9 S. Aydin, Development of a high-temperature-resistant mortar by
using slag and pumice, Fire Safety Journal, 43 (2008), 610–617,
doi:10.1016/j.firesaf.2008.02.001
10 Ý. Yüksel, R. Siddique, Ö. Özkan, Influence of high temperature on
the properties of concretes made with industrial by-products as fine
aggregate replacement, Construction and Building Materials, 25
(2011), 967–972, doi:10.1016/j.conbuildmat.2010.06.085
11 G. Yuan, Q. Li, The use of surface coating in enhancing the mecha-
nical properties and durability of concrete exposed to elevated
temperature, Construction and Building Materials, 95 (2015),
375–383, doi:10.1016/j.conbuildmat.2015.07.120
12 A. Nadeem, S. A. Memon, T. Yiu Lo, Mechanical performance,
durability, qualitative and quantitative analysis of microstructure of
fly ash and Metakaolin mortar at elevated temperatures, Construction
and Building Materials, 38 (2013), 338–347, doi:10.1016/
j.conbuildmat.2012.08.042
13 T. Harun, C. Ahmet, An experimental investigation of bond and
compressive strength of concrete with mineral admixtures at high
temperature, Arab. J. Sci. Eng., 33 (2008) 2B, 443–449
14 V. ^erný, [. Keprdová, Usability of fly ashes from Czech Republic
for sintered artificial aggregate, Advanced Materials Research,
(2014), 805–808
15 ^SN EN 1504-3, Products and systems for the protection and repair
of concrete structures – Definitions, requirements, quality control
and evaluation of conformity – Part 3: Structural and non-structural
repair, ^NI, 2006
16 ^SN EN 1363-1, Fire resistance tests – Part 1: General requirements,
^NI, 2013
17 ^SN 73 1326, including Z1, Resistance of cement concrete surface
to water and defrosting chemicals, ^NI, 2003
18 ^SN EN 12190, Products and systems for the protection and repair
of concrete structures – Test methods – Determination of com-
pressive strength of repair mortar, ^NI, 1999
19 ^SN EN 196-1, Methods of testing cement – Part 1: Determination
of strength, ^NI, 2005
T. MELICHAR et al.: DURABILITY OF MATERIALS BASED ON A POLYMER-SILICATE MATRIX ...
758 Materiali in tehnologije / Materials and technology 51 (2017) 5, 751–758
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
759–768
INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE
GRAIN-GROWTH BEHAVIOUR OF NEW Fe-Al BASED ALLOYS
WITH FINE Al2O3 PRECIPITATES
VPLIV TERMOMEHANSKE OBDELAVE FeAl ZLITIN S FINIMI
Al2O3 IZLO^KI NA RAST ZRN
Bohuslav Ma{ek1, Omid Khalaj1, Hana Jirková1, Jiøí Svoboda2,
Dagmar Bublíková1
1University of West Bohemia, Regional Technology Institute, Univerzitní 22, 306 14 Pilsen, Czech Republic
2Institute of Physics of Materials, Academy of Sciences of the Czech Republic, @i`kova 22, 616 62 Brno, Czech Republic
svobj@ipm.cz
Prejem rokopisa – received: 2016-07-20; sprejem za objavo – accepted for publication: 2017-04-20
doi:10.17222/mit.2016.232
To obtain the superior-high-temperature creep strength, a transformation of a fine-grained structure to large grains due to
abnormal grain growth or recrystallization is an important process in oxide-dispersion-strengthened (ODS) alloys. The
processing of steel is enabled with powder metallurgy, which utilizes powders consisting of a Fe-Al metal matrix with a large O
content, prepared with mechanical alloying, and their hot consolidation due to rolling. The thermomechanical characteristics of
new ODS alloys with a Fe-Al matrix are investigated in terms of the changes in the grain-size distribution. The recrystallization
and grain growth were quantified after heating up to 1200 °C, which is the typical consolidation temperature for standard
nanostructured ferritic steels. The results show that new ODS alloys are significantly affected by the thermomechanical
treatment leading to microstructural changes. Recrystallization is mostly affected by decreasing the deformation and increasing
the holding time, which leads to a growth of the grain size.
Keywords: grain growth, ODS alloys, steel, Fe-Al, Al2O3
Da bi pridobili najvi{jo temperaturo mo~i lezenja, je transformacija drobnozrnate strukture v ve~ja zrna ali celo v abnormalno
velika oziroma rekristalizacija pomemben proces pri zlitinah, oja~anih z oksidno disperzijo (angl. ODS). Obdelava jekla je
omogo~ena z metalurgijo v prahu, ki pomaga prahom, ki vsebujejo metalno matrico Fe-Al z veliko vsebnostjo kisika,
pripravljeno z mehanskim legiranjem in njihovo vro~o konsolidacijo pri valjanju. Termomehanske karakteristike novih
ODS-zlitin s Fe-Al matrico so bile preiskovane, ker je pri{lo do sprememb v velikosti porazdelitve zrn. Rekristalizacija in rast
zrn sta bili ovrednoteni po segretju do 1200 °C, ki je tipi~na temperatura konsolidacije za nanostrukturirana feritna jekla.
Rezultati ka`ejo, da na ODS-zlitine zelo vpliva termomehanska obdelava, ki pripelje do sprememb v mikrostrukturi. Na
rekristalizacijo najve~krat vpliva deformacija in pove~anje ~asa zadr`anosti, ki povzro~i rast velikost zrn.
Klju~ne besede: rast zrn, ODS-zlitine, jeklo, Fe-Al, Al2O3
1 INTRODUCTION
Historically, ODS alloys have been employed to
improve high-temperature mechanical properties. In
1910, the first utilization of an oxide dispersion was
reported by W. D. Coolidge. Using the classical powder
metallurgy, a tungsten-based alloy reinforced with tho-
rium oxides was developed to impede high-temperature
grain growth and therefore increase the life span of a
tungsten-filament lamp.1 After this first development,
several other applications with various metallic matrices
such as aluminium or nickel were implemented over the
decades. A notable improvement in the field was made
by J. S. Benjamin at the International Nickel Company
(INCO) laboratory: he proposed a new process based on
high-energy-milling powder metallurgy – later called
mechanical alloying.2 This process was introduced to
obtain a fine and homogeneous oxide dispersion within a
nickel matrix. Its aim was to produce high-temperature-
resistant materials for gas-turbine applications.3
Currently, mechanical alloying is still considered to
be the most effective process for obtaining fine and
homogeneously distributed particles. The volume frac-
tion of dispersed spherical oxides (usually Y2O3) is
typically below 1 % and the oxides typically have a
mean size of 5–30 nm. The mechanically alloyed powder
is then consolidated at high temperatures and pressures
to produce the bulk material in the form of a bar or tube
stock. Subsequently, different thermomechanical treat-
ments are applied to optimize its microstructure and
mechanical properties.
Oxides are much more resistant to coarsening in the
coarse-grained ferrite than the ’-precipitates in super-
alloys, and the limiting temperature for a long-term
operation is between 1000 °C and 1100 °C when the
mean size of oxides is 20–30 nm. This clearly indicates
that oxides are extremely stable and the microstructural
stability of ODS alloys is much higher than that of
nickel-based superalloys. However, a loss of mechanical
properties due to the coarsening of oxides was also
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 759
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
UDK 67.017:621.78:669.15 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 51(5)759(2017)
observed in fine-grained ODS steels at temperatures of
about 800 °C for the systems with an extremely fine
oxide dispersion (about 5 nm).4 Thus, the size of oxides
as well as the grain size are also important for the
stability of the microstructure – coarser oxides are much
more stable than fine ones, which is in line with the
cubic law for coarsening kinetics.
