/j.compositesa.
2011.01.017
2 M. S. Huda, L. T. Drzal, A. K. Mohanty, M. Misra, Chopped glass
and recycled newspaper as reinforcement fibers in injection molded
poly(lactic acid) (PLA) composites: A comparative study, Compo-
sites Science and Technology, 66 (2006), 1813–1824, doi:10.1016/
j.compscitech.2005.10.015
3 K. Oksman, M. Skrifvars, J. F. Selin, Natural fibres as reinforcement
in polylactic acid (PLA) composites, Composites Science and
Technology, 63 (2003), 1317–1324, doi:10.1016/S0266-3538(03)
00103-9
4 N. Graupner, A. S. Herrmann, J. Müssig, Natural and man-made
cellulose fibre-reinforced poly(lactic acid) (PLA) composites: An
overview about mechanical characteristics and application areas,
Composites: Part A, 40 (2009), 810–821, doi:10.1016/j.compositesa.
2009.04.003
5 A. K. Bledzki, A. Jaszkiewicz, D. Scherzer, Mechanical properties of
PLA composites with man-made cellulose and abaca fibres, Com-
posites: Part A, 40 (2009), 404–412, doi:10.1016/j.compositesa.
2009.01.002
6 S. Ochi, Mechanical properties of kenaf fibers and kenaf/PLA
composites, Mechanics of Materials, 40 (2008), 446–452,
doi:10.1016/j.mechmat.2007.10.006
7 B. Bax, J. Müssig, Impact and tensile properties of PLA/Cordenka
and PLA/flax composites, Composites Science and Technology, 68
(2008), 1601–1607, doi:10.1016/j.compscitech.2008.01.004
U. ÖZMEN, B. OKUTAN BABA: PREDICTION OF THE ELASTIC MODULUS OF CHICKEN-FEATHER-REINFORCED PLA ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 141–146 145
8 M. P. Ho, K. T. Lau, H. Wang, D. Bhattacharyya, Characteristics of a
silk fibre reinforced biodegradable plastic, Composites: Part B, 42
(2011), 117–122, doi:10.1016/j.compositesb.2010.10.007
9 H. Y. Cheung, K. T. Lau, X. M. Tao, D. Hui, A potential material for
tissue engineering Silkworm silk/PLA biocomposite, Composites:
Part B, 39 (2008), 1026–1033, doi:10.1016/j.compositesb.2007.11.
009
10 H. Y. Cheung, K. T. Lau, Y. F. Pow, Y. Q. Zhao, D. Hui, Bio-
degradation of a silkworm silk/PLA composite, Composites: Part B,
41 (2010), 223–228, doi:10.1016/j.compositesb.2009.09.004
11 S. Huda, Y. Yang, Composites from ground chicken quill and poly-
propylene, Composites Science and Technology, 68 (2008),
790–798, doi:10.1016/j.compscitech.2007.08.015
12 M. Zhan, R. P. Wool, J. Q. Xiao, Electrical properties of chicken
feather fiber reinforced epoxy composites, Composites: Part A, 42
(2011), 229–233, doi:10.1016/j.compositesa.2010.11.007
13 A. K. Kaw, Mechanics of Composite Materials, 2nd ed., Taylor &
Francis Group, New York 2006, 457
14 R. M. Christensen, F. M. Waals, Effective Stiffness of Randomly
Oriented Fiber Composites, Journal of Composite Materials, 6
(1972), 518–532, doi:10.1177/002199837200600307
15 N. Pan, The elastic constants of randomly oriented fiber composite:
A new approach to prediction, Science and Engineering of Composite
Materials, 5 (1996) 2, 63–72, doi:10.1515/SECM.1996.5.2.63
16 M. Manera, Elastic properties of randomly oriented short fiber-glass
composites, Journal of Composite Materials, 11 (1977), 235–247,
doi:10.1177/002199837701100208
17 E. Vannan, P. Vizhia, Prediction of the Elastic Properties of Short
Basalt Fiber Reinforced Al Alloy Metal Matrix Composites, Journal
of Minerals and Materials Characterization and Engineering, 2
(2014), 61–69, doi:10.4236/jmmce.2014.21010
18 J. E. Ashton, J. C. Halpin, P. H. Petit, Primer on Composite Mate-
rials: Analysis, Technomic, Stamford, Conn. 1969
19 J. C. Halpin, Stiffness and Expansion Estimates for Oriented Short
Fiber Composites, Journal of Composite Materials, 3 (1969),
732–734, doi:10.1177/002199836900300419
U. ÖZMEN, B. OKUTAN BABA: PREDICTION OF THE ELASTIC MODULUS OF CHICKEN-FEATHER-REINFORCED PLA ...
146 Materiali in tehnologije / Materials and technology 50 (2016) 1, 141–146
A. DUFKA, T. MELICHAR: COMPOSITES BASED ON INORGANIC MATRICES FOR EXTREME EXPOSURE CONDITIONS
147–151
COMPOSITES BASED ON INORGANIC MATRICES FOR
EXTREME EXPOSURE CONDITIONS
KOMPOZITI Z ANORGANSKO OSNOVO ZA IZPOSTAVITEV
EKSTREMNIM RAZMERAM
Amos Dufka, Tomá{ Melichar
Brno University of Technology, Faculty of Civil Engineering, Institute of Building Materials and Components, Veveøí 95, 602 00 Brno, Czech
Republic
dufka.a@fce.vutbr.cz, melichar.t@fce.vutbr.cz
Prejem rokopisa – received: 2014-08-18; sprejem za objavo – accepted for publication: 2015-01-06
doi:10.17222/mit.2014.205
The vast majority of reinforced-concrete structures are exposed to aggressive substances from the environment during their
exploitation. The consequence is the degradation of the materials and reduction in the lifetime of the structures. Therefore, it is
natural to make an effort to develop and apply the materials that are maximally resistant to such conditions. Materials based on
alkali-activated matrices show a high resistance against chemically aggressive environments. The article is focused on the use of
the materials based on alkali-activated substances as repair materials for the structures exposed to aggressive groundwater.
Keywords: alkali-activated substances, aggressive surrounding, durability of structures
Velika ve~ina betonskih konstrukcij je med uporabo izpostavljena agresivnim snovem iz okolja. Posledica je razpadanje mate-
riala in skraj{anje trajnostne dobe konstrukcije. Zato je naravno, da si prizadevamo razviti in uporabiti materiale, ki so najbolj
zdr`ljivi v danih razmerah. Materiali, ki imajo z alkalijami aktivirano osnovo, izkazujejo veliko odpornost pri izpostavitvi
kemijsko agresivnemu okolju. ^lanek obravnava uporabo materialov, ki temeljijo na snoveh, aktiviranih z alkalijami, kot mate-
rialih za popravilo konstrukcij, ki so izpostavljene agresivni talni vodi.
Klju~ne besede: z alkalijami aktivirane snovi, agresivno okolje, zdr`ljivost konstrukcij
1 INTRODUCTION
A lot of structures are exposed to the environments
that are, due to their chemistry, quite aggressive to rein-
forced concrete. The material’s resistance against the
aggressive substances from the external environment is
mainly determined by the features of its matrix.1 Com-
pared to concrete stone, the materials with the matrices
based on alkali-activated substances have a significantly
higher resistance to chemical substances. The article
deals with the development and, especially, the applica-
tion of the repair materials based on alkali-activated
materials used for the repair of the reinforced-concrete
lining of an underground collector used in a chemically
aggressive environment.
2 SPECIFICATIONS OF THE COLLECTOR AND
THE CONDITIONS OF ITS EXPLOITATION
The excavation of the underground collector was
done using the machine-excavation technology, while the
final scraping of the excavation area was carried out with
the classical, manual excavation method. The excavation
area is from 13 m2 to 14 m2 (by stationing). The length of
the underground part of the collector is 7.87 km; the
lining and casing of the tunnels are made of shotcrete,
whose strength within this project is determined to be 20
MPa. The lining of the tunnels is reinforced with welded
mesh panels (Ø 6–100/100 mm) and a rigid mine casing.
The thickness of the shotcrete is 100 mm. Dry shotcrete
was applied.
Already one year after the construction, significant
defects began to occur. These were mainly the penetra-
tion of water through the lining, new formations of solid
carbonate on the surface of the lining, the formation of
waste products associated with the reinforcement
corrosion, etc. A typical state of the damaged lining after
one year of operation is shown in Figures 1 and 2.
Materiali in tehnologije / Materials and technology 50 (2016) 1, 147–151 147
UDK 691.32:620.193:624.012.4 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 50(1)147(2016)
Figure 1: View of the location, in which a massive penetration of
moisture occurred: new solid-carbonate formations are visible on the
surface of the lining
Slika 1: Prikaz podro~ja, kjer se je pojavila mo~na penetracija vlage;
viden je nastanek trdnega karbonata na povr{ini podlage
The state of the structure continued to deteriorate
very quickly and, therefore, it was necessary to take cor-
rective measures. It was found that before the construc-
tion of the collector no hydrogeological survey was per-
formed, or it was performed unprofessionally and
insufficiently.
To properly assess the causes of the failures, a
detailed construction and technical investigation was
carried out in the places where the state of the concrete
was being evaluated, and an analysis of the chemical
composition of the groundwater in various locations of
the collector was performed. The results of the chemical
analysis of the groundwater are given in Table 1.
The foregoing shows that in terms of the durability of
the concrete the amount of sulphate ions in the water is
essential. The aggressiveness of the environment in diffe-
rent parts of the collector can be classified as relevant in
accordance with EN 206-1 (Table 2).
Table 1: Results of the chemical analysis of groundwater
Tabela 1: Kemijska analiza talnice
Chemical
composition
Stationing
0.57 km –
tertiary
sediments
Stationing
1.56 km –
quaternary
sediments
Stationing
2.68 km –
tertiary
sediments
Alkalinity 7.68 7.50 7.70
Sulphates 3450 mg L–1 148 mg L–1 1369 mg L–1
Chlorides 170 mg L–1 152 mg L–1 260 mg L–1
Nitrates 146 mg L–1 124 mg L–1 180 mg L–1
Ammonia ions 0.17 mg L–1 0.13 mg L–1 0.08 mg L–1
Chemical
composition
Stationing
4.42 km –
tertiary
sediments
Stationing
5.93 km –
tertiary
sediments
Stationing
6.67 km –
quaternary
sediments
Alkalinity 7.15 7.25 7.38
Sulphates 3825 mg L–1 1740 mg L–1 193 mg L–1
Chlorides 156 mg L–1 78 mg L–1 178 mg L–1
Nitrates 160 mg L–1 112 mg L–1 134 mg L–1
Ammonia ions 4.12 mg L–1 0.17 mg L–1 16. mg L–1
Table 2: Evaluation of groundwater aggression
Tabela 2: Ocena po{kodb zaradi talne vode
Stationing
0.57 km – tertiary sediments
Highly aggressive
environment
Stationing
1.56 km – quaternary sediments No aggression
Stationing
2.68 km – tertiary sediments
Moderately aggressive
chemical environment
Stationing
4.42 km – tertiary sediments
Highly aggressive
environment
Stationing
5.93 km – tertiary sediments
Moderately aggressive
chemical environment
Stationing
6.67 km – quaternary sediments No aggression
On the basis of the results of the chemical analysis of
the water samples it was possible to conclude that the
water that passes through quaternary sediments does not
cause the sulphate corrosion of the concrete. A com-
pletely different situation occurs when the water passes
through tertiary sediments. These waters show moderate
and, in some areas, even high sulphate aggressiveness.
The results of the chemical analysis clearly demonstrated
a high level of aggressiveness of the groundwater affect-
ing the collector. Underestimating this fact or failing to
carry out a hydrogeological research before the construc-
tion of a collector is, therefore, shown to be a serious
deficiency.
To assess the actual state after two years of the ope-
ration of the collector, a construction and technical
research was performed, aimed especially to:
• assess the state of the concrete, in terms of both
physical and mechanical parameters (i.e., particularly
in terms of compressive strength) and in terms of its
chemical and mineralogical compositions;
• assess the state of the reinforcement and the degree
of its corrosion, especially in connection to the
concrete’s ability to passivate the reinforcement
corrosion with its natural alkalinity.
The process of the construction and technical re-
search was carried out in accordance with the provisions
of the Technical Conditions for Reinforced Concrete
Structure Reinstalment1 and the relevant technical stan-
A. DUFKA, T. MELICHAR: COMPOSITES BASED ON INORGANIC MATRICES FOR EXTREME EXPOSURE CONDITIONS
148 Materiali in tehnologije / Materials and technology 50 (2016) 1, 147–151
Figure 2: Collection of core bores from the lining. Significant car-
bonate efflorescence is visible on the surface of the wall.
Slika 2: Zbiranje jeder iz izvrtine v podlagi. Na povr{ini stene je vi-
den mo~an razcvet karbonatov.
Table 3: Concrete’s compressive strength depending on the aggres-
siveness of groundwater in different parts of the collector
Tabela 3: Trdnost betona v odvisnosti od po{kodb zaradi talne vode
na razli~nih pozicijah kolektorja
Stationing
0.57 km – tertiary sediments 8 MPa
Stationing
1.56 km – quaternary sediments 25 MPa
Stationing
2.68 km – tertiary sediments 14 MPa
Stationing
4.42 km – tertiary sediments 10 MPa
Stationing
5.93 km – tertiary sediments 16 MPa
Stationing
6.67 km – quaternary sediments 27 MPa
dards. The sampling for testing the compressive strength
and for the physico-chemical analysis was carried out
with core bores.
Within the project, the strength class of the concrete
was defined as 20 MPa. The results for the compressive
strength determined on the samples and collected within
the research are summarized in Table 3.
The mineralogical composition of the concrete was
evaluated with an X-ray diffraction analysis, its results
are presented in Table 4.
Table 4: Results of the mineralogical analysis of the concrete
Tabela 4: Rezultati mineralo{ke sestave betona
Sampling point Identified minerals
Stationing
0.57 km – tertiary sediments
Calcite, ettringite, gypsum,
quartz, feldspar
Stationing
1.56 km – quaternary
sediments
Calcite, portlandite, calcium
silicate hydrate II, mono-
sulphate, quartz, feldspar
Stationing
2.68 km – tertiary sediments
Calcite, portlandite,
monosulphate, quartz, feldspar
Stationing
4.42 km – tertiary sediments
Calcite, ettringite, gypsum,
monosulphate, quartz, feldspar
Stationing
5.93 km – tertiary sediments
Calcite, portlandite, calcium
silicate hydrate II, mono-
sulphate, quartz, feldspar
Stationing
6.67 km –quaternary
sediments
Calcite, portlandite, calcium
silicate hydrate II, carbonate
complex, monosulphate,
quartz, feldspar
It is clear that in certain locations, already after two
years of the groundwater penetration, a massive degra-
dation of the concrete occurred. The concrete’s compres-
sive strength in the affected areas is only about half of its
original value. This finding corresponds completely with
the results of the physical and chemical analyses show-
ing massive corrosion formations (gypsum, secondary
ettringite) in the microstructure of the affected concrete.
In addition, the physical and chemical analyses showed
that in some areas the water penetrating the lining caused
quite an intense decomposition ("a washout") of the
cement matrix. This, of course, resulted in a decrease in
the strength of the concrete.
So, massive and negative symptoms occurred after
two years of the operation of the collector. On the basis
of the experience with similar types of structures, sup-
ported by a mathematical simulation, an assumption was
formulated, according to which in approximately two
years, in some parts of the collector the rate of degra-
dation of the lining will be so high that the static load of
the collector will be affected.
Therefore, it was necessary to repair the structure.
The first phase of the repair work related to the most
affected areas. In these locations, the entire layer of the
concrete along the lining was removed and the reinforce-
ment affected by extreme corrosion was replaced. The
removed concrete was then replaced with the repair ma-
terial. Due to the highly aggressive environment, a mate-
rial with a matrix based on alkali-activated substances
was used.
3 REPAIR MATERIAL BASED ON
ALKALI-ACTIVATED MATERIALS
When developing a formula based on alkali-activated
materials, we used the results from earlier researches.2–7
In terms of the properties of the resulting mixture, the
compositions of individual components are important.
