Int J Adv Manuf Technol (2014) 70:787–796

DOI 10.1007/s00170-013-5292-3

ORIGINAL ARTICLE

Laser surface hardening of 42CrMo cast steel for obtaining

a wide and uniform hardened layer by shaped beams
Peipei Sun & Shaoxia Li & Gang Yu & Xiuli He &
Caiyun Zheng & Weijian Ning

Received: 5 March 2013 / Accepted: 4 September 2013 / Published online: 29 September 2013
# Springer-Verlag London 2013

Abstract For laser surface hardening (LSH) of large-sized time, small deformation after hardening, ease of automation,
workpieces, a wide and uniform hardened layer of a single track and selective hardening of parts [1–3]. Temperature field
is pursued. In this study, two kinds of shaped laser beams were computation of overlapping zone or specific workpiece [4,
used in LSH of 42CrMo cast steel to obtain the required hard- 5], optimization and quantitative evaluation of the LSH pro-
ened layer. One is a stripy spot with uniform-intensity array cess [6], and LSH by shaped beams [7–9] are the current study
spots and the other a stripy spot with intensity blowup in the hot issues of LSH. Back tempering is one of the most critical
edge of the whole array spots. As a comparison, a Gaussian laser problems in laser surface hardening of extended surfaces.
beam was also adopted. A three-dimensional finite element Temper softening in overlapping zone [5, 10, 11] will
model was used to simulate the thermal history of specific points exist in multi-track LSH of large-sized workpieces such
by the latter shaped beam and the Gaussian laser beam. The as cold roll and tug boat of rotary kiln. To solve the problem,
surface morphology, microstructure, microhardness, and unifor- a wide and uniform hardened layer of the single-track LSH is
mity of hardened layers were studied. The results showed that a pursued.
wider and more uniform hardened layer could be obtained using To get a relatively wider hardened layer, using defocusing
the latter shaped beam at relative higher scanning velocities and laser beams, increasing laser power, and decreasing the laser
laser power. The thermal history of a material has an important scanning velocity are usual methods [12]. But the most effec-
effect on the microstructure and microhardness finally formed. tive method is increasing laser beam size at the focal plane.
Due to the high peak temperature and heating rate caused by the Using laser beam shaping technology, a large-sized Gaussian
latter shaped beam, a higher value of microhardness in the beam [13], a wide-band beam [8], and a rectangular beam [7]
transformation hardened zone was found. were designed to get a wider single hardened layer. To get a
relatively more uniform hardened layer, a customized beam
Keywords Laser surface hardening (LSH) . 42CrMo cast with a slope intensity profile [14] and an M-shaped laser beam
steel . Laser beam shaping . Strip array spots profile [9] were proposed and theoretically investigated.
However, little literature has been reported with detailed ex-
perimental results.
1 Introduction In this paper, a new stripy laser beam with one-dimensional
array spots produced by a self-designed diffractive optical
Laser surface hardening (LSH) has been widely used in the element was used. This stripy beam is composed of a row of
automobile and machine tool industries due to short hardening circular spots. According to different laser output powers, and
the material properties and size of the workpieces, the number
of spots and intensity distribution of the whole laser beam can
P. Sun (*)
be designed to get a relatively more uniform hardened layer.
Basic Technology Research Center, AVIC Engine Establishment,
Beijing 100028, People’s Republic of China As an example, two differently shaped laser beams were
e-mail: pipnus@yahoo.cn used in LSH of 42CrMo cast steel in this paper. Considering
the maximum power of the used Nd:YAG laser, shaped laser
S. Li : G. Yu : X. He : C. Zheng : W. Ning
beam 1 is a stripy spot with uniform-intensity array spots of
Key Laboratory of Mechanics in Advanced Manufacturing, Institute
of Mechanics, Chinese Academy of Sciences, Beijing 100190, 37; shaped laser beam 2 is a stripy spot with intensity blowup
People’s Republic of China in the edge of the whole array spots which also has 37 spots. A
788 Int J Adv Manuf Technol (2014) 70:787–796

Fig. 1 Schematic diagram of the

optical system for laser beam
shaping and application

Gaussian laser beam was used as the third one. The surface 2 Experimental procedures
morphology, microstructure, microhardness, and uniformity
of the hardened layers were investigated for the three laser The laser source used was a fiber-coupled 1,000-W Nd:YAG
beams. In order to explain the results, a mathematical model of laser equipped with a self-designed diffractive optical ele-
laser surface hardening of 42CrMo cast steel was established ment. A schematic diagram of the optical system for laser
and the temperature history at specific points was calculated. beam shaping and application is shown in Fig. 1.

