Laser welding of low carbon steel and thermal stress analysis
Content
ARTICLE IN PRESS
Optics & Laser Technology 42 (2010) 760–768
Contents lists available at ScienceDirect
Optics & Laser Technology
journal homepage: www.elsevier.com/locate/optlastec
Laser welding of low carbon steel and thermal stress analysis
B.S. Yilbas n, A.F.M. Arif, B.J. Abdul Aleem
Mechanical Engineering Department, KFUPM, Saudi Arabia
a r t i c l e in f o
Article history: Received 13 November 2008 Received in revised form 17 November 2009 Accepted 26 November 2009 Available online 21 December 2009 Keywords: Laser welding Temperature Finite element model
a b s t r a c t
Laser welding of mild steel sheets is carried out under nitrogen assisting gas ambient. Temperature and stress fields are computed in the welding region through the finite element method. The residual stress developed in the welding region is measured using the XRD technique and the results are compared with the predictions. Optical microscopy and the SEM are used for the metallurgical examination of the welding sites. It is found that von Mises stress attains high values in the cooling cycle after the solidification of the molten regions. The residual stress predicted agreed well with the XRD results. & 2009 Elsevier Ltd. All rights reserved.
1. Introduction Laser is widely used as a thermal source for industrial applications. This is because of the local treatment, precise operation, and short processing time. One of the important industrial applications of laser processing is the laser welding, which offers considerable advantages over the conventional welding methods. High intensity laser beam melts and partially evaporates the welded material during the process. Attainment of high temperature gradient during the heating and cooling periods results in the development of high thermal stresses in the welding zone. Once the cooling period ends, the residual stress in the welding zone is resulted. This, in turn, influences the mechanical performances of the resulting welds. Consequently, investigation into thermal stress development and residual stress formation in the weld zone becomes essential. Considerable research studies were carried out to examine the laser welding process. An extensive review on laser welding and related process was carried out by Mackwood and Crafer [1]. They presented the applications laser welding processes under different welding categories such as laser spot welding, laser butt welding, etc. Nd:YAG laser repair welding of tool steels and microstructural changes in the welded region were examined by Vedani[2]. They indicated that heat affected zone was narrow and carbides dissolve during the heating phase of the welding process. Li et al. [3] studied the laser forming process using the finite element method. They showed that thermal expansion and
n
Corresponding author. Tel.: + 966 3 860 4481; fax: +966 3 860 2949. E-mail address: [email protected] (B.S. Yilbas).
contraction took place during the laser processing, which in turn allowed the thermo-mechanical forming of complex shapes. Laser welding and heating analysis were carried out by Papadakis and Tobias [4]. They examined the lap and fillet seam welds and measured the distortion during the processing. Moreover, the experimental findings were compared with the model predictions. A high speed laser pulse welding of metallic substrates was examined by Holtz et al. [5]. They demonstrated that an improvement in the structural integrity of the weld site was resulted for high speed pulsed contour welding than the seam welding process. The transient effects on the formation of laser produced weld pool were studied by Ehlen et al. [6]. They accommodated the Marangoni effect using a semi-empirical model for the temperature dependent surface tension gradient. The thermal stress developed during the lap welding of thermoplastic films was examined by Coelho et al. [7]. They showed that the influence of thermal stresses and expansion and contraction forces played an important role on the achievement of strong welds; in which case, tensile strength improved 80% as compared to the original thermo-plastic substrate. The high power laser welding of construction steel was investigated by Engstron et al. [8]. They indicated that laser welding offered new design opportunities, which improved the manufacturing process by shortening lead and production times as well as reducing the amount of materials used. The metallurgical and residual stress evaluation of CO2 laser welded super-austenitic stainless steel was carried out by Zambon et al. [9]. They showed that the residual stress was tensile and close to the yielding strength of the substrate material in the longitudinal direction in the weld bead while the stresses were compressive in the transverse direction in the base material. Laser welding of steel and residual stress
0030-3992/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.optlastec.2009.11.024
ARTICLE IN PRESS
B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768 761
distribution were examined by Olabi et al. [10]. They examined the effect of laser parameters on the residual stress developed through the statistical analysis. Laser welding and the simulation of temperature field were carried out by Zeng-rong et al. [11]. They indicated that the finite element method predictions were in agreement with the experimental findings. Laser welding characteristics of cold rolled carbon steel were examined by Shin et al. [12]. They showed that the optimal welding conditions, which resulted in no defect sites in the vicinity of the welded area, were possible for a certain combination of laser output power, welding speed, and focus setting of focusing lens. The influence of shielding condition on the weld quality of underwater laser welding was investigated by Zhang et al. [13]. They showed that there existed the relationship between shielding conditions and the quality of the weld bead. The porosity formation and penetration in pulsed laser welding were examined by Zhou and Tsai [14]. They indicated that the porosity formation was strongly related to the aspect ratio (depth to width) of the key hole formed during the welding process. The hybrid laser beam welding and its applications were studied by Mahrle and Beyer [15]. They demonstrated some practical examples of potential hybrid technologies, which would be used to extend the application spectrum of the laser beam welding. Laser microwelding of copper and aluminum was investigated by Ihor and Schmidt [16]. They showed that an application of suitable filler material helped in avoiding the brittle intermetallic phases at the interface between copper and the solidified melt in the welded joints. The experimental study on welding characteristics of CO2 laser TIG hybrid welding process was carried out by Chen et al. [17]. They indicated that when the current was increased to a critical value, the laser induced key hole disappeared and the arc expanded, which, in turn, decreased the penetration depth. The mechanical and metallurgical aspects of tailored welded blanks were examined by Bayraktar et al. [18]. They showed that the fracture surfaces tested at low temperatures presented the fracture topography for cleavage in the notched section for steel samples. However, at high temperatures, the fractured always appeared in the ductile mode in the specimens. In the present study, laser welding of low carbon steel plates is carried out. Temperature and thermal stress fields are modeled using the finite element method. The residual stress developed in the welding zone is measured using the XRD technique and compared with the predictions. The metallurgical and morphological changes in the weld zone are examined using the optical microscopy, the SEM, the EDS and the XRD.
