AISTech 2019 — Proceedings of the Iron & Steel Technology Conference

6–9 May 2019, Pittsburgh, Pa., USA

DOI 10.1000.377.233

Effect of Solution Treatment on Grain Size and Toughness of Lightweight Fe-Mn-Al-C Steel

D.M. Field1, and K.R. Limmer1

1
U.S. Army Research Laboratory
Address: 6300 Rodman Rd. MD 21005,
Email: [email protected]

Keywords: Lightweight Steel, austenitic steel, precipitation hardening, Grain Size, Toughness, κ-carbide,

INTRODUCTION
A significant body of work has been produced on advanced high strength steels to reduce weight and fuel consumption of
modern vehicles. Steels with increased concentrations of both aluminum and manganese, termed Fe-Mn-Al-C steel, have
been developed to fit this need. Alloys that contain 0.3 to 1.2 wt. pct. carbon and 5 to 11 wt. pct. aluminum are also age
hardenable in the temperature range of 840 °F to 1290 °F (450 to 700 °C) [1-9]. Age-hardening will produce a microstructure
containing coherent nano-sized κ-carbide precipitate, (Fe,Mn)3AlC. The work by Kayak, and Kim et al. [8,9] showed that
through precipitation strengthening high strength and ductile alloys were a prime candidate for springs and landing gears.
After age hardening these steels have been shown to have strength and toughness equivalent to quench and tempered steels
[2, 3, 8-10]. Alloying with aluminum has a secondary benefit, when concentrations exceed 12 wt. pct. a density reduction of
16-17 pct. can also be obtained [1-3, 11, 12]. This reduction in density is attributed to two root causes: first a dilation of the
lattice and the second is the substitution of iron atoms for the less dense aluminum atoms. These steels therefore are an
especially attractive candidate for weight reduction in transportation due to the associated decrease in density and a match in
strength with current materials.
The deformation mechanisms of these steels are very complicated. A number of hardening mechanisms have been reported
for manganese steels including: transformation induced plasticity (TRIP), twinning induced plasticity (TWIP), or slip band
refinement (SBR). The activated deformation mechanism is determined by the stacking fault energy (SFE) of the steel. TRIP
is reported to be active when the SFE is below 20 mJ/m2 [13-16] and is identified as a transformation of austenite to  or α-
martensite leading to high ultimate tensile strengths and increased total ductility. Medium SFE steels (20-40 mJ/m2) are
characterized as TWIP steels, where deformation twins are produced during straining leading to high work hardening rates.
Steel alloys formulated with SFE in the range of 40-90 mJ/m2 undergo slip as the dominant deformation mechanism. High
SFE alloys ≥ 90 mJ/m2 are described as steels that do not mechanically twin but undergo slip; these steels also display a
strong planar glide and have been termed as steels which exhibit slip band refinement [17-19]. With the higher SFE, cross
slip should be easier causing a “wavy” glide from the activation of multiple slip systems, however, the addition of aluminum
causes these steels to exhibit planar glide from short range ordering. κ-carbide has also been shown to form from this short
range ordering reaction [2, 5].
As is typical of most age-hardening systems a solutionizing step is utilized before subsequent age hardening treatments. For
most works on both as-cast and wrought austenitic Fe-Mn-Al-C steels, a solution treatment, (referred to as “STQ” for
simplicity), has consistently been performed at 1922 °F (1050 °C) [1-6, 8, 9, 20] regardless of the compositional variations
between the works cited. This is typically rationalized and cited as being a temperature regime at which a single phase
austenite can be obtained due to thermodynamic calculations [5, 7]. The nominal alloy composition for these lightweight
steels is typically 1C – 30Mn – 9Al – 1Si – bal.Fe (in wt. pct.). It can be shown from thermodynamic calculations (see
Figure 1) that between 1600 to 1920 °F (870 – 1050 °C) a single phase γ-austenite is obtained. The increase in time at the
STQ temperature has been shown to have a pronounced effect on the grain size and can be deleterious to both impact
properties and precipitation structure.

