j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/jmrt

Original Article

Evaluation of the macroscopic formability of

metallic materials using a cylinder compression
test

S.M. Ji a, J.M. Choi b, M.S. Joun c,*

a
School of Mechanical and Aerospace Engineering, Gyeongsang National University, Jinju, 52828, South Korea
b
Jinhap Co., Ltd, Daejeon, 34302, Republic of Korea
c
Engineering Research Institute, School of Mechanical and Aerospace Engineering, Gyeongsang National University,
Jinju, 52828, South Korea

article info abstract

Article history: Several flow features revealed by hot cylinder compression tests are macroscopically


Received 9 November 2022 analyzed and compared. Materials including AISI 1025 (100e500 C), AISI 52100 (900
Accepted 24 January 2023 e1200
C), AA6082 (20e300
C and 350e550
C), Tie6Ale4V (500e800
C) and AZ61A (250
Available online 30 January 2023 e400
C) were grouped in terms of formability. The shapes of cylinder-compressed speci-
mens were analyzed and related to formability. The addition of the maximum radius in-
Keywords: crease near the mid-plane and the averaged minimum radius increase at the two
Comparative study specimen-tool interfaces, i.e., the “cylinder compression evaluation factor” (CCEF), serves
Formability as an index of formability. To quantify the effect of temperature on formability, several
Steel metrics were used, including the CCEF and temperature softening formability index (TSFI).
Aluminum It has been shown that the CCEF and TSFI have utility for evaluating the formability of
Magnesium tested materials and that temperature softening greatly affects the formability of metallic
Titanium materials.
© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC
BY license (http://creativecommons.org/licenses/by/4.0/).

compositions. Experimental methods vary in terms of accu-

1. Introduction racy, and increase engineering costs and time.
Many previous studies have used artificial intelligence [1]
Flow behavior during metal-forming is important because, to analyze flow behavior. However, it is important to analyze
together with the law of friction, it macroscopically governs global observations, and their correlations, to enhance un-
the process and product quality. Many papers have addressed derstanding of important factors and identify practical uni-
these topics for a variety of reasons. However, practical un- versal rules. Many researchers have analyzed correlations
derstanding of flow behaviors has not greatly improved. The among materials within the same family. Wahabi et al. [2]
key problem faced by application engineers is that successful compared the hot deformation behavior of two austenitic
manufacturing is very reliant on material chemical

* Corresponding author.
E-mail address: [email protected] (M.S. Joun).
https://doi.org/10.1016/j.jmrt.2023.01.173
2238-7854/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://
creativecommons.org/licenses/by/4.0/).
 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9 2799