Laser Beam Laser Irradiated Spot z x y Weld section 2 mm 30 mm
Fig. 1. View of laser melting process.
Nitrogen
x=0 y=0 z=0
1 3 5 7 9 11 13
40 mm
U
In the case of a moving heat source along the z-axis with a constant velocity U, energy gain by the substrate material yields
15 17 19
r
DE @E @E ¼ r ÀrU Dt @t @z
ð2Þ
or DE @ðCpTÞ @ðCpTÞ r ¼r -rU Dt @t @z Combining equations (1) and (3) yields ð3Þ
21 23 25 27 29 31 33 35
@ðCpTÞ @ðCpTÞ r ð4Þ ¼ ðrðkrTÞÞ þ rU þSo @t @z At the free surfaces of the welded workpiece (the irradiated surface, Fig. (1)), the convective boundary is assumed and at the rear side of the workpiece, the convective and radiative boundary conditions are considered. Therefore, the corresponding boundary condition is: At the irradiated surface @T h ¼ ðTs ÀTamb Þ @z k and at the rear side of the surface: @T h es 4 4 ðT ÀTamb Þ ¼ ðTs ÀTamb Þ þ k s @y k and @T h ¼ ðTs ÀTamb Þ @z k ð5Þ
ð6Þ
37 39 41 43 45 47 49 51 53 55 57
where h is the heat transfer coefficient due to natural convection, and Ts and Tamb are the surface and ambient temperatures, respectively, e is the emissivity (e =0.63 is considered [19]), and s is the Stefan–Boltzmann constant (s =5.67 Â 10 À 8 W/m2 K). At far away boundary (at edges of the solution domain) constant temperature boundary is assumed (T= 293 K), i.e. x ¼ 1; y ¼ 1; z ¼ 1-T ¼ 293K Initially (prior to laser treatment), the substrate material is assumed to be at constant ambient temperature, i.e. T =Tamb, which is considered as constant (Tamb = 293 K). Equation (4) is solved numerically with the appropriate boundary conditions to predict the temperature field in the substrate material. However, to analyze the phase change problem, a non-linear transient thermal analysis is performed employing the enthalpy method. To account for latent heat evolution during phase change, the enthalpy of the material as a function of temperature is incorporated in the energy equation.
2. Heating analysis Since the welding situation is involved with the transient heating during laser scanning, a transient equation for heat transfer needs to be considered. In this case, the transient heat transfer equation in the Cartesian coordinates is
r
DE ¼ ðrðkrTÞÞ þ So Dt
ð1Þ
where D represents the material derivative or substantial D @ @ @ @ derivative Dt ¼ u @x þ v @y þ w @z þ @t , E is the energy gain by the substrate material and So is the volumetric heat source term and So ¼ Io dð1Àrf ÞeÀdy eðÀx
2
3. Modeling of thermal stresses 59 For structural response, the finite element formulation is based on the principle of virtual work. From the principle of virtual work (PVW), a virtual (very small) change in the internal strain energy (dU) must be offset by an identical change in external work due to the applied loads (dV). Considering the strain energy due to thermal stresses resulting from the constrained motion of a body 61 63 65
þ z2 =a2 Þ
Io is laser peak intensity, d is the absorption depth, rf is the surface reflectivity, r is the density, Cp is the specific heat capacity, k is the thermal conductivity, a is the Gaussian parameter, and x and z are the axes while the laser beam scans the surface along the z-axis and the laser beam axis is the y-axis (Fig. (1)).
ARTICLE IN PRESS
762 B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768
during a temperature change, PVW yields Z Z T T ½BT ½D½Bdvfug ¼ fdug ½BT ½Dfeth gdv fdug
vol vol
residual stress (s) is given [20] ð7Þ
s¼
E ðdn Àdo Þ ð1 þ uÞSin2 c do
ð8Þ
Noting that the {du}T vector is a set of arbitrary virtual displacements common in all of the above terms; the condition required to satisfy above equation reduces to ½Kfug ¼ fF th g where R ½K ¼ vol ½BT ½D½Bdv ¼ element stiffness matrix R fF th g ¼ vol ½BT ½Dfeth gdv ¼ element thermal load vector feth g ¼ fagDT ¼ thermal strain vector fag ¼ vector of coefficients of thermal expansion In the present study, the effect of mechanical deformation on heat flow has been ignored and the thermo-mechanical phenomenon of melting is idealized as a sequentially-coupled unidirectional problem. For thermal analysis, the given structure is modeled using the thermal element (SOLID70). SOLID70 has a 3D thermal conduction capability (ANSYS code). The element has eight nodes with a single degree of freedom, and temperature, at each node. The element is applicable to a 3D, steady-state or transient thermal analysis. Since the model containing the conducting solid element is also to be analyzed structurally, the element is replaced by an equivalent structural element (such as SOLID45) for the structural analysis. SOLID45 is used for the 3D modeling of solid structures. The element is defined by eight nodes having three degrees of freedom at each node: translations in the nodal x, y, and z directions. The element has plasticity, creep, swelling, stress stiffening, large deflection, and large strain capabilities. The thermal and structural properties used in the current simulations are given in the Table 1. It should be noted that the conditions for the current simulations resemble the actual experiments carried out in the present study.
where E is the Young’s modulus, n is the Poisson’s ratio, c is the tilt angle, and di are the d spacing measured at each tilt angle. If there are no shear strains present in the specimen, the d spacing changes linearly with Sin2 c.