© 2019 by the Association for Iron & Steel Technology. 2271

Figure 1. Pseudo-binary phase diagram of the Fe-Al system with a fixed manganese concentration of 30 wt. pct. and a fixed
carbon and silicon content of 1.0 wt. pct.
Grain refinement has been shown to typically improve the strength and toughness of steels, work on microalloying of the Fe-
Mn-Al-C systems to pin grains are of interest when cold work and annealing cannot be utilized as a means of grain
refinement. Moon et al. [21] investigated three low-density steels. The base composition was a nominal 0.9C – 30Mn – 9Al
– 1Si – balFe, a second steel was micro-alloyed with 0.5 wt. pct. vanadium, and a third steel was micro-alloyed with a
combination of both niobium (0.03 wt. pct.) and vanadium (0.5 wt. pct). The processing of the steel involved hot rolling at an
unspecified temperature to a plate of 8 mm in thickness followed by air cooling. Solution treatment was performed at the
typical 1922 °F (1050 °C) for 2 hours followed by water quenching to room temperature and subsequent aging at 1020 °F
(550 °C) for up to 103 minutes. It was noted by the authors that a significant grain refinement was obtained from the addition
of vanadium to the steel leading to a reported solution treated grain size of 17 μm compared to the 80 μm of the base
composition; Moon et al. however does not state if these values consider twin boundaries. From the reported data and
assuming an inverse root grain size dependence on strength (Hall-Petch relationship), the grain boundary hardening term, k,
can be calculated to be 1170 MPa√𝜇𝑚 with a friction stress, σ0, of 560 MPa, which is converted from Vickers Hardness.
Interestingly, the aging kinetics of the alloys varied significantly with grain size but not micro-alloying addition. The steels
micro-alloyed with V and V + Nb aged at similar rates and had nearly identical grain sizes, 17 ± 0.2 μm. The base
composition with an 80 μm grain size aged much slower and did not obtain an equivalent hardness to the micro-alloyed steels
until aging was performed for longer times, however this effect was not discussed by the authors.
It can be rationalized that with a finer grain structure and greater grain boundary area the kinetics of precipitation should be
increased from the increased mobility of solute elements along grain boundaries. Thermo-Calc2017a with the precipitation
module, PRISMA, using the thermodynamic TCFE9 steel database and MOBFE2 mobility database were utilized to model
the transformation start curve of κ-carbide precipitation in an 1C – 30Mn – 9Al – 1Si – balFe (in wt. pct.) steel with a fine
and coarse grain structure. It is shown in Figure 2 that with the finer grain size the time to form 0.01 vol. pct. precipitate is
decreased compared to the coarse grained steel.

2272 © 2019 by the Association for Iron & Steel Technology.

Figure 2. Time Temperature Transformation (TTT) diagram of the κ-carbide start for a 1C – 30Mn – 9Al – 1Si – balFe with
coarse and fine grained structures
Work on understanding the recrystallization behavior of strip cast Fe-Mn-Al-C steels was performed by Ji et al. [22] with a
composition of 1.3C – 28Mn – 9Al – 1Si – balFe (in wt. pct.). The steel under investigation was cold worked 50 pct. and then
annealed at a range of temperatures for 20 minutes. It was shown that when the steel was annealed at a temperature of 1650
°F (900 °C) the best balance of strength and ductility could be obtained. Ji et al. notes that increasing the annealing
temperature from 1470 °F (800 °C) to 1650 °F (900 °C) produced a larger uniform grain structure (8.5 μm vs. 16.3 μm)
however this effect on softening was offset by the increase in twin boundary frequency. Twin boundaries were observed to
increase by 10 pct. leading to a negligible change in the mean linear intercept distance (7 μm vs. 10 μm). This significant
increase in twin boundaries, and grain size leads to a significant loss in strength of 100 MPa, and Ji et al reports a Hall-Petch
relationship for the ultimate strength of the materials tested where the grain hardening term, k, is 760 MPa√𝜇𝑚 and the
frictional stress, σ0, is 725 MPa.
Little work has investigated the effect of grain size on toughness of wrought Fe-Mn-Al-C steels. Many works published have
shown that these steels obtain excellent quasi-static strength and ductility combinations, however, the energy absorption
capability of the material must be investigated. Little work has also been performed to optimize the heat treatment parameters
of solution treatment before aging on the toughness of the steel. The purpose of this work is to determine the impact
toughness of a wrought Fe-Mn-Al-C plate and the optimum heat treatment parameters.