In the consolidation step, the processing temperatures
are critical in order to retain the nanocrystalline structure
generated during the mechanical alloying and to impede
the particle coarsening and grain growth.5–9 The Ni- and
F-based ODS alloys rely on the formation of slowly
growing and strongly adherent chromium and aluminium
scales for their high temperature oxidation/corrosion
resistance. In the present study, the ODS alloys consist of
a ferritic Fe-Al matrix strengthened with about 4 %
volume fraction of Al2O3 particles.10–15 In order to obtain
a more detailed insight into these new groups of
materials, this paper will concentrate on both the general
microstructure and mechanical properties of ODS steels,
the phenomenon of recrystallization and the related
microstructural evolutions.
2 EXPERIMENTAL PART
2.1 Material preparation
The classical processing route used to produce Fe-Al
based alloys with fine Al2O3 precipitates is highly
dependent on powder metallurgy (Figure 1). The first
step of the elaboration is the mechanical alloying of
powders consisting of Fe-11 w/% Al matrix (90 w/% Fe
and 10 w/% Al) and 1 w/% of O2 in gas, which is
absorbed, in a low-energy ball mill developed by the
authors, enabling evacuation and filling with oxygen. It
has two steel containers (each holding 24 L) and each
container is filled with 80 steel balls with a diameter of
40 mm. The revolution speed varies between 20–75
min–1. However, the speed for our research was kept
constant (75 min–1) for all the material preparations.
This step allows the powder to be forced into a solid
solution. The milled powder is then deposited into a steel
container with a diameter of 70 mm, evacuated
(degassed) and sealed by welding. The steel container is
then heated up to a temperature of 750 °C and rolled in a
hot rolling mill to a thickness of 20 mm in the first
rolling step, then heated up to a temperature of 900 °C
and rolled to a thickness of 8 mm in the second step. A
6-mm-thick sheet of the ODS alloy covered on both
sides with a 1-mm-thick scale from the rolling container
was produced in this way. Afterwards, the specimens
were cut with a water jet (Figure 5).
Eight types of material were used in this research as
described in Table 1. They are all based on a Fe-10w/%
Al ferritic matrix with different particle sizes and 4 %
volume fraction of Al2O3.
Al2O3 powder was added to prepare the composite;
fine oxides in the ODS alloys were obtained with the
internal oxidation during mechanical alloying and
precipitated during hot consolidation. It should be noted
that in the case of Material 1, the Al2O3 particles were
added, but in all the other materials, oxides were pro-
duced due to internal oxidation as shown in Figure 1.
Different sizes of the oxides in ODS steels are due to
different heat treatments after hot rolling. SEM obser-
vations indicated several inhomogeneities due to the
material sticking to the walls of the milling container
during mechanical alloying.
Table 1: Material parameters
Material
No.
Material
type
Milling
time
(hours)
Ferritic
matrix
Vol.%
of
Al2O3
Typical
particle
size
(nm)
1 ODS Alloy 100 Fe10wt%Al 10 300
2 ODS Alloy 100 Fe10wt%Al 6 50-200
3 ODS Alloy 150 Fe10wt%Al 6 50-150
4 ODS Alloy 200 Fe10wt%Al 6 30-150
5 ODS Alloy 245 Fe10wt%Al 7 20-50
6* ODS Alloy 245 Fe10wt%Al 7 20-50
7* ODS Alloy 245 Fe10wt%Al 7 20-50
8* ODS Alloy 245 Fe10wt%Al 7 20-50
* Different rolling conditions
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
760 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 1: Material-preparation process These inhomogeneities can also influence the mecha-
nical properties of the material, but the mechanical
alloying process is steadily optimized with respect to the
homogeneity of the materials.
2.2 Specimen preparation
Two specimen types (Figure 2) were selected from
several examples, which exhibited the most homo-
geneous temperature fields for specimen type 1 and
simplicity of tests for specimen type 2. Specimen type 1
was chosen from six types of specimen shapes with
different active parts in the middle. All the samples were
carefully monitored using a thermal camera to see how
the temperature field is distributed within the active part
when heated up to 1200 °C. So, regarding the results,
specimen type 1 (5a) has the best homogeneity regarding
the temperature distribution. On the other hand, speci-
men type 2 is designed to have four different deforma-
tions (5, 8, 20 and 50) % at the same time. The shortest
part (9 mm) is designed to have no deformation during
the pressing and the rest have appropriate values of
deformation. This specimen is held in a hydraulic
forging press with two positioning side plates (10 mm
thick) (Figure 3) in order not to apply more deformation
than required. The prepared containers were annealed at
1000 °C over 16 h. After normal cooling at room tem-
perature, all the specimens were cut by a water-jet ma-
chine in the longitudinal direction (Figure 4) and then all
the specimens were removed from the steel containers.
After grinding, the thickness of the specimens was
approximately 6 mm.
2.3 Testing equipment
In order to speed up the tests, a hydraulic forging
press (Figure 5) was used to apply four different defor-
mations at the same time on the type-2 specimens. Also,
in order to investigate the thermomechanical treatment of
the specimens, a servo-hydraulic MTS thermomecha-
nical simulator (Figure 6) was used, which allows
various temperature-deformation paths to be run in order
to find the conditions leading to, e.g., the most effective
grain coarsening due to recrystallization. The thermo-
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 761
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 4: Positions of specimens on a rolled semi-product: a) type 1,
b) type 2
Figure 2: Specimen dimensions in mm: a) type 1, b) type 2
Figure 3: Positions of specimens between holding plates Figure 5: Hydraulic forging press
observed in fine-grained ODS steels at temperatures of
about 800 °C for the systems with an extremely fine
oxide dispersion (about 5 nm).4 Thus, the size of oxides
as well as the grain size are also important for the
stability of the microstructure – coarser oxides are much
more stable than fine ones, which is in line with the
cubic law for coarsening kinetics.
In the consolidation step, the processing temperatures
are critical in order to retain the nanocrystalline structure
generated during the mechanical alloying and to impede
the particle coarsening and grain growth.5–9 The Ni- and
F-based ODS alloys rely on the formation of slowly
growing and strongly adherent chromium and aluminium
scales for their high temperature oxidation/corrosion
resistance. In the present study, the ODS alloys consist of
a ferritic Fe-Al matrix strengthened with about 4 %
volume fraction of Al2O3 particles.10–15 In order to obtain
a more detailed insight into these new groups of
materials, this paper will concentrate on both the general
microstructure and mechanical properties of ODS steels,
the phenomenon of recrystallization and the related
microstructural evolutions.
2 EXPERIMENTAL PART
2.1 Material preparation
The classical processing route used to produce Fe-Al
based alloys with fine Al2O3 precipitates is highly
dependent on powder metallurgy (Figure 1). The first
step of the elaboration is the mechanical alloying of
powders consisting of Fe-11 w/% Al matrix (90 w/% Fe
and 10 w/% Al) and 1 w/% of O2 in gas, which is
absorbed, in a low-energy ball mill developed by the
authors, enabling evacuation and filling with oxygen. It
has two steel containers (each holding 24 L) and each
container is filled with 80 steel balls with a diameter of
40 mm. The revolution speed varies between 20–75
min–1. However, the speed for our research was kept
constant (75 min–1) for all the material preparations.