stainless steels (ASSs; AISI 304-type) with different carbon macroscopic stability. We thus analyzed data from hot
contents (0.02% and 0.087%). Wei et al. [3] investigated 0.26C- compressed cylinder experiments of the six materials
1.56Mn-1.72Si (wt%)-based and 0.23C-1.50Mn-1.79Al (wt (Tie6Ale4V, AA6082 (20e300
C), AA6082 (350e550
C), AISI
%)-based microalloyed high-strength steels using hot cylinder 52100, AZ61A and AISI 1025) under the wide range of tem-
compression tests, and compared the flow behaviors at peratures and performed a comparative study based on the
different temperatures (900e1100
C) and strain rates theoretical effects of heat transfer during plastic deforma-
(0.01e30 s1) to reveal the effects of aluminum and silicon. tion on cylinder compression. Several novel geometrical and
Guo and Li [4] experimentally studied the effects of common thermal parameters, including CCEF and TSFI, respectively,
alloying elements (C, Mn, Si, and Al) on the warm-deformation were presented to represent material flow characteristics
behaviors of high-Mn TRIP steels with a martensitic structure and macroscopic formability, and rules were established for
at various temperatures (550e650
C) and strain rates material evaluation in terms of process stability and product
(0.001e0.1 s 1). Gao et al. [5] investigated the effects of the quality.
initial austenite grain size of boron microalloyed steel with
three different boron levels (20, 40, and 60 ppm) at 1150, 1100,
and 1050
C, and performed hot cylinder compression tests 2. Summary of hot cylinder compression
over wide temperature (900e1100
C) and strain rate (0.1e10 s tests (five materials)
1
) ranges. Spezzapria et al. [6] studied the effects of prior
microstructure and heating rate on the kinetics of austenitic In previous studies, five materials including AISI 1025
transformation of the 39NiCrMo3 steel using different prior (20e500
C) [16], AISI 52100 (900e1200
C) [17], AA6082
microstructures in the wide range of heating rates. Menapace (20e300
C and 350e550
C) [18], the Tie6Ale4V titanium alloy
[7] compared the hot deformation behaviors of four different (500e800
C) [19], and the AZ61A magnesium alloy
steels (under as-cast conditions) using hot cylinder compres- (250e400
C) [20] were characterized by hot cylinder
sion tests at temperatures of 1100e1200
C and strain rates of compression tests using a Gleeble 3800 (AISI 52100) or 3500
0.12e2.4 s1. Zheng [8] investigated the combined effects of Nb tester. The height: radius ratios of all cylinder specimens were
and B on the hot ductility of 25CrMo alloy steel at tempera- the same, although the initial radii of AISI 1025, AISI 52100,
tures of 700e1100
C at a strain rate of 0.5 s1. Gao [9] exam- AA6082, Tie6Ale4V, and AZ61A are 5.0, 5.0, 5.0, 3.9, and
ined the effects of titanium on the hot deformation behaviors 5.0 mm, respectively. Several specimens that were cylinder-
of titanium-free and titanium-treated boron microalloyed compressed at a strain rate of 1 s1 are shown in the left
steel at temperatures of 850e1100
C and strain rates of panel of Fig. 1, whereas their corresponding predicted shapes
0.1e10 s 1. Hu [10] compared two medium-carbon steels (in are given on their right side. Our experience of the experi-
terms of activation energy at different strains) by fitting data ments and flow characterizations suggests that temperature
from cylinder compression tests. Wei et al. [11] compared low- compensation may significantly affect the final flow curves
and medium-carbon niobium microalloyed steels at temper- depending on the material properties, and that the effect of
atures of 900e1100
C and strain rates of 0.01e10 s1 using hot friction cannot be neglected when temperature compensation
cylinder compression tests, to reveal softening behaviors. is minimal or the thermal conductivity of the material (the k
Note that the above literature is not exhaustive; there are value) is large (Table 1). This is the case for AA6082, for
many other relevant papers. example. Temperature effects are especially crucial for the
Other comparative studies focused on flow behaviors to AZ61A and Tie6Ale4V alloys. The AA6082 alloy exhibits very
identify optimal materials (among several candidates) for different behavior. Thus, we analyzed the effects of temper-
particular applications [11,12]. For example, Park et al. [13] ature on flow behavior and formability in detail.
studied materials with cryogenic applications, including ASSs, Fig. 2 shows flow curves at selected sample strain rates.
aluminum alloys, and nickel alloys, at various temperatures The full, well-validated flow data can be found in related
(163 to 20
C) and strain rates (0.00016e0.01 s1), and re- studies [16e20]. The thermal properties of the test materials
ported the temperature- and strain rate-dependent properties are summarized in Table 1 including the k values and volu-
revealed by tensile tests. Notably, numerous paramteric metric heat capacities (rc values). The friction coefficients
studies on the formability of the materials have been between the specimens and tools were assumed to be 0.1 for
accomplished to find the optimal working conditions. AISI 1025 and AA6082 at the temperature range of cold metal
However, few researches have been made to evaluate the forming [16] and 0.3 for all other materials at elevated tem-
formability or macroscopic instability of the bulk materials, peratures [18e21]. It is noted that the friction effects in the
based on metallurgical [14,15] and mechanical [16] back- cylinder compression tests may not be considerable because
grounds. Prasad et al. [14] employed the strain rate sensitivity the friction-induced deformation heterogeneity causing the
to partition the stress power generation into either heat or compression load can lead to its decrease owing to the
microstructural change and obtained the processing map decreased contact area between the material and tool [18].
using on the energy dissipation information from the flow Finite element (FE) predictions corresponding to the
function. Joun et al. [16] calculated the macroscopic instability experimentally deformed shapes in Fig. 1 were obtained using
index with an emphasis on the stroke softening of the mate- an axisymmetric thermoviscoplastic FE method [22], based on
rial weighted by the stress power using the predicted state the flow curves in Fig. 2 and thermal conditions and properties
variables during metal forming. listed in Tables 1 and 2 (see the right panel of Fig. 1). The colors
It is noted that there have been few comparative studies in Fig. 1 indicate the effective strains. Comparison of the
on material flow behaviors in terms of forgeability and experimental and predicted (deformed) shapes indicated that
2800 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9

Fig. 1 e Experimentally deformed selected cylindrical specimens and predicted shapes with effective strain at the final
stroke. (a) Tie6Ale4V at 650
C (reduction 41%); (b) AISI 1025 at 200
C (reduction 42%); (c) AA6082 at 400
C (reduction 40%);
(d) AZ61A at 325
C (reduction 50%).