EXPERIMENTAL PROCEDURE
A lightweight steel with nominal composition of 1C – 30Mn – 9Al – 1Si – 0.5Mo – balFe (in wt. pct.) was commercially cast
into 12,000 lb ingots. The stacking fault energy of the alloy was calculated to be 89 mJ/m2 using a regular solution model
according to the work of Pisarik and Van Aken [2]. After cooling the ingot was conditioned by grinding and then forged to
obtain a 50 pct. reduction in thickness at a temperature above 2100 °F (1150 °C) followed by air cooling. The forged slab was
then reheated above 2000 °F (1090 °C) and rolled to a plate thickness of 0.5 in (12.7 mm) to obtain a total thickness
reduction of 95 pct. Composition of the steel was confirmed using inductively couple plasma after dissolution in nitric acid,
and carbon content was determined using Leco gas combustion analysis. Sectioning was performed using a water-jet to
obtain coupons for heat treatments. Solutionizing was performed in a range of temperatures from 1650 – 1960 °F (900 – 1075
°C). Samples for solution treatment were placed in stainless steel bags, to minimize decarburization and oxidation, and heated
to the solution treating temperature for 2 hours followed by water quenching to room temperature. The temperature of the
sample was tracked using an external thermocouple and treatment time was based upon surface temperature obtaining the
targeted solution treatment value; heating rates to the targeted solution treat temperature ranged from 60-50 °F/min (33-28
°C/min). Age hardening was performed at 930 °F (500 °C) with a measured heating rate of 29 ° F/min (16 °C/min). After
aging, samples were air cooled on a refractory brick. A schematic of the processing is shown in Figure 3(a). Charpy V-notch
(CVN) testing was performed in accordance with the ASTM E23 [24] using a Type-A specimen and tested at room
temperature in an instrumented Tinius Olsen Charpy pendulum impact machine. All test samples were milled to the 10 x 10 x
55 mm3 dimensions and the v-notch was cut using an appropriate woodruff key to obtain the appropriate root radius and
depth as defined by the standard. The test coupons were of the T-L orientation relative to the final rolling direction of the
plate as shown in Figure 3(b), with the long axis of the coupon parallel to the transverse direction and the notch parallel to the
rolling direction. Hardness was measured by Rockwell C and B techniques and converted to Brinell hardness for reporting
according to ASTM E18-17 [25].

© 2019 by the Association for Iron & Steel Technology. 2273

Figure 3. (a) Schematic diagram of the heat treatment utilized for CVN testing. (b) Sample location of the CVN test coupons
relative to the rolling, transverse and normal direction. (Dimensions not to scale)
Microstructural examination was performed in the transverse plane (perpendicular to both the normal direction and rolling
direction). Samples were mechanically polished to 0.02 μm with a colloidal silica solution and etched with a 10 pct. nital
solution of nitric acid and methanol. Scanning electron microscopy images were taken using a Phenom XL operating at 15
kV in backscatter mode. Grain size was measured according to the ASTM E112-13 using the Heyn Lineal Intercept method
and included twin boundaries [26]. Fracture surfaces were examined using a Phenom XL operating at 15 kV to determine
fracture failure mode.

RESULTS
Solution treating temperature, grain size, and hardness are shown in Table I. The Brinell hardness number was converted to
an ultimate tensile strength according to Eq (2) as reported in [27], and has been plotted assuming a Hall-Petch relationship in
Figure 4. It should be noted that Hall-Petch grain size relationship shown in Figure 4 is very similar to the values calculated
earlier from Moon et al. [21] with similar grain boundary hardening terms, k, (1170 MPa√𝜇𝑚 vs. 1306 MPa√𝜇𝑚) and
friction stress, σ0, (560 MPa vs. 610 MPa).

𝜎 𝑀𝑃𝑎 3.38 ∗ 𝐵𝐻𝑁 (2)

Table I. Grain size, hardness, and converted ultimate strength for the alloys solutionized at different temperatures. (95%
confidence level reported in the error)
Solution Treatment Calculated Ultimate Tensile
Grain size (μm) Hardness (BHN)
Temperature °F (°C) Strength (MPa)

1650 (900) 30 ± 9 252 ± 4 853 ± 14

1740 (950) 43 ± 1 236 ± 8 798 ± 26
1785 (975) 82 ± 2 225 ± 4 761 ± 12
1830 (1000) 156 ± 2 218 ± 8 737 ± 27
1920 (1050) 201 ± 1 203 ± 5 686 ± 17
1965 (1075) 327 ± 1 201 ± 7 679 ± 23

2274 © 2019 by the Association for Iron & Steel Technology.

