P. JIMBERT et al.: CARBIDE DISTRIBUTION BASED ON AUTOMATIC IMAGE ANALYSIS ...
609–611
CARBIDE DISTRIBUTION BASED ON AUTOMATIC IMAGE
ANALYSIS FOR CRYOGENICALLY TREATED TOOL STEELS
PRIKAZ PORAZDELITVE KARBIDNIH DELCEV V ORODNIH
JEKLIH, OBDELANIH S PODHLAJEVANJEM S POMO^JO
AVTOMATSKE ANALIZE SLIK
Pello Jimbert, Maider Iturrondobeitia, Julen Ibarretxe,
Roberto Fernandez-Martinez
University of the Basque Country (UPV/EHU), Faculty of Engineering, Paseo Rafael Moreno Pitxitxi, 348013, Bilbao, Spain
pello.jimbert@ehu.eus
Prejem rokopisa – received: 2016-07-14; sprejem za objavo – accepted for publication: 2016-09-29
doi:10.17222/mit.2016.165
Particle distribution in steels has been widely reported in the scientific literature in terms of quantity and size of these particles.
However, in these studies there is barely any information about the homogeneity or heterogeneity of the particle distribution. A
cryogenically treated AISI A8 tool steel was compared against a conventionally treated one, not only in terms of the number and
size of precipitated carbides, but also their distribution. Scanning-electron-microscopy images were automatically analyzed for
this quantification. The results revealed an increase of 100 % in the number of carbides and a reduction of 35 % in the mean size
of the precipitates for the cryogenically treated sample. The distribution of the particles – which turned out to be more
homogeneous for the cryogenically treated sample – was characterized using the methodology described here, which applies
automatic image analysis. These results may lead to an improvement in the wear resistance of a cryogenically treated AISI A8
alloy.
Keywords: particle distribution, automatic image analysis, tool steel, cryogenic treatment
Porazdelitev karbidnih delcev v jeklih glede na koli~ino in velikost je ob{irno predstavljena v znanstveni literaturi. Toda te
{tudije prakti~no ne podajajo informacije o homogenosti oz. heterogenosti porazdelitve teh delcev. [tevilo, velikost in
porazdelitev karbidnih delcev v podhlajenem orodnem jeklu AISI A8 smo primerjali z vzorci konvencionalno pridobljenega
jekla. Slike, posnete z vrsti~nim elektronskim mikroskopom, smo za potrebe kvantifikacije avtomatsko analizirali. Rezultati
ka`ejo 100 % pove~anje {tevila karbidnih delcev in zmanj{anje povpre~ne velikosti karbidnih precipitatov za 35 % v
podhlajenih vzorcih. Avtomatska analiza slik, razvita za potrebe te {tudije, podaja informacijo z izra~unom porazdelitve delcev.
Predstavljena {tudija nakazuje na bolj homogeno porazdelitev karbidnih delcev v podhlajenih vzorcih. Na{i rezultati bi lahko
pripomogli k izbolj{anju obrabne odpornosti podhlajenih AISI A8 zlitin.
Klju~ne besede: porazdelitev delcev, avtomatska analiza slik, orodno jeklo, obdelava s podhlajevanjem
1 INTRODUCTION
There have been reports of carbide1 or oxide2,3
precipitate quantity and size distribution in steels using
image-processing software, even for particle sizes at the
nanoscale4 but not so many about the homogeneity of the
distribution of these particles in the microstructure.
Some studies show the distribution of these particles
qualitatively using different techniques without a clear
explanation of the methodology they followed.5 One of
the shortcomings of this kind of particle-distribution
calculation is the large amount of particles (and therefore
images as well) needed to have a statistically represen-
tative quantification. In this work, an automatic
methodology was developed to quantitatively analyze the
microstructure of AISI A8 tool steels, by analyzing a
statistically representative number of images of the bulk
material.
2 MATERIALS AND METHODOLOGY
The methodology for characterizing the carbide
distribution presented in this work was developed for the
AISI A8 tool steel. It is a widely used tool-steel alloy
that exhibits a high hardenability, good wear resistance,
resistance to dynamic loading and toughness. Some of its
applications include cold forming, blanking and bending
dies, knurling tools, forming rolls, master dies and gages.
Its chemical composition can be seen in Table 1.