3000 2400 1800 1200 600 0 0.0E+00 2000 t=2s TEMPERATURE (K) 1500 y = 0 mm; z = 15 mm y = 0.8 mm; z = 15 mm y = 1.6 mm; z = 15 mm t=1s y = 0 mm; z = 15 mm y = 0.8 mm; z = 15 mm y = 1.6 mm; z = 15 mm
TEMPERATURE (K)
2.0E-03
4.0E-03 6.0E-03 DISTANCE (m)
8.0E-03
1.0E-02
1000
500
4. Residual stress measurements XRD Technique: The measurement relies on the stresses in fine grained polycrystalline structure. The position of the diffraction peak undergoes shifting as the specimen is tilted by an angle c. The magnitude of the shift is related to the magnitude of the residual stress. The relationship between the peak shift and the
Table 1 Mechanical and thermal properties of mild steel used in the simulations. Temperature (K) E (GPa) Temperature (K) a  10 À 6 (1/K) k (W/mK) Cp (J/kgK) u r (kg/m3) 294 203 273 11.2 51.9 486 0.3 7700 373 11.2 50.7 486 366 199 473 12.1 48.2 515 477 191 573 13.0 45.6 548 589 184 673 13.6 41.9 586 644 176 773 14.0 38.1 649 700 167 873 14.6 33.9 708 973 14.8 30.1 770 755 154 1073 11.8 24.7 624 811 141 1273 13.6 26.8 548 866 124 1473 13.6 29.7 548
0 0.0E+00
2.0E-03
4.0E-03 6.0E-03 DISTANCE (m)
8.0E-03
1.0E-02
Fig. 2. Temperature distribution along the x-axis at different depths below the surface and for two time periods.
Table 2 . Laser welding conditions. Feed Rate (mm/min) 1000 Power (W) 2000 Frequency (Hz) 1500 Nozzle Gap (mm) 1.5 Nozzle Diameter (mm) 1.5 Focus setting (mm) 127 N2 Pressure (kPa) 600
ARTICLE IN PRESS
B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768 763
Fig. 3. Three-dimensional view of temperature distribution during welding process for two time periods. Temperature is in K.
5. Experimental The CO2 laser (LC-ALPHAIII) delivering nominal output power of 2 kW at pulse mode with different frequencies is used to irradiate the workpiece surface. The nominal focal length of the focusing lens is 127 mm. Nitrogen assisting gas emerging from the conical nozzle and co-axially with the laser beam is used. The welding conditions are given in Table 2. The workpiece accommodated is mild steel at 2 mm in thicknesses. Material characterization of the laser nitrided surfaces is carried out using the SEM, the XRD and the XPS. The Jeol 6460 electron microscopy is used for the SEM examinations and the Broker D8 Advanced having MoKa radiation is used for the XRD analysis. A typical setting of the XRD was 40 kV and 30 mA. Microphotonics digital microhardness tester (MP-100TC) was
used to obtain microhardness across the weld zone (parallel to the workpiece surface). The standard test method for Vickers indentation hardness of advanced ceramics (ASTM C1327–99) was adopted. The measurements were repeated three times at each location.
6. Results and discussions Laser welding of low carbon steel sheets is carried out. Temperature and stress fields developed during the welding are computed using the finite element method. The residual stress developed in the weld region is measured using the XRD technique while metallurgical and morphological changes in the
ARTICLE IN PRESS
764 B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768
1.0E+09
STRESS (Pa)
1.0E+08 t=1s 1.0E+07
y = 0 mm; z = 15 mm y = 0.8 mm; z = 15 mm y = 1.6 mm; z = 15 mm
1.0E+06 0.0E+00 1.0E+09
2.0E-03
4.0E-03
6.0E-03
8.0E-03
1.0E-02
DISTANCE (m)
1.0E+08 STRESS (Pa)
1.0E+07 t=2s 1.0E+06
y = 0 mm; z = 15 mm y = 0.8 mm; z = 15 mm y = 1.6 mm; z = 15 mm
1.0E+05 0.0E+00
2.0E-03
4.0E-03
6.0E-03
8.0E-03
1.0E-02
DISTANCE (m)
Fig. 4. von Mises stress distribution along the x-axis at different depths below the surface and for two time periods.
weld zone are examined using the optical microscopy, the SEM and the EDS. Fig. (2) shows temperature variation along the x-axis at three different locations in the y-axis while Fig. (3) shows 3D view of temperature distribution. The z-axis location is 15 mm from the welding starting point (Fig. (1)) in Fig. (2). It should be noted that the time t= 1 s corresponds to the laser beam spot center at the location x = 0, y= 0, and z= 15 mm. Temperature decays sharply in the surface region and as the depth below the surface increases temperature decay becomes sharp and further increase in the depth results in gradual decay of temperature. In the region, where the sharp delay occurs, temperature gradient attains high values. This is more pronounced at some depth below the surface. Temperature in the surface region exceeds the melting temperature of the substrate material. Although temperature at the melting–solid interface remains the same – due to the influence of latent heat of fusion on the energy transfer in the vicinity of melting–solid interface – temperature gradient in the molten region becomes different in the solid phase. Moreover, the decay rate of temperature as well as its magnitude changes at different y-axis locations in the melting zone. The y-axis location represents different depths below the surface inside the substrate material. The difference in temperature in the molten zone is attributed to the heat conduction in this region. In the case of t =2 s heating duration, the laser beam scans over this region and energy transfer from the surface through convection and
conduction reduces temperature significantly in this region. The heat conduction becomes the sole mechanism after this time in this region. In addition, temperature profiles along the depth below the surface becomes almost self similar. Fig. (4) shows von Mises stress along the x-axis at tree depths below the surface for the same z-axis location as shown in Fig. (1) for two heating periods while Fig. (5) shows 3D view of von Mises stress in the substrate material. von Mises stress attains low values in the surface region because of the melting; in which case, molten metal is free to expand in this region. von Mises stress increases sharply as the depth below the surface increases along the y-axis. Moreover, it remains high as the distance increases further. This sharp rise of von Mises stress is related to the sharp increase in temperature gradient at some depth below the surface. The variation of von Mises stress along the x-axis is similar at different y-axis locations, since the difference in the stress magnitude is negligibly small. This situation is associated with the similar behavior of temperature distribution at different y-axis locations along the x-axis. In the case of time period t =2 s, von Mises stress rises with increase in distance along the x-axis. This is true for all the y-axis locations. This increase is attributed to the strain developed at early periods when the laser peak intensity scans over the z-axis locations corresponding to the figure. Therefore, thermal stress developed during high temperature heating contributes to the stress field developed during the cooling period for t = 2 s. Consequently, von Mises stress for this time period does not follow exactly the temperature distribution as shown in Fig. (2). Moreover, the maximum stress level along the x-axis is below the elastic limit of the substrate material. Fig. (6) shows temperature distribution along the z-axis (laser scanning direction) at three different y-axis locations and for three time periods. It should be noted that the x-axis location is x= 0 (Fig. (1)). The location of peak temperature changes along the z-axis for