 Figure 4. Hall-Petch relationship for the 1C – 30Mn – 9Al – 1Si – 0.5Mo – balFe steel solution treated and quenched to
obtain various grain sizes.
Optical micrographs using differential interference contrast (DIC) of the alloy after solution treating at 1650, 1740, and 1830
°F (900, 950, and 1050 °C) are shown in Figure 5(a-c). It is noted that significant grain growth occurs when the solution
treatment temperature is raised above 1740 °F (950 °C). The hardness and CVN toughness aging response of the alloys is
shown as a function of aging time in Figure 6, with values summarized in Table II. It is noted that the alloys which were
solution treated at 1740 °F (950 °C) aged at an accelerated rate compared to alloys solution treated at 1920 °F (1050 °C),
which is consistent with the results from Moon et al. [21].

Figure 5. Optical microscopy images of the Fe-Mn-Al-C samples taken with differential interference contrast after solution
treatment and water quenching for two hours at (a) 1650 °F (900 °C), (b) 1740 °F (950 °C), and (c) 1920 °F (1050 °C)

Figure 6. Charpy V-notch Toughness and Brinell hardness aging response of the Fe-Mn-Al-C alloy after solution treating at
various temperatures.

© 2019 by the Association for Iron & Steel Technology. 2275

 Table II. Hardness and CVN toughness of samples throughout the aging process for samples solution treated at varying
temperature.
Aging Time: 5 hours 20 hours 66 hours 120 hours
Hardness CVN Hardness CVN Hardness CVN Hardness CVN
STQ Temp.
(BHW) (ftꞏlbs) (BHW) (ftꞏlbs) (BHW) (ftꞏlbs) (BHW) (ftꞏlbs)
1740 °F (950 °C) 265 ± 9 46 ± 5 327 ±7 18 ± 2 363 ± 7 7.4 ± 2.0 392 ± 5 10 ± 3
1830 °F (1000 °C) 222 ± 11 102 ± 1 280 ± 22 14 ± 8 345 ± 6 2.4 ± 0.2 364 ± 19 3.0 ± 0.7
1920 °F (1050 °C) 201 ± 11 163 ± 10 210 ± 22 64 ± 5 296 ± 15 3.2 ± 1.4 312 ± 14 2.8 ± 0.3

To better elucidate the effect of grain size on toughness the CVN toughness is mapped directly as a function of hardness in
Figure 7(a). A magnified view of the toughness at elevated hardness levels (≥ 250 BHW) is included in Figure 7(b) to resolve
the lower toughness values. It is noted that for equivalent hardness values the toughness is increased when solution treatment
temperatures is decreased. Backscatter electron (BSE) images of the alloy age hardened to 328 ± 17 BHN are shown in
Figure 8; it should be noted that no grain growth occurs at the low temperature utilized for age hardening of these lightweight
steels. Fractography was performed on broken CVN bars to determine the fracture type, and are shown in Figure 9(a-f).

Figure 7. Toughness as a function of hardness for the steel under investigation.

Figure 8. BSE image of the 1C – 30Mn – 9Al – 1Si – 0.5Mo – balFe steel after (a) solution treatment at 1740 °F (950 °C)
followed by 20 hours of aging, (b) solution treatment at 1830 °F (1000 °C) followed by 66 hours of aging, and (c) solution
treatment at 1920 °F (1050 °C) followed by 120 hours of aging

2276 © 2019 by the Association for Iron & Steel Technology.

 Figure 9. Fracture surfaces of samples solution treated at 1830 °F (1000 °C) and aged for (a) 5 hours, (c) 20 hours, and (e)
120 hours. Samples solution treated at 1740 °F (950 °C) and aged for (b) 20 hours and (d) 66 hours. (f) Solution treated at
1920 °F (1050 °C) and aged for 120 hours. Tup travel is from right to left in all images.