Two samples of the AISI A8 were analyzed. Sample
1 underwent a conventional heat-treatment process of
austenization at 1050 °C for 40 min, air quenching and
Materiali in tehnologije / Materials and technology 51 (2017) 4, 609–611 609
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
UDK 620.1:546.261:62-712 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 51(4)609(2017)
Table 1: Chemical composition of the AISI A8 steel
Material C Mn Si Cr Ni Mo W
AISI A8 0.5-0.6 0.5 0.75-1.1 4.75-5.5 0.3 1.15-1.65 1-1.5
two tempering processes of 90 min at 500 °C. Sample 2
underwent an additional final-cyclic cryogenic treatment
of 6 cycles from –73 °C to –172 °C taking 15 h after the
conventional heat treatment. The cryogenic treatment is
applied to different tool steels for the improvement of the
wear resistance and this improvement is related to the
precipitation of carbides.6 The influence of the cryogenic
treatment on the precipitation of carbides of the AISI A8
medium-alloy tool steel was investigated for the first
time in this work.
Scanning electron microscope (SEM) images were
used to calculate the carbide distribution and microstruc-
tural parameters such as the number of particles, their
mean sizes and the total area. The images were acquired
with a JEOL JSM-6400F SEM model at 10 KV and a
500× magnification. The pixel size was 9.2 pixel/μm.
Ten images of 200 μm × 200 μm, acquired in different
areas, were analyzed for each sample. Considering the
resulting average sizes and their corresponding
deviations (Figure 1), the number of analyzed particles
(900 for sample 1 and 1800 for sample 2) is sufficient to
statistically ensure that the average values are accurate
within ± 0.1 μm at a 95 % confidence.
The automatic image analysis was carried out using
Fiji.7 First of all, the images were binarized using the
Auto Local Threshold plug-in with different algorithms
(Bernsen, Mean, MidGrey, Niblack and Sauvola). The
Sauvola algorithm was chosen since it gave the most
suitable binarization of the carbides over the martensitic
matrix, based on our experience in this field as can be
seen in Figure 1. The SEM images were then automa-
tically analyzed using the Sauvola algorithm and the
following information for the carbides was obtained:
centroid positions, the number of particles and sizes. To
calculate the distribution of the carbides, their centroids
were used, computing the distances to the four closest
neighbors and their averages using an in-house made
script in Matlab.8
3 RESULTS AND DISCUSSION
The density of carbide particles increases for the
cryogenically treated sample (sample 2) from 0.026
particles/μm2 to 0.054 particles/μm2. This indicates an
increase of around 100 % in the number of particles.
This result is in good agreement with the results obtained
for different tool steels, which show an increase in the
number of carbides when they are cryogenically treated.9
This increase in the number of carbides is related, in the
literature, with an increase in the wear resistance.10 The
observed effect is very promising regarding the
possibility to include the cryogenic treatment of the AISI
A8 tool steel when used for severe friction applications.
Also, the mean particle length decreases from
6.2±1.3 to 4.1±1.5 μm. This result is in concordance
with those reported for other tool steels after being
cryogenically treated.10 D. Das et al.11 concluded that the
cryogenic treatment creates new carbide segregates
during the cooling and this segregates grow during the
heating from the cryogenic temperature to the ambient
temperature creating small secondary carbides. The
small size of these new carbides reduces the mean size of
the precipitates present in the material.
Regarding the distribution of the particles (Figure 2),
the distribution of the newly precipitated carbides is
more homogeneous for sample 2. The percentage of the
area occupied by the carbides increases from 0.4 % for
the conventionally treated sample to 0.6 % for the cryo-
genically treated sample. The final microstructure of the
P. JIMBERT et al.: CARBIDE DISTRIBUTION BASED ON AUTOMATIC IMAGE ANALYSIS ...
610 Materiali in tehnologije / Materials and technology 51 (2017) 4, 609–611
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 2: Carbide distribution for the conventionally treated (sample
1) and for the cryogenically treated (sample 2) samplesFigure 1: a) Original SEM image and b) binarized image
cryogenically treated sample has more carbides, they
cover a bigger area and they are more homogeneously
distributed than in the conventionally treated one.
4 CONCLUSION
The microstructure is quantitatively measured using a
methodology based on automatic image analysis. The
methodology used for the carbide-distribution calcula-
tion developed in this study was satisfactorily applied to
tool steels and the obtained results support the suitability
of this methodology.