different periods, which is because of the location of laser beam intensity along the scanning direction (z-axis). In this case, the laser peak intensity is at z= 1.5 Â 10 À 2 m for t =1 s, z= 2.25 Â 10 À 2 m for t =1.5 s, and z = 3 Â 10 À 2 m for t = 2 s. The small irregular behavior is observed in temperature profile at about 1600 K. This is because of the phase change process (melting), which suppresses the rate of rise of temperature in the solid due to the latent heat of fusion. However, the sharp increase in temperature occurs at a location where the laser beam intensity is the maximum. This is due to the absorption of the laser beam energy by the substrate material, which enhances significantly the internal energy gain in this region. Therefore, temperature increases sharply in the molten material. This situation is true for different y-axis locations, provided that temperature gradient around the peak temperature reduces as the y-axis location increases below the surface. The decrease in time derivative of temperature (qT/qt) is because of less energy transfer through conduction from the surface in this region where the laser energy absorption is significantly high. The rise and delay of temperature around the peak temperature is not the same. This suggests that temperature remains high at the surfaces during the heating period when the laser scans the workpiece surface. However, once the welding is over, temperature delay is sharp in the cooling cycle. This is true for all the y-axis locations and for all periods in the cooling cycle. Fig. (7) shows von Mises stress along the z-axis for three y-axis locations and three time periods. It should be noted that the locations and time periods are the same as Fig. (6). von Mises stress attains high values immediately after temperature reaches its peak value (Fig. (6)). This is associated with the temperature gradient, which is higher after the peak temperature. It should be noted that material in the liquid phase, where temperature
ARTICLE IN PRESS
B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768 765
Fig. 5. Three-dimensional view of von Mises stress distribution during welding process for two time periods. Stress is in Pa.
reaches its peak, expands freely and stress level becomes very low. However, upon decaying from the peak temperature, the solidification takes place rapidly and conduction cooling from the heated region results in attainment of high temperature gradient. This causes a sharp increase in von Mises stress in this region. von Mises stress level rises sharply and decays gradually for all time periods and the y-axis locations, despite the fact that temperature decays sharply after its peak value. Consequently, strain developed during the heating period plays an important role on the stress formation after the laser beam passes over the concerned location along the z-axis (laser scanning direction). The maximum value of von Mises stress attains almost 195 MPa, which is less than the elastic limit of the substrate material. This is true for all the y-axis locations and time periods. Moreover, the
decay in von Mises stress is observed before rising to its maximum. This may be associated with the initial solidification and temperature variation immediately after the solidification, i.e. temperature change is small in the region of solidification resulting in attainment of low temperature gradient while lowering von Mises stress in this region. Fig. (8) shows optical photographs of welding surface and different regions in the weld cross-section while Fig. (9) shows the SEM micrographs of the weld cross-section. The heat affected zone (HAZ) is evident from the SEM micrograph and the heat affected zone extends both sides of the weld section almost at a same size. This indicates the plane symmetry of the heating during the actual welding process, which is also adapted in the simulations. The grain size coarsening is evident in the HAZ due to
ARTICLE IN PRESS
766 B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768
3000 TEMPERATURE (K)
1.00E+09
x = 0 mm and y = 0 mm
1800 1200 600 0 0.0E+00 3000 t=1s t=2s 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
STRESS (Pa)
2400
1.00E+08 1.00E+07 1.00E+06 1.00E+05 0.0E+00 1.00E+09
t=1s t=2s
x = 0 mm and y = 0 mm 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
DISTANCE (m) TEMPERATURE (K) x = 0 mm and y = 0.8 mm
DISTANCE (m) t=1s t=2s
1800 1200 600 0 0.0E+00 3000 6.0E-03 t=1s t=2s 1.2E-02 1.8E-02 2.4E-02 3.0E-02
STRESS (Pa)
2400
1.00E+08 1.00E+07 1.00E+06 1.00E+05 0.0E+00 1.00E+09
x = 0 mm and y = 0.8 mm 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
DISTANCE (m) TEMPERATURE (K)
DISTANCE (m) t=1s STRESS (Pa) 1.00E+08 1.00E+07 1.00E+06 1.00E+05 0.0E+00 t=2s
2400 1800 1200 600
x = 0 mm and y = 1.6 mm
t=1s t=2s 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
x = 0 mm and y = 1.6 mm 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
0 0.0E+00
DISTANCE (m)
Fig. 6. Temperature distribution along the z-axis at different depths below the surface and for two time periods.
DISTANCE (m)
Fig. 7. von Mises stress distribution along the z-axis at different depths below the surface and for two time periods.
the slow cooling rates in this region. However, some fine grains are observed in the region away from the weld interface. The grain refining is formed because of high cooling rates due to high temperature gradient. In general, the fine grains consist of circular ferrite and fine pearlite structures (Fig. (9)), in which case, the light region is ferrite while dark region is pearlite. Moreover, the nucleation of pearlite results in partial formation of cementite lamellae structure. The pearlitic modulus grows from the nuclei along the austenite boundaries and the scattered colonies of alternating ferrite and cementite lamellae are formed within the pearlitic structure. In the weld core, the austenite phase allowing a considerable grain growth is observed. In this case, grain boundary ferrite and ferrite/carbide aggregate appear to be fine pearlite. In addition, in the fusion zone, very fine grains, as well as locally scattered large grains, where crystallization takes place, are evident. Fig. (10) shows microhardness distribution across the welding cross-section along the line parallel to the free surface (transverse to the weld section). The hardness decays sharply with increase in distance across the weld. The peak hardness is almost 40% higher than the base material hardness. It should be noted that the increase in hardness is one of the indications of formation of fine structures and the residual stress developed during the welding process. The residual stress measured using the XRD technique
and observed from the simulations are 105 MPa and 90 MPa, respectively. It should be noted that the residual stress is measured 2.5 mm away from the welding line and at the midway of the welded sample along the welding direction. Moreover, the prediction of the residual stress is compared at the same location of the measurement. It can be observed that both results are in agreement (in the order of 100 MPa), provided that a small discrepancy occurs, which is because of the assumptions made in the simulations such as uniform structure and mechanical properties are considered.
7. Conclusions Laser welding of mild steel sheets is carried out and stress field developed in the welding zone is modeled using the finite element method (FEM). The residual stress developed in the weld region is measured using the XRD technique and compared with the FEM predictions. The morphological and metallurgical changes in the welding region are examined using the optical microscopy and the SEM. Microhardness distribution across the welding zone is measured. It is found that temperature decay rate in the molten zone is lower than in the solid. This is because of the absorption and dissipation of the laser energy in the molten zone,
ARTICLE IN PRESS
B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768 767
2 mm Top view of laser welded section
Fusion Zone Fusion Zone Heat Affected Zone Base Material Partially formed pearlite
1 mm Cross-section of laser welded section. Heat Affected Zone
Fig. 9. SEM micrographs of welding cross-section.
200 180 HARDNESS (HV) 160 140 120 100 0.0E+00
0.2 mm Fusion Zone and Heat Affected Zone Interface.
Fig. 8. Optical photographs of welding cross-section.
6.0E-04
1.2E-03
1.8E-03
2.4E-03
3.0E-03
which is generated in the surface region. This, in turn, increases the internal energy gain in this region. In the solid region, temperature variation along the x-axis becomes similar for all the y-axis locations. This results in similar von Mises stress distribution along the x-axis provided that at large y-axis locations (at some depth below the surface); the strain developed in this region modifies this behavior slightly due to temperature increase, which reaches its peak at the z-axis location when the laser beam intensity is the maximum. Moreover, once the laser beam scans over this region, temperature decays sharply. This causes sharp increase in von Mises stress due to the attainment of high temperature gradient in this region. This situation is true for all the y-axis locations and time periods considered in this study. However, the maximum stress level is less than the elastic limit of the substrate material. The residual stress predicted from the FEM agrees well with the XRD results, which is in the order of 100 MPa. The grain coarsening occurs in the HAZ and grain refining and recrystallization takes place in the fusion zone as observed from the optical and the SEM micrographs. The closed examination of welding section reveals that the weld section is free from cracks and voids.
DISTANCE (m)
Fig. 10. Microhardness across the weld cross-section and parallel to the free surface.
Acknowledgements The authors acknowledge the support of King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia, for the funded project, Project #SB080003.
References
[1] Mackwood. AP, Crafer RC. Thermal modeling of laser welding and related processes: a literature review. Optics and Laser Technology, 2005;37:99–115. [2] Vedani M. Microstyrctural evolution of tool steels after Nd-YAG laser repair welding. Journal of Materials Science 2004;39:241–9. [3] Li LQ, Chen YB, Wang WY, Lin SY. FEM simulation for laser forming processing. Acta Metallurgica Sinica (English Letters) 2004;17:317–22. [4] Papadakis L, Tobias H. Interaction between laser beam, process effects, and structural properties during welding using models based on the finite element analysis. Journal of Laser Applications 2007;19(3):189–96.
ARTICLE IN PRESS
768 B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768
[5] R. Holtz, M. Jokiel,, D.Lavoie, U. Duerr, High-speed single-pulse welding with Nd:YAG lasers. Proceedings of the SPIE—the international society for optical engineering, 5713, photon processing in microelectronics and photonics IV, January 24–27, 2005, pp. 471–78. [6] Ehlen G, Ludwig. A, Sahm PR. Simulation of time-dependent pool shape during laser spot welding: transient effects. Metallurgical and Materials Transactions A (Physical Metallurgy and Materials Science) 2003;34A(12): 2947–2961. [7] M.P. Coelho, A.M. Joao, F.Abreu, C. Carvalho, Influence of thermal stresses in modeling the lap welding of thermoplastic films. Conference on lasers and electro-optics Europe, Munich, Germany, June 12–7, 2005. [8] H. Engstrom, K. Nilsson, J. Flinkfeldt, T. Nilsson, A. Skirffors, M. Miller, High power laser welding of construction steels. Proceedings of 18th international congress on applications of lasers and electro-optics (ICALEO’99), 15–18 November, San Diego, CA, USA, 1999. [9] Zambon A, Ferro P, Bonollo F. Microstructural, compositional and residual stress evaluation of CO2 laser welded superaustenitic AISI 904L stainless steel. Materials Science and Engineering A 2006;424:117–27. [10] Olabi AG, Benounis KY, Hashmi MSJ. Application of response surface methodology in describing the residual stress distribution in CO2 laser welding of AISI 304. Strain 2007;43(1):37–46. [11] Zeng-rong H, Jian-zhong Z, Hua-feng G, Jian-jun D. Simulation of temperature field of laser welding by ABAQUS. Laser Technology 2007;31(3): 326–329.