the flow models obtained in previous studies can be used for phenomena complicate warm forging of the Tie6Ale4V alloy
further comprehensive study. [19] and hot forging of the AZ61A alloy [20], as shown in Fig. 3.
Fig. 1 shows that the Tie6Ale4V alloy at 650
C and AZ61A On the contrary, AISI 1025 at 200
C shows typical barreling
alloy at 325
C exhibited excessive barreling, especially around during cylinder compression near room temperature and
the mid-plane of symmetry. Barreling is associated with clear favorable formability. The smooth barrel shape reflects the
macroscopic instability. The “maximum radius points” are not effects of both friction and strain-hardening on flow behavior.
near the mid-plane of symmetry and the radii of the upper These factors tend to expand the plastic deformation region to
and lower specimen-tool interfaces are distinctly different, distances beyond the major plastic-deforming region (indi-
reflecting poor formability of both materials attributable to the cated by the shear bands arising on cylinder compression).
flow behavior. The shear band connecting the central spot to Such flow features increase the macroscopic stability of the
the edge, where folding may occur, is severely deformed in the material under plastic deformation. The deformed pattern of
AZ61A alloy, which is associated with excessive deformation AA6082 at 400
C is between the Tie6Ale4V (650
C) or AZ61A
along the shear band and fracture of the barreled region. Such (325
C) alloy and AISI 1025 (200
C), as shown in Fig. 1.
Fig. 2(a) shows that the Tie6Ale4V alloy does not exhibit
strong flow stress dependence on strain. However, all flow
curves (except those at small strain rates and 500
C) exhibit
Table 1 e Thermal properties of the tested materials. distinct peak stress points over a relatively wide range of strains.
Material Temperature (
C) rc (Ws/(mm3
C) k (W/(mm
C) On the contrary, as shown in Fig. 2(b), AA6082 does not exhibit
Tie6Ale4V 650 0.00301 0.016 typical strain softening after strain hardening at elevated tem-
AA6082 400 0.00244 0.160 perature. Notably, some strain softening is apparent near room
AISI 52100 1000 0.00510 0.027 temperature, leading to some macroscopic instability during
AZ61A 325 0.00191 0.082 cold forging or cylinder compression [23,24].
AISI 1025 200 0.00416 0.047 Fig. 2(c) shows that the flow curve of AZ61A is highly
AA6082 100 0.00244 0.160
dependent on temperature (because the peak stress is
 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9 2801

Fig. 2 e Experiment and temperature-corrected flow curves of the materials. (a) Tie6Ale4V; (b) AA6082; (c) AZ61A; (d) AISI
1025.
2802 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9

those at 250
C and 325
C. Notably, both strain hardening and

Table 2 e Thermal conditions and properties of the
strain softening are very strong. This flow behavior, combined
cylinder compression test [19].
with the severe damage to the specimen in Fig. 3, implies that
Heat transfer coefficient to ambient (W/mm2$
C) 2.95ⅹ106
flow stress at a certain point changes to an extent that com-
Heat radiation coefficient to ambient (W/mm2$
C4) 3.97ⅹ1014
Heat transfer coefficient between material and die (W/ 0.01
promises local stability and decreases material ductility or
mm2$
C) formability. Such effects increase macroscopic instability
Volumetric heat capacity of dies WC-24Co (Ws/(mm3
C) 0.0050 during hot-forging of AZ61A at elevated temperatures, even-
Coefficient of thermal conductivity of dies WC-24Co 0.0950 tually causing severe fractures during hot cylinder compres-
(W/(mm
C) sion [20], as shown in Figs. 1(d) and 3(b).
Based on these experiences, we believe that assuming a
barrel profile during cylinder compression testing is macro-
sensitive to temperature). This flow behavior differs greatly scopically important. To qualitatively evaluate the barreling,
from those of other materials [Fig. 2(a) and (b)]. AZ61A exhibits FE analysis of the cylinder compression of test materials at
relatively high strain hardening at 250
C and 325
C, and very selected temperatures (650
C for Tie6Ale4V, 100
C and
high peak strain (0.25) at 250
C under a 20 s1 strain rate. 400
C for AA6082, 1000
C for AISI 52100, 325
C for AZ61A, and
However, the flow patterns at 400
C differ dramatically from 200
C for AISI 1025) was performed using the flow and

