Kovove Mater.

55 2017 133–139 133

DOI: 10.4149/km 2017 2 133

Effect of microstructure on corrosion behavior of dual-phase steels

E. Salamci1 *, S. Candan2 , F. Kabakci3

1
Department of Mechanical Engineering, Gazi University, 06570 Ankara, Turkey
2
Department of Mechanical and Manufacturing Engineering, Bilecik University, Bilecik, Turkey
3
Alapli Vocational High School, Bulent Ecevit University, Zonguldak, Turkey

Received 16 December 2015, received in revised form 7 April 2016, accepted 15 April 2016

Abstract
Dual phase (DP) steels with different volume fractions of martensite and epitaxial ferrite
were produced in different intercritical annealing temperatures and cooling rates. Corrosion
behaviors of four different DP steels were investigated in 3.5 % NaCl solution by using po-
tentiodynamic polarization and immersion tests. The results showed that both amounts of
martensite and epitaxial ferrite have an influence on the corrosion behavior of DP steels. It
was observed that corrosion resistance increased considerably with the presence of epitaxial
ferrite as compared to dual phase steel containing non-epitaxial ferrite. On the other hand, the
corrosion resistance of DP steels decreased with an increase in the martensite volume fraction
(MVF).

K e y w o r d s : dual-phase steel, epitaxial ferrite, martensite, corrosion

1. Introduction tion products which can be formed from the austen-

ite subsequent intercritical annealing by varying inter-
Low carbon Dual Phase (DP) steels developed in critical annealing temperature and cooling rate. The
the 1970s have aroused considerable attention in the austenite entirely transforms to martensite resulting
last three decades. Compared to the High Strength in the duplex structure of ferrite and martensite at
Low Alloy (HSLA) steels, they display improved the higher cooling rates. During the slow cooling rates,
strength, ductility and formability characteristics the austenite firstly transforms to new ferrite which
[1–7]. These properties make them attractive for grows epitaxially on the retained or intercritical fer-
weight-saving applications in automotive industry. By rite without a nucleation step. Then, depending on the
using dual phase steel sheets, vehicle weight can be alloy content and cooling rate, the remaining austen-
reduced, and significant savings in fuel consumption ite transforms to martensite, bainite, and/or pearlite
can be obtained. One another application area might with continued cooling [10]. Transformed ferrite from
be the construction sector where reinforcing steels in the austenite during the slow cooling is also called
concrete would be desirable to have both superior me- epitaxial ferrite. It has been shown that there is no
chanical properties and corrosion resistance. structural interface between the two types of ferrite,
DP steel is delineated by the composite microstruc- and the epitaxial ferrite differs from the retained fer-
ture comprising 20–25 % hard martensite particles in rite in terms of composition [11, 12]. Previous investi-
a soft ferritic matrix. They are mainly produced by gations have shown that ductility of dual phase steels
heating the low carbon steel between AC1 and AC3 increases with a small decrease in the strength, as the
temperatures in the austenite plus ferrite phase field epitaxial ferrite content increases [13–15].
followed by rapid cooling in order to transform the Ferritic-martensitic steels which possess excellent
austenite to martensite. The strength of DP steels in- mechanical properties can respond to higher corrosion
creases with increasing the volume fraction of marten- resistance requirements which cannot be obtained in
site phase [8, 9]. ferritic-pearlitic steels [16]. In addition to their incom-
It is possible to control the range of transforma- parable combination of mechanical properties, the cor-
*Corresponding author: tel.: +90 312 5823452; fax: +90 312 2319810; e-mail address: [email protected]
134 E. Salamci et al. / Kovove Mater. 55 2017 133–139

T a b l e 1. Chemical composition of the steel (wt.%)