This step allows the powder to be forced into a solid
solution. The milled powder is then deposited into a steel
container with a diameter of 70 mm, evacuated
(degassed) and sealed by welding. The steel container is
then heated up to a temperature of 750 °C and rolled in a
hot rolling mill to a thickness of 20 mm in the first
rolling step, then heated up to a temperature of 900 °C
and rolled to a thickness of 8 mm in the second step. A
6-mm-thick sheet of the ODS alloy covered on both
sides with a 1-mm-thick scale from the rolling container
was produced in this way. Afterwards, the specimens
were cut with a water jet (Figure 5).
Eight types of material were used in this research as
described in Table 1. They are all based on a Fe-10w/%
Al ferritic matrix with different particle sizes and 4 %
volume fraction of Al2O3.
Al2O3 powder was added to prepare the composite;
fine oxides in the ODS alloys were obtained with the
internal oxidation during mechanical alloying and
precipitated during hot consolidation. It should be noted
that in the case of Material 1, the Al2O3 particles were
added, but in all the other materials, oxides were pro-
duced due to internal oxidation as shown in Figure 1.
Different sizes of the oxides in ODS steels are due to
different heat treatments after hot rolling. SEM obser-
vations indicated several inhomogeneities due to the
material sticking to the walls of the milling container
during mechanical alloying.
Table 1: Material parameters
Material
No.
Material
type
Milling
time
(hours)
Ferritic
matrix
Vol.%
of
Al2O3
Typical
particle
size
(nm)
1 ODS Alloy 100 Fe10wt%Al 10 300
2 ODS Alloy 100 Fe10wt%Al 6 50-200
3 ODS Alloy 150 Fe10wt%Al 6 50-150
4 ODS Alloy 200 Fe10wt%Al 6 30-150
5 ODS Alloy 245 Fe10wt%Al 7 20-50
6* ODS Alloy 245 Fe10wt%Al 7 20-50
7* ODS Alloy 245 Fe10wt%Al 7 20-50
8* ODS Alloy 245 Fe10wt%Al 7 20-50
* Different rolling conditions
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
760 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 1: Material-preparation process These inhomogeneities can also influence the mecha-
nical properties of the material, but the mechanical
alloying process is steadily optimized with respect to the
homogeneity of the materials.
2.2 Specimen preparation
Two specimen types (Figure 2) were selected from
several examples, which exhibited the most homo-
geneous temperature fields for specimen type 1 and
simplicity of tests for specimen type 2. Specimen type 1
was chosen from six types of specimen shapes with
different active parts in the middle. All the samples were
carefully monitored using a thermal camera to see how
the temperature field is distributed within the active part
when heated up to 1200 °C. So, regarding the results,
specimen type 1 (5a) has the best homogeneity regarding
the temperature distribution. On the other hand, speci-
men type 2 is designed to have four different deforma-
tions (5, 8, 20 and 50) % at the same time. The shortest
part (9 mm) is designed to have no deformation during
the pressing and the rest have appropriate values of
deformation. This specimen is held in a hydraulic
forging press with two positioning side plates (10 mm
thick) (Figure 3) in order not to apply more deformation
than required. The prepared containers were annealed at
1000 °C over 16 h. After normal cooling at room tem-
perature, all the specimens were cut by a water-jet ma-
chine in the longitudinal direction (Figure 4) and then all
the specimens were removed from the steel containers.
After grinding, the thickness of the specimens was
approximately 6 mm.
2.3 Testing equipment
In order to speed up the tests, a hydraulic forging
press (Figure 5) was used to apply four different defor-
mations at the same time on the type-2 specimens. Also,
in order to investigate the thermomechanical treatment of
the specimens, a servo-hydraulic MTS thermomecha-
nical simulator (Figure 6) was used, which allows
various temperature-deformation paths to be run in order
to find the conditions leading to, e.g., the most effective
grain coarsening due to recrystallization. The thermo-
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 761
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 4: Positions of specimens on a rolled semi-product: a) type 1,
b) type 2
Figure 2: Specimen dimensions in mm: a) type 1, b) type 2
Figure 3: Positions of specimens between holding plates Figure 5: Hydraulic forging press
mechanical simulator also allows a combination of
tensile and compressive deformation, thus accumulating
a high plastic deformation (and a high dislocation den-
sity) in a specimen.
2.4 Testing programme
The testing programme involved two different
groups. The tests are summarized in Table 2.
Tests in group A were carried out to investigate the
thermomechanical behaviour of different materials (1–8)
at different temperatures, looking at a single tensile and
compression loading with a constant strain rate of 1 s–1
(Figure 7). In order to give a clearer comparison of the
results, only the results at room temperature (RT), (800,
1000 and 1200) °C are presented.
Tests in group B were performed to investigate the
effects of the holding time at 1000 °C and the defor-
mation percentage on materials 7 and 8 with the type-2
specimen shape (Figure 2b). The specimen was de-
signed in order to apply four different deformations (5, 8,
20 and 50 %) at the same time (Figure 8), and sub-
sequently one specimen from each material was
quenched in water for immediate cooling and the rest
were kept in the furnace at three different holding times
(10, 20, 40) h, after which they were cooled slowly to
room temperature.
Treatment No. 1 was designed to apply three
different single deformations on each sample at different
temperatures. The specimen is first heated up to the
desired temperature in 5 min and is then held for another
five minutes. Immediately after these 10 min, the first
deformation (5 % compression) is applied to the speci-
men and then the temperature is held for another five
minutes. The second deformation (3 % tension) is
applied immediately after 5 min and again the tempe-
rature is held for five minutes and after that the last
deformation (50 % tension) is applied and the specimen
is left to cool down to room temperature.
Treatment No. 2 was designed to apply four different
deformations (5, 8, 20 and 50) % at the same time on a
specimen. First, the specimen is heated to the desired
temperature in 5 minutes and immediately after that, the
deformation is applied. The temperature is kept for
different holding times (0, 10, 20, 40) h and then the
specimen is left to cool down to room temperature.
3 RESULTS AND DISCUSSION
3.1 Test group A
Figure 9 shows the stress-strain curves for all the
materials at different temperatures, with 5 % com-
pression, corresponding to Treatment No. 1. Material 2
exhibits better strength at 30 °C (almost 500 MPa) and
800 °C (almost 300 MPa); at 1200 °C, material 1 still
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
762 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
Figure 7: Treatment No. 1
Figure 6: Treatment with a thermomechanical simulator
Table 2: Parameters of the testing programme
Test group MaterialNo.
Treatment
No.
Holding
time (h)
Maximum temperature
(°C)
Number of
tests Purpose of tests
A 1, 2, 3, 4, 5,6, 7, 8 1 –
1200, 1100, 1000, 900,
800, RT 48
Investigation of
thermomechanical behaviour
B 7,8 2
0, 10,
20, 40
1000 40 Investigation of grain-size growth
Figure 8: Treatment No. 2
shows better strength (almost 60 MPa), but not very
different from material 2 (almost 50 Mpa). It seems that
the difference in the compression strength between these
two materials is almost 37 % at RT; however, this
difference increased to 137 % at 800 °C. On the other
hand, all the materials have almost the same elastic
modulus and none of them fails at the 5 % compression.
However, material 4 shows a strange behaviour at
800 °C. The microstructure analysis shows that this
occurred because of an inhomogeneous formation of the
particles in the specimen.