Fig. 2 Relative intensity profiles

of shaped beam 1 (a) and shaped
beam 2 (b)
Int J Adv Manuf Technol (2014) 70:787–796 789

Shaped laser beams were used in this study. Shaped beam 1 nominal chemical composition (in weight percent) is 0.42C,
is a stripy spot with uniform-intensity array spots which has 0.39Si, 0.73Mn, 0.014P, 0.004S, 1.01Cr, 0.29Mo, and bal-
37 single-spots, as shown in Fig. 2a. The diameter of every anced Fe. The fabricated sample was cross-sectioned by wire-
single spot is about 0.11 mm. Shaped beam 2 is a stripy spot electrode cutting, polished using standard metallographic
with intensity blowup in the edge of the whole array spots techniques, and then chemically etched in a solution of
which also has 37 single-spots, as shown in Fig. 2b. The HNO3 and C2H5OH. The morphology and microstructure of
intensity proportion of diffractive orders is: the hardened layers were characterized with optical microsco-
py (OM). Microhardness along the height centerline of the
I ð0Þ : I ð1Þ : … : I ð13Þ : I ð14Þ : I ð15Þ : I ð16Þ : I ð17Þ : I ð18Þ ¼ transverse section and in the longitudinal direction at 0.05 mm
1 : 1 : … : 1 : 1:1 : 1:2 : 1:3 : 1:4 : 1:5 below the irradiated surface was measured by an automatic
microhardness tester (HXD-1000B) using a 100-gf test force
Which means that from zero order to the 13th order, the for a dwell time of 15 s.
intensity is the same, which is 1; from the 14th order on, the
intensity grows 10 % every one order.
Figure 3 shows a schematic diagram of laser surface hard-
ening with shaped beams. Shielding gas was not used during 3 Mathematical modeling
the whole process. Two thermocouples were used to record
the temperature at points A and B during the process. As In this section, a three-dimensional finite element model
shown in Fig. 3, the coordinate values of point A were: X = (FEM) was used to simulate the transient temperature field
20 mm, Y =48 mm, Z =8 mm; the coordinate values of point B during the LSH process of the 42CrMo cast steel.
were: X =60 mm, Y =53 mm, Z =8 mm. The dimensions of the
workpiece to be hardened were 100 mm (length)×85 mm 3.1 Assumptions
(width)×8 mm (height).
To compare with the shaped beams, a Gaussian laser beam To simplify the calculation, the following assumptions are
whose defocusing distances were from 3 to 40 mm was also made:
used in the experiments. If the defocusing distance is small, a A1. A combined heat transfer coefficient, h c , is used to
high laser power density will easily lead to vaporization for simulate the effect of radiation and convection. It can
the Gaussian laser beam; when the laser power is appropriate be calculated from the relation as follows:
for LSH, the hardened layer will narrow due to the small size
of the Gaussian laser beam. If the defocusing distance is large hc ¼ 2:41  10−3 εT 1:61 ð1Þ
enough to get a relative wider hardened layer, the output
where ε is the emissivity and T is the temperature.
power would be high enough due to the edge energy loss of
A2. The thermal physical properties of the materials are
the Gaussian laser beam. The selected defocusing distance of
temperature-dependent and piecewise linear with
the Gaussian laser beam is 26 mm to compare with the shaped
temperature.
beams. A laser power of 700 W and a laser scanning velocity
of 10 mm/s were chosen to ensure that the highest temperature
obtained was just below the melting point in the case of the 3.2 Governing equation
Gaussian laser beam.
42CrMo cast steel was used in the experiment. The micro- The transient temperature distribution T(x,y,z,t) can be cal-
structure of the 42CrMo cast steel is cementite and pearlite; its culated from the heat transfer equation with appropriate

Fig. 3 Schematic diagram of

LSH with shaped beam
790 Int J Adv Manuf Technol (2014) 70:787–796

where a is the absorptivity and q m represents the average

thermal flux density.