C Mn P S Si Cu Cr Ni

0.09 1.65 0.015 0.013 0.55 0.02 0.07 0.69

rosion resistance of DP steels is critically important for 10 × 10 mm2 size were heat treated. Each speci-
structural and constructional purposes. It is inevitable men, namely WQ715, FC725, WQ725, and FC752,
that the reinforcing steels in concrete suffer from cor- was coded according to the applied heat treatment
rosion over time. For this reason, the investigation of procedures. The left-hand letters represent the cool-
the steels with higher corrosion resistance has been ing media, and the right-hand numbers represent the
attracting the attention of researchers. Although ex- intercritical annealing temperatures.
tensive researches concerning microstructure and me- The heat treated samples were ground, polished
chanical properties exist in the literature, the investi- and etched in 2 % nital solution of 2 % nitric acid in
gations on corrosion behavior of DP steels are quite a 98 % ethanol for microstructural characterization by
few. Trejo et al. [17] have studied corrosion susceptibil- using a light microscope. After etching with 2 % nital,
ity of a DP steel in concrete and observed that the DP the samples were further etched in a solution of 10 %
steel bar has better atmospheric corrosion resistance sodium metabisulfite in 90 % pure water in order to
than a standard billet bar consisting of ferrite and reveal the martensite structure. The boiling alkaline
pearlite phases. Zhang et al. [18] also observed better chromate etching method developed by Lawson et al.
corrosion resistance in the DP weathering steel. Sarkar was used to differentiate the epitaxial ferrite from the
et al. [19] have investigated electrochemical corrosion retained ferrite [23]. This method is a two-step pro-
behavior of micro-alloyed DP steel in 3.5 % NaCl so- cedure consisting of a first etch in a solution of 4 %
lution and determined that an increase in marten- picral for 40–60 s, followed by a second etch in a boil-
site content and refinement of martensite morphol- ing alkaline chromate solution for 4–10 min. The vol-
ogy decreases corrosion resistance. Osorio et al. [20] ume fraction of martensite and epitaxial ferrite were
have also studied electrochemical corrosion behavior determined by using a point counting on etched met-
of low carbon steel in three different heat treated and allographic sections. Almost 1000 points were counted
as-received conditions by carrying out electrochemical by using Swift mark point counter.
impedance spectroscopy (EIS) and polarization test. The immersion tests were carried out by suspend-
They found that normalized and annealed steels have ing the square shaped samples (10 × 10 mm2 and
the highest and lowest corrosion resistances respecti- 1.5 mm thick) in a still solution of 3.5 wt.% NaCl in
vely and the DP steel has slightly smaller corrosion re- deionized water exposed to atmospheric air. The re-
sistance than annealed and as-received steels. Bhaga- sults of the corrosion tests were evaluated by mass
vathi et al. [21] have recently reported that DP steels loss measurements, performed following the ASTM
are more resistant to corrosion than the ferrite-pearlite G31 recommended practice. The surfaces of the spec-
steel. imens were abraded on 1200 mesh emery paper, pol-
The main purpose of this study is to explore the ished with 1 µm diamond paste, washed and ultrason-
effect of the epitaxial ferrite on the corrosion behavior ically rinsed in distilled water prior to the immersion
of DP steel which has not been thoroughly assessed in tests. Exposure time for the tests was 1, 3, 7, 14, and
the literature yet. In this work, corrosion behaviors of 28 days. The surfaces of the specimens were gently
DP steel microstructures which are produced at two brushed before ultrasonically cleaning in distilled wa-
different intercritical annealing temperatures followed ter at the end of each exposure time. After cleaning,
by rapid and slow cooling rates are investigated. they were dried and weighed. Mass loss of the sam-
ples was then normalized in the unit of mg cm−2 , by
considering the total surface area of the samples. For
2. Experimental procedure the potentiodynamic polarization measurements, ma-
chined samples of 15 mm in length and 10 mm in di-
The chemical composition of the steel used in the ameter were connected to the copper wire and embed-
experiments is given in Table 1. The material was pro- ded in an epoxy resin holder. The potentiodynamic
duced in an induction furnace and poured into sil- curves were performed by means of a Gamry model
ica sand mold as a block of 26 × 300 × 300 mm3 in PC4/300 mA potentiostat/galvanostat controlled by
size. The block was then cut into smaller pieces of a computer with DC105 corrosion analysis software.
26 × 26 × 300 mm3 . These pieces were hot rolled in The specimens were prepared in the same way as that
which the thickness of each was reduced to 4 mm. for the mass loss evaluation. Embedded specimens in
The critical annealing temperatures, AC1 and epoxy resin were utilized as working electrodes. A car-
AC3 , were identified as 711 and 862 ◦C by using bon rod (6 mm in diameter) and a saturated calomel
Andrew’s empirical formulas [22]. The samples of electrode (SCE) were used as a counter electrode and
 E. Salamci et al. / Kovove Mater. 55 2017 133–139 135