DISCUSSION
It is noted from Table I that the grain size doubled (82 μm vs. 156 μm) when the solution treatment temperature increased
from 1786 °F (975 °C) to 1830 °F (1000 °C), suggesting the dissolution of a grain pinning particle in this temperature range.
This allowed grains to grow at an accelerated pace. The micro-alloying addition of molybdenum was made for the intent of

© 2019 by the Association for Iron & Steel Technology. 2277

grain refinement during hot rolling of the steel however, it is noted from thermodynamic calculations using ThemoCalc
2017a with the TCFE9 database that the molybdenum rich M6C solvus temperature is 1394 °F (757 °C), well below the
utilized hot rolling temperature and lowest STQ temperature of 1650 °F (900 °C). The κ-carbide solvus is 1484 °F (807 °C)
as shown in Table III. This provides a 305-400 F° (170-220 C°) difference in the observed and modeled temperature for
carbide dissolution based on the grain size response of these experiments. This discrepancy may be due to kinetic restraints
for dissolution of κ-carbide or M6C. Moon et al. showed that utilization of V and Nb are effective grain pinning agents after
solution treatment at 1920 °F (1050 °C) for an alloy system of comparable composition. The CalPhad calculations for the
solvus of VC and/or (V,Nb)C carbides in the alloy class under investigation is shown in Table III. It is shown that the
carbide solvus temperatures of 1904 °F (1040 °C) and 2168 °F (1187 °C), respectively are obtained, which is in good
agreement with the insignificant change in grain size observed after solution treatment at 1920 °F (1050 °C). It is noted that
the introduction of vanadium reduces the stability of the κ-carbide by 134 F° (75 C°), and is rationalized as due to the loss of
carbon to the VC and (Nb,V)C precipitates. It can be shown from the conjunction of this work and the work by Moon et al.
[21] that molybdenum is not effective at grain pinning during solution treatment above 1785 °F (975 °C).

Table III. Thermodynamic calculations performed using ThermoCalc 2017a of carbide solvus of the steel under investigation
and the micro alloyed steels reported by Moon et al. [21]
Solvus Temperature °F (°C)
Alloy
κ-carbide M6C VC (V,Nb)C
This Study 1484 (807) 1394 (757) - -
V alloy [21] 1350 (732) - 1904 (1040) -
V + Nb Alloy [21] 1348 (731) - - 2168 (1187)

The aging kinetics are strongly correlated to the solution treatment temperature and associated austenite grain size as shown
in Figure 6. Samples solution treated at 1920 °F (1050 °C) did not exhibit a statistically significant change in hardness until
aging hardening was performed for greater than 20 hours. Samples solution treated at 1740 °F (950 °C) exhibited a higher
initial hardness, due to the finer grain size as shown in Figure 4, and faster aging kinetics. The 1740 °F (950 °C) STQ
samples showed a significant increase in hardness after aging for 20 hours (236 to 327 BHW) with a ~90 BHW increase. The
increase in kinetics is currently attributed to the grain structure as was shown in Figure 2 and the results from Moon et al.
[15]. The effect of grain size on the precipitation of κ-carbide is significant and is effective in reducing the time required for
precipitation to a target strength value.
The increase in hardness leads to a decrease in toughness for all solution treatment temperatures. This response has been
shown by previous works [28-31]. Bartlett and Van Aken [29] reported a significant loss in toughness as aging was
performed on a cast steel of similar composition. Acselrad showed a 250 MPa√𝑚 decrease in toughness of a Fe-28Mn-8.5Al-
1C after aging from 15 hours at 1020 °F (550 °C) [30]. A metric for the loss in toughness with the increase in strength can be
calculated as the slope between points from Figure 6 and is summarized in Table IV. It is noted that for the 1740 °F (950 °C)
STQ samples the loss in toughness is minimized compared to the 1830 °F and 1920 °F (1000, and 1050 °C) STQ specimens.
The greatest loss in toughness occurred in the 1920 °F (1050 °C) STQ samples after a short aging time of 20 hours. The 1740
°F (950 °C) STQ samples had the lowest toughness loss and this is associated with the reduced grain size leading to an
improvement in toughness. It is noted that for the 1740 °F (950 °C) STQ sample there is a slight increase in toughness from
the 66 hour to 120 hour age hardening treatment, however it should be realized that the toughness of the 66 hour treated
sample was 7.4 ± 2.0 ft∙lbs and the toughness of the 120 hour sample was 10 ± 3 J/cm2 where the difference in the values are
statistically insignificant. The toughness values of the 1830 °F and 1920 °F (1000 °C and 1050 °C) show a similar behavior
with no statistically significant change in toughness after age hardening is performed beyond 66 hours.