3.4 Numerical method

A three-dimensional finite element model based on the above

mathematical model was developed with the ANSYS para-
metric design language. As the geometry, boundary condi-
tions, and the heat source were symmetric, half of the geom-
etry was chosen for calculation. The model was shown in
Fig. 4, containing three-dimensional volume elements
Fig. 4 Finite element mesh and geometry of the workpiece
SOLID70 with eight nodes and thermal surface effect element
SURF152 defined by four nodes from the basic body and an
boundary conditions. The governing equation based on the extra node away from the base element. The three-
heat transfer equation is expressed as: dimensional element was used for the basic body structure
      and the thermal surface effect element for the boundary be-
∂ðρcT Þ ∂ ∂T ∂ ∂T ∂ ∂T tween the structure and the environment. A denser mesh was
¼ k þ k þ k ð2Þ
∂t ∂x ∂x ∂y ∂y ∂z ∂z used in the area along the laser line and a coarser mesh for the
where ρ,c, t and k are the density, specific heat capacity, time, rest of the structure.
and thermal conductivity, respectively.

3.3 Boundary conditions

4 Results and discussion
The initial temperature of the materials is room temperature.
4.1 Temperature history of specific points
Therefore, the initial condition is
T ðx; y; z; 0Þ ¼ T a ð3Þ Temperature readings at points A and B (Fig. 3) were used to
evaluate the three-dimensional finite element model. The cal-
where T a is the room temperature. culated and measured temperature history at points A and B
The convection and radiation boundary condition can be by the Gaussian laser beam are shown in Fig. 5. The laser
considered together as power and scanning velocity used were 700 W and 10 mm/s,
∂T respectively. It can be seen that the calculation results show a
k ¼ hc ðT −T a Þ: ð4Þ good agreement with the experimental ones.
∂n
Figure 6 shows a comparison of temperatures for shaped
The effect of the moving laser beam can be considered as a
laser beam 2 and the Gaussian laser beam at node O shown in
surface heat source as follows:
Fig. 3, which is in the middle of the workpiece. The laser
  power and scanning velocity used were 700 W and 10 mm/s,
∂T respectively. The highest temperature in the case of the
k ¼ aqm ð5Þ
∂n Gaussian beam is 1393 °C, which is just below the melting

Fig. 5 Comparison between the measured and simulated temperature history at points A (a) and B (b)
Int J Adv Manuf Technol (2014) 70:787–796 791

4.2 Morphology

To compare the morphology of the hardened layers by the

Gaussian laser beam and the shaped laser beams, cross-
sections of the hardened layers using laser power of 700 W
and scanning velocity of 10 mm/s by the three beams are shown
in Fig. 7. The width of the hardened layer is defined as the width
at the top surface; the depth is defined as the maximum depth of
the hardened layer. It is clear to see that the widths of the
hardened layers by the shaped laser beams are much larger than
that by the Gaussian beam. This means that the shaped beam
has a higher energy efficiency along the width of the hardened
layer due to its large length-to-width ratio.
It can be seen from Fig. 7 that the hardened layer of the
sample hardened by the Gaussian laser beam is crescent-
Fig. 6 Temperature versus time at O for shaped beam 2 and Gaussian beam
shaped, while shaped beams produce a relatively flat-based
or even hardened zone. Kock [15] found similar shapes of the
point; in the case of shaped beam 2, melting occurs. The hardened layer and explained the phenomenon as the energy
heating rate and the peak temperature for shaped beam 2 are absorbed per unit area across the surface perpendicular to the
much higher than the Gaussian beam, and the duration of direction of travel of the laser beam being different. More
austenite transformation is much shorter. All the factors will detailed results about the uniformity will be discussed in
significantly affect the microstructure and microhardness fi- Section 4.5. The width and depth, microhardness, and the
nally formed. microstructure of the hardened layers are very similar for

Fig. 7 Cross-section of the

hardened layers using laser power
of 700 W and scanning velocity of
10 mm/s by the Gaussian beam
(a), shaped beam 1 (b), and
shaped beam 2 (c)
792 Int J Adv Manuf Technol (2014) 70:787–796