The hot-working behaviour of alloys is generally
reflected by flow curves, which are a direct consequence
of microstructural changes: the nucleation and growth of
new grains, dynamic recrystallization (DRX), the gene-
ration of dislocations, work hardening (WH), the
rearrangement of dislocations and their dynamic
recovery (DRV). In deformed materials, DRX seems to
be the prominent softening mechanism at high tempe-
ratures. DRX occurs during the straining of metals at
high temperatures, characterized by a nucleation of low-
dislocation-density grains and their posterior growth,
producing a homogeneous grain structure when dynamic
equilibrium is reached. Material 4 showed a strange
curve shape at 800 °C. The test was repeated several
times and similar behaviour was observed each time. It
was concluded that this happens because of the
inhomogeneity of the microstructure of this material.
Figure 10 shows the stress-strain curves at different
temperatures corresponding to the 3 % tension of Treat-
ment No. 1. As can be seen in Figure 10, material 2
shows a higher strength at 30 °C (almost 650 MPa) and
800 °C (almost 270 MPa), but at 1200 °C, material 1
again shows a better strength (almost 65 MPa). This
value was 35 % lower when the temperature was in-
creased. In Figure 10, the strain at the maximum tension
is almost 2 % at RT (Figure 10a), then it increases to
almost 3 % at 800 °C (Figure 10b); however, it de-
creases to almost 1 % at 1200 °C (Figure 10c). All the
materials have almost the same elastic modulus and none
of them failed at the 3 % deformation. The yield stress
and the shape of the flow curves are sensitive to tempe-
rature. Comparing all these curves, it is found that
decreasing the deformation temperature increases the
yield-stress level. In other words, it prevents softening
due to dynamic recrystallization (DRX) and dynamic
recovery (DRV) and allows the deformed metals to
exhibit work hardening (WH). For every curve, after a
rapid increase in the stress to the peak value, the flow
stress decreases monotonically towards a steady-state
regime with a varying softening rate, which typically
indicates the onset of DRX, Figure 10c.
Figure 11 shows the stress-strain curves at different
temperatures for the 50 % tension of Treatment No. 1.
All eight materials failed at RT, but only four materials
failed below the 50 % tension at higher temperatures.
Material 2 failed at the 34 % strain, materials 4 and 5
failed at around 45 % strain and material 8 failed at
around 50 % at 800 °C. At 1200 °C, only material 1
failed at 41 % and material 2 failed at the 45 % strain.
From these curves, it can also be seen that the stress evo-
lution with strain exhibits three distinct stages.
Work hardening (WH) predominates in the first stage
and causes dislocations to polygonise into stable sub-
grains. The flow stress exhibits a rapid increase with the
increasing strain up to the critical value. Then DRX
occurs due to a large difference in the dislocation density
within subgrains or grains. When the critical driving
force of DRX is attained, new grains are nucleated along
the grain boundaries, deformation bands and disloca-
tions, resulting in the formation of equiaxed DRX grains.
In the second stage, the flow stress exhibits a smaller
and smaller increase until the peak value, or an inflection
of the work-hardening rate is reached. This shows that
thermal softening becomes more and more important due
to DRX and dynamic recovery (DRV) and it exceeds
WH.
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 763
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 10: Stress-strain curves (3 % tension) for: a) RT, b) 800 °C, c)
1000 °C, d) 1200 °C
Figure 9: Stress-strain curves (5 % compression) for: a) RT, b)
800 °C, c) 1000 °C, d) 1200 °C
mechanical simulator also allows a combination of
tensile and compressive deformation, thus accumulating
a high plastic deformation (and a high dislocation den-
sity) in a specimen.
2.4 Testing programme
The testing programme involved two different
groups. The tests are summarized in Table 2.
Tests in group A were carried out to investigate the
thermomechanical behaviour of different materials (1–8)
at different temperatures, looking at a single tensile and
compression loading with a constant strain rate of 1 s–1
(Figure 7). In order to give a clearer comparison of the
results, only the results at room temperature (RT), (800,
1000 and 1200) °C are presented.
Tests in group B were performed to investigate the
effects of the holding time at 1000 °C and the defor-
mation percentage on materials 7 and 8 with the type-2
specimen shape (Figure 2b). The specimen was de-
signed in order to apply four different deformations (5, 8,
20 and 50 %) at the same time (Figure 8), and sub-
sequently one specimen from each material was
quenched in water for immediate cooling and the rest
were kept in the furnace at three different holding times
(10, 20, 40) h, after which they were cooled slowly to
room temperature.
Treatment No. 1 was designed to apply three
different single deformations on each sample at different
temperatures. The specimen is first heated up to the
desired temperature in 5 min and is then held for another
five minutes. Immediately after these 10 min, the first
deformation (5 % compression) is applied to the speci-
men and then the temperature is held for another five
minutes. The second deformation (3 % tension) is
applied immediately after 5 min and again the tempe-
rature is held for five minutes and after that the last
deformation (50 % tension) is applied and the specimen
is left to cool down to room temperature.
Treatment No. 2 was designed to apply four different
deformations (5, 8, 20 and 50) % at the same time on a
specimen. First, the specimen is heated to the desired
temperature in 5 minutes and immediately after that, the
deformation is applied. The temperature is kept for
different holding times (0, 10, 20, 40) h and then the
specimen is left to cool down to room temperature.
3 RESULTS AND DISCUSSION
3.1 Test group A
Figure 9 shows the stress-strain curves for all the
materials at different temperatures, with 5 % com-
pression, corresponding to Treatment No. 1. Material 2
exhibits better strength at 30 °C (almost 500 MPa) and
800 °C (almost 300 MPa); at 1200 °C, material 1 still
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
762 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
Figure 7: Treatment No. 1
Figure 6: Treatment with a thermomechanical simulator
Table 2: Parameters of the testing programme
Test group MaterialNo.
Treatment
No.
Holding
time (h)
Maximum temperature
(°C)
Number of
tests Purpose of tests
A 1, 2, 3, 4, 5,6, 7, 8 1 –
1200, 1100, 1000, 900,
800, RT 48
Investigation of
thermomechanical behaviour
B 7,8 2
0, 10,
20, 40
1000 40 Investigation of grain-size growth
Figure 8: Treatment No. 2
shows better strength (almost 60 MPa), but not very
different from material 2 (almost 50 Mpa). It seems that
the difference in the compression strength between these
two materials is almost 37 % at RT; however, this
difference increased to 137 % at 800 °C. On the other
hand, all the materials have almost the same elastic
modulus and none of them fails at the 5 % compression.
However, material 4 shows a strange behaviour at
800 °C. The microstructure analysis shows that this
occurred because of an inhomogeneous formation of the
particles in the specimen.
The hot-working behaviour of alloys is generally
reflected by flow curves, which are a direct consequence
of microstructural changes: the nucleation and growth of
new grains, dynamic recrystallization (DRX), the gene-
ration of dislocations, work hardening (WH), the
rearrangement of dislocations and their dynamic
recovery (DRV). In deformed materials, DRX seems to
be the prominent softening mechanism at high tempe-
ratures. DRX occurs during the straining of metals at
high temperatures, characterized by a nucleation of low-
dislocation-density grains and their posterior growth,
producing a homogeneous grain structure when dynamic
equilibrium is reached. Material 4 showed a strange
curve shape at 800 °C. The test was repeated several
times and similar behaviour was observed each time. It
was concluded that this happens because of the
inhomogeneity of the microstructure of this material.