The basic characteristics of the materials used for pro-
ducing the mixtures with alkali-activated matrices are
summarized in the following tables (Tables 5 to 7).
Table 5: Chemical composition of metakaolin
Tabela 5: Kemijska sestava metakaolina
Component Representation (%)
Al2O3 41.90
SiO2 52.90
K2O 0.77
Fe2O3 1.08
TiO2 1.80
MgO 0.18
CaO 0.13
Table 6: Chemical composition of slag
Tabela 6: Kemijska sestava `lindre
Component Representation (%)
Al2O3 7.42
SiO2 38.51
CaO 36.26
MgO 10.11
K2O 0.43
Fe2O3 0.74
TiO2 0.026
Table 7: Chemical composition of sodium-silica water glass
Tabela 7: Kemijska sestava vodnega stekla Na-Si
Component Representation (%)
SiO2 50.82
Na2O 26.25
Water 22.12
Silicate module 1.92
Table 8: Formula of the repair mixture based on alkali-activated mate-
rials
Tabela 8: Sestava reparaturne me{anice na osnovi z alkalijo aktivi-
ranega materiala
Ingredient Ingredient dosage of concrete,kg/m3
Slag 430
Sodium-silica glass 95
Metakaolin 50
Water 180
Aggregates 1750
Aggregates with fractions of 0–4 mm and 4–8 mm
were used as the filler. The particle-size distribution
A. DUFKA, T. MELICHAR: COMPOSITES BASED ON INORGANIC MATRICES FOR EXTREME EXPOSURE CONDITIONS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 147–151 149
curve was optimized according to the Empa II proce-
dure.5
The mixture composition is given in Table 8.
This mixture was applied in the locations of the
lining, where it was necessary to remove the original
concrete. The effectiveness of this measure was then
evaluated using the procedure described in the following
section.
4 PERFORMANCE ANALYSIS OF THE NEW
MATERIAL
The efficiency, or durability, of the repair material
was evaluated as follows: One year after the repairs,
control samples were taken from the localities in which
the repair mixture in question was applied. A set of
physico-mechanical and physico-chemical parameters of
these samples was determined. Attention was focused
especially on the following parameters:
• the tensile strength of the surface layers;
• the compressive strength;
• the microstructure, analysed with physico-chemical
methods. The X-ray diffraction analysis and the
scanning electron microscopy were used.
The tensile-strength test of the surface layers was
made "in situ", the sampling for testing the compressive
strength and for the physico-chemical analysis was
carried out with core bores.
These analyses were performed with the procedures
that are in accordance with the provisions of the relevant
technical standards. The samples representing the repair
material were then subjected to the same set of experi-
ments after being stored for about 28 d and 360 d in
standard laboratory conditions (i.e., t = (20 ± 2) °C,  =
(60 ± 5) %). These values were considered as the refe-
rences. The comparison of the values recorded on the
reference bodies and the samples of the repair materials
exposed to the environment of the collector for one year
was the essential criterion in the evaluation of the
effectiveness of the repairs carried out. The data found
with the set of experiments are shown below (Tables 9
and 10).
Table 9: Results of the strength-parameter tests of the repair material
Tabela 9: Rezultati preizkusa trdnosti reparaturnega materiala
sample identification
Bulk
density
(kg/m3)
Tensile
strength of
surface
layers
(MPa)
Compres-
sive
strength
(MPa)
Reference set – stored for
28 d in standard
laboratory conditions
2450 2.1 35.6
Reference set – stored for
one year in standard
laboratory conditions
2400 2.3 36.2
Stationing 0.57 km –
significant sulphate
aggressiveness, annual
exposure
2420 1.9 34.3
Stationing 4.42 km –
tertiary sediments, annual
exposure
2380 2.0 34.0
Table 10: Results for the mineralogical composition of the repair
material
Tabela 10: Mineralo{ka sestava reparaturnega materiala
Sample identification Identified minerals
Reference set – stored for
28 d in standard laboratory
conditions
Calcium silicate hydrate II,
goethit, quartz, feldspar
Reference set – stored for one
year in standard laboratory
conditions
Calcium silicate hydrate II,
goethit, quartz, feldspar
Stationing 0.57 km – signi-
ficant sulphate aggressiveness,
annual exposure
Calcium silicate hydrate II,
goethit, quartz, feldspar
Stationing 4.42 km – tertiary
sediments, annual exposure
Calcium silicate hydrate II,
goethit, quartz, feldspar
The microstructure of the material was also analysed
using the scanning electron microscopy (Figure 3).
On the basis of the above findings we can say that in
terms of both the strength parameters and mineralogy,
the parameters of the repair material exposed to the
highly corrosive environment are quite comparable with
the reference parameters.
A. DUFKA, T. MELICHAR: COMPOSITES BASED ON INORGANIC MATRICES FOR EXTREME EXPOSURE CONDITIONS
150 Materiali in tehnologije / Materials and technology 50 (2016) 1, 147–151
Figure 3: Detailed analysis of the microstructure showed a: a)
de-facto amorphous phase of the matrix for both the reference material
and b) the exposed material of the collector
Slika 3: Podrobnej{a analiza mikrostrukture poka`e: a) amorfno
osnovo tako v primerjalnem materialu kot tudi v b) izpostavljenem
materialu v kolektorju
(a)
(b)
5 CONCLUSIONS
The article deals with the way of repairing an under-
ground collector affected, in some areas, by underground
waters with a highly aggressive chemical effect. Already
in one or two years of the collector’s operation these wa-
ters caused distinct disorders. In the affected areas there
was a significant reduction in the mechanical properties
of the concrete, and its ability to protect the reinforce-
ment against the corrosion was significantly reduced.
Thus, local repairs were carried out, whereby a material
with its matrix based on alkali-activated materials was
used as the repair material.
The matrix of the repair material was composed of a
mixture of blast-furnace slag and metakaolin. Water
glass was used as the activator. The matakaolin added to
the mixture was to reduce the negative effects associated
with the autogenous shrinkage that usually accompanies
the aging of the materials with pure-slag matrices.3,6
The control tests that were conducted one year after
the repair work demonstrated that the aggressive effects
of the water did not result in a decrease in the strength of
the repair material. Neither were negative changes found
in its microstructure. It was also established that the rein-
forcement was very well protected against the corrosion
due to the repair material.
It is clear that the period (one year of application) is
short in terms of the durability of the structures, and for a
precise verification of our findings, we will need to con-
tinue with the monitoring of the concerned construction.
However, despite this fact, it can be noted that so far the
results have indicated a high potential of the material
with a matrix based on alkali-activated materials in the
repair of the reinforced concrete structures exposed to
chemically aggressive environments.
Acknowledgements
This research was done with the financial help of
project GA^R 14-25504S, Research of Behaviour of
Inorganic Matrix Composites Exposed to Extreme Con-
ditions, and EU project The Research and Development
for Innovation, reg. number CZ.1.05/2.1.00/03.0097,
through the activities of regional centre AdMaS –
Advanced Materials, Structures and Technologies.
6 REFERENCES
1 R. Drochytka, J. Dohnálek, J. Byd`ovský, V. Pumpr, A. Dufka, P.
Dohnálek, Technické podmínky pro sanace betonových konstrukcí
TP SSBK III., 1. edition, Sdru`ení pro sanace betonových
konstrukcí, (The specifications for the reinstalment of concrete
structures TP SSBK III, 1st edition, Association of Concrete
Structure Reinstalment), Brno 2013, 262 pages
2 M. Palacios, F. Puertas, Effect of shrinkage reducing admixtures on
properties of alkali-activated slags, mortars and pastes, Cement and
Concrete Research, 37 (2007) 5, 691–702, doi:10.1016/j.cemconres.