Fig. 1. Optical micrographs of (a) WQ715, (b) FC725, (c) WQ725, (d) FC752 DP steel samples (etched in nital 2 %).

a reference electrode, respectively. Experiments were pending on cooling rate. Optical micrographs of the
performed at room temperature in a glass cell con- WQ715, FC725, WQ725, and FC752 etched with ni-
taining 3.5 % NaCl solution, pH ∼= 7. For electrochem- tal 2 % are displayed in Figs. 1a-d, respectively. In
ical measurements, the specimen was immersed in the these micrographs, martensite appears as dark color
solution for 60 min. prior to polarization, by which in a light ferrite matrix. It can be seen that martensite
the time for a stable potential (open circuit potential particles distribute along the ferrite grain boundaries.
“OCP”) was monitored. With reference to saturated In the specimens WQ715 and WQ725, most of the
calomel electrode (SCE), potentiodynamic polariza- austenite transformed to martensite by rapid quench-
tion curves were generated by sweeping the potential ing from 715 and 725 ◦C. Martensite was replaced
from cathodic to the anodic direction at a scan rate by epitaxial ferrite with the decreasing cooling rates.
of 1 mV s−1 , starting from ∼ 0.40 up to 0.80 V. Each Slower cooling from 725 and 752 ◦C resulted in the for-
data point for both immersion and potentiodynamic mation of epitaxial ferrite in addition to martensite.
polarization tests represents at least average of three Retained and epitaxial ferrites were distinguished by
different measurements. Corrosion potentials (Ecorr ) using Lawson etchant. Figures 2a,b show the micro-
and the corrosion current densities (Icorr ) were cal- graphs of the FC725 and FC752 etched with 4 % picral
culated by using instantaneous Tafel-type fit Gamry- + hot alkaline chromate revealing martensite as black,
DC105 Corrosion analysis software. epitaxial ferrite white and retain ferrite gray. It can be
seen from Figs. 2a,b that martensite is surrounded by
a ring of epitaxial ferrite.
3. Results and discussion In the present study, the critical cooling rate for
martensite formation is quite low when compared to
3.1. Micro structure previous studies [23–27]. Figures 1b,d show that a sig-
nificant amount of austenite transformed to marten-
Austenite at intercritical annealing temperatures site even at the lower cooling rates (furnace cooling).
transforms to martensite and epitaxial ferrite de- The reason for this may be due to high Mn and Ni
136 E. Salamci et al. / Kovove Mater. 55 2017 133–139

T a b l e 2. Applied heat treatment procedure and quantitative metallographic data for DP steel samples

Specimen code Intercritical annealing temperature ( ◦C) Cooling media Martensite (%) Epitaxial ferrite (%)