Figure 10 shows the stress-strain curves at different
temperatures corresponding to the 3 % tension of Treat-
ment No. 1. As can be seen in Figure 10, material 2
shows a higher strength at 30 °C (almost 650 MPa) and
800 °C (almost 270 MPa), but at 1200 °C, material 1
again shows a better strength (almost 65 MPa). This
value was 35 % lower when the temperature was in-
creased. In Figure 10, the strain at the maximum tension
is almost 2 % at RT (Figure 10a), then it increases to
almost 3 % at 800 °C (Figure 10b); however, it de-
creases to almost 1 % at 1200 °C (Figure 10c). All the
materials have almost the same elastic modulus and none
of them failed at the 3 % deformation. The yield stress
and the shape of the flow curves are sensitive to tempe-
rature. Comparing all these curves, it is found that
decreasing the deformation temperature increases the
yield-stress level. In other words, it prevents softening
due to dynamic recrystallization (DRX) and dynamic
recovery (DRV) and allows the deformed metals to
exhibit work hardening (WH). For every curve, after a
rapid increase in the stress to the peak value, the flow
stress decreases monotonically towards a steady-state
regime with a varying softening rate, which typically
indicates the onset of DRX, Figure 10c.
Figure 11 shows the stress-strain curves at different
temperatures for the 50 % tension of Treatment No. 1.
All eight materials failed at RT, but only four materials
failed below the 50 % tension at higher temperatures.
Material 2 failed at the 34 % strain, materials 4 and 5
failed at around 45 % strain and material 8 failed at
around 50 % at 800 °C. At 1200 °C, only material 1
failed at 41 % and material 2 failed at the 45 % strain.
From these curves, it can also be seen that the stress evo-
lution with strain exhibits three distinct stages.
Work hardening (WH) predominates in the first stage
and causes dislocations to polygonise into stable sub-
grains. The flow stress exhibits a rapid increase with the
increasing strain up to the critical value. Then DRX
occurs due to a large difference in the dislocation density
within subgrains or grains. When the critical driving
force of DRX is attained, new grains are nucleated along
the grain boundaries, deformation bands and disloca-
tions, resulting in the formation of equiaxed DRX grains.
In the second stage, the flow stress exhibits a smaller
and smaller increase until the peak value, or an inflection
of the work-hardening rate is reached. This shows that
thermal softening becomes more and more important due
to DRX and dynamic recovery (DRV) and it exceeds
WH.
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 763
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 10: Stress-strain curves (3 % tension) for: a) RT, b) 800 °C, c)
1000 °C, d) 1200 °C
Figure 9: Stress-strain curves (5 % compression) for: a) RT, b)
800 °C, c) 1000 °C, d) 1200 °C
In the third stage, three types of curves can be re-
cognized:
• A gradual decrease to a steady state with DRX soft-
ening (Figure 11c).
• A continuous increase with significant work-harden-
ing (Figure 11a).
• A continuous decrease with significant DRX soften-
ing (Figure 11b).
3.2 Test group B
Figures 12 to 19 show the microstructures of
materials 7 and 8 without deformation and with the 50 %
deformation at different annealing times of (0, 10, 20 and
40) h at 1000 °C. It can be seen that by increasing the
annealing time, the number of low-angle grain boun-
daries (<15°) decreased. This indicates that recrystalli-
zation does not occur at the zones with shorter annealing
times (Figure 12a). A few banding zones with a high
density of low-angle grain boundaries are observed in
Figures 14a and 16a showing that recrystallization
occurs in most zones. Figure 18a shows that the banding
zones with a high density of low-angle grain boundaries
cannot be observed, indicating that almost completely
recrystallized ferrite grains can be obtained after
annealing at 1000 °C for 40 h.
Similar results were achieved for the rest of the
specimens; however, there is a major difference between
the microstructures. Material 7 is partially recrystallized,
but material 8 retained fine grains. Both materials were
prepared using the same procedure, but with different
rolling pressures during the hot-rolling process. During
the rolling process, a specimen passes through different
stages. In the first stage of rolling, the material is under a
considerable mechanical stress, resulting from an in-
ternally balanced elastic strain. This elastic strain is
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
764 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 11: Stress-strain curves (50 % tension) for: a) RT, b) 800 °C,
c) 1000 °C, d) 1200 °C
Figure 12: Microstructure for material 7 without annealing a) without
deformation, b) with 50 % deformation
Figure 13: Microstructure of material 8 without annealing: a) without
deformation, b) with 50 % deformation
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 765
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 16: Microstructure for material 7 with 20 h annealing: a) with-
out deformation, b) with 50 % deformation
Figure 14: Microstructure for material 7 after 10 h annealing: a) with-
out deformation, b) with 50 % deformation
Figure 17: Microstructure for material 8 h with 20 h annealing:
a) without deformation, b) with 50 % deformation
Figure 15: Microstructure for material 8 after 10 h annealing: a) with-
out deformation, b) with 50 % deformation
In the third stage, three types of curves can be re-
cognized:
• A gradual decrease to a steady state with DRX soft-
ening (Figure 11c).
• A continuous increase with significant work-harden-
ing (Figure 11a).
• A continuous decrease with significant DRX soften-
ing (Figure 11b).
3.2 Test group B
Figures 12 to 19 show the microstructures of
materials 7 and 8 without deformation and with the 50 %
deformation at different annealing times of (0, 10, 20 and
40) h at 1000 °C. It can be seen that by increasing the
annealing time, the number of low-angle grain boun-
daries (<15°) decreased. This indicates that recrystalli-
zation does not occur at the zones with shorter annealing
times (Figure 12a). A few banding zones with a high
density of low-angle grain boundaries are observed in
Figures 14a and 16a showing that recrystallization
occurs in most zones. Figure 18a shows that the banding
zones with a high density of low-angle grain boundaries
cannot be observed, indicating that almost completely
recrystallized ferrite grains can be obtained after
annealing at 1000 °C for 40 h.
Similar results were achieved for the rest of the
specimens; however, there is a major difference between
the microstructures. Material 7 is partially recrystallized,
but material 8 retained fine grains. Both materials were
prepared using the same procedure, but with different
rolling pressures during the hot-rolling process. During
the rolling process, a specimen passes through different
stages. In the first stage of rolling, the material is under a
considerable mechanical stress, resulting from an in-
ternally balanced elastic strain. This elastic strain is
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
764 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 11: Stress-strain curves (50 % tension) for: a) RT, b) 800 °C,
c) 1000 °C, d) 1200 °C
Figure 12: Microstructure for material 7 without annealing a) without
deformation, b) with 50 % deformation
Figure 13: Microstructure of material 8 without annealing: a) without
deformation, b) with 50 % deformation
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 765
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 16: Microstructure for material 7 with 20 h annealing: a) with-
out deformation, b) with 50 % deformation
Figure 14: Microstructure for material 7 after 10 h annealing: a) with-
out deformation, b) with 50 % deformation
Figure 17: Microstructure for material 8 h with 20 h annealing:
a) without deformation, b) with 50 % deformation
Figure 15: Microstructure for material 8 after 10 h annealing: a) with-
out deformation, b) with 50 % deformation
added to the jamming of the dislocation, which occurred
during cold forming.
In Figures 12, 14, 16 and 18, there is a visible alter-
nation in the distorted shape of the cold-work crystals at
the stress-relief stage (Figures 12a, 14a, 16a and 18a).