[12] Shin BH, Yoo YT, Shin HJ, Kim JH. laser welding characteristics of cold rolled carbon steel utilize continuous wave Nd:YAG laser. Key Engineering Materials 2007;345-346(pt. 2):1461–4. [13] Zhang X, Ashida E, Shono S, Matsuda F. Effect of shielding conditions of local dry cavity on weld quality in underwater Nd:YAG laser welding. Journal of Materials Processing Technology 2006;174(1-3):34–41. [14] Zhou J, Tsai H. ‘‘Porosity formation and prevention in pulsed laser welding’’, Transactions of the ASME. Journal of Heat Transfer 2007;128(9):1014–24. [15] Mahrle A, Beyer E. Hybrid laser beam welding-classification, characteristics, and applications. Journal of Laser Applications 2006;18(3):169–80. [16] M. Ihor, M. Schmidt, Laser micro welding of copper and aluminum. Proceedings of the SPIE—the international society for optical engineering, laser based micropackaing, San Jose, CA, USA, 6107, January 25–26, 2006, 610703. [17] Chen YB, Lei ZL, Li LQ, Wu L. Experimental study on welding characteristics of CO2 laser TIG hybrid welding process. Science and Technology of Welding and Joining 2007;11(4):403–11. [18] Bayraktar E, Kaplan D, Yilbas BS. Comparative study: mechanical and metallurgical aspects of tailored welded blanks (TWBs). Journal of Materials Processing Technology 2007. [19] /www.engineeringtoolbox.com/emissivity-coefficients-d_447.htmlS. [20] Totemeier TC, Wright RN. Residual stress determination in thermally sprayed coatings—a comparison of curvature models and X-ray techniques. Surface and Coatings Technology 2006;vol. 200(12–13):3962–95.
Optics & Laser Technology 42 (2010) 760–768
Contents lists available at ScienceDirect
Optics & Laser Technology
journal homepage: www.elsevier.com/locate/optlastec
Laser welding of low carbon steel and thermal stress analysis
B.S. Yilbas n, A.F.M. Arif, B.J. Abdul Aleem
Mechanical Engineering Department, KFUPM, Saudi Arabia
a r t i c l e in f o
Article history: Received 13 November 2008 Received in revised form 17 November 2009 Accepted 26 November 2009 Available online 21 December 2009 Keywords: Laser welding Temperature Finite element model
a b s t r a c t
Laser welding of mild steel sheets is carried out under nitrogen assisting gas ambient. Temperature and stress fields are computed in the welding region through the finite element method. The residual stress developed in the welding region is measured using the XRD technique and the results are compared with the predictions. Optical microscopy and the SEM are used for the metallurgical examination of the welding sites. It is found that von Mises stress attains high values in the cooling cycle after the solidification of the molten regions. The residual stress predicted agreed well with the XRD results. & 2009 Elsevier Ltd. All rights reserved.
1. Introduction Laser is widely used as a thermal source for industrial applications. This is because of the local treatment, precise operation, and short processing time. One of the important industrial applications of laser processing is the laser welding, which offers considerable advantages over the conventional welding methods. High intensity laser beam melts and partially evaporates the welded material during the process. Attainment of high temperature gradient during the heating and cooling periods results in the development of high thermal stresses in the welding zone. Once the cooling period ends, the residual stress in the welding zone is resulted. This, in turn, influences the mechanical performances of the resulting welds. Consequently, investigation into thermal stress development and residual stress formation in the weld zone becomes essential. Considerable research studies were carried out to examine the laser welding process. An extensive review on laser welding and related process was carried out by Mackwood and Crafer [1]. They presented the applications laser welding processes under different welding categories such as laser spot welding, laser butt welding, etc. Nd:YAG laser repair welding of tool steels and microstructural changes in the welded region were examined by Vedani[2]. They indicated that heat affected zone was narrow and carbides dissolve during the heating phase of the welding process. Li et al. [3] studied the laser forming process using the finite element method. They showed that thermal expansion and
n
Corresponding author. Tel.: + 966 3 860 4481; fax: +966 3 860 2949. E-mail address: [email protected] (B.S. Yilbas).
contraction took place during the laser processing, which in turn allowed the thermo-mechanical forming of complex shapes. Laser welding and heating analysis were carried out by Papadakis and Tobias [4]. They examined the lap and fillet seam welds and measured the distortion during the processing. Moreover, the experimental findings were compared with the model predictions. A high speed laser pulse welding of metallic substrates was examined by Holtz et al. [5]. They demonstrated that an improvement in the structural integrity of the weld site was resulted for high speed pulsed contour welding than the seam welding process. The transient effects on the formation of laser produced weld pool were studied by Ehlen et al. [6]. They accommodated the Marangoni effect using a semi-empirical model for the temperature dependent surface tension gradient. The thermal stress developed during the lap welding of thermoplastic films was examined by Coelho et al. [7]. They showed that the influence of thermal stresses and expansion and contraction forces played an important role on the achievement of strong welds; in which case, tensile strength improved 80% as compared to the original thermo-plastic substrate. The high power laser welding of construction steel was investigated by Engstron et al. [8]. They indicated that laser welding offered new design opportunities, which improved the manufacturing process by shortening lead and production times as well as reducing the amount of materials used. The metallurgical and residual stress evaluation of CO2 laser welded super-austenitic stainless steel was carried out by Zambon et al. [9]. They showed that the residual stress was tensile and close to the yielding strength of the substrate material in the longitudinal direction in the weld bead while the stresses were compressive in the transverse direction in the base material. Laser welding of steel and residual stress
0030-3992/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.optlastec.2009.11.024
ARTICLE IN PRESS
B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768 761
distribution were examined by Olabi et al. [10]. They examined the effect of laser parameters on the residual stress developed through the statistical analysis. Laser welding and the simulation of temperature field were carried out by Zeng-rong et al. [11]. They indicated that the finite element method predictions were in agreement with the experimental findings. Laser welding characteristics of cold rolled carbon steel were examined by Shin et al. [12]. They showed that the optimal welding conditions, which resulted in no defect sites in the vicinity of the welded area, were possible for a certain combination of laser output power, welding speed, and focus setting of focusing lens. The influence of shielding condition on the weld quality of underwater laser welding was investigated by Zhang et al. [13]. They showed that there existed the relationship between shielding conditions and the quality of the weld bead. The porosity formation and penetration in pulsed laser welding were examined by Zhou and Tsai [14]. They indicated that the porosity formation was strongly related to the aspect ratio (depth to width) of the key hole formed during the welding process. The hybrid laser beam welding and its applications were studied by Mahrle and Beyer [15]. They demonstrated some practical examples of potential hybrid technologies, which would be used to extend the application spectrum of the laser beam welding. Laser microwelding of copper and aluminum was investigated by Ihor and Schmidt [16]. They showed that an application of suitable filler material helped in avoiding the brittle intermetallic phases at the interface between copper and the solidified melt in the welded joints. The experimental study on welding characteristics of CO2 laser TIG hybrid welding process was carried out by Chen et al. [17]. They indicated that when the current was increased to a critical value, the laser induced key hole disappeared and the arc expanded, which, in turn, decreased the penetration depth. The mechanical and metallurgical aspects of tailored welded blanks were examined by Bayraktar et al. [18]. They showed that the fracture surfaces tested at low temperatures presented the fracture topography for cleavage in the notched section for steel samples. However, at high temperatures, the fractured always appeared in the ductile mode in the specimens. In the present study, laser welding of low carbon steel plates is carried out. Temperature and thermal stress fields are modeled using the finite element method. The residual stress developed in the welding zone is measured using the XRD technique and compared with the predictions. The metallurgical and morphological changes in the weld zone are examined using the optical microscopy, the SEM, the EDS and the XRD.