Fig. 8 Widths and depths of the hardened layers under different scanning velocities (a) and laser powers (b)

shaped beams 1 and 2, so only the comparison results of shaped The widths and depths of the hardened layers under differ-
beam 2 and the Gaussian beam will be given. ent scanning velocities (laser power, 700 W) using shaped

Fig. 9 OM macrographs (a) and

micrographs of the top (b),
middle-upper (c), lower-middle
(d), and lower (e) zones of the
laser-hardened 42CrMo cast steel
by shaped beam 2
Int J Adv Manuf Technol (2014) 70:787–796 793

beam 2 and the Gaussian beam are shown in Fig. 8a. Figure 8b
shows the widths and depths of the hardened layers under
different laser powers (laser scanning velocity, 20 mm/s).
Both Fig. 8a, b show an obvious increase in the width of the
hardened layer by shaped beam 2. The widths and depths of
the layers decrease with the scanning velocity increasing for a
given laser power and increase with the laser power increasing
for a given scanning velocity by both beam shapes. As shown
in Fig 8a, the widths of the hardened layers in the case of
shaped beam 2 decrease more slowly with the scanning ve-
locity increasing than in the case of the Gaussian laser beam,
and it seems that a wider and no shallow hardened layer by the
shaped laser beam can be obtained at relative higher scanning
velocities. It can be seen from Fig 8b that the depths are nearly
the same using both laser beams when the scanning velocity is
20 mm/s. Fig. 11 Microhardness profile along the depth of the hardened layer by
shaped beam 2 and the Gaussian beam

4.3 Microstructure
It can be seen from Fig. 9d that complete austenization has
Figure 9 shows the microstructure of the cross-section zone of occurred following a high enough peak temperature and a
the hardened layer by shaped beam 2. It can be seen that the rapid quenching process in the laser transformation hardened
layer consists of a laser-melted thin surface layer of about zone. A mixture of needle, lath martensite, and some areas of
100 μm, a laser transition zone, a laser transformation hard- retained austenite can be found. In the lower region of the
ened zone, and a heat-affected zone. In the top area of the heat-affected zone, as shown in Fig. 9e, there are a mixed
hardened layer (Fig. 9b), butterfly martensite grows radially structure of tempered martensite, retained austenite, bainite,
perpendicular to the surface of laser irradiation. In the laser and untransformed pearlite.
transition zone (Fig. 9c), a coarse structure of lath martensite The cross-section microstructure of the hardened layer by
can be found, and the layer is only about 20 μm. This zone the Gaussian beam is shown in Fig. 10; the laser transforma-
presents as a relative dark line in Fig. 9a. tion hardened zone and the heat-affected zone can be

Fig. 10 OM macrographs (a)

and micrographs of the top (b)
and lower (c) zones of the laser-
hardened 42CrMo cast steel by
the Gaussian beam
794 Int J Adv Manuf Technol (2014) 70:787–796