In this stage, new crystals begin to grow in the deformed
crystals. In the next stage (Figures 12b, 14b, 16b and
18b), the small crystals that formed in the previous stage
gradually grow into bigger crystals by absorbing each
other in the cannibal fashion, thus making the structure
relatively coarse grained.
Figures 12 to 19 show that with the increasing strain,
a cellblock structure gradually develops and the sizes of
the cellblocks and the cells decrease. In other words, in
material 7 (Figures 12, 14, 16 and 18), there is a great
transformation. It can be seen that with the increasing
strain, there are changes in the spacing of the dense
dislocation walls and micro-bands (DDW-MBs) and in
the cell size. It is seen that after a 50 % deformation, the
spacing of the DDW–MBs decreased to almost 50 μm,
close to the cell size. In addition, the rate of the decrease
in spacing with the increasing strains is much larger for
geometrically necessary boundaries (GNBs) than for
incidental dislocation boundaries (IDBs). However, in
material 8 (Figures 13, 15, 17 and 19), there was no
serious transformation and the structure still shows
ferrite and pearlite. The grain size is relatively similar for
both deformations. The grain boundaries can no longer
be seen clearly, as the ferrite precipitated with the pear-
lite, creating a grey shade instead of a clear black-and-
white contrast.
Figure 20 gives an overview of the grain sizes of
materials 7 and 8 with different deformations and
annealing times. It can be seen that as material 7 is
almost recrystallized, it has a bigger grain size than
material 8. From Figure 20 it is clear that by increasing
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
766 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 20: Grain size for: a) material 7, b) material 8
Figure 19: Microstructure for material 8 h with 40 h annealing: a)
without deformation, b) with 50 % deformation
Figure 18: Microstructure for material 7 with 40 h annealing: a) with-
out deformation, b) with 50 % deformation
the annealing time, the grain size increases significantly
for material 7 (Figure 20a), however the rate of increase
in grain size is lower for material 8 (Figure 20b). The
grain size reached almost 120 μm after 40 hours
annealing without any deformation in material 7, while
the grains in material 8 reached almost 10 μm under the
same conditions. On the other hand, by increasing the
applied deformation, the grain size of both materials
decreased smoothly. Material 7 reached 50 μm by apply-
ing 50 % deformation without annealing while material 8
reached almost 3 μm under the same conditions. One can
thus conclude that the stability of grain microstructure
can be significantly influenced by the processing. The
analysis of the reasons for this will be the topic of future
work by the team.
4 CONCLUSIONS
If you mean that why we didn’t present the micro-
structe of material 1-6, because all the materials except 7
has fine microstructure similar to material 8, that is why
we decide to present only the microstructure results from
material 7 and 8 and compare them in this view point.
This paper outlines the influence of thermomecha-
nical treatment on the grain growth of new Fe-Al based
alloys with fine Al2O3 precipitates. Eight materials
differing in the amount and size of the oxides embedded
in the ferritic matrix were tested under different condi-
tions. The advantages of all the materials are their
low-cost, simplicity of preparation and significant me-
chanical properties together with micro structures, due to
the Fe-Al based ferritic matrix of the ODS alloy. The
results from material 1-6 described in the relevant
section (3.1. test group A) and the results from material
7-8 in case of microstructure described in section (3.2
test group B). As all the materials except 7 has fine
microstructure similar to material 8, only the micro-
structure results from material 7 and 8 compared in this
view point. It can be concluded that in general the oxide
dispersion significantly strengthens the material.
However, the typical form of the flow curve with DRX
softening, including a single peak followed by a steady
state flow as a plateau, is more recognizable at high
temperatures than at low temperatures. This is because at
high temperatures the DRX softening compensates for
the work hardening (WH), and both the peak stress and
the onset of steady state flow are therefore shifted to
lower strain levels. The characteristics of softening flow
behaviour coupled with DRX were investigated for eight
materials and can be summarized as follows:
Decreasing deformation temperature causes the flow
stress level to increase. In other words, it prevents the
occurrence of softening due to DRX and dynamic
recovery (DRV) and causes the deformed metals to
exhibit work hardening (WH).
For every curve, after a rapid increase in the stress to
a peak value, the flow stress decreases monotonically
towards a steady state regime (a steady state flow as a
plateau due to DRX softening is more recognizable at
higher temperatures). A varying softening rate typically
indicates the onset of DRX, and the stress evolution with
strain exhibits three distinct stages.
At higher temperatures, a higher DRX softening
compensates the WH, and both the peak stress and the
onset of steady state flow are therefore shifted to lower
strain levels.
The elastic part of the total strain amplitude is always
higher than the plastic part in all specimens tested, even
for the highest total strain amplitudes of 15 %. This is
further confirmation of the strong strengthening effect of
oxide particles.
Material 7 is more crystallised than material 8. Thus,
it has a larger grain size compared to the fine grains of
material 8.
Acknowledgements
This paper includes results from projects 14-24252S
Preparation and Optimization of Creep Resistant
Submicron-Structured Composite with Fe-Al Matrix and
Al2O3 Particles subsidised by the Czech Science Founda-
tion.
5 REFERENCES
1 M. Mohan, R. Subramanian, Z. Alam, P. C. Angelo, Evaluation of
the Mechanical Properties OF A Hot Isostatically Pressed Yttria-
Dispersed Nickel-Based Superalloy, Mater. Tehnol., 48 (2014) 6,
899–904
2 W. Quadakkers, Oxidation of ODS alloys, Journal de Physique IV, 03
(1993) C9, 177–186, doi:10.1051/jp4:1993916
3 F. Pedraza, Low Energy-High Flux Nitridation of Metal Alloys:
Mechanisms, Microstructures and High Temperatures Oxidation
Behaviour, Mater. Tehnol., 42 (2008) 4, 157–169
4 O. Khalaj, B. Ma{ek, H. Jirkova, A. Ronesova, J. Svoboda, Investi-
gation on New Creep and Oxidation Resistant Materials, Mater.
Tehnol., 49 (2015) 4, 173–179, doi:10.17222/mit.2014.210
5 F. D. Fischer, J. Svoboda, P. Fratzl, A thermodynamic approach to
grain growth and coarsening, Journal of Philosophical Magazine, 83
(2003) 9, 1075–1093, doi:10.1080/0141861031000068966
6 M. J. Alinger, G. R. Odette, D. T. Hoelzer, On the role of alloy com-
position and processing parameters in nanocluster formation and
dispersion strengthening in nanostuctured ferritic alloys, Acta
Material, 57 (2009) 2, 392–406, doi:10.1016/j.actamat.2008.09.025
7 P. Unifantowicz, Z. Oksiuta, P. Olier, Y. de Carlan, N. Baluc,
Microstructure and mechanical properties of an ODS RAF steel
fabricated by hot extrusion or hot isostatic pressing, Fusion Engi-
neering and Design, 86 (2011), 2413–2416, doi:10.1016/j.fusengdes.