Fig. 3 e Failure cases of materials. (a) fractured Tie6Ale4V warm-forged at 650
C [19]; (b) a fractured AZ61A specimen
cylinder-compressed at a 20 s¡1-strain rate and 400
C.
 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9 2803

Fig. 4 e Predictions of deformation and metal flow lines. (a) Tie6Ale4V at 650
C; (b) AA6082 at 400
C; (c) AISI 52100 at
1000
C; (d) AZ61A at 325
C; (e) AISI 1025 at 200
C; (f) AA6082 at 100
C.

simulation data employed to obtain the FE predictions in where folding may occur, while the metal flow lines of all
Fig. 1. The sample temperatures were based on the tempera- other materials show no discontinuities. A macroscopic dif-
tures for cold forging (AA6082, 0e200
C; AISI 1025, 0e400
C), ference in the sample strain rate causes the temperature to
warm forging (Tie6Ale4V, 500e800
C), and hot forging increase greatly, especially in severely deformed regions,
(AA6082, 300e500
C; AISI 52100, 900e1150
C; AZ61A, because of the lower thermal conductivity or smaller ratio of
250e400
C). the volumetric heat capacity to flow stress.
The deformed shapes and metal flow lines predicted at To quantitatively evaluate the deformed profiles of cylin-
strain rates of 1 and 10 s1 are compared in Fig. 4. The drical specimens (height (H0 ) ¼ 15 mm, radius (R0 ) ¼ 5 mm),
Tie6Ale4V [Fig. 4(a)] and AZ61A [Fig. 4(d)] alloys exhibit the normalized barrel profiles of the six test cases of Fig. 5
distinct discontinuities in the metal flow lines at corners (predicted at a 10 s1-strain rate) are compared in Fig. 5,
2804 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9

Fig. 5 e Comparison of the barrel profile patterns.

Fig. 6 e Geometrical parameters used to quantify the barrel profile. (a) initial; (b) final.
 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9 2805

separate the two former materials from the other materials

Table 3 e Quantification of the barrel profiles using the
tested. The CCEF is thus a key parameter when evaluating
cylinder compression test.
macroscopic instability/formability during plastic deforma-
Temp. Strain Rmax Rmin rmax rmin CCEF
tion of metallic materials. Notably, three parameters (rmax ,
(
C) rate
rmin and the CCEF) did not change as the sample strain rate (1,
(s1)
10, 20 s1) varied for AISI 1025, AA6082 (100
C) or AISI 52100,
Tie6Al 650 1 6.637 5.493 0.327 0.099 0.426
which can easily be forged at the given temperatures.
e4V 10 6.548 5.822 0.310 0.164 0.474
20 6.499 6.069 0.300 0.214 0.514
AA6082 400 1 6.580 5.437 0.316 0.087 0.403
10 6.552 5.471 0.310 0.094 0.404 3. Key parameters of thermal stability and
20 6.546 5.488 0.309 0.098 0.407 formability
AISI 52100 1000 1 6.565 5.622 0.313 0.124 0.437
10 6.566 5.634 0.313 0.127 0.440
Heat generation and transfer are very important when eval-
20 6.566 5.628 0.313 0.126 0.439
uating formability. The heat conduction equation used to
AZ61A 325 1 6.569 5.556 0.314 0.111 0.425
10 6.578 5.975 0.316 0.195 0.511 calculate heat transfer during metal forming is:
20 6.580 6.038 0.316 0.208 0.524

v4
AISI 1025 200 1 6.405 5.912 0.281 0.182 0.463 k4;i þ asε_ ¼ rc (1)
;i vt
10 6.406 5.914 0.281 0.183 0.464
20 6.409 5.913 0.282 0.183 0.465 _ a, and rc are the coefficient of thermal conduc-
where k, s, ε,
AA6082 100 1 6.537 5.732 0.307 0.146 0.453 tivity, effective stress, effective strain rate, heat generation
10 6.561 5.703 0.312 0.141 0.453
ratio, and volumetric heat capacity, respectively. Assuming
20 6.566 5.699 0.313 0.140 0.453
that the coefficient of thermal conductivity (k) is constant and
the heat generation ratio (a) is unity (all plastic work is
transferred to heat), and that the effective strain rate and
together with the ideal lines (radius ¼ 6.26 mm) obtained from stress are the constants of the hot cylinder compression test
*
(i.e., ε_ ¼ ε_ ¼ ε_ and s ¼ s* ), the following approximate equa-
j
homogeneous cylinder compressions of the same specimens.
All profiles exhibit unique characteristics that provide infor- tion for heat transfer is finally obtained:
mation on the flow behaviors.
 *
To quantify the deformation pattern, geometrical param- k s* ε_ v4
4;ii þ ¼ (2)
eters including P1 , P2 , Rmax , and Rmin ¼ ðRmin U þRmin L Þ=2 were rc rc vt
defined, as shown in Fig. 6. P1 and P2 are the intersection
where ε_ is the j-th test strain rate and s* is the mean flow
j
points of the profile with the ideal line of homogeneous cyl-
stress over the strain interval of interest [0, 0.45] at ε_ . The flow
j
inder compression. Rmax and Rmin are the maximum radius
and averaged minimum radius at the two specimen-tool in- curves in Fig. 2 show that the flow stress at a strain of 0.3
terfaces (Radius of the upper interface: Rmin U ; radius of the approximates the average flow stress over the strain range.
*
lower interface: Rmin L ), respectively. The approximate values of s* and ε_ at fixed temperatures are
Table 3 summarizes the maximum radius, averaged min- listed in Table 4.
imum radius, ratio of maximum radius increase near the mid- We now define ND and NV , as follows:
plane and ratio of averaged minimum radius increase at two
specimen-tool interfaces, defined by rmax ¼ ðRmax  R0 Þ= R0 , k
ND ¼ (3)
and rmin ¼ ðRmin  R0 Þ=R0 , respectively. The sum of the two rc
ratios is the cylinder compression evaluation factor (CCEF).
*
The CCEFs of each material were calculated at a selected s* ε_
NV ¼ (4)
temperature, with three strain rates (1, 10 and 20 s1), and the rc
same stroke (35% reduction, average strain ~0.45). The CCEF ND represents the extent of heat diffusion; the larger that
thus quantitatively describes the barreling phenomenon. number, the greater the heat generation and conduction.
Table 3 shows that the rmax values of Tie6Ale4V and
AA6082 (400
C) vary considerably with strain rate, while the
rmin values of Tie6Ale4V and AZ61A vary greatly by the strain
rate. However, only Tie6Ale4V and AZ61A exhibit large CCEF
values and major changes in strain rate. On the contrary, the Table 4 e Flow stresses (in MPa) for selected state
variables at a fixed strain of 0.3.
CCEFs of other materials are within a relatively narrow range
(0.438e0.456). The rmin value is the main factor responsible for Material Temp. (
C) Sample (test) strain rate
CCEF non-uniformity, which is greatly affected by viscous 1 s1 10 s1 20 s1
heating around the corner where folding may occur. Tie6Ale4V 650 663 707 736
Previous studies using our test materials revealed that AA6082 400 61 78 88
Tie6Ale4V and AZ61A were unstable during warm and hot AISI 52100 1000 152 203 221
forgings, as shown in Fig. 3(a) and (b), respectively, while the AZ61A 325 139 158 163
other test materials were stable in terms of large plastic AISI 1025 200 495 516 529
AA6082 100 322 326 329
deformation. We conclude that the CCEF can be used to
2806 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9

Table 5 e Quantitative comparison of the test materials.

Material 4 (
C) *
ε_ (s1) ND (mm2/s) krc (N2/(mm2s
C2)) NV (
C/s) 4m (
C)/M4 () NV NV 4D
M4
Tie6Ale4V 650 1 5.2 47 220 429 343
10 2346 1620/0.51 4578 3660
20 4882 9527 7615
AA6082 400 1 65.7 391 25 134 7
10 321 555/0.19 1715 87
20 721 3852 195
AISI 52100 1000 1 5.4 139 30 120 16
10 397 1424/0.25 1589 214
20 867 3470 468
AZ61A 325 1 42.8 157 72 535 88
10 826 418/0.13 6137 1000
20 1700 12,631 2057
AISI 1025 200 1 11.4 197 119 164 23
10 1242 1450/0.73 1712 236
20 2545 3508 484
AA6082 100 1 65.6 390 132 240 65
10 1338 555/0.55 2435 656
20 2696 4906 1321