WQ715 715 Water 25 –

FC725 725 Furnace 25 10
WQ725 725 Water 35 –
FC752 752 Furnace 37 15

a higher manganese and lower silicon content when it

forms from austenite cooling. These speculations were
experimentally confirmed by Santofimia et al. recently
[11]. They investigated the distribution of alloying el-
ements in the present phases after intercritical anneal-
ing followed by cooling by using electron probe micro-
analysis in the low carbon steel. They reported that
the manganese concentration is 1.6 wt.% in retained
ferrite, 2.2 wt.% in epitaxial ferrite, and 2.6 wt.% in
M/A (martensite/austenite). The concentration pro-
file of silicon is 0.44 wt.% in retained ferrite, 0.38 wt.%
in epitaxial ferrite, 0.44 wt.% in retained ferrite and
0.28 wt.% in M/A. The carbon concentration is close
to zero in both ferrite and 1.2 wt.% in M/A.
The results of the quantitative metallographic
measurements are summarized in Table 2. In addi-
tion to composition, intercritical annealing tempera-
ture (ICAT) and cooling rate affect hardenability of
DP steels. The martensite (austenite) volume fraction
(MVF) increases with increasing the ICAT, as shown
in Table 2. This can be explained by lever rule. MVF of
WQ715 is found to be 25 %, whereas MVF of WQ725
is higher (35 %).
The cooling rate is another factor for increasing
MVF at a constant ICAT. High cooling rate (wa-
ter quenching) causes an increase in MVF at a con-
stant ICAT. MVF of FC725 (slow cooling) is found to
be 25 %, whereas MVF of WQ725 (rapid cooling) is
higher (35 %) at 725 ◦C.
As can be seen from Table 2, the specimens FC752
and WQ725 have almost constant amount of marten-
Fig. 2. Optical micrographs of (a) FC725, (b) FC752 site; even the specimen FC725 was produced by slow
DP steel samples (etched in 4 % picral + hot chromate cooling. This is probably due to high hardenability
reagent).
of the present steel and ICAT. The specimen FC752
contains 37 % and 15 % MVF and epitaxial ferrite, re-
spectively. The specimen WQ725 contains almost the
contents. The concentrations of austenite stabilizing same amount of martensite (35 %) and is free of epi-
elements (1.65 % Mn and 0.69 % Ni) are higher in the taxial ferrite. Although both specimens have the al-
steel used in the present work than those used in pre- most constant amount of martensite, no epitaxial fer-
vious studies. It is well known that Mn and Ni alloying rite formation is observed in WQ725. The slow cooling
elements decrease the critical cooling rate and increase rate for FC752 allows the epitaxial ferrite to grow on
the hardenability of steel. the existing ferrite and there is no grain boundary
The difference in the etching response is attributed between the epitaxial and retained ferrite. Similarly,
to the different content of alloying elements of the re- while WQ715 and FC725 contain the same amount of
tained and the epitaxial ferrite. Korzekwa et al. [12] martensite (25 %), WQ715 does not contain epitax-
indicated that manganese and silicon might partition ial ferrite, FC725 contains 10 % epitaxial ferrite. As
between the austenite and ferrite during intercriti- shown in Table 2, the amount of epitaxial ferrite in-
cal annealing and thus epitaxial ferrite may inherit creases from 10 % for FC725 to 15 % for FC752. At
 E. Salamci et al. / Kovove Mater. 55 2017 133–139 137

T a b l e 3. Summary of the results obtained from corrosion tests performed in 3.5 % NaCl solution

Specimen code Ecorr Icorr CR from Mass loss from CR from

(mV) (µA cm−2 ) polarization test immersion test immersion test
(mm year−1 ) after 7 days (g) (mm year−1 )

WQ715 –73.20 2.11 (± 0.2) 0.050 1.6 × 10−3 0.106

FC725 –18.90 2.39 (± 0.4) 0.055 1.3 × 10−3 0.086
WQ725 –167.9 14.5 (± 1.0) 0.336 1.8 × 10−3 0.119
FC752 +37.15 3.11 (± 1.4) 0.072 1.1 × 10−3 0.073

Fig. 3. Mass loss of the corroded DP steel samples from Fig. 4. Potentiodynamic polarization curves of the DP steel
the immersion tests. samples.