2011.01.022
8 M. A. Auger, V. de Castro, T. Leguey, A. Muñoz, R. Pareja, Micro-
structure and mechanical behavior of ODS and non-ODS Fe-14Cr
model alloys produced by spark plasma sintering, Journal of Nuclear
Materials, 436 (2013) 5, 68–75, doi:10.1016/j.jnucmat.2013.01.331
9 M. Kos, J. Fer~ec, M. Brun~ko, R. Rudolf, I. An`el, Pressing of
Partially Oxide-Dispersion-Strenghtened Copper using the ECAP
Process, Mater. Tehnol., 48 (2014) 3, 379–384, UDK
621.777.2:669.35’71
10 B. Ma{ek, O. Khalaj, Z. Nový, T. Kubina, H. Jirkova, J. Svoboda, C.
[tádler, Behaviour of New ODS Alloys under Single and Multiple
Deformation, Mater. Tehnol., 50 (2016) 6, 891–898, doi:10.17222/
mit.2015.156
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 767
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
added to the jamming of the dislocation, which occurred
during cold forming.
In Figures 12, 14, 16 and 18, there is a visible alter-
nation in the distorted shape of the cold-work crystals at
the stress-relief stage (Figures 12a, 14a, 16a and 18a).
In this stage, new crystals begin to grow in the deformed
crystals. In the next stage (Figures 12b, 14b, 16b and
18b), the small crystals that formed in the previous stage
gradually grow into bigger crystals by absorbing each
other in the cannibal fashion, thus making the structure
relatively coarse grained.
Figures 12 to 19 show that with the increasing strain,
a cellblock structure gradually develops and the sizes of
the cellblocks and the cells decrease. In other words, in
material 7 (Figures 12, 14, 16 and 18), there is a great
transformation. It can be seen that with the increasing
strain, there are changes in the spacing of the dense
dislocation walls and micro-bands (DDW-MBs) and in
the cell size. It is seen that after a 50 % deformation, the
spacing of the DDW–MBs decreased to almost 50 μm,
close to the cell size. In addition, the rate of the decrease
in spacing with the increasing strains is much larger for
geometrically necessary boundaries (GNBs) than for
incidental dislocation boundaries (IDBs). However, in
material 8 (Figures 13, 15, 17 and 19), there was no
serious transformation and the structure still shows
ferrite and pearlite. The grain size is relatively similar for
both deformations. The grain boundaries can no longer
be seen clearly, as the ferrite precipitated with the pear-
lite, creating a grey shade instead of a clear black-and-
white contrast.
Figure 20 gives an overview of the grain sizes of
materials 7 and 8 with different deformations and
annealing times. It can be seen that as material 7 is
almost recrystallized, it has a bigger grain size than
material 8. From Figure 20 it is clear that by increasing
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
766 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 20: Grain size for: a) material 7, b) material 8
Figure 19: Microstructure for material 8 h with 40 h annealing: a)
without deformation, b) with 50 % deformation
Figure 18: Microstructure for material 7 with 40 h annealing: a) with-
out deformation, b) with 50 % deformation
the annealing time, the grain size increases significantly
for material 7 (Figure 20a), however the rate of increase
in grain size is lower for material 8 (Figure 20b). The
grain size reached almost 120 μm after 40 hours
annealing without any deformation in material 7, while
the grains in material 8 reached almost 10 μm under the
same conditions. On the other hand, by increasing the
applied deformation, the grain size of both materials
decreased smoothly. Material 7 reached 50 μm by apply-
ing 50 % deformation without annealing while material 8
reached almost 3 μm under the same conditions. One can
thus conclude that the stability of grain microstructure
can be significantly influenced by the processing. The
analysis of the reasons for this will be the topic of future
work by the team.
4 CONCLUSIONS
If you mean that why we didn’t present the micro-
structe of material 1-6, because all the materials except 7
has fine microstructure similar to material 8, that is why
we decide to present only the microstructure results from
material 7 and 8 and compare them in this view point.
This paper outlines the influence of thermomecha-
nical treatment on the grain growth of new Fe-Al based
alloys with fine Al2O3 precipitates. Eight materials
differing in the amount and size of the oxides embedded
in the ferritic matrix were tested under different condi-
tions. The advantages of all the materials are their
low-cost, simplicity of preparation and significant me-
chanical properties together with micro structures, due to
the Fe-Al based ferritic matrix of the ODS alloy. The
results from material 1-6 described in the relevant
section (3.1. test group A) and the results from material
7-8 in case of microstructure described in section (3.2
test group B). As all the materials except 7 has fine
microstructure similar to material 8, only the micro-
structure results from material 7 and 8 compared in this
view point. It can be concluded that in general the oxide
dispersion significantly strengthens the material.
However, the typical form of the flow curve with DRX
softening, including a single peak followed by a steady
state flow as a plateau, is more recognizable at high
temperatures than at low temperatures. This is because at
high temperatures the DRX softening compensates for
the work hardening (WH), and both the peak stress and
the onset of steady state flow are therefore shifted to
lower strain levels. The characteristics of softening flow
behaviour coupled with DRX were investigated for eight
materials and can be summarized as follows:
Decreasing deformation temperature causes the flow
stress level to increase. In other words, it prevents the
occurrence of softening due to DRX and dynamic
recovery (DRV) and causes the deformed metals to
exhibit work hardening (WH).
For every curve, after a rapid increase in the stress to
a peak value, the flow stress decreases monotonically
towards a steady state regime (a steady state flow as a
plateau due to DRX softening is more recognizable at
higher temperatures). A varying softening rate typically
indicates the onset of DRX, and the stress evolution with
strain exhibits three distinct stages.
At higher temperatures, a higher DRX softening
compensates the WH, and both the peak stress and the
onset of steady state flow are therefore shifted to lower
strain levels.
The elastic part of the total strain amplitude is always
higher than the plastic part in all specimens tested, even
for the highest total strain amplitudes of 15 %. This is
further confirmation of the strong strengthening effect of
oxide particles.
Material 7 is more crystallised than material 8. Thus,
it has a larger grain size compared to the fine grains of
material 8.
Acknowledgements
This paper includes results from projects 14-24252S
Preparation and Optimization of Creep Resistant
Submicron-Structured Composite with Fe-Al Matrix and
Al2O3 Particles subsidised by the Czech Science Founda-
tion.
5 REFERENCES
1 M. Mohan, R. Subramanian, Z. Alam, P. C. Angelo, Evaluation of
the Mechanical Properties OF A Hot Isostatically Pressed Yttria-
Dispersed Nickel-Based Superalloy, Mater. Tehnol., 48 (2014) 6,
899–904
2 W. Quadakkers, Oxidation of ODS alloys, Journal de Physique IV, 03
(1993) C9, 177–186, doi:10.1051/jp4:1993916
3 F. Pedraza, Low Energy-High Flux Nitridation of Metal Alloys:
Mechanisms, Microstructures and High Temperatures Oxidation
Behaviour, Mater. Tehnol., 42 (2008) 4, 157–169
4 O. Khalaj, B. Ma{ek, H. Jirkova, A. Ronesova, J. Svoboda, Investi-
gation on New Creep and Oxidation Resistant Materials, Mater.
Tehnol., 49 (2015) 4, 173–179, doi:10.17222/mit.2014.210
5 F. D. Fischer, J. Svoboda, P. Fratzl, A thermodynamic approach to
grain growth and coarsening, Journal of Philosophical Magazine, 83
(2003) 9, 1075–1093, doi:10.1080/0141861031000068966
6 M. J. Alinger, G. R. Odette, D. T. Hoelzer, On the role of alloy com-
position and processing parameters in nanocluster formation and
dispersion strengthening in nanostuctured ferritic alloys, Acta
Material, 57 (2009) 2, 392–406, doi:10.1016/j.actamat.2008.09.025
7 P. Unifantowicz, Z. Oksiuta, P. Olier, Y. de Carlan, N. Baluc,
Microstructure and mechanical properties of an ODS RAF steel
fabricated by hot extrusion or hot isostatic pressing, Fusion Engi-
neering and Design, 86 (2011), 2413–2416, doi:10.1016/j.fusengdes.