Table IV. Loss of toughness with the increase in hardness between aging steps for the three solution treated samples.
STQ Sample 5 → 20 20 → 66 66 → 120
1740 °F (950 °C) -0.46 ft∙lbs/BHN -0.29 ft∙lbs/BHN 0.10 ft∙lbs/BHN
1830 °F (1000 °C) -1.5 ft∙lbs/BHN -0.18 ft∙lbs/BHN 0.04 ft∙lbs/BHN
1920 °F (1050 °C) -11.0 ft∙lbs/BHN -0.71 ft∙lbs/BHN -0.02 ft∙lbs/BHN

From the fractography shown in Figure 9 it was noted that the fracture surface transitioned as aging progressed. This is best
observed in the 1830 °F (1000 °C) samples. It is noted that at the low hardness of the initially aged material, a dimple rupture

2278 © 2019 by the Association for Iron & Steel Technology.

fracture surface was observed as shown in Figure 9(a). As aging progressed and κ-carbides are further precipitated within the
grain interior this led to a faceted transgranular fracture surface as shown in Figure 9(c and e) this is due to the short range
ordering associated with κ-carbide formation. Similar trends in the fracture morphology transition are observed: the 1740 °F
(950 °C) and 1920 °F (1050 °C) STQ samples shown in Figure 9(b, d, and f). The transgranular failure surface is due to κ-
carbide being a weak dislocation barrier leading to glide plane softening. It has been reported [2, ,19, 28, 32] that the coherent
precipitate is not a strong dislocation barrier and as the particle is cut by a moving dislocation the particle becomes less
effective at resisting further dislocations. This can then cause the formation of planes of easy glide and localized shearing will
occur. This behavior has been well documented in Fe-Mn-Al-C alloys [3, 19, 31, 33]. Acselrad et al. [30] and Bartlett et al.
[31] describe the behavior as glide plane “slipping off” along slip planes of the type {111}γ and a fracture feature which is
observed as linear fracture facet markings. As the age hardening treatment time is increased the fracture surface is dominated
by an intergranular fracture associated with a thin film of κ-carbide or M6C on the grain boundaries. These results are
consistent that what Feng et al., Welsch et al., and Bartlett et al. [31, 32, 34,] report. Feng et al. showed that during aging κ-
carbide initially precipitated at the grain interior and concluded by forming at the grain boundaries leading to an initial
faceted transgranular fracture surface and once κ-carbide formed at the grain boundaries an intergranular fracture pattern was
observed. Bartlett et al. reported that broken CVN bars with intergranular fracture facets had coarse intergranular κ-carbide
precipitated at grain boundaries. However it is noted that the work by Bartlett et al [29, 31, 33] was performed on cast
material, and the precipitate size is much coarser.

CONCLUSION
It was found that lower solution treatment temperatures produced microstructures with finer grains due to reduced grain
growth. The M6C and κ-carbide were effective at reducing the grain growth up to 1248 K (975 °C). The finer grain structure
obtained after the 1223 K (950 °C) STQ treatment provided the best combination of hardness and toughness, and was more
insensitive to toughness loss as hardness increased during aging. The fracture morphology of the material produced a
transgranular fracture when hardness exceeded 300 BHN for the 1273 K and 1323 K (1000 and 1050 °C) STQ alloys, and
intergranular fracture was observed when hardness exceeded 350 BHN for all samples.

ACKNOWLEDGEMENTS
This research was supported in part by Congressional support through the Combat Vehicle Weight Reduction Initiative. Dr.
D. M. Field was supported by a research appointment at the U.S. Army Research Laboratory accomplished under
Cooperative Agreement Number W911NF-18-2-0046. The authors would also like to thank Mr. Edward Horwath, Mr.
Steven Marsh, and Mr. Mike Aniska for their assistance in this project. The views and conclusions contained in this
document are those of the authors and should not be interpreted as representing the official policies, either expressed or
implied, of the U.S. Army Research Laboratory of the U.S. Government.