the same MVF, more epitaxial ferrite was present at in g, A is an area of exposure (1 cm2 ), T is a time of
higher ICAT compared to lower one. This increase exposure in h and D is density in g cm−3 .
in volume fraction of epitaxial ferrite is due to in- The potentiodynamic polarization behaviors of the
crease in ICAT or decrease in cooling rate. In DP heat investigated specimens in 3.5 wt.% NaCl solution are
treatment, ICAT controls the carbon concentration of given in Fig. 4. Corrosion potential, Ecorr , and cor-
austenite. Austenite (martensite) carbon content de- rosion current density, Icorr values were obtained
creases with the increase of ICAT as predicted by the from potentiodynamic polarization curves, and the
lever rule. The lower carbon concentration of austen- CR (based on Icorr ) is calculated by using the follow-
ite at higher ICAT results in lower hardenability. Thus ing formula [28]:
more epitaxial ferrite forms instead of martensite. EW
CR = 3.27 × 10−3 Icorr , (2)
D
3.2. Corrosion behavior
where CR is the corrosion rate in mm year−1 . The
The mass loss of the samples increased with in- equivalent weight, EW, is approximated from the
creasing exposure time as shown in Fig. 3. It is worth atomic weight of iron (55.845 g mol−1 ), and D is the
noting that mass loss of all the specimens remained density of steel (7.88 g cm−3 ).
almost similar for the early stages (i.e., up to 3 days) The calculated corrosion rates (CR) based on both
after which the mass loss was deviated as the immer- the potentiodynamic polarization curves and the mass
sion time prolonged. For example, the mass loss is ap- loss are summarized in Table 3. The trend of the CR
proximately 30 % less for FC752 in 28-days exposure values obtained from the mass loss measurements af-
time as compared to WQ725. ter the immersion tests relatively agrees with the CR
From immersion test results (mass loss), the corro- values obtained by the polarization method in Table 3.
sion rate is estimated by using the following formula Ecorr values of WQ715, FC725, WQ725, and FC752
[28]: samples are –73.20, –18.90, –167.9, and +37.15 mV,
W respectively (Table 3). It was noted that the Ecorr val-
CR = K , (1) ues of epitaxial ferrite containing samples (i.e., FC725
AT D
and FC752) had shifted to a more positive region
where CR is the corrosion rate in mm year−1 , K is a (Fig. 4). It was also observed that the Icorr values de-
constant having value of 8.76 × 104 , W is a mass loss viated in the range between 2.11 and 3.11 µA cm−2 for
138 E. Salamci et al. / Kovove Mater. 55 2017 133–139