Table 5 indicates that the values for AA6082 and AZ61A are conditions. The values of NV =M4 (the viscous heating insta-
larger than those for the other test materials. The ND value of bility index; VHII) are also summarized in Table 5, where
AA6082 is high because of the elevated thermal conductivity, AZ61A (325
C) and Tie6Ale4V (650
C) show larger VHII values
whereas that of AZ61A is high because the volumetric heat than the other materials. Notably, the VHII hierarchy of the
capacity is the lowest among all materials tested. We also test materials was almost the same as that of the CCEF (Table
derived the thermal conductivity values multiplied by the 3).
volumetric specific heat capacities (k rc values). Only the Fig. 7 compares the predicted temperature distributions
AA6082 alloy had a high value, emphasizing that it generates at 35% reduction, strain rates of 1 and 10 s1, and selected
less heat during plastic deformation than the other test ma- temperatures. Only the AA6082 alloy showed no distinct
terials, although AA6082 thermal conduction increases shear band connecting the central spot to the corner of the
rapidly. The relatively low ND values of Tie6Ale4V and AISI specimen-tool interface, where folding may occur. This
52100 are attributable to their low thermal conductivities. phenomenon is related to the very high krc of the AA6082
NV refers to weighted heat generation. This weight is the alloy (Table 5); krc is a measure of thermal diffusion or
inverse of the volumetric heat capacity, such that it can be temperature uniformity. At the 10 s1 strain rate, the tem-
referred to as “accelerated heat generation”. Table 5 shows perature patterns of Tie6Ale4V, AISI 52100, and AISI 1025 are
that the NV values can be divided into three classes. Only the identical. However, the difference between the minimum
Tie6Ale4V alloy belongs to the highest class. The AZ61A and maximum temperatures of the Tie6Ale4V alloy is very
alloy, AISI 1025 at 200
C, and AA6082 at 100
C are in the large, while the flow stress decrement per unit temperature
middle class. The other materials have low NV values. The rise (FSPUT) (4D ) is very high (Table 6). This triggers local
remarkable NV value of Tie6Ale4V is attributable to the folding and thus creates a larger maximum radius than that
relatively high flow stress that developed during warm of Tie6Ale4V [Fig. 7(a)].
forming. The NV value of a useable material must be high at On the contrary, the temperature distribution patterns of
room temperature, or in the cold working temperature range AA6082 and AZ61A, at both 100
C and 400
C, are almost the
(AISI 1025, 200
C; AA6082, 100
C). However, AZ61A has a same, and are characterized by weak shear bands. However,
large NV value even in the hot forming temperature range, for the AZ61A alloy, a large difference in flow stress between
especially compared with AA6082 and AISI 52100 at elevated the near-dead metal zone and other regions is evident
temperatures. because of the high FSPUT value and the relatively large NV
Here, we define the “adjacency coefficient (M4 )” of the compared to the AA6082 alloy (Tables 5 and 6).
temperature of a material as the melting temperature (4m ), as It is apparent that FSPUT greatly influences material
follows: formability. Therefore, this parameter was used to weight the
NV value when defining an index of formability. The values of
j4  4m j this 4D NV , termed the temperature softening formability
M4 ¼ (5)
4m
index (TSFI), are summarized in Table 5. A comparison of
This is a measure of the “hot shortness” probability and 4D NV and NV =M4 reveals some similarity in terms of the order
difference between metalworking conditions according to of magnitude (Table 5). The former value can replace the VHII
temperature. The inverse serves as a weight discriminating value. Also, in Table 3, the TSFI value exhibits the same
hot or warm forming conditions from cold metalworking pattern as the CCEF value.
 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9 2807

Fig. 7 e Predicted temperatures during cylinder compression at selected strain rates. (a) Tie6Ale4V at 650
C; (b) AA6082 at
400
C; (c) AISI 52100 at 1000
C; (d) AZ61A at 325
C; (e) AISI 1025 at 200
C; (f) AA6082 at 100
C.

Table 6 e Average flow stress decrement per unit temperature rise at a fixed strain of 0.3.
Material (Temp. range,
C) Strain rate (s1) Flow stress (MPa) 4D (MPa/
C)
Lowest temperature Middle temperature Highest temperature
Tie6Ale4V (650e820) 10 707 566 e 1.56
AA6082 (400e420) 10 78 76 73 0.27
AISI 52100 (1000e1030) 10 203 194 186 0.54
AZ61A (325e365) 10 158 134 110 1.21
AISI 1025 (200e265) 10 516 510 504 0.19
AA6082 (100e175) 10 326 308 290 0.49
2808 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9

AZ61A causes from its high thermal softening or FSPUT

4. Concluding remarks despite considerable thermal conductivity.

Two indices that evaluate material formability, the CCEF and

TSFI, were presented in this study, according to the cylinder
compression test results of six materials: AISI 1025 Declaration of competing interest
(100e500
C), AISI 52100 (900e1200
C), AA6082 (20e300
C),
AA6082 (350e550
C), Tie6Ale4V (500e800
C), and AZ61A The authors declare that they have no known competing
(250e400
C). The followings were the remarks of this study. financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
 The CCEF value was used to geometrically evaluate tem-
perature softening-mediated instability or formability
during cylinder compression. The CCEF could distinguish Acknowledgement
the materials with low formability (including Tie6Ale4V
and AZ61A alloys, with distinct formability limits being This work was partly supported by Korea Evaluation Institute
demonstrated in previous studies [19,20]) from those with of Industrial Technology (KEIT) (10081334, Development of
good or moderate formability (AA6082, AISI 52100, and AISI Commercialization Technology for High Strength Ti Alloy
1025). Aircraft Fasteners using Warm Forming) and Korea Institute
 Notably, the CCEF values of Tie6Ale4V and AZ61A were of Energy Technology Evaluation and Planning (KETEP)
very large over the test strain rate range compared to the (20214000000520, Human Resource Development Project in
other materials. It is also noted that they changed consid- Circular Remanufacturing Industry) grant funded by the Korea
erably with the strain rate while the other materials did not government (MOTIE).
exhibit a distinct change of the CCEF value with the strain
rate. Therefore, the CCEF value can be a practical measure to
references
evaluate the formability or macroscopic instability of the
materials during metal forming.
 The CCEF value of the Tie6Ale4V and AZ61A alloys are
[1] Ashtiani HRR, Shahsavari P. A comparative study on the
greatly affected by the radius increases of the specimens at phenomenological and artificial neural network models to
the two specimen-tool interfaces. As the ratio of the aver- predict hot deformation behavior of AlCuMgPb alloy. J Alloys
aged minimum radius increase at the two specimen-tool Compd 2016;687:263e73. https://doi.org/10.1016/
interfaces is greatly affected by temperature softening, j.jallcom.2016.04.300.
the CCEF value reflects the effect of such softening. The [2] El Wahabi M, Cabrera JM, Prado JM. Hot working of two AISI 304
steels: a comparative study. Mater Sci Eng A 2003;343:116e25.
CCEF can thus be used as a phenomenological measure of
https://doi.org/10.1016/S0921-5093(02)00357-X.
the effect of temperature softening on metallic materials.
[3] Wei HL, Liu GQ, Zhao HT, Kang RM. Hot deformation
 The TSFI is based on the theory of thermomechanics, i.e., behavior of two CeMneSi based and CeMneAl based
the equation for heat conduction in solids and can be microalloyed high-strength steels: a comparative study.
described as the stress power per the volumetric heat ca- Mater Des 2013;50:484e90. https://doi.org/10.1016/
pacity weighted by the FSPUT. The TSFI characterizes the j.matdes.2013.03.043.
effects of viscous heating on the formability or macro- [4] Guo Z, Li L. Influences of alloying elements on warm
deformation behavior of high-Mn TRIP steel with martensitic
scopic stability of plastic deformation during cylinder
structure. Mater Des 2016;89:665e75. https://doi.org/10.1016/
compression. As the FSPUT value reflects the strength of j.matdes.2015.10.010.
temperature softening and the stress power per volumetric [5] Gao YL, Xue XX, Yang H. Influence of Boron on initial
heat capacity means the strength of heat generation, the austenite grain size and hot deformation behavior of Boron
TSFI is an index of temperature softening. microalloyed steels. Crystals 2015;5:592e607. https://doi.org/
 It was found that the CCEF and TSFI were similarly effec- 10.3390/cryst5040592.
tive for evaluating material formability. Temperature [6] Spezzapria M, Settimi AG, Pezzato L, Novella MF, Forzan M,
Dughiero F, Bruschi S, Ghiotti A, Brunelli K, Dabala M. Effect
softening is one of the major factors affecting the form-
of prior microstructure and heating rate on the
ability of metallic materials. Our study has shed light on austenitization kinetics of 39NiCrMo3 steel. Steel Res Int
the formability characteristics of metallic materials. This 2017;88:1600267.
will optimize process design when using novel or difficult- [7] Menapace C, Sartori N, Pellizzari M, Straffelini G. Hot
to-form materials. deformation behavior of four steels: a comparative study. J
 Other parameters discussed in this study include the co- Eng Mater Technol 2018;140:021006. https://doi.org/10.1115/
1.4038670.
efficient of heat conduction, multiplied or divided by the
[8] Zheng Y, Wang F, Li C, Cheng J, Li Y. Effect of compound
volumetric heat capacity. The former parameter discrimi-
addition of Nb-B on hot ductility of Cr-Mo alloy steel. Mater
nated the AZ61A and AA6082 alloys from other materials, Sci Eng A 2018;715:194e204. https://doi.org/10.1016/
and the latter identified the unique thermal characteristics j.msea.2018.01.001.
of AA6082, i.e., low viscous heating and fast conduction. [9] Gao YL, Xue XX, Yang H. Effect of titanium on hot
 It has been revealed that macroscopic instability of the deformation behaviors of boron microalloyed steel. ISIJ Int
Tie6Ale4V alloy in warm forming is owing to its low 2016;56:619e27. https://doi.org/10.2355/
isijinternational.ISIJINT-2015-605.
thermal conductivity and high flow stress while that of
 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 0 2 3 ; 2 3 : 2 7 9 8 e2 8 0 9 2809