2006.11.021
3 P. Rovnaník, Vliv pùsobení vysokých teplot na stavební materiály na
bázi alkalicky aktivovaných pojiv, habilita~ní práce (The influence of
high temperatures on building materials based on alkali-activated
binders, habilitation thesis), University of Technology in Brno, 2012
4 W. Brylicki, J. Malolepszy, S. Stryczek, Alkali activated cementi-
tious material for drilling operation, 9th International Congress on
the Chemistry of Cement, New Delhi, India, 3 (1992), 312–318
5 D. Wu, Y. Pei, B. Huang, Slag-mug mixtures improve cementing
operations in China, Oil and Gas Journal, (1996), 95–100
6 X. Chen, N. Yang, Influence of polymeric structure of granulated
blast furnace slag on their hydraulic activities, 2nd Beijing Interna-
tional Symposium on Cement and Concrete, Beijing, 1989, 346–351
7 J. Malolepszy, J. Deja, W. Brylicky, Industrial application of slag
alkaline in concretes, 9th international conference on alkaline ce-
ments and concretes, Kiev, 2 (1994), 989–1001
A. DUFKA, T. MELICHAR: COMPOSITES BASED ON INORGANIC MATRICES FOR EXTREME EXPOSURE CONDITIONS
Materiali in tehnologije / Materials and technology 50 (2016) 1, 147–151 151

M. BASIAGA et al.: THE EFFECT OF EO AND STEAM STERILIZATION ON THE MECHANICAL ...
153–158
THE EFFECT OF EO AND STEAM STERILIZATION ON THE
MECHANICAL AND ELECTROCHEMICAL PROPERTIES OF
TITANIUM GRADE 4
VPLIV EO IN STERILIZACIJE S PARO NA MEHANSKE IN
ELEKTROKEMIJSKE LASTNOSTI TITANA GRADE 4
Marcin Basiaga, Witold Walke, Zbigniew Paszenda, Anita Kajzer
Silesian University of Technology, Faculty of Biomedical Engineering, Zabrze, Poland
marcin.basiaga@polsl.pl
Prejem rokopisa – received: 2014-09-23; sprejem za objavo – accepted for publication: 2015-03-09
doi:10.17222/mit.2014.241
Currently, various modifications to surfaces are made more and more frequently in order to improve implants’ haemocom-
patibility. The main criterion determining the applicability of the respective surface-modification method is obtaining a product
featuring suitable functional properties. These properties depend to a great extent on the corrosion resistance in the environment
of human blood. Subject-matter literature does not devote much attention to the sterilisation process for titanium and cpTi alloys
with surface modifications. A problem that still remains unsolved is the selection of a proper test showing the full characteristics
of their behaviour contact with a blood environment during the time that the implant is used. Therefore, the authors of this study
made an attempt to evaluate the impact of medical sterilisation methods, i.e., the ethylene oxide anodic oxide and SiO2 layer, by
means of the sol-gel method. The efficiency of the suggested technology for oxide layer application was evaluated on the basis
of mechanical and electrochemical tests. Sterilisation in ethylene oxide and steam had a favourable influence on the
electrochemical and mechanical properties of cpTi, irrespective of the method of surface preparation. In order to simulate real
conditions, the tests were performed in artificial plasma at a temperature of T = 37 ± 1 °C and pH = 7.0 ± 0.2. The results proved
the diversification of electrochemical properties of the oxide layers, depending on the technological parameters of its
application. The suggestion of proper variants of the surface modification with the application of electrochemical and chemical
methods is of long-range importance and will contribute to the development of technological conditions with specific
parameters for the creation of oxide layers on metallic implants made of cpTi.
Keywords: cpTi (Grade 4), SiO2, TiO2, mechanical properties, electrochemical properties
Vedno pogosteje se opravljajo razli~ne modifikacije povr{ine, da bi se izbolj{ala hemokompatibilnost vsadkov. Glavni kriterij, ki
dolo~a uporabnost metode za modifikacijo povr{ine je, da proizvod poka`e primerne funkcionalne lastnosti. Te lastnosti so v
veliki meri odvisne od korozijske odpornosti v ~love{ki krvi. Obstoje~a literatura ne posve~a velike pozornosti postopku
sterilizacije titana in cpTi zlitin z modificirano povr{ino. Problem, ki {e ni re{en, je izbira primernega preizkusa, ki bi pokazal
vse zna~ilnosti o obna{anju stika s krvjo med uporabo vsadka. Zato so avtorji v tej {tudiji poizkusili oceniti vpliv medicinskih
metod sterilizacije, kot je etilen oksid anodni oksid in SiO2 plast izdelano s pomo~jo metode sol-gel. Predlagana tehnologija
uporabe oksidnega sloja je bila ocenjena z mehanskimi in elektrokemijskimi preizkusi. Sterilizacija v etilen oksidu in pari je
imela ugodne vplive na elektrokemijske in mehanske lastnosti cpTi, ne glede na na~in priprave povr{ine. Za simulacijo realnih
pogojev so bili preizkusi izvr{eni v umetni plazmi pri temperaturi T = 37 ± 1 °C in pH = 7,0 ± 0,2. Dobljeni rezultati so potrdili
razli~nost v elektrokemijskih lastnostih oksidnih plasti, odvisno od tehnolo{kih parametrov njene uporabe. Predlog ustreznega
na~ina spremembe povr{ine z uporabo elektrokemijskih in kemijskih metod je dolgoro~no pomemben in bo prispeval k razvoju
tehnolo{kih pogojev s specifi~nimi parametri nastajanja plasti oksidov na kovinskem implantatu iz cpTi.
Klju~ne besede: cpTi (Grade 4), SiO2, TiO2, mehanske lastnosti, elektrokemijske lastnosti
1 INTRODUCTION
Literature does not devote much attention to the steri-
lisation process for titanium and cpTi alloys with surface
modification.1,2 In recent years there has been a dynamic
development in pure titanium coating methods, aimed at
improving the hemocompatibility.3,4 Surface-layer modi-
fication methods must be selected in a way that does not
cause phase or structural transitions, or precipitation
processes in the base material. It has a particular
significance for the geometry of implants having small
cross-sectional areas or small-radii edges (stents or heart
valves). For this reason, surface-layer modification for
titanium and its alloys uses low-temperature methods,
and especially anodic oxidation coating and the sol-gel
method. The properties of TiO2 layers obtained through
the process of anodic oxidation and SiO2 layers obtained
with the sol-gel method depend on the properties of the
titanium surface and the process parameters. The deve-
lopment and verification of conditions for coating cpTi
with layers of TiO2 and SiO2 characterized by high
hemocompatibility was the topic of previous papers pub-
lished by the authors.5–7 The safety of a medical device
involving contact with blood is also related to the ne-
cessity of observing the appropriate procedures, pre-
venting the transfer of pathogenic microorganisms into
the human body. Such procedures are aimed at the remo-
val and efficient elimination of microorganisms and the
obtaining of sterile medical devices complying with
certain specified quality requirements. A medical device
is considered sterile when it achieves a Sterility Assur-
Materiali in tehnologije / Materials and technology 50 (2016) 1, 153–158 153
UDK 669.295:544.6:532.6 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 50(1)153(2016)
ance Level (SAL) of 10–6. Therefore, in order to deter-
mine the impact of the procedure on the properties of the
proposed surface layers, the authors subjected cpTi
samples to steam sterilization and ethylene oxide sterili-
zation. These two methods of medical sterilization are
currently the most frequently used for devices having
contact with blood. Taking the sterilization process into
consideration will allow the characterization of the
physical and chemical properties of the proposed surface
layers in a more complete way, which has a significant
influence on reducing the incidence of failures in the
treatment of cardiovascular diseases.
2 MATERIALS AND METHODS
The study material was Grade 4 cpTi in the form of
disks of the following dimensions: diameter, d = 14 mm,
thickness, g = 2 mm. The samples were subjected to
metal finishing consisting of mechanical grinding and
electrolytic polishing, and then coating with layers of
SiO2 and TiO2 based on two methods. The SiO2 layer was
applied using the sol-gel method with the following
process parameters: v = 3.0 cm/min, T = 430 °C, t = 60
min. The silica precursors were: tetraethyl orthosilicate
(TEOS), Si(OC2H5)4 and tetramethyl orthosilicate
(TMOS), Si(OCH3)4. Other reagents included ethyl
alcohol (EtOH) and water. On the other hand, the TiO2
layer was applied with the use of anodic oxidation at a
potential of 100 V in an electrolyte based on phosphoric
and sulphuric acids. Later, the prepared samples were
subjected to ethylene oxide sterilization and steam
sterilization. Ethylene oxide sterilization was conducted
during a 12-hour cycle of exposure to ethylene oxide at
30 °C. After the process, the samples were ventilated for
2 h with the use of an EOGas series sterilizer from An-
dersen Products. The cycle yielded a Sterility Assurance
Level SAL of 10–6. The process was controlled by means
of a chemical indicator, a biological indicator and an
ethylene oxide exposure indicator. The steam steriliza-
tion was performed in a Basic Plus autoclave at tempe-
rature, T = 134 °C, pressure, p = 2.1 bar, and time, t = 12
min.