WQ715, FC725, and FC752 samples while Icorr value rite and martensite may result in increased potentials
of WQ725 was around 14.5 µA cm−2 . Although more thus propagating micro galvanic corrosion. As shown
positive Ecorr and lower Icorr values indicate that the in Figs. 2a,b, epitaxial ferrite surrounds the marten-
corrosion resistance of the samples increases, the cor- site phase in the form of rings. It is expected that
rosion rate is directly proportional to Icorr according the potential difference between the phases should
to Faraday’s law. In the present study, the values of be lowered as the concentrations of the phases are
Icorr of all samples used are lower than those obtained similar. From this point of view, lower potential dif-
by Sarkar et al. and Bhagavathi et al. [19, 21] indicat- ferences should be presented between epitaxial fer-
ing much better corrosion resistance. Sarkar et al. [19] rite and martensite as well as between epitaxial and
reported that the Icorr value of DP steel having 33 % retained ferrite. Therefore, the presence of the epi-
MVF was 32.29 µA cm−2 and it was 17.40 µA cm−2 taxial ferrite may hinder the corrosion by smooth-
for DP steel with 24 % MVF in the work of Bhaga- ing the compositional fluctuations between the re-
vathi et al. [21]. tained ferrite and the martensite. On the other hand,
The corrosion of DP steels may be governed by lower CR for WQ715 sample which has no epitax-
four relative mechanisms: chemical composition, the ial ferrite could be explained by having lower MVF.
volume fraction of martensite phase (i.e. MVF), the Bhagavathi et al. [21] reported that the Icorr val-
presence of epitaxial ferrite, and the presence of ues were similar for all the DP steels having 21,
stresses. 24, and 28 % MVF concluding that the MVF up to
It is known [28] that the presence of Si, Ni, and 28 % had little effect on the corrosion rate of DP
lower amount of S have a beneficial effect on the cor- steels.
rosion resistance of carbon steels. It is important to In addition, stresses developed during rapid cool-
point out that the steel used in this work contained ing may be another reason for the increased corrosion
Ni and Si as well as lower S which may result in im- rate. Our previous TEM study [13] showed that the
proved corrosion resistance of the samples. dislocation density of the specimen WQ725 was higher
Martensite volume fractions, MVF, of WQ715 and than that of the specimen FC752. This means that
WQ725 samples are 25 and 35 %, respectively. The higher dislocation density causes higher stress levels
higher CR for WQ725 sample is attributed to MVF. in the WQ725 specimens. By considering the stress-
Based on galvanic corrosion principles, the total sur- induced corrosion, it can be speculated that stresses
face area of the cathode (martensite) in relation to due to higher dislocation density may trigger corrosion
the surface area of the anode (ferrite) governs the gal- in the specimen WQ725.
vanic corrosion. Therefore, any relative change in the
amount of the microstructural constituents would lead
to alteration in the ratio of anode to cathode areas. In 4. Conclusions
literature [17, 19, 20], the corrosion rate of DP steels
increased with increasing MVF which was reported to 1. The martensite volume fraction increases with
be due to the adverse area ratio of martensite phase an increase in intercritical annealing temperature.
to ferrite phase in accordance with the present work. 2. The slow cooling rate for FC752 allows the epi-
However, the MVF of FC752 is the highest (i.e., taxial ferrite to grow on the existing ferrite attributed
37 %), its CR is much lower as compared to WQ725 to increasing in intercritical annealing temperature or
(i.e. 35 % MVF). This was attributed to the pres- decreasing in cooling rate.
ence of epitaxial ferrite present in FC752 sample. Al- 3. Mass loss of all the investigated specimens re-
though the presence of epitaxial ferrite in DP steels mained almost similar for the early stages (i.e., up to
and its effect on mechanical properties have been stud- 3 days) after which the mass loss was deviated as the
ied by many researchers in the literature [13–15], its immersion time prolonged. The mass loss is approxi-
influence on corrosion behavior has not been reported mately 30 % less for FC752 at 28-day exposure time
yet. As discussed earlier in microstructure section, as compared to WQ725.
compositional differences between retained ferrite and 4. The Icorr values deviated in the range be-
epitaxial ferrite are evident. The Mn concentration tween 2.11 and 3.11 µA cm−2 for WQ715, FC725, and
increases sequentially from retained ferrite, epitaxi- FC752 samples while Icorr value of WQ725 was around
ally ferrite to martensite while the content of Si de- 14.5 µA cm−2 .
creases in the same manner, reported by Korzekwa et 5. Corrosion rate increased with increasing marten-
al. [12] and recently confirmed by Santofimia et al. site volume fraction. However, this phenomenon was
[11]. Thus, the galvanic couple formed between re- inhibited in the epitaxial ferrite containing samples,
tained ferrite and epitaxial ferrite is expected to be attributed to the presence of epitaxial ferrite hinder-
weaker than the galvanic couple formed between re- ing corrosion by smoothing the compositional fluctu-
tained ferrite and martensite. Different contents of al- ations between retained ferrite and martensite as well
loying elements (i.e., Mn, Si, C) between retained fer- as reduced dislocation density.
 E. Salamci et al. / Kovove Mater. 55 2017 133–139 139