2011.01.022
8 M. A. Auger, V. de Castro, T. Leguey, A. Muñoz, R. Pareja, Micro-
structure and mechanical behavior of ODS and non-ODS Fe-14Cr
model alloys produced by spark plasma sintering, Journal of Nuclear
Materials, 436 (2013) 5, 68–75, doi:10.1016/j.jnucmat.2013.01.331
9 M. Kos, J. Fer~ec, M. Brun~ko, R. Rudolf, I. An`el, Pressing of
Partially Oxide-Dispersion-Strenghtened Copper using the ECAP
Process, Mater. Tehnol., 48 (2014) 3, 379–384, UDK
621.777.2:669.35’71
10 B. Ma{ek, O. Khalaj, Z. Nový, T. Kubina, H. Jirkova, J. Svoboda, C.
[tádler, Behaviour of New ODS Alloys under Single and Multiple
Deformation, Mater. Tehnol., 50 (2016) 6, 891–898, doi:10.17222/
mit.2015.156
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768 767
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
11 Marmy, P., Kruml, T., Low cycle fatigue of Eurofer 97, Journal of
Nuclear Materials, 377 (2008) 1, 52–58, doi:10.1016/j.jnucmat.
2008.02.054
12 M. Mi{ovi}, N. Tadi}, M. Ja}imovi}, M. Janji}, Deformations and
Velocities during the Cold Rolling of Aluminium Alloys, Mater.
Tehnol., 50 (2016) 1, 59-67, doi:10.17222/mit.2014.250
13 A. Grajcar, Microstructure Evolution of Advanced High-Strength
Trip-Aided Bainitic Steel, Mater. Tehnol., 49 (2015) 5, 715–720,
doi:10.17222/mit.2014.154
14 B. [u{tar{i~, I. Paulin, M. Godec, S. Glode`, M. [ori, J. Flasker, A.
Koro{ec, S. Kores, G. Abramovi~, DSC/TG of Al-based Alloyed
Powders for P/M Applications, Mater. Tehnol., 48 (2014) 4, 439–450
15 F. Tehovnik, J. Burja, B. Podgornik, M. Godec, F. Vode, Micro-
structural evolution of Inconel 625 during hot rolling, Mater. Tehnol.,
49 (2015) 5, 899–904, doi:10.17222/mit.2015.274
B. MA[EK et al.: INFLUENCE OF THERMOMECHANICAL TREATMENT ON THE GRAIN-GROWTH BEHAVIOUR ...
768 Materiali in tehnologije / Materials and technology 51 (2017) 5, 759–768
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
K. MATUS et al.: ANALYSIS OF PRECIPITATES IN ALUMINIUM ALLOYS WITH THE USE OF HIGH-RESOLUTION ...
769–773
ANALYSIS OF PRECIPITATES IN ALUMINIUM ALLOYS WITH
THE USE OF HIGH-RESOLUTION ELECTRON MICROSCOPY
AND COMPUTER SIMULATION
RAZISKAVE OBORIN V ALUMINIJEVIH ZLITINAH Z
VISOKORESOLUCIJSKO ELEKTRONSKO MIKROSKOPIJO IN
RA^UNALNI[KO SIMULACIJO
Krzysztof Matus, Anna Tomiczek, Klaudiusz Go³ombek, Miros³awa Pawlyta
Silesian University of Technology, Institute of Engineering Materials and Biomaterials, 18A Konarskiego Str.,
44100 Gliwice, Poland
krzysztof.matus@polsl.pl
Prejem rokopisa – received: 2016-07-27; sprejem za objavo – accepted for publication: 2017-01-27
doi:10.17222/mit.2016.226
This paper presents the results of the tests using high-resolution transmission electron microscopy (HRTEM) with both
transmission and scanning modes, as well as energy-dispersive-spectroscopy (EDS) investigation of the AlSi9Cu alloy after
laser surface remelting. The possibility of using a computer simulation to identify precipitates in the analysed alloy was also
explored. The obtained results and computer simulations were compared. Moreover, this article presents the advantages of the
computer aid in solving the diffraction patterns and precipitates in supercell simulations.
Keywords: precipitation, crystallography, TEM, computer simulations
V prispevku so predstavljeni rezultati preiskave AlSi9Cu zlitine z laserskim pretaljevanjem povr{in z uporabo transmisijske
elektronske mikroskopije z visoko lo~ljivostjo (angl. HRTEM) na dva na~ina: s transmisijskim skeniranjem, kot tudi z
energijsko disperzijsko spektroskopijo (angl. EDS). Z uporabo ra~unalni{ke simulacije je bila preiskana mo`nost identifikacije
delcev v analizirani zlitini. Dobljeni rezultati in ra~unalni{ke simulacije so bili primerjani. Poleg tega ~lanek predstavlja
prednosti ra~unalni{ke simulacije pri {tudiji vzorcev difrakcije in oborin v posameznih stanjih celic.
Klju~ne besede: oborine, kristalografija, TEM, ra~unalni{ke simulacije
1 INTRODUCTION
Aluminium alloys are the most widely used alloys in
modern technology, mainly due to high specific strength
and low density. Other factors behind the widespread
distribution of aluminium alloys are their excellent
electrical conductivity and high corrosion resistance.1,2
The use of laser surface treatment to improve the utility
properties of aluminium alloys allows the remelting of
the surface layer of the material or its enrichment with
alloy elements. The remelting of the material surface and
rapid crystallisation allow an improvement of the mecha-
nical properties of alloys, mainly in the field of abrasion
resistance and tribological properties of the surface.
Through the generation of plenty of fine precipitates, the
strength is increased as well. The influence of alloying
element precipitates on aluminium alloys is a complex
issue and it is still one of the most promising topics in
materials science.3–7
Laser surface treatment ensures that the processed
material obtains new properties due to the rapid disper-
sion of the heat from the melted zone. This phenomenon
enables the crystallisation of very fine precipitates to
occur. Laser remelting can be used for small and large
elements. The most common alloyed layers have a thick-
ness from 0.3 3 mm to about 3 mm. A layer formed by
remelting usually has a high homogeneity as well as a
fine crystalline structure. The use of the laser technology
makes it possible to obtain surface layers with high con-
tents of alloying elements and a unique combination of
elements that conventional alloys rarely contain.8–11
Generally, the laser surface treatment is accom-
plished with two methods, which differ from one another
based on how the alloying addition is introduced to the
surface layer (which is schematically presented in
Figure 1). When the surface of a material is subjected to
a laser beam and then melted, the process is called
remelting (Figure 1a). When an alloying material is
applied to a surface and then melted with a laser beam
(most of the materials are added in the forms of tapes,
pastes or powders), the process is called alloying (Figure
1b).12,13
This article aims to identify the precipitates in an alu-
minium alloy after the laser treatment using transmission
electron microscopy and computer simulations.
2 EXPERIMENTAL PART
For the experimental procedure, cast AlSi9Cu alumi-
nium alloy was used. Its average chemical composition
is shown in Table 1. This alloy was subjected to
Materiali in tehnologije / Materials and technology 51 (2017) 5, 769–773 769
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
UDK 620.1:669.715:621.385.833:004.942 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 51(5)769(2017)