The tests were performed on samples coated with
SiO2 and TiO2 (cpTi+SiO2 and cpTi+TiO2), coated with
SiO2 and TiO2 and subjected to steam sterilization
(cpTi+SiO2+steam and cpTi+TiO2+steam), and coated
with SiO2 and TiO2 and subjected to ethylene oxide
sterilization (cpTi+SiO2+EO and cpTi+TiO2+EO). In
order to assess the impact of a given type of medical
sterilization on the mechanical and electrochemical
properties of the proposed cpTi surface modifications,
the authors selected electrochemical potentiodynamic
and impedance tests. The assessment of the mechanical
properties, in turn, consisted of a measurement of the
layer adhesion to the base and a measurement of the
hardness.
The investigation of the electrochemical properties
began with potentiodynamic tests and registration of the
polarization curves. The measurement facility consisted
of a potentiostat with a PGP-201 generator, an electro-
chemical cell with a set of electrodes (platinum PtP-201
electrode the auxiliary and calomel electrode as the
reference), and a solution (250 mL) functioning as artifi-
cial plasma (pH = 7.0 ± 0.2). The artificial plasma during
the test had a temperature T = 37.0 ± 1 °C. Corrosion
tests started by determining the open-circuit potential
EOCP for no current flowing. Anodic polarization curves
were registered from the initial potential value, Estart =
EOCP – 100 mV. The potential change took place in the
anodic direction at a speed of 0.16 mV/s.8–10 After
obtaining an anodic current density of 1 mA/cm2 or the
measurement range +4000 mV, the polarization direction
was changed in order to register the formation of any
hysteresis loop. Further, electrochemical impedance
spectroscopy (EIS) tests were conducted using an
AutoLab PGSTAT 302N system equipped with an FRA2
module. The measurement system was used in the fre-
quency range 104–10–3 Hz, with a sinusoidal voltage
amplitude of the stimulating signal of 10 mV. The impe-
dance spectra were determined and the measurements
matched to an equivalent circuit by means of the
least-squares method. The impedance spectra of the
studied system were presented in the form of Nyquist
plots for various frequencies and Bode plots.11
The mechanical properties were then tested, with the
properties of interest being the adhesion of the layers to
the base and their hardness. The adhesion of the layers to
the base was tested using the scratch-test method on an
open platform equipped with a CSM Micro-Combi-
Tester in compliance with the standard.12 The test in-
volved generating a scratch with an indenter (Rockwell-
type diamond cone), gradually increasing the load on the
indenter. The assessment of the critical load, Lc, was
made on the basis of registered acoustic emission
changes, friction force and coefficient, as well as
observations from the light microscope integrated into
the platform. The tests were performed at increasing
loads in the range 0.03–20 N and for the following para-
meters: load rate 10 N/min, table speed 1.5 mm/min,
scratch length 3 mm. The instrumental hardness
measurement was performed using the Oliver-Pharr
method, with the use of a Berkovich indenter on the open
platform Micro-Combi-Tester by CSM Instruments. The
loading and unloading rate amounted to 0.40 mN/min,
while the loading force exerted on the indenter was 0.20
mN.13
3 RESULTS
Polarization curves registered during the potentiody-
namic tests were the basis for a determination of the
characteristic quantities describing the pitting-corrosion
resistance of cpTi with a modified surface layer before
and after sterilization (steam and EO) (Figure 1 and
Table 1).
M. BASIAGA et al.: THE EFFECT OF EO AND STEAM STERILIZATION ON THE MECHANICAL ...
154 Materiali in tehnologije / Materials and technology 50 (2016) 1, 153–158
The corrosive potential registered for the samples
before sterilization assumed the following values:
cpTi+TiO2 – Ecorr = –235 mV and cpTi+SiO2 – Ecorr =
–178 mV. In the case of the samples subjected to anodic
oxidation and sterilization, the corrosive potential was:
cpTi+TiO2+steam – Ecorr = –142 mV and cpTi+TiO2+EO
– Ecorr = –170 mV, while for the samples coated with
silica and sterilized it was: cpTi+SiO2+steam – Ecorr =
–121 mV and cpTi+SiO2+EO – Ecorr = –121 mV.
Regardless of the surface-modification method and the
medical sterilization method, no hysteresis loop was
observed, which shows that the analysed samples
polarized up to potential value of +4000 mV are fully
resistant to pitting corrosion (Figure 1 and Table 1).
Table 1: Results of potentiodynamic test (mean measurement values)
Tabela 1: Rezultati potenciodinami~nega preizkusa (srednje izmerje-
ne vrednosti)
Surface Ecorr, mV Rp, M cm2
cpTi+TiO2
inital state –235 4.02
EO –170 43.70
steam –142 30.58
cpTi+SiO2
inital state –178 0.47
EO –121 9.20
steam –121 1.30
M. BASIAGA et al.: THE EFFECT OF EO AND STEAM STERILIZATION ON THE MECHANICAL ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 153–158 155
Figure 3: Impedance spectra for the sample cpTi(SiO2): a) Nyquist
plot, b) Bode diagram
Slika 3: Spekter impedance za vzorec cpTi(SiO2): a) Nyquist diagram,
b) Bode diagram
Figure 1: Anodic polarisation curves of sample: a) cpTi(TiO2), b)
cpTi(SiO2)
Slika 1: Krivulje anodne polarizacije vzorcev: a) cpTi(TiO2), b)
cpTi(SiO2)
Figure 2: Impedance spectra for the sample cpTi(TiO2): a) Nyquist
plot, b) Bode diagram
Slika 2: Spekter impedance za vzorec cpTi(TiO2): a) Nyquist
diagram, b) Bode diagram
The impedance spectra determined during the EIS
tests indicate diversified kinetics during the process
taking place in the surface layers formed on the cpTi
base before and after medical sterilization (Figures 2
and 3). The impedance spectra were analysed with the
use of equivalent circuits, and their resultant impedances
described (Figure 4 and Table 2). The best conformity
between the model spectra and the experimental spectra
obtained from the artificial plasma was achieved in the
following circuits:5,6,11
• circuit indicating the existence of a double layer,
where Rs is the resistance of the artificial plasma,
CPEp is a constant phase element modelling the
capacity of the surface zone of the material, with a
highly developed surface, Rp – an element modelling
the solution resistance in this zone, Rct and CPEdl –
resistance and capacity of the oxide layer (Figure
4a),
• circuit including also CPEpore – an element represent-
ing the capacity of a double (porous) layer and an
Rpore resistor representing the electrolytic resistance in
the pores (Figure 4b).
In the case of the cpTi+TiO2+steam sample, the layer
was triple. This was a result of the penetration of the
artificial plasma into the surface layer, resulting from the
fact that the oxide layer created during anodic oxidation
was destroyed by the steam, and recreated deeper inside,
up to the layer adjacent to the base. In the remaining
samples, the presence of a layer with high surface
development was observed. The layer functioned as an
additional barrier protecting the material against the
corrosive environment.