The distribution of the carbides precipitated during
the cryogenic treatment of the AISI A8 steel is more
homogenous, indicating a more homogeneous final
microstructure.
The AISI A8 tool steel showed a potential to be cryo-
genically treated since there is an increase in the number
of precipitated carbides, reducing their mean sizes and
covering a 43 % larger area, which may lead to an
increase in the wear resistance.
Acknowledgements
This work was supported by the Department of
Industry, Innovation, Trade and Tourism of the Basque
Country Government through the S-PE12UN080
SAIOTEK project. The authors thank for the technical
and human support provided by SGIker of UPV/EHU
and European funding (ERDF and ESF).
5 REFERENCES
1 C. Revilla, B. López, J. M. Rodriguez Ibabe, Carbide size refinement
by controlling the heating rate during induction tempering in a low
alloy steel, Materials and Design, 62 (2014), 296–304, doi:10.1016/
j.matdes.2014.05.053
2 S. F. Li , Z. J. Zhou, P. H. Wang, H.Y. Sun, M. Wang, G. M. Zhang,
Long-term thermal-aging stability of a 16Cr-oxide dispersion
strengthened ferritic steel at 973 K, Materials and Design, 90 (2016),
318–329, doi:10.1016/j.matdes.2015.10.138
3 M. Nagini, R. Vijay, M. Ramakrishna, A.V. Reddy, G. Sundararajan,
Influence of the duration of high energy ball milling on the
microstructure and mechanical properties of a 9Cr oxide dispersion
strengthened ferritic–martensitic steel, Materials Science & Engi-
neering A, 620 (2015), 490–499, doi:10.1016/j.msea.2014.10.050
4 C. Solenthaler, M. Ramesh, P. J. Uggowitzer, R. Spolenak, Preci-
pitation strengthening of Nb-stabilized TP 347 austenitic steel by a
dispersion of secondary Nb (C,N) formed upon a short-term
hardening heat treatment, Materials Science & Engineering A, 647
(2015), 294–302, doi:10.1016/j.msea.2015.09.028
5 S. Amirkhanlou, B. Niroumand, Effects of reinforcement distribution
on low and high temperature tensile properties of Al356/SiCp cast
composites produced by a novel reinforcement dispersion technique,
Materials Science and Engineering A, 528 (2011), 7186–7195,
doi:10.1016/j.msea.2011.06.013
6 P. Baldissera, C. Delprete, Deep cryogenic treatment: a bibliographic
review, The Open Mechanical Engineering Journal, 2 (2008), 1–11,
doi:10.2174/1874155X00802010001
7 J. Schindelin, I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair,
T. Pietzsch, S. Preibisch, C. Rueden, S. Saalfeld, B. Schmid, J.-Y.
Tinevez, D. J. White, V. Hartenstein, K. Eliceiri, P. Tomancak, A.
Cardona, Fiji: an open-source platform for biological-image analysis,
Nat. Methods, 9 (2012) 7, 676–82, doi:10.1038/nmeth.2019
8 MATLAB Release 2014b, The MathWorks, Inc., Natick, Massachu-
setts, USA
9 J. Y. Huang, Y. T. Zhu, X. Z. Liao, I. J. Beyerlein, M. A. Bourke, T.
E. Mitchell, Microstructure of cryogenically treated M2 tool steel,
Materials Science and Engineering A, 339 (2003), 241–244,
doi:10.1016/S0921-5093(02)00165-X
10 S. Akincioðlu, H. Gökkaya, I. Uygur, A review of cryogenic
treatment on cutting tools, Int. J. Adv. Manuf. Technol, 78 (2015),
1609–1627, doi:10.1007/s00170-014-6755-x
11 D. Das, A. K. Dutta, K. K. Ray, Sub-zero treatments of AISI D2
steel: Part I. Microstructure and hardness, Materials Science and
Engineering A, 527 (2010), 2182–2193, doi:10.1016/j.msea.2009.
10.070
P. JIMBERT et al.: CARBIDE DISTRIBUTION BASED ON AUTOMATIC IMAGE ANALYSIS ...
Materiali in tehnologije / Materials and technology 51 (2017) 4, 609–611 611
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS