Materials and Manufacturing Processes, 31: 838–844, 2016

Copyright # Taylor & Francis Group, LLC

ISSN: 1042-6914 print=1532-2475 online
DOI: 10.1080/10426914.2015.1070431

An Investigation of Machining-Induced Residual Stresses and

Microstructure of Induction-Hardened AISI 4340 Steel
W. Jomaa, V. Songmene, and P. Bocher

Department of Mechanical Engineering, École de technologie supérieure, ÉTS, Montreal, QC, Canada

Excessive induction hardening treatment may result in deep-hardened layers, combined with tensile or low compressive residual stresses.
This can be detrimental to the performance of mechanical parts. However, a judicious selection of the finishing process that possibly follows
the surface treatment may overcome this inconvenience. In this paper, hard machining tests were performed to investigate the residual stres-
ses and microstructure alteration induced by the machining of induction heat-treated AISI 4340 steel (58–60 HRC). The authors demon-
strate the capacity of the machining process to enhance the surface integrity of induction heat-treated parts. It is shown how cutting
conditions can affect the residual stress distribution and surface microstructure. On the one hand, when the cutting speed increases, the
residual stresses tend to become tensile at the surface; and on the other hand, more compressive stresses are induced when the feed rate
is increased. A microstructural analysis shows the formation of a thin white layer less than 2 mm and severe plastic deformations beneath
Downloaded by [Bibliothèque ÉTS] at 16:18 12 March 2016

the machined surface.

Keywords AISI4340; Induction; Machining; Microstructure; Residual; Stresses.

INTRODUCTION thick white layer [4]. The impact of machining white

Many aerospace and automotive systems require layer on parts performances is a subject of debate.
structural parts presenting high in-service perfor- Grinding white layer has been found to reduce fatigue
mance. Gears and crankshafts are among these compo- strength, while hard-machined parts have been found
nents whose surface and subsurface characteristics play to have greater fatigue strength than ground ones,
a key role in controlling their service life. Hence, although the former presents a white layer [5]. Guo
additional surface treatments are used in order to et al. [6] demonstrated that the white layer has negative
enhance their mechanical behavior by introducing effects on the fatigue life of hard-turned parts in rolling
favorable compressive residual stresses, combined with contact applications.
high surface hardness. In fact, it is generally agreed Unlike machining processes (turning and milling),
that surface hardness improves wear resistance and residual stresses have been widely studied in the case of
residual stresses significantly affect the life and perfor- induction surface-hardened materials grinding [2]. Other
mance of structural components. Hashimoto et al. [1] works [7] have found that ceramic cutting tools have excel-
stated that the fatigue strength can be increased by lent performance in the hard machining of AISI D2 and
inducing compressive residual stresses in machined AISI 4340 steels. Therefore, this paper investigates the
parts. Induction heat treatment is one of the surface hard machining of induction heat-treated AISI 4340 steel
processes used, which can produce mechanical parts (58–60 HRC) using mixed ceramic inserts. The work
with high surface hardness and compressive residual focuses on the effects of tool=work material interactions
stress. This process presents good repeatability in terms on residual stresses and microstructure alterations.
of the hardened layer quality, is a short heat-treatment
process (seconds), and is easily incorporated into auto- MATERIALS AND METHODS
mated manufacturing processes [2]. Furthermore, com- The work material is the AISI 4340 steel. It was
pressive residual stresses can be generated along with heat-treated using an induction hardening process on
proper hardened depths. To ensure suitable dimensions an EFD machine equipped with a medium and high
and surface roughness, induction hardening is most frequency generator (1.2 MW total power). Power of
often followed by grinding. However, additional grind- 80 kW was used, at a frequency of 10 kHz, over a
ing always induces tensile residual stresses [3] and a 1.2 s time period. A mixture of water and polymer
(12%) was used as a quenching medium. A finer mar-
tensitic microstructure with a typical hardness of
Received January 29, 2015; Accepted June 23, 2015
Address correspondence to V. Songmene, Department of Mechan-
58–60 HRC was obtained along the case depth of
ical Engineering, École de technologie supérieure, ÉTS, 1100 2 mm after quenching (Fig. 1(a) and (b)).
Notre-Dame Street West, Montreal, QC H3C-1K3, Canada; E-mail: Orthogonal machining was carried out on a MAZAK
[email protected] 3-axis CNC machine (Fig. 2(a)) under dry cutting
Color versions of one or more of the figures in the article can be conditions using PVD TiN-coated mixed ceramic insert.
found online at www.tandfonline.com/lmmp. Its geometry was TNGA 120408-E (KY4400 grade

838
 MACHINING-INDUCED STRESSES AND HARDENED AISI 4340 STEEL 839
Downloaded by [Bibliothèque ÉTS] at 16:18 12 March 2016

FIGURE 1.—(a) Depth of hardened layer, (b) micrograph of hardened layer, (c) axial cross section, and (d) microstructure of bulk material.

Kennametal Inc., Montreal, Canada). A right-hand tool RESULTS AND DISCUSSION

holder DTFNR2525M16KC04 was used in the experi- Figure 3 presents the residual stress distributions
ments. A new cutting edge was used for each test in beneath machined surfaces. These distributions were
order to avoid a possibly tool wear effects on machining analyzed based on surface residual stress (SRS), max-
residual stresses. The workpiece geometry and dimen- imum compressive stress (MCS) level and position,
sions are presented in Fig. 2(b). Different cutting con- and the area of the compressive zone (ACZ). The feed
ditions were tested: cutting speed: V ¼ 54, 128, and rate influences the residual stress distributions sig-
202 m=min; feed rate: f ¼ 0.01, 0.08, and 0.15 mm=rev; nificantly (Fig. 3(a) and (b)). The MCS level increased,
rake angle: c ¼  5 ; depth of cut: DOC ¼ 0.5 mm; and and its position was located deeper beneath the
width of cut: w ¼ 4 mm. A Kistler type 9255B dyna- machined surface as the feed rate increased. Compared
mometer was used for cutting force measurements. to the axial SRS (Fig. 3(b)), the hoop SRS (Fig. 3(a))
The residual stresses were measured in depth using the was somewhat more sensitive to cutting feed
X-ray diffraction technique (Proto iXRD1 system) and variations.
electrochemical polishing process. X-ray parameters The cutting speed affects the residual stress profile in
used in the present work are the following: radiation, different ways, as shown in Fig. 3(c) and (d). The hoop
Cr Ka ; Bragg angle (2h), 156 ; and X-ray elastic SRS tends toward tensile stress when the cutting speed is
constant, 171 MP. The microstructure changes were increased and the inverse occurs for the axial SRS. The
examined using a HITACHI S-3600 N scanning micro- highest SRS and MCS levels in the hoop direction were
scope (SEM; Hitachi, Japan) and an optical microscope, achieved for a low cutting speed (54 m=min), as shown in
model Olympus GX51 (Olympus, Canada). Fig. 3(c), while the highest axial MCS values were

FIGURE 2.—(a) Experimental setup and (b) geometry of specimens.

 840 W. JOMAA ET AL.
Downloaded by [Bibliothèque ÉTS] at 16:18 12 March 2016

FIGURE 3.—Residual stress distributions below the surface. (a) Hoop stress at V ¼ 128 m=min. (b) Axial stress at V ¼ 128 m=min. (c) Hoop stress at
f ¼ 0.08 mm=rev. (d) Axial stress at f ¼ 0.08 mm=rev.

achieved for a high cutting speed (202 m=min), as shown initially, by Loewen and Shaw [8]. This model was
in Fig. 3(d). Furthermore, the MCS location moved dee- modified to take into account friction forces operating
per below the surface as the speed increased. Obviously, on the tool=workpiece interface, and a new equation
the residual stress state below the machined surface is for the temperature of the machined surface was pro-
very sensitive to the feed rate change. Figure 4 indicates posed.
that the hoop ACZ is wider than the axial one. Besides, Based on the work of Waldorf et al. [9], the authors
observing the trends of the ACZ, one can establish a assume a stable dead zone formation during machin-
proportionality relationship between the hoop and the ing since a negative rake angle was adopted in the
axial directions. In fact, during orthogonal cutting, a present study. Thus, three sources of heat generation
plane strain state is established in the cutting zone and were considered: the primary shear zone PSZ (Zone
the hoop and axial residual stresses can be considered I), the secondary shear zone SSZ (zone II), and the ter-
as principal stresses. tiary shear zone TSZ (zone III), as shown in Fig. 5.
In an attempt to understand how residual stress attri- The workpiece and chip temperatures are the result
butes were affected by machining, results relative to of the interaction between these three heating sources.
surface and subsurface alterations were discussed based The physical properties of the work material and the
on cutting forces and temperatures. The latter were cal- tool insert used in the experiments are summarized
culated thanks to a modified analytical model proposed, in Table 1.

FIGURE 4.—Area of compressive zone ACZ versus (a) cutting feed and (b) cutting speed.
 MACHINING-INDUCED STRESSES AND HARDENED AISI 4340 STEEL 841

V c lc
L2 ¼ ð7Þ
2Dw
where r and Ø are the chip thickness ratio and shearing
angle, respectively, which are calculated from the
measured chip thickness tc and the undeformed chip
thickness f. Fc and Ft are the measured cutting and
thrust forces, respectively. cprim is the shear strain calcu-
lated from the shearing Ø and rake a angles, as in [8]. Dw
is the thermal diffusivity of the work material. Vc is the
chip velocity, and lc is the measured chip=tool contact
length.
FIGURE 5.—Geometry of the cutting zone.
The temperature on the machined surface Tworkpiece
can be written as

The temperature on the chip Tchip can be written as Tworkpiece ¼ DTprim þ DTrub þ T0 ð8Þ

Tchip ¼ T0 þ DTprim þ DTsec ð1Þ The temperature rise on the machined surface due to
rubbing, DTrub, is estimated based on a rectangular heat
where DTprim and DTsec are the temperature rises in the source (tool) of uniform strength moving over a semi-
Downloaded by [Bibliothèque ÉTS] at 16:18 12 March 2016

primary and secondary shear zones, respectively, and are infinite body (workpiece). The DTrub is determined as
defined as follows [8]:
qrub lflank A
DTrub ¼ ð1  RÞ ð9Þ
ð1  R1 Þqprim f cscðØÞ Dw
DTprim ¼ 0:754 pffiffiffiffiffiffi ð2Þ
2Dw L1
where
and
Fr V
qrub ¼ ð10Þ
R2 qsec lc wlflank
DTsec ¼ 0:377 pffiffiffiffiffiffi ð3Þ
K w L2
is the heat intensity from rubbing heat source along the
where R1 is the fraction of the heat dissipated into the tool=workpiece interface. The rubbing force Fr is
chip, and is defined as computed as follows [9]:
p a
1 Fr ¼ ster wre tan þ ð11Þ
R1 ¼ 0:754 qffiffiffiffiffiffiffiffiffiffiffi ð4Þ 4 2
Dw cprim
1 þ 1:328 Vf
where lflank is the length of the lower interface between the
qshear is the heat intensity generated on the primary shear tool (built-up region) and the work material (Fig. 5), and
zone, and is defined as is measured on the inserts. ster is the shear stress acting on
the flank face, and is assumed to be equal to the shear
ðFc cos Ø  Ft sin ØÞV cosa flow stress on the primary shear zone kprim:
qshear ¼ ð5Þ
wtc r cosðØ  aÞ csc Ø Fc cos Ø  Ft sin Ø
kprim ¼ sin Ø ð12Þ
and L1 and L2 the velocity-diffusivity factors relative to wf
the primary and secondary shear zones, respectively: The calculated temperatures are displayed in Fig. 6. It
Vf cprim is shown that as the feed rate increases, two opposing
L1 ¼ ð6Þ phenomena occur: the chip temperature increases and
4Dw the workpiece temperature is roughly constant or tends
to decrease at a high feed rate. In fact, the fraction of
the heat evacuated by the chip increases with an increase
TABLE 1.—Properties of the work material and tool insert.
in the chip volume, leading to a reduced workpiece
Workpiece Tool insert temperature. Thus, the thermal effect decreases and
the residual stresses obtained are compressive. More-
Material AISI 4340 [10] Mixed ceramic [11] over, the increase of the cutting and thrust forces
q (kg=m3) 7850 4650
(Fig. 7(a)) when the cutting feed increases favors plastic
K (W=m C) 44 21 deformation, leading to an increase in MCS and ACZ.
Cp (J=kg C) 475 645 As no side spread was observed under the tested cutting
feeds, we postulated that a plane strain state was
 842 W. JOMAA ET AL.

established. These conditions suggested that these

machining tests were performed under roughly the same
degree of constraint [12] and workpiece temperature
(Fig. 6(a)), which resulted in comparable axial SRS.
The variability of the hoop SRS with the cutting feed
can be explained by the fact that under orthogonal cut-
ting conditions, the shearing action is most severe, and
occurs mainly in the cutting direction.
Two opposing phenomena counteract to create the
final residual stress state when varying the cutting speed
and feed. The first is the mechanical effect, which comes
from the increase in the strain rate and results in higher
cutting forces, leading to a larger compressive stress.
The second is the thermal effect, which may result in
excessive heat that can produce tensile residual stress.
Moreover, metallurgical phenomena can take place if
the cutting temperature attains a critical value. This
induces compressive residual stresses thanks to volume
expansion accompanying an eventual phase change [6].
Hence, the actual residual stress state is the result of a
Downloaded by [Bibliothèque ÉTS] at 16:18 12 March 2016

competition among these opposing phenomena (mechan-

ical, thermal, and metallurgical). In the present work, the
observed compressive residual stress at high feeds
indicates that the mechanical (and possibly, the metallur-
gical) effect is predominant. Inversely, when the cutting
speed increases, cutting forces decrease (Fig. 7(b)) while
cutting temperatures increase (Fig. 6(b)), leading to
FIGURE 6.—Calculated workpiece and chip temperatures versus (a) cutting high-intensity thermal effects and tensile residual stresses.
feed and (b) cutting speed. The SEM images revealed the formation of a thin
white layer (less than 2 mm) and plastic deformation
beneath the surface (Fig. 8). The observed white layer
was etch-resistant and has a featureless structure.
Figure 9 shows an increasing tendency of the white layer
and plastic deformed layer thicknesses when the cutting
speed and feed rate increase. Research works [3, 13] have
shown that the white layer forms based on a phase trans-
formation process. However, in the present study, the
workpiece temperature was found to be less than the
phase change temperature of AISI4340 steel (800 C).
This result is in agreement with those obtained in [6].
In fact, a previous study [3] demonstrated that, thanks
to high stresses and strains induced by hard machining,
the phase change process can start at a temperature
lower than the nominal phase change temperature (esti-
mated at 800 C for 4340 steel).
The measurements of residual stresses on a surface
presenting a white layer have been debated within the
scientific community. On the one hand, Warren and
Guo [14] stated that the white layer was associated with
a tensile residual stress state after machining of hard
bearing steel. On the other hand, Ramesh [15] showed,
thanks to a machining model, that white layer formation
induced additional compressive stresses due to phase
change. In this work, the X-ray penetration depth is
thicker than the white layer (5–10 mm versus less than
2 mm). Therefore, the measured SRS is the average of
residual stresses measured over the X-ray penetration
depth, which is a combination of a white layer and the
FIGURE 7.—Cutting Fc and thrust Ft forces versus (a) cutting feed and (b) plastic deformed layer. So, the measured SRSs cannot
cutting speed. be related to only the white layer. Conversely, one does
 MACHINING-INDUCED STRESSES AND HARDENED AISI 4340 STEEL 843

FIGURE 8.—White layer and deformed layer obtained at cutting speed of (a) 56 m=min and (b) 202 m=min and for feed rate ¼ 0.08 mm=rev.

not need to involve white layers to explain residual stress terms of tool life and chip formation is around
trends in the present work as no specific trends are 100–130 m=min, and as a result, no further increase
observed close to the surface. in cutting speed was recommended during hard
Based on the results obtained, we can argue that machining [17].
in the machining of induction-hardened 4340 steel
Downloaded by [Bibliothèque ÉTS] at 16:18 12 March 2016

(58–60 HRC), using a mixed ceramic insert at a cutting CONCLUSIONS

speed equal to 128 m=min and feed rate equal to In this study, the dry orthogonal machining of induc-
0.08 mm=rev can be suitable machining conditions in tion heat-treated AISI 4340 steel was conducted using a
terms of residual stresses and productivity (tool life). mixed ceramic tool. On the one hand, the results showed
For the latter point, Sahin [16] found that the tool life that a compressive residual stress state can be achieved
decreased significantly during the machining of har- after dry hard machining, and the levels of these stresses
dened bearing steel at cutting speeds above 140 m=min. were strongly affected by the machining conditions. On
min. It was found that the optimal cutting speed in the other hand, a thin white layer was formed on all
machined surfaces. The authors suggest that a suitable
selection of cutting speed and feed rate can lead to good
surface integrity characteristics.

REFERENCES
1. Hashimoto, F.; Guo, Y.B.; Warren, A.W. Surface integrity
difference between hard turned and ground surfaces and its
impact on fatigue life. CIRP Annals—Manufacturing Tech-
nology 2006, 55, 81–84.
2. Grum, J. A review of the influence of grinding conditions on
resulting residual stresses after induction surface hardening
and grinding. Journal of Materials Processing Technology
2001, 114, 212–226.
3. Kundrák, J.; Gácsi, Z.; Gyáni, K.; Bana, V.; Tomolya, G.
X-ray diffraction investigation of white layer development
in hard-turned surfaces. The International Journal of
Advanced Manufacturing Technology 2012, 62, 457–469.
4. Guo, Y.B.; Sahni, J. A comparative study of hard turned and
cylindrically ground white layers. International Journal of
Machine Tools and Manufacture 2004, 44, 135–145.
5. König, W.; Berktold, A.; Koch, K.F. Turning versus grinding—
a comparison of surface integrity aspects and attainable accura-
cies. CIRP Annals—Manufacturing Technology 1993, 42, 39–43.
6. Guo, Y.B.; Warren, A.W.; Hashimoto, F. The basic relation-
ships between residual stress, white layer, and fatigue life of hard
turned and ground surfaces in rolling contact. CIRP Journal of
Manufacturing Science and Technology 2010, 2, 129–134.
7. Çiçek, A.; Kara, F.; Kivak, T.; Ekici, E. Evaluation of
machinability of hardened and cryo-treated AISI H13 hot
FIGURE 9.—Machining-affected layer thickness versus (a) cutting feed and work tool steel with ceramic inserts. International Journal of
(b) cutting speed. Refractory Metals and Hard Materials 2013, 41, 461–469.
 844 W. JOMAA ET AL.

8. Loewen, E.G.; Shaw, M.C. On the analysis of cutting tool cryogenic cooling conditions. The International Journal of
temperature. Transactions of ASME 1954, 76 (2), 217. Advanced Manufacturing Technology 2013, 64, 633–642.
9. Waldorf, D.J. A simplified model for ploughing forces in 14. Warren, A.; Guo, Y. Characteristics of residual stress profiles
turning. Journal of Manufacturing Processes 2006, 8, 76–82. in hard turned versus ground surfaces with and without a
10. Mabrouki, T.; Rigal, J.-F. A contribution to a qualitative white layer. Journal of Manufacturing Science and Engineering
understanding of thermo-mechanical effects during chip for- 2009, 131.
mation in hard turning. Journal of Materials Processing Tech- 15. Ramesh, A.; Melkote, S.N. Modeling of white layer forma-
nology 2006, 176, 214–221. tion under thermally dominant conditions in orthogonal
11. Brito, R.F.; Carvalho, S.R.d.; Lima e Silva, S.M.M.d.; machining of hardened AISI 52100 steel. International Journal
Ferreira, J.R. Thermal analysis in coated cutting tools. Inter- of Machine Tools and Manufacture 2008, 48, 402–414.
national Communications in Heat and Mass Transfer 2009, 36, 16. Sahin, Y. Comparison of tool life between ceramic and cubic
314–321. boron nitride (CBN) cutting tools when machining hardened
12. Liu, C.R.; Barash, M.M. Variables governing patterns of steels. Journal of Materials Processing Technology 2009, 209,
mechanical residual stress in a machined surface. Journal of 3478–3489.
Engineering for Industry 1982, 104, 1–10. 17. Poulachon, G.; Moisan, A.; Jawahir, I. Tool-wear mechan-
13. Umbrello, D. Analysis of the white layers formed during isms in hard turning with polycrystalline cubic boron nitride
machining of hardened AISI 52100 steel under dry and tools. Wear 2001, 250, 576–586.
Downloaded by [Bibliothèque ÉTS] at 16:18 12 March 2016