M. BASIAGA et al.: THE EFFECT OF EO AND STEAM STERILIZATION ON THE MECHANICAL ...
156 Materiali in tehnologije / Materials and technology 50 (2016) 1, 153–158
Figure 4: Physical models of an electrical equivalent system for a corrosion system: metal – solution5,6,11
Slika 4: Fizikalni modeli ekvivalentnega elektri~nega sistema za korozijski sistem: kovina – raztopina5,6,11
Table 2: EIS analysis results
Tabela 2: Rezultati EIS-analize
Surface
Rs/
cm2
Rpore/
cm2
CPEpore Rp/
kcm2
CPEp Rct/
Mcm2
CPEdl
Y0/–1cm–2s–n n Y0/–1cm–2s–n n Y0/–1cm–2s–n n
cpTi+TiO2
inital state 19 – – – 968 0.2083E-6 0.89 4.26 0.1852E-5 0.81
EO 18 – – – 700 0.1460E-6 0.96 4.40 0.1901E-6 0.93
steam 18 112 0.2092E-6 0.94 817 0.5091E-6 0.90 3.66 0.2417E-5 0.82
cpTi+SiO2
inital state 18 – – – 188 0.6605E-5 0.93 1.59 0.2065E-4 0.88
EO 19 – – – 224 0.1730E-5 0.83 3.00 0.6317E-5 0.75
steam 17 – – – 160 0.9806E-5 0.85 2.40 0.2664E-4 0.85
Table 3: The results of adhesion of the layer on the cpTi substrate
Tabela 3: Rezultati adhezije plasti na podlago iz cpTi
Failure of the layer
The value of registered indenter load Fn/N
cpTi+SiO2 cpTi+TiO2
inital state steam EO inital state steam EO
Measurment 1 Delamination Lc1 2.59 2.89 2.79 2.71 1.66 1.87
Complete break Lc2 7.58 4.32 4.05 5.35 2.67 2.58
Measurement 2 Delamination Lc1 3.78 3.26 1.95 2.96 1.67 1.65
Complete break Lc2 5.76 4.55 2.59 4.95 2.56 2.37
Measurement 3 Delamination Lc1 3.38 2.41 2.01 3.33 1.82 1.78
Complete break Lc2 6.16 4.88 2.89 5.08 2.89 2.12
Average Delamination Lc1 3.25 2.85 2.25 3.00 1.71 1.76
Complete break Lc2 6.50 4.58 3.08 5.12 2.70 2.35
Standard
deviation
Delamination Lc1 ±0.60 ±0.42 ±0.46 ±0.31 ±0.09 ±0.11
Complete break Lc2 ±0.95 ±0.28 ±0.77 ±0.20 ±0.16 ±0.23
The results of the tests of adhesion of the analysed
layers to the base material are presented in Table 3 and
Figures 5 and 6. Regardless of the sterilization method,
the results indicate the impairment of adhesion of the
layers coated by both anodic oxidation and the sol-gel
method in comparison to the baseline samples. This can
be seen from the values of the parameters determined on
the basis of the measurements (Table 3). It was observed
that for the samples not subjected to sterilization, the
critical load causing layer delamination inwards and
outwards was Lc2 = 6.50 N (cpTi+SiO2 – sol-gel method)
and Lc2 = 5.12 N (cpTi+TiO2 – anodic oxidation). After
sterilization (both steam and EO sterilization), the cri-
tical load value decreased to: Lc2 = 4.58 N (cpTi+SiO2+
Steam), Lc2 = 3.08 N (cpTi+SiO2+EO) for the sol-gel
method, and Lc2 = 2.70 N (cpTi+TiO2+Steam), Lc2 =
2.35 N (cpTi+TiO2+EO) for the anodic oxidation. Re-
gardless of the type of sample, no acoustic emission
signal was registered, which indicated that the bond
energy between the base and the coating was too low. No
significant changes between steam and EO sterilization
were observed.
The next step involved the hardness testing of the
layers. The results of the measurements are shown in
Table 4. The results do not show any significant diffe-
rences between the samples after steam and ethylene
oxide sterilization and the baseline samples.
Table 4: The results of nanohardness of the layers
Tabela 4: Rezultati meritve mikrotrdote plasti
i.c.
Nanohardness HIT, MPa
cpTi+SiO2 cpTi+TiO2
inital
state steam EO
inital
state steam EO
Measurement 1 1181 1304 1082 1236 1434 1171
Measurement 2 1236 1278 923 913 1247 1241
Measurement 3 1021 1189 1120 1021 1156 1089
Average 1146 1257 1041 1056 1279 1167
Standard
deviation ±152 ±60 ±104 ±164 ±141 ±76
4 CONCLUSIONS
The correct selection of physical and chemical pro-
perties is a significant issue in the process of adjusting
the functionalities of cardiac implants,14 and has a direct
impact on the final quality of medical devices. Physical
and chemical devices are shaped by means of various
types of metal finishing.14,15 The influence of various
types of titanium alloy finishing on orthopaedic implants
has been demonstrated by Szewczenko et al.16. He has
shown that the method of base preparation has a crucial
influence on the physical and chemical properties of the
anodic oxide coating. Paszenda has proposed chemical
surface modifications of the Cr-Ni-Mo steel intended for
cardiac implants in order to achieve a higher hemo-
compatibility.17 The literature review indicates that there
are no comprehensive works devoted to a possible
change of properties during medical sterilization.18–20
Therefore, the authors of this paper focused on an
assessment of the possible changes to the mechanical
and electrochemical properties of a coating subjected to
pressurized steam sterilization and ethylene oxide
sterilization. The tests on the samples following medical
sterilization processes have shown differences in relation
to those samples that did not undergo sterilization.
Pressurized-steam sterilization caused a change in the
corrosive potential value towards positive values and
increased the polarization resistance, regardless of the
surface-modification method. For the TiO2-coated sam-
ples, it stopped the diffusion processes stemming from
the anodic oxidation and caused the formation of an
additional layer with a highly developed surface, which
functioned as an additional barrier protecting the mate-
rial against the corrosive environment and increasing the
ion-transfer resistance. In the case of the SiO2 layer, this
type of sterilization did not cause significant changes to
M. BASIAGA et al.: THE EFFECT OF EO AND STEAM STERILIZATION ON THE MECHANICAL ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 153–158 157
Figure 6: Results of the adhesion tests of the sample cpTi+SiO2+EO
Slika 6: Rezultati preizkusa adhezije vzorca cpTi+SiO2+EO
Figure 5: Results of the adhesion tests of the sample cpTi+TiO2+EO
Slika 5: Rezultati preizkusa adhezije vzorca cpTi+TiO2+EO
the corrosive resistance of cpTi. The mechanical proper-
ties, on the other hand, remained unchanged, regardless
of the type of metal finishing. However, the adhesion to
the base was weaker in comparison to those samples not
subjected to sterilization, which may be caused by the
restructuring of the surface oxides under the influence of
a sterilizing agent, which has a direct impact on the
strength of the bonds between the chemical compounds,
such as between TiO2/Ti2O3 and the base. Ethylene oxide
sterilization of the samples with a modified surface also
caused changes to their electrochemical and mechanical
properties. As in the case of pressurized-steam steriliza-
tion, the samples coated with TiO2 and SiO2 had a better
resistance to corrosion in contact with the artificial
plasma. On the surface of the samples coated with TiO2,
the presence of an additional porous layer has been ob-
served. This is the result of the partial degradation of the
oxide layer and an increase in its porosity, which, as a
consequence, leads to much weaker adhesion to the base.
In the scratch test, a decrease in the critical load causing
delamination of the layer towards the base was observed.
In conclusion, the study has shown the significant
influence of medical sterilization on the physical and
chemical properties of the TiO2 and SiO2 layers. It has
been proved that, regardless of the surface layer type, the
choice of the sterilizing method is an important issue. A
SiO2 layer applied with the sol-gel method exhibited a
higher electrochemical stability. In the future, the authors
plan further work to help identify the chemical compo-
sition and chemical compounds formed as a result of the
sterilization processes and possible changes to the layer
thickness.
Acknowledgements
The project was funded by the National Science
Centre allocated on the basis of the decision No.
2011/03/B/ST8/06499.