Laser Beam Laser Irradiated Spot z x y Weld section 2 mm 30 mm
Fig. 1. View of laser melting process.
Nitrogen
x=0 y=0 z=0
1 3 5 7 9 11 13
40 mm
U
In the case of a moving heat source along the z-axis with a constant velocity U, energy gain by the substrate material yields
15 17 19
r
DE @E @E ¼ r ÀrU Dt @t @z
ð2Þ
or DE @ðCpTÞ @ðCpTÞ r ¼r -rU Dt @t @z Combining equations (1) and (3) yields ð3Þ
21 23 25 27 29 31 33 35
@ðCpTÞ @ðCpTÞ r ð4Þ ¼ ðrðkrTÞÞ þ rU þSo @t @z At the free surfaces of the welded workpiece (the irradiated surface, Fig. (1)), the convective boundary is assumed and at the rear side of the workpiece, the convective and radiative boundary conditions are considered. Therefore, the corresponding boundary condition is: At the irradiated surface @T h ¼ ðTs ÀTamb Þ @z k and at the rear side of the surface: @T h es 4 4 ðT ÀTamb Þ ¼ ðTs ÀTamb Þ þ k s @y k and @T h ¼ ðTs ÀTamb Þ @z k ð5Þ
ð6Þ
37 39 41 43 45 47 49 51 53 55 57
where h is the heat transfer coefficient due to natural convection, and Ts and Tamb are the surface and ambient temperatures, respectively, e is the emissivity (e =0.63 is considered [19]), and s is the Stefan–Boltzmann constant (s =5.67 Â 10 À 8 W/m2 K). At far away boundary (at edges of the solution domain) constant temperature boundary is assumed (T= 293 K), i.e. x ¼ 1; y ¼ 1; z ¼ 1-T ¼ 293K Initially (prior to laser treatment), the substrate material is assumed to be at constant ambient temperature, i.e. T =Tamb, which is considered as constant (Tamb = 293 K). Equation (4) is solved numerically with the appropriate boundary conditions to predict the temperature field in the substrate material. However, to analyze the phase change problem, a non-linear transient thermal analysis is performed employing the enthalpy method. To account for latent heat evolution during phase change, the enthalpy of the material as a function of temperature is incorporated in the energy equation.
2. Heating analysis Since the welding situation is involved with the transient heating during laser scanning, a transient equation for heat transfer needs to be considered. In this case, the transient heat transfer equation in the Cartesian coordinates is
r
DE ¼ ðrðkrTÞÞ þ So Dt
ð1Þ
where D represents the material derivative or substantial D @ @ @ @ derivative Dt ¼ u @x þ v @y þ w @z þ @t , E is the energy gain by the substrate material and So is the volumetric heat source term and So ¼ Io dð1Àrf ÞeÀdy eðÀx
2
3. Modeling of thermal stresses 59 For structural response, the finite element formulation is based on the principle of virtual work. From the principle of virtual work (PVW), a virtual (very small) change in the internal strain energy (dU) must be offset by an identical change in external work due to the applied loads (dV). Considering the strain energy due to thermal stresses resulting from the constrained motion of a body 61 63 65
þ z2 =a2 Þ
Io is laser peak intensity, d is the absorption depth, rf is the surface reflectivity, r is the density, Cp is the specific heat capacity, k is the thermal conductivity, a is the Gaussian parameter, and x and z are the axes while the laser beam scans the surface along the z-axis and the laser beam axis is the y-axis (Fig. (1)).
ARTICLE IN PRESS
762 B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768
during a temperature change, PVW yields Z Z T T ½BT ½D½Bdvfug ¼ fdug ½BT ½Dfeth gdv fdug
vol vol
residual stress (s) is given [20] ð7Þ
s¼
E ðdn Àdo Þ ð1 þ uÞSin2 c do
ð8Þ
Noting that the {du}T vector is a set of arbitrary virtual displacements common in all of the above terms; the condition required to satisfy above equation reduces to ½Kfug ¼ fF th g where R ½K ¼ vol ½BT ½D½Bdv ¼ element stiffness matrix R fF th g ¼ vol ½BT ½Dfeth gdv ¼ element thermal load vector feth g ¼ fagDT ¼ thermal strain vector fag ¼ vector of coefficients of thermal expansion In the present study, the effect of mechanical deformation on heat flow has been ignored and the thermo-mechanical phenomenon of melting is idealized as a sequentially-coupled unidirectional problem. For thermal analysis, the given structure is modeled using the thermal element (SOLID70). SOLID70 has a 3D thermal conduction capability (ANSYS code). The element has eight nodes with a single degree of freedom, and temperature, at each node. The element is applicable to a 3D, steady-state or transient thermal analysis. Since the model containing the conducting solid element is also to be analyzed structurally, the element is replaced by an equivalent structural element (such as SOLID45) for the structural analysis. SOLID45 is used for the 3D modeling of solid structures. The element is defined by eight nodes having three degrees of freedom at each node: translations in the nodal x, y, and z directions. The element has plasticity, creep, swelling, stress stiffening, large deflection, and large strain capabilities. The thermal and structural properties used in the current simulations are given in the Table 1. It should be noted that the conditions for the current simulations resemble the actual experiments carried out in the present study.
where E is the Young’s modulus, n is the Poisson’s ratio, c is the tilt angle, and di are the d spacing measured at each tilt angle. If there are no shear strains present in the specimen, the d spacing changes linearly with Sin2 c.