REFERENCES
1. G. Frommeyer, U. Brux: Steel Research Int. (2006), vol. 77, pp. 627-633
2. L. N. Bartlett, D.C. Van Aken, J. Medvedeva, D. Isheim, N.I. Medvedeva, and K. Song: Met Trans A. (2014) vol. 45A,
2421-2435
3. R. Howell, T. Weerasooriya, and D.C. Van Aken: AFS Trans. (2009), vol. 117, pp. 751-763
4. G.L. Kayak: Metal Sci Heat Treat, (1969), vol. 11, pp. 95-97
5. O. Acselrad, I.S. Kalashnikov, E.M. Silva, R.A. Simao, C.A. Achete, and L.C. Pereira: Met Trans A. (2002) vol. 33A,
pp. 3569-3573
6. I.S. Kalashnikov, O. Acselrad, A. Shalkevich, L.D. Chumakova, and L.C. Pereira: Jour. of Mater. Processing Tech.
(2003), vol. 136, pp. 72-79
7. B. Hallstedt, A. V. Khvan, B.B. Lindahl, M. Selleby, and S. Liu: CALPHAD (2017), vol. 56, pp. 49-57
8. J.L. Kayak: Met Sci Heat Treat (1969) vol. 22, pp. 95-97
9. Y.G. Kim, Y.S. Park, and J.K. Han, Met Trans A. (1985) vol. 16A, pp. 1689-1693
10. S.A. Buckholz, D.C. Van Aken, and L.N. Bartlett AFS Transactions (2013)
11. H. Kim, D.W. Suh, and N.J. Kim: Sci. Technol. Adv. Mater. (2013) vol. 14, pp. 1-11

© 2019 by the Association for Iron & Steel Technology. 2279

12. L.N. Bartlett, R. Rahman, and A. Torres: Inter Jour Met Cast (2018), vol. 12, pp.164-181
13. L. Remy and A. Pineau: Mater Sci Eng, (1976) vol. 26, pp. 123-132
14. S. Allain, J.P. Chateau, and O. Bouaziz: Mater Sci Eng. (2004), vol. 387, pp. 143-147
15. T.H. Lee, E. Shin, C.S. Oh, H.Y. Ha, and S.J. Kim: Acta Mater, (2010) vol. 58, pp. 3173-3186
16. D.M. Field, J. Qing, and D.C. Van Aken Met Trans A, (2018) vol. 49, pp. 4615-4632
17. J.D. Yoo and K.T. Park: Mater Sci Eng. A (2008) vol. 2, pp. 417-424
18. K.T. Park: Scripta Met, (2013) vol. 68, pp. 375-379
19. V. Gerold and H.P. Karnthaler: Acta Met. vol . 37, pp. 2177-2183
20. C. Zhao, R. Song, L. Zhang, F. Yang, and T. King: Mater. & Design (2016), vol. 91, pp. 348-360
21. J. Moon, S-J. Park, C. Lee, H. N. Han, T-H. Lee, and C-H Lee: Met Trans A (2017), vol 48A, pp. 4500-4510
22. F. Ji, W. Song, Y. Ma, C. Li, W. Bleck, and G. Wang: Mater. Sci & Eng. A (2018), vol. 733, pp. 87-97.
23. S.T. Pisarik and D.C. Van Aken, Met Trans A., Vol. 47A 2016 pp1009-1018
24. ASTM E23 Notched Bar Impact Testing of Metallic Materials
25. ASTM E18-17e1 Standard Test Methods for Rockwell Hardness of Metallic Materials
26. ASTM E112-13 Standard Test Methods for Determining Average Grain Size
27. ASM Metals Refernce Book 3rd ed.. ASM international 1993
28. S. Chen, R. Rana, A. Haldar, R.K. Ray: Progress in Materials Sci. (2017) vol. 89, pp. 345-391
29. L.N. Bartlett, and D.C. Van Aken: AFS Transactions vol. 118 pp. (2013) 511-533
30. O. Acselrad, J. Dille, L.C. Pereira, and J. Delplancke: Met Trans A (2004), vol. 35A, pp. 3863-3866
31. L.N. Bartlett, A. Dash, D.C. Van Aken, V. Rishards, and K. Peaslee: Inter Jour Met Cast (2013), vol. 7 pp. 17-33
32. Y. Feng, R. Song, Z. Pei, R. Pei, R. Song, and G. Dou: Metal and Mater. Int. (2018) vol. 24, pp. 1012-1023
33. L. N. Bartlett, D.C. Van Aken, J. Medvedeva, D. Isheim, N.I. Medvedeva, and K. Song: Met Trans A. (2017) vol. 48,
pp. 5500-5515
34. E. Welsch, D. Ponge, S.M. Hafez Haghighat, S. Sandlobes, P. Choi, M. Herbig, S. Zaefferer, and D. Raabe: Acta
Materialia (2016), vol. 116, pp. 188-199

2280 © 2019 by the Association for Iron & Steel Technology.