5 REFERENCES
1 P. Karasiñski, Opt. Appl., 35 (2005), 117–128
2 J. Szewczenko, J. Jaglarz, M. Basiaga, J. Kurzyk, E. Skoczek, Prz.
Elektrot., 88 (2012) 12B, 228–231
3 A. Sadeq, Z. Cai, R. D. Woody, A. W. Miller, J. Prosth. Dent., 90
(2003) 1, 10–17, doi:10.1016/S0022-3913(03)00263-4
4 B. Surowska, J. Bieniaœ, M. Walczak, K. Sangwal, A. Stoch, Appl.
Surf. Sci., 238 (2004), 288–294, doi:10.1016/j.apsusc.2004.05.219
5 M. Basiaga, W. Walke, Z. Paszenda, P. Karasiñski, J. Szewczenko,
Biomatt., 4 (2014), 1–6, doi:10.4161/biom.28535
6 M. Basiaga, Z. Paszenda, W. Walke, P. Karasiñski, J. Marciniak, Inf.
Technol. Biomed., 284 (2014), 411–420, doi:10.1007/978-3-
319-06596-0_39
7 W. Walke, Z. Paszenda, M. Basiaga, P. Karasiñski, M. Kaczmarek,
Inf. Technol. Biomed., 284 (2014), 403–410, doi:10.1007/978-
3-319-06596-0_38
8 ASTM F2129-08 Standard Test Method for Conducting Cyclic
Potentiodynamic Polarization Measurements to Determine the
Corrosion Susceptibility of Small Implant Devices, 2008,
doi:10.1520/F2129-08
9 W. Kajzer, A. Kajzer, Prz. Elektrot., 12 (2013), 275–279
10 A. Kajzer, W. Kajzer, J. Semenowicz, A. Mroczka, Sol. St. Phen.,
227 (2015), 523–526, doi:10.4028/www.scientific.net/SSP.227.523
11 M. Kaczmarek, W. Walke, Z. Paszenda, Prze. Elektrot., 12b (2011),
74–77
12 PN-EN 1071-3. Advanced technical ceramics. Determination of
adhesion and other mechanical failure modes by a scratch test, 2007
13 EN ISO 14577-1 Metallic materials-Instrumented indentation test for
hardness and materials parameters-Part1: Test method, 2015
14 J. G³uszek, In¿. Mat., 5 (2002), 351–354
15 L. Gan, J. Wang, A. Tache, N. Valiquette, D. Deporter, R. Pilliar,
Biomat., 25 (2004), 5313–5321, doi:10.1016/j.biomaterials.2003.
12.039
16 J. Szewczenko J. Jaglarz, M. Basiaga, J. Kurzyk, Z. Paszenda, Optica
Applicata, 43 (2013) 1, 173–180, doi:10.5277/oa130121
17 Z. Paszenda, Inf. Tech. Biomed., Adv. Soft Comp., 47 (2008), 15–27,
doi:10.1007/978-3-540-68168-7_2
18 N. Kuromoto, R. Simao, G. Soares, Materials Characterization, 58
(2007), 114–121, doi:10.1016/j.matchar.2006.03.020
19 B. Yang, M. Uchida, H. M. Kim, X. Zhang, T. Kokubo, Biomat, 25
(2004), 1003–1010, doi:10.1016/S0142-9612(03)00626-4
20 H. Song, S. Park, S. Jeong, Y. Park, Journal of Mater. Proc. Techn.,
209 (2009), 864–870, doi:10.1016/j.jmatprotec.2008.02.055
M. BASIAGA et al.: THE EFFECT OF EO AND STEAM STERILIZATION ON THE MECHANICAL ...
158 Materiali in tehnologije / Materials and technology 50 (2016) 1, 153–158
J. DLOUHY et al.: INFLUENCE OF THE CARBIDE-PARTICLE SPHEROIDISATION PROCESS ON THE MICROSTRUCTURE ...
159–162
INFLUENCE OF THE CARBIDE-PARTICLE SPHEROIDISATION
PROCESS ON THE MICROSTRUCTURE AFTER THE
QUENCHING AND ANNEALING OF 100CrMnSi6-4 BEARING
STEEL
VPLIV PROCESA SFEROIDIZACIJE KARBIDNIH DELCEV NA
MIKROSTRUKTURO JEKLA 100CrMnSi6-4 ZA LE@AJE PO
KALJENJU IN POPU[^ANJU
Jaromir Dlouhy, Daniela Hauserova, Zbysek Novy
COMTES FHT, Prumyslova 995, 334 41 Dobrany, Czech Republic
jdlouhy@comtesfht.cz
Prejem rokopisa – received: 2014-12-12; sprejem za objavo – accepted for publication: 2015-01-21
doi:10.17222/mit.2014.303
Bearings are used mostly in the quenched and tempered state, e.g., steel 100CrMnSi6-4 with a microstructure of low-tempered
martensite and carbide particles undissolved during the quenching austenitization. The size and density of the particles depend
on the spheroidisation annealing which is the standard operation at the beginning of the bearing manufacturing. The particles
decrease the grain growth of the austenite and determine the grain size after the quenching. The process of accelerated carbide
spheroidisation and refinement (ASR) was developed and is used as a replacement of the conventional spheroidisation soft
annealing. The ASR process produces the structure of a ferritic matrix and fine globular carbides. The carbide size is several
times smaller in comparison with the conventional soft annealing. This microstructure is better for quenching and tempering and
ensures a better bearing performance. The article compares the structures and properties of the quenched and tempered
100CrMnSi6-4 steel pre-treated with the conventional soft annealing and ASR. The smaller ASR particle size allows the use of
lower quenching temperatures, ensuring the desired final hardness. Samples in the hardened state were compared, considering
the prior-austenite grain size, the carbide-particle size and distribution as well as the hardness.
Keywords: accelerated spheroidisation, carbide-particle morphology, hardening, bearing steel
Le`aji se obi~ajno uporabljajo v kaljenem in popu{~enem stanju, na primer jeklo 100CrMnSi6-4 z mikrostrukturo nizko
popu{~enega martenzita in s karbidnimi delci, ki se ne raztopijo pri avstenitizaciji pred kaljenjem. Velikost in pogostost delcev
je odvisna od sferoidizacijskega `arjenja, ki je standardna operacija na za~etku izdelovanja le`aja. Delci zavirajo rast avstenitnih
zrn in dolo~ajo velikost zrn po kaljenju. Razvit je bil postopek pospe{ene sferoidizacije in udrobnjenja karbidov (ASR), ki je bil
uporabljen namesto obi~ajnega sferoidizacijskega mehkega `arjenja. Pri ASR procesu nastane feritna osnovna mikrostruktura in
drobni globularni karbidi. Velikost karbidov je nekajkrat manj{a v primerjavi z obi~ajnim mehkim `arjenjem. Taka
mikrostruktura je bolj{a za kaljenje in popu{~anje in zagotavlja bolj{e lastnosti le`aja. V ~lanku so primerjane mikrostrukture in
lastnosti kaljenega in popu{~anega jekla 100CrMnSi6-4 predhodno mehko `arjenega in ASR. Manj{a velikost delcev pri ASR
omogo~a uporabo ni`je temperature kaljenja za doseganje `eljene trdote. Vzorci v utrjenem stanju so bili primerjani z
upo{tevanjem velikosti prvotnih avstenitnih zrn, velikosti in razporeditve karbidnih delcev, kot tudi trdote.
Klju~ne besede: pospe{ena sferoidizacija, morfologija karbidnih zrn, kaljivost, jeklo za le`aje
1 INTRODUCTION
Bearing steels are well studied in terms of the effect
of the technological parameter on the final micro-
structural properties.1 The final properties of a hardened
product depend on the parameters of hardening, e.g., the
austenitization temperature and or the tempering tempe-
rature. Their influence on the hardness and prior-
austenite grain size was studied properly. Much less
attention was paid to the influence of the initial material
microstructure on the final material properties. The soft
annealed microstructure, used as the standard material
state for hardening, consists of globular carbide particles
with a size mostly from 0.5 μm to 1 μm dispersed in the
ferritic matrix. Two initial states were used for the
hardening – the conventional soft-annealed state and the
state after accelerated carbide spheroidisation and
refinement (ASR).2 The ASR state exhibits the same
microstructural morphology, but the globular carbides
are about 3-times smaller than after the soft annealing
and are also more densely spread in the ferritic matrix.3,4
Such a significantly finer structure provides for a faster
Materiali in tehnologije / Materials and technology 50 (2016) 1, 159–162 159
UDK 620.186:669.056:621.785 ISSN 1580-2949
Professional article/Strokovni ~lanek MTAEC9, 50(1)159(2016)
Table 1: Chemical composition of the 100CrMnSi6-4 bearing steel in mass fractions (w/%)
Tabela 1: Kemijska sestava 100CrMnSi6-4 jekla za le`aje v masnih dele`ih (w/%)
C Si Mn P S Cr Mo Ni Al Cu Fe
0.98 0.54 1.14 0.011 0.011 1.50 0.006 0.02 0.018 0.017 bal.
carbide-particle dissolution during the austenitization at
the quenching temperature3 and a stronger pinning of the
grain boundaries, thus leading to a smaller austenite
grain size due to the pinning effect and a martensitic
structure after the quenching. These microstructural
features can lead to a higher hardness and toughness
with the same hardening parameters. Another possibility
is to reduce the quenching temperature or austenitization
time, but still with good final mechanical properties.5,6
2 EXPERIMENTAL WORK
2.1 Material
The experimental material was the 100CrMnSi6-4
bearing steel grade with the chemical composition shown
in Table 1. The material was supplied as hot-rolled
21 mm-diameter bars, with a microstructure of pearlite
and a small amount of secondary cementite along the
austenite boundaries and a hardness of 383 HV10. The
samples were cut in the form of 30 mm long cylinders.