In the case of the Gaussian beam, the martensite zone

hardness is 50–100 HV 0.1, lower than that in the case of
shaped beam 2, besides at the upper part of the zone whose
hardness is nearly the same as the transformation hardened
zone by shaped beam 2. Safdar et al. [16] have reported that
both the heating and cooling rates influence the outcome of
hardening achieved during LSH and that higher hardness was
observed with lower heating rates. Obviously, the result here
is different; the possible reason is the heating rate and peak
temperature both being high enough in the case of shaped
beam 2, as shown in Fig. 6. The high heating rate elevates the
values of the final austenitization and homogenization tem-
peratures, but the peak temperature is high enough, and both
the high heating rate and peak temperature can accelerate the
austenitization and homogenization process.
Fig. 12 Microhardness profile along the width of the hardened layer by Figure 12 shows the microhardness profiles along the
shaped beam 2 and the Gaussian beam at 0.05 mm below the surface
width of the hardened layer by shaped beam 2 and the
Gaussian laser beam at 0.05 mm below the surface. The
distinguished. No melting occurs. The highest temperature is
profiles show a uniform hardness which is about 800 HV
1,393 °C, which is merely below the melting point.
0.1 in the case of shaped beam 2 and about 700 HV 0.1 in
The top region (Fig. 10b) is exactly the same martensitic
the case of the Gaussian beam. Due to the Gaussian distribu-
structure as that found in the hardened layer by shaped beam
tion of the laser beam, hardness in the edge of the layer is a
2, including needle and lath martensite. The heat-affected
little lower than in the center. The high hardness area is
zone has a mixed structure of needle martensite, retained
narrower in the case of the Gaussian beam, which confirmed
austenite, bainite, and untransformed pearlite, as shown in
the different widths of the hardened layers by the Gaussian
Fig. 10c.
beam and the shaped beam well. The higher hardness which
appeared in the case of shaped beam 2 is caused by the
4.4 Microhardness
different microstructure discussed above.
Figure 11 shows the microhardness profiles along the depth of
4.5 Uniformity of the hardened layer
the hardened layers by shaped beam 2 and the Gaussian beam.
The results confirmed the microstructure evolution discussed
As shown in Fig. 13, there are two different hardened layer
above. In the micro-melting zone of shaped beam 2, the
profiles. It is clear to see that the hardened layer profile b is
hardness is about 800 HV 0.1, while in the transition zone
more uniform than profile a. But there is no definition of
the hardness reduces to about 730 HV 0.1 due to the appear-
uniformity of a hardened layer, as yet. To evaluate the
ance of coarse lath martensite. The hardness of the mixed zone
uniformity, a uniformity coefficient R was defined, which is
is in the range of 700–750 HV 0.1, which consists of needle,
related as
lath martensite, and some areas of retained austenite. The
microhardness decreases along the depth with martensite de- s1 s1
R¼ ¼
creasing and the retained austenite increasing. s LH

Fig. 13 Two different shapes of

the hardened layer
Int J Adv Manuf Technol (2014) 70:787–796 795

Fig. 14 Cross-section of the

hardened layers (700 W; 5, 10,
15, and 20 mm/s) by the Gaussian
beam (a), shaped beam 1 (b), and
shaped beam 2(c)

where S 1 denotes the area of the hardened layer profile, S hardened layer profile, L denotes the width of the hardened
denotes the area of the rectangle which exactly contains the layer, and H denotes the depth of the hardened layer. The

Fig. 15 Uniformity of the hardened layers under different scanning velocities (a) and laser powers (b)
796 Int J Adv Manuf Technol (2014) 70:787–796