[10] Hu Z, Wang K, Yang Y. A comparative study on the activation [18] Park JH, Ji SM, Choi JM, Joun MS. Accurate flow
energy for hot deformation of 5CrNiMoV and S34MnV steel. characterization of A6082 for precision simulation of a hot
Curr Mater Sci Former Recent Pat Mater Sci 2021;14:70e9. metal forming process. Materials 2022;15:8656.
https://doi.org/10.2174/2666145413999201113144419. [19] Ji SM, Jang SM, Lee YS, Kwak HM, Choi JM, Joun MS.
[11] Wei H, Deng X, Zhou H, Pan H. A comparative study on hot Characterization of Ti-6Al-4V alloy in the temperature
deformation behaviors of niobium microalloyed low-carbon range of warm metal forming and fracture analysis of the
and medium-carbon steels by physical constitutive analysis. warm capping process. J Mater Res Technol
Steel Res Int 2022;93:2100784. https://doi.org/10.1002/ 2022;18:1590e606. https://doi.org/10.1016/
srin.202100784. j.jmrt.2022.03.066.
[12] Bertolini R, Simonetto E, Pezzato L, Fabrizi A, Ghiotti A, [20] Joun MS, Ji SM, Yoo JD, Chung SH, Moon HK, Kim EJ, Yoon DJ,
Bruschi S. Mechanical and corrosion resistance properties of Choi JM, Babu A. Characterization of AZ31B, AZ61A and
AA7075-T6 sub-zero formed sheets. J Adv Manuf Technol AZ80A magnesium alloys with an emphasis on temperature
2021;115:2801e24. compensation for their application to a hot forging. J Manuf
[13] Park WS, Yoo SW, Kim MH, Lee JM. Strain-rate effects on the Process 2022;84:764e85. https://doi.org/10.1016/
mechanical behavior of the AISI 300 series of austenitic j.jmapro.2022.10.054.
stainless steel under cryogenic environments. Mater Des [21] Razali MK, Kim SW, Irani M, Kim MC, Joun MS. Practical
2010;31:3630e40. https://doi.org/10.1016/j.matdes.2010.02.041. quantification of the effects of flow stress, friction,
[14] Prasad YVRK, Seshacharyulu T. Modelling of hot microstructural properties, and the tribological environment
deformation for microstructural control. Int Mater Rev on macro-and micro-structure formation during hot forging.
1998;43:243e58. https://doi.org/10.1179/imr.1998.43.6.243. Tribol Int 2021;164:107226.
[15] Prasad YVRK. Processing maps: a status report. J Mater Eng [22] Joun MS, Lee MC, Eom JG. Intelligent metal-forming
Perform 2003;12:638e45. https://doi.org/10.1361/ simulation. Int Manuf Sci Eng Conf 2011;44304:161e8. https://
105994903322692420. doi.org/10.1115/MSEC2011-50128.
[16] Joun MS, Lee HJ, Lim SG, Lee KH, Cho GS. Dynamic strain [23] Joun MS, Razali MK, Jee CW, Byun JB, Kim MC, Kim KM. A
aging of an AISI 1025 steel coil and its relationship with review of flow characterization of metallic materials in
macroscopic responses during the upsetting process. Int J the cold forming temperature range and its major
Mech Sci 2021;200:106423. https://doi.org/10.1016/ issues. Materials 2022;15:2751. https://doi.org/10.3390/
j.ijmecsci.2021.106423. ma15082751.
[17] Moon HK, Lee JS, Yoo SJ, Joun MS, Lee JK. Hot deformation [24] Jin HT, Choi SD, Joun MS. Experimental and numerical
behavior of bearing steels. J Eng Mater Technol studies on cold forging of an aluminum auto part. Korean
2007;129:349e55. https://doi.org/10.1115/1.2744392. Soc Technol Plast Conf Fall; 2015. p. 109e12.