2.2 Spheroidisation annealing
The initial pearlitic microstructure was spheroidised
and samples with fine and coarse globular pearlite were
prepared. The coarse structure was obtained with the
conventional soft annealing in an atmospheric furnace at
a temperature of 805 °C for 11 h and air cooling. Fully
spheroidised pearlite consisted of cementite globules
with diameters of 0.3–1 μm in the ferritic matrix. The
fine structure was obtained with the ASR heat treatment
consisting of three temperature cycles. Each cycle
consisted of the induction heating to 780 °C, a 15-second
dwell period at this temperature and cooling in air to a
temperature of 650 °C. The overall annealing duration
was 5 min. The structure after the ASR process consisted
of globular carbides with a size of 0.1–0.3 μm dispersed
in the ferritic matrix.
2.3 Quenching and tempering
The heating to the quenching temperature was
performed in an electric atmospheric furnace. The
samples were held at the austenitization temperature for
30 min. Austenitization temperatures of (800, 820, 840
and 860) °C were used. The quenching was performed in
an oil bath, immediately followed by tempering. All the
samples were tempered for 4 h at a temperature of
240 °C.
2.4 Sample analyses
All the samples were cut longitudinally and
metallographic specimens were prepared by mechanical
grinding, polishing and Nital etching. The microstructure
was examined with SEM JEOL 7400F. The prior-auste-
nite grain size (PAGS) was assessed on the samples
etched in a picric-acid-saturated aqueous solution at a
temperature of 80 °C and the average grain diameter was
determined with a linear intercept procedure. The
hardness was measured with the Vickers method at a
load of HV10.
3 RESULTS AND DISCUSSION
3.1 Microstructure analyses
The microstructures after the soft annealing and ASR
treatment are shown in Figures 1 and 2. Much finer
carbide globules were formed during the ASR treatment.
Figures 3 and 4 show quenched-sample microstructures
(quenched from 800 °C). There is a clear difference in
the carbide density between the samples. This difference
is pronounced with higher austenitizing temperatures.
The ASR-treated samples retained a much denser
carbide distribution even after the quenching from 860 °C
(Figures 5 and 6). A higher dispersion strengthening
could cause a hardness increase in comparison with the
conventionally soft-annealed samples. Apparently, a
denser carbide distribution causes a pronounced pinning
effect on the prior-austenite grain boundaries. There
were coarse carbides retained in the soft-annealed
samples after the quenching from higher temperatures
J. DLOUHY et al.: INFLUENCE OF THE CARBIDE-PARTICLE SPHEROIDISATION PROCESS ON THE MICROSTRUCTURE ...
160 Materiali in tehnologije / Materials and technology 50 (2016) 1, 159–162
Figure 2: ASR-treated sample with finer carbides
Slika 2: ASR-obdelan vzorec z drobnimi karbidi
Figure 1: Conventionally soft-annealed sample
Slika 1: Obi~ajno mehko `arjen vzorec
(840 and 860 °C). Smaller carbides with a size of up to
0.5 μm were almost completely dissolved.
3.2 Hardness and PAGS
Different austenitization temperatures resulted in
different hardness values found after the quenching and
tempering and also in different PAGS values. There were
significant differences between the samples with coarse
and fine structures as shown in Figure 7.
There is a clear trend in the hardness value in the
case of the conventionally soft-annealed samples with
coarse carbides in the microstructure. A higher austeni-
tization temperature caused a hardness increase due to
the dissolution of a larger amount of carbon. On the
other hand, there is no clear trend in the case of the hard-
ness of the quenched ASR samples. The hardness after
the tempering increased with the quenching temperature
and was significantly higher with all the quenching
temperatures in comparison with the conventionally
soft-annealed samples.
J. DLOUHY et al.: INFLUENCE OF THE CARBIDE-PARTICLE SPHEROIDISATION PROCESS ON THE MICROSTRUCTURE ...
Materiali in tehnologije / Materials and technology 50 (2016) 1, 159–162 161
Figure 7: Hardness and PAGS of samples after quenching and tem-
pering
Slika 7: Trdota in PAGS vzorcev po kaljenju in po popu{~anju
Figure 5: Soft-annealed sample quenched from 800 °C
Slika 5: Mehko `arjen vzorec po kaljenju iz 800 °C
Figure 6: ASR-treated sample quenched from 800 °C
Slika 6: ASR-obdelan vzorec po kaljenju iz 800 °C
Figure 3: Soft-annealed sample quenched from 800 °C
Slika 3: Mehko `arjen vzorec po kaljenju iz 800 °C
Figure 4: ASR-treated sample quenched from 800 °C
Slika 4: ASR-obdelan vzorec po kaljenju iz 800 °C
The PAGS is much smaller in the case of the
ASR-treated samples. Its values grew with the increasing
quenching temperature for both sample types – the
ASR-treated and conventionally soft-annealed samples.
The finer-structured ASR-treated samples exhibited a
more pronounced PAGS increase. Finer carbides
dissolved at higher quenching temperatures and thus
their pinning effect on the austenite grain boundaries was
diminished. The difference between the PAGS values of
the ASR and soft-annealed samples decreased with the
rising quenching temperatures.
4 CONCLUSION
The microstructures and hardness values of hardened
samples with different initial microstructures were
compared. The influence of the carbide-particle density
is clearly visible in terms of the prior-austenite grain size
and hardness. The samples with fine carbides after the
ASR process had higher hardness values after the
quenching and tempering for the whole range of
examined quenching temperatures from 800 to 860 °C.
The hardness increase was about 40 HV10 in
comparison with the conventionally soft-annealed
samples with coarser carbide particles and was probably
caused by the dispersion strengthening and a possible
higher dissolution of the fine carbides during the
austenitization.
Acknowledgment
This paper was created by project Development of
West-Bohemian Centre of Materials and Metallurgy No.:
LO1412, financed by the MEYS of the Czech Republic.
5 REFERENCES
1 H. K. D. H. Bhadeshia, Steels for bearings, Progress in Materials
Science, 57 (2012), 268–435, doi:10.1016/j.pmatsci.2011.06.002
2 D. Hauserova, J. Dlouhy, Z. Novy, Microstructure Development of
Bearing Steel during Accelerated Carbide Spheroidisation, Materials
Science Forum, 782 (2014), 123–128, doi:10.4028/
www.scientific.net/MSF.782.123
3 J. H. Kang, P. E. J. Rivera-Díaz-del-Castillo, Carbide dissolution in
bearing steels, Computational Materials Science, 67 (2012),
364–372, doi:10.1016/j.commatsci.2012.09.022
4 D. Hauserova, J. Dlouhy, Z. Novy, J. Zrnik, Accelerated carbide
spheroidization and refinement (ASR) of the C45 steel during
induction heating, Mater. Tehnol., 47 (2013) 6, 701–705
5 H. Jirkova, D. Hauserova, L. Kucerova, B. Masek, Energy- and
time-saving low-temperature thermomechanical treatment of low-
carbon plain steel, Mater. Tehnol., 47 (2013) 3, 335–339
6 A. I. Katsamas, A computational study of austenite formation
kinetics in rapidly heated steels, Surface & Coatings Technology,
201 (2007), 6414–6422, doi:10.1016/j.surfcoat.2006.12.014
J. DLOUHY et al.: INFLUENCE OF THE CARBIDE-PARTICLE SPHEROIDISATION PROCESS ON THE MICROSTRUCTURE ...
162 Materiali in tehnologije / Materials and technology 50 (2016) 1, 159–162