uniformity coefficient R actually means the geometrical unifor- the laser power is high enough, the uniformity coefficient
mity of the hardened layer. It can be seen that the larger the will approach a fixed value.
uniformity coefficient R, the more uniform is the hardened layer. 3. The thermal history of a material has an important effect
The morphology of the cross-section of the hardened layers on the microstructure and microhardness finally formed.
by the three beams is shown in Fig. 14. The laser power used Shaped beam has a higher hardness in the transformation
was 700 W; the scanning velocities used were 5, 10, 15, and zone due to the high heating rate and peak temperature.
20 mm/s, which correspond to 1, 2, 3, and 4 in Fig. 14.
According to the definition of uniformity of the hardened Acknowledgments This work was supported by the National Natural
layer, values of R are in the range of 62–72 % in the case of Science Foundation of China under grant no. 10832011.
the Gaussian beam (Fig. 14a) and are in the range of 77–85 %
in the case of shaped beam 1 (Fig. 14b); in the case of shaped
beam 2, the values are from 82 to 91 % (Fig. 14c). Values of R
References
in the case of shaped beam 1 are higher than those in the case
of Gaussian beam; the reason is that this beam shape has a
higher energy efficiency along the width of hardened layer, 1. Shin HJ, Yoo YT, Ahn DG, Im K (2007) Laser surface hardening of
S45C medium carbon steel using ND:YAG laser with a continuous
and the reason why values of R in the case of shaped beam 2 wave. J Mater Process Technol 187–188:467–470
are higher than those in the case of shaped beam 1 is that more 2. Wu W, Liang NG, Gan CH, Yu G (2006) Numerical investigation on
energy is supplied in the edge of shaped beam 2. It reveals that laser transformation hardening with different temporal pulse shapes.
laser beam geometry and intensity distribution have an impor- Surf Coat Technol 8:2686–2694
3. Dinesh Babu P, Buvanashekaran G, Balasubramanian KR (2013)
tant influence on the uniformity of the hardened layer. Experimental investigation of laser transformation hardening of low
Uniformity of the hardened layers under the Gaussian alloy steel using response surface methodology. Int J Adv Manuf
beam and shaped beams 1 and 2 is shown and compared in Technol 67:1883–1897
Fig. 15a. Figure 15b shows the values of R under different 4. Xu H, Chen WW, Zhou K, Huang Y, Wang QJ (2010) Temperature
field computation for a rotating cylindrical workpiece under laser
laser powers (laser scanning velocity, 20 mm/s). As illustrated quenching. Int J Adv Manuf Technol 47:679–686
in Fig. 15a, the variety of R seems to have no rules with 5. Giorleo L, Previtali B, Semeraro Q (2011) Modelling of back tem-
scanning velocity increasing. It can be seen from Fig. 15b that pering in laser hardening. Int J Adv Manuf Technol 54:969–977
a plateau exists with laser power increasing, which means that 6. Chen SL, Shen D (1999) Optimization and quantitative evaluation of
the qualities for Nd-YAG laser transformation hardening. Int J Adv
the value of R cannot increase unlimitedly. When the laser Manuf Technol 15:70–78
power is high enough, uniformity of the hardened layer will 7. Zhou P, Ma YX, Wang XL, Ma HT, Xu XJ, Liu ZJ (2009) Average
approach a fixed value. intensity of a partially coherent rectangular flat-topped laser array
propagating in a turbulent atmosphere. Appl Optics 48:5251–5258
8. Li CW, Wang Y, Han B (2011) Microstructure, hardness and stress in
melted zone of 42CrMo steel by wide-band laser surface melting. Opt
5 Conclusion Lasers Eng 49:530–535
9. Hagino H, Shimizu S, Ando H, Kikuta H (2010) Design of a
The process of laser surface hardening of 42CrMo cast steel computer-generated hologram for obtaining a uniform hardened
profile by laser transformation hardening with a high-power diode
by shaped beams has been analyzed. A three-dimensional
laser. Precis Eng 34:446–452
FEM has been developed to calculate the temperature history 10. Hwang J-H, Lee Y-S, Kim D-Y, Youn J-G (2002) Laser surface
of special points to help in understanding the microstructure hardening of gray cast iron used for piston ring. J Mater Eng
and microhardness achieved. The main conclusions are sum- Perform 11:294–300
11. Pantelis D, Bouyiouri E, Kouloumbi N, Vassiliou P, Koutsomichalis
marized as follows:
A (2002) Wear and corrosion resistance of laser surface hardened
1. Due to the large length-to-width ratio for shaped beams, structural steel. Surf Coat Technol 161:125–134
12. Sandven O (1981) Laser surface transformation hardening. Metals
the widths of the hardened layers by these beams were handbook, 9th edn. ASM, Metals Park, OH
much larger than that of the Gaussian laser beam. The 13. Soriano C, Leunda J, Lambarri J, Garcia Navas V, Scanz C (2011)
shaped beam was more suitable for relatively higher laser Effect of laser hardening on the microstructure, hardness and residual
scanning velocities to get a wider and no shallow hard- stresses of austempered ductile iron grades. Appl Surf Sci 257:7101–
7106
ened layer. 14. Leung MKH, Man HC, Yu JK (2007) Theoretical and experimental
2. Laser beam geometry and intensity distribution affect the studies on laser transformation hardening of steel by customized
uniformity of the hardened layer significantly. The uni- beam. Int J Heat Mass Transfer 50:4600–4606
formity of the hardened layers by shaped beam 2 was 15. Kock DJ (2004) Lasers offer unique heat treating capabilities.
www.industrialheating.com
better than the other two beams due to its beam shape and 16. Safdar S, Li L, Sheikh MA, Liu Z (2006) An analysis of the effect of
intensity blowup in the edge. But the uniformity of the laser beam geometry on laser transformation hardening. J Manuf Sci
hardened layers cannot increase unlimitedly, and when E-T ASME 12:8659–8667