References [15] Jeong, W. C., Kim, C. H.: Metall. Trans. A, 18, 1987,
p. 629. doi:10.1007/BF02649478
[1] Rashid, M. S.: In: Formable HSLA and Dual-Phase [16] Demir, B., Candan, S., Salamci, E., Erdoğan, M.: In:
Steels. Ed.: Davenport, A. T. Chicago, Illinois, AIME IXth Int. Corrosion Symp. and Exhibition. Ed.: Koc,
1977, p. 1. T. Ankara, The Corrosion Association 2004, p. 495.
[2] Koo, J. Y., Thomas, G.: In: Formable HSLA and Dual- [17] Trejo, D., Monterio, P., Thomas, G., Wang, X. Y.:
Phase Steels. Ed.: Davenport, A. T. Chicago, Illinois, Cem. Concr. Res., 24, 1994, p. 1245.
AIME 1977, p. 40. doi: 10.1016/0008-8846(94)90109-0
[3] Morrow, J. W., Tither, G., Buck, R. M.: In: Formable [18] Zhang, C., Cai, D., Liao, B., Zhao, T., Fan, Y.: Mater.
HSLA and Dual-Phase Steels. Ed.: Davenport, A. T. Lett., 58, 2004, p. 1524.
Chicago, Illinois, AIME 1977, p. 151. doi:10.1016/j.matlet.2003.10.018
[4] Hayami, S., Furukawa, T., Gondoh, H., Takechi, H.: [19] Sarkar, P. P., Kumar, P., Manna, M. K., Chakraborti,
In: Formable HSLA and Dual-Phase Steels. Ed.: Dav- P. C.: Mater. Lett., 59, 2005, p. 2488.
enport, A. T. Chicago, Illinois, AIME 1977, p. 167. doi:10.1016/j.matlet.2005.03.030
[5] Davies, R. G., Magee, C. L.: In: Structure and Prop- [20] Osorio, W. R., Peixoto, L. C., Garcia, L. R., Garcia,
erties of Dual-Phase Steels. Eds.: Kot, R. A., Morris, A.: Mater. Corros., 60, 2009, p. 804.
J. W. New Orleans, AIME 1977, p. 1. doi: 10.1002/maco.200805181
[6] Hayami, S., Furukawa, T.: In: Proceedings of Microal- [21] Bhagavathi, L. R., Chaudhari, G. P., Nath, S. K.:
loying 75. Vanitec, London 1975, p. 78. Mater. Des., 32, 2011, p. 433. doi:10.1007/BF02660523
[7] Rashid, M. S.: GM 980X-A Unique High Strength doi:10.1016/j.matdes.2010.06.025
Sheet Steel with Superior Fromability. SAE Preprint [22] Andrews, K. W.: J. Iron and Steel Inst., 203, 1965, p.
No. 760206. 1976, p. 938. 721.
[8] Davies, R. G.: Metall. Trans. A, 9A, 1978, p. 671. [23] Lawson, R. D., Matlock, D. K., Krauss, G.: Metallogr.,
doi:10.1007/BF02659924 13, 1980, p. 71. doi: 10.1016/0026-0800(80)90023-3
[9] Speich, G. R., Miller, R. L.: In: Structure and Prop- [24] Lawson, R. D., Matlock, D. K., Krauss, G.: In: Funda-
erties of Dual-Phase Steels. Eds.: Kot, R. A., Morris, mentals of Dual Phase Steels. Eds.: Kot, R. A., Bram-
J. W. New Orleans, AIME 1977, p. 145. fitt, B. L. New York, AIME 1981, p. 347.
[10] Huppi, G. S., Matlock, D. K.: Scripta Metall., 14, [25] Jeong, W. C., Kim, C. H.: Scripta Metall., 19, 1985,
1980, p. 1239. doi:10.1016/0036-9748(80)90264-1 p. 37. doi:10.1016/0036-9748(85)90261-3
[11] Santofimia, M. J., Kwakernaak, C., Sloof, W. G., [26] Geib, M. D., Matlock, D. K., Krauss, G.: Met. Trans.
Zhao, L., Siestma, J.: Mater. Charact., 61, 2010, p. A, 11, 1980, p. 1683. doi:10.1007/BF02660523
937. doi:10.1016/j.matchar.2010.06.006 [27] Huppi, G. S., Matlock, D. K., Krauss, G.: Scripta Met-
[12] Korzekwa, D. A., Matlock, D. K., Krauss, G.: Metall. all., 14, 1980, p. 1239.
Trans. A, 13, 1982, p. 2061. doi: 10.1007/BF02645952 doi: 10.1016/0036-9748(80)90264-1
[13] Salamci, E., Kabakci, F.: J. Fac. Eng. Arch. Gazi [28] Cramer, S. D., Covino, B. S., Jr. (Eds.): ASM Interna-
Univ., 26, 2011, p. 263. doi:10.17341/gummfd.76139. tional Metals Handbook. Corrosion. 9th Edition. Met-
[14] Sarwar, M., Ahmad, E., Qureshi, K. A., Manzoor, T.: als Park, ASM International 1987.
Mater. Des., 28, 2007, p. 335.
doi:10.1016/j.matdes.2005.05.019