3000 2400 1800 1200 600 0 0.0E+00 2000 t=2s TEMPERATURE (K) 1500 y = 0 mm; z = 15 mm y = 0.8 mm; z = 15 mm y = 1.6 mm; z = 15 mm t=1s y = 0 mm; z = 15 mm y = 0.8 mm; z = 15 mm y = 1.6 mm; z = 15 mm
TEMPERATURE (K)
2.0E-03
4.0E-03 6.0E-03 DISTANCE (m)
8.0E-03
1.0E-02
1000
500
4. Residual stress measurements XRD Technique: The measurement relies on the stresses in fine grained polycrystalline structure. The position of the diffraction peak undergoes shifting as the specimen is tilted by an angle c. The magnitude of the shift is related to the magnitude of the residual stress. The relationship between the peak shift and the
Table 1 Mechanical and thermal properties of mild steel used in the simulations. Temperature (K) E (GPa) Temperature (K) a  10 À 6 (1/K) k (W/mK) Cp (J/kgK) u r (kg/m3) 294 203 273 11.2 51.9 486 0.3 7700 373 11.2 50.7 486 366 199 473 12.1 48.2 515 477 191 573 13.0 45.6 548 589 184 673 13.6 41.9 586 644 176 773 14.0 38.1 649 700 167 873 14.6 33.9 708 973 14.8 30.1 770 755 154 1073 11.8 24.7 624 811 141 1273 13.6 26.8 548 866 124 1473 13.6 29.7 548
0 0.0E+00
2.0E-03
4.0E-03 6.0E-03 DISTANCE (m)
8.0E-03
1.0E-02
Fig. 2. Temperature distribution along the x-axis at different depths below the surface and for two time periods.
Table 2 . Laser welding conditions. Feed Rate (mm/min) 1000 Power (W) 2000 Frequency (Hz) 1500 Nozzle Gap (mm) 1.5 Nozzle Diameter (mm) 1.5 Focus setting (mm) 127 N2 Pressure (kPa) 600
ARTICLE IN PRESS
B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768 763
Fig. 3. Three-dimensional view of temperature distribution during welding process for two time periods. Temperature is in K.
5. Experimental The CO2 laser (LC-ALPHAIII) delivering nominal output power of 2 kW at pulse mode with different frequencies is used to irradiate the workpiece surface. The nominal focal length of the focusing lens is 127 mm. Nitrogen assisting gas emerging from the conical nozzle and co-axially with the laser beam is used. The welding conditions are given in Table 2. The workpiece accommodated is mild steel at 2 mm in thicknesses. Material characterization of the laser nitrided surfaces is carried out using the SEM, the XRD and the XPS. The Jeol 6460 electron microscopy is used for the SEM examinations and the Broker D8 Advanced having MoKa radiation is used for the XRD analysis. A typical setting of the XRD was 40 kV and 30 mA. Microphotonics digital microhardness tester (MP-100TC) was
used to obtain microhardness across the weld zone (parallel to the workpiece surface). The standard test method for Vickers indentation hardness of advanced ceramics (ASTM C1327–99) was adopted. The measurements were repeated three times at each location.
6. Results and discussions Laser welding of low carbon steel sheets is carried out. Temperature and stress fields developed during the welding are computed using the finite element method. The residual stress developed in the weld region is measured using the XRD technique while metallurgical and morphological changes in the
ARTICLE IN PRESS
764 B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768
1.0E+09
STRESS (Pa)
1.0E+08 t=1s 1.0E+07
y = 0 mm; z = 15 mm y = 0.8 mm; z = 15 mm y = 1.6 mm; z = 15 mm
1.0E+06 0.0E+00 1.0E+09
2.0E-03
4.0E-03
6.0E-03
8.0E-03
1.0E-02
DISTANCE (m)
1.0E+08 STRESS (Pa)
1.0E+07 t=2s 1.0E+06
y = 0 mm; z = 15 mm y = 0.8 mm; z = 15 mm y = 1.6 mm; z = 15 mm
1.0E+05 0.0E+00
2.0E-03
4.0E-03
6.0E-03
8.0E-03
1.0E-02
DISTANCE (m)
Fig. 4. von Mises stress distribution along the x-axis at different depths below the surface and for two time periods.
weld zone are examined using the optical microscopy, the SEM and the EDS. Fig. (2) shows temperature variation along the x-axis at three different locations in the y-axis while Fig. (3) shows 3D view of temperature distribution. The z-axis location is 15 mm from the welding starting point (Fig. (1)) in Fig. (2). It should be noted that the time t= 1 s corresponds to the laser beam spot center at the location x = 0, y= 0, and z= 15 mm. Temperature decays sharply in the surface region and as the depth below the surface increases temperature decay becomes sharp and further increase in the depth results in gradual decay of temperature. In the region, where the sharp delay occurs, temperature gradient attains high values. This is more pronounced at some depth below the surface. Temperature in the surface region exceeds the melting temperature of the substrate material. Although temperature at the melting–solid interface remains the same – due to the influence of latent heat of fusion on the energy transfer in the vicinity of melting–solid interface – temperature gradient in the molten region becomes different in the solid phase. Moreover, the decay rate of temperature as well as its magnitude changes at different y-axis locations in the melting zone. The y-axis location represents different depths below the surface inside the substrate material. The difference in temperature in the molten zone is attributed to the heat conduction in this region. In the case of t =2 s heating duration, the laser beam scans over this region and energy transfer from the surface through convection and
conduction reduces temperature significantly in this region. The heat conduction becomes the sole mechanism after this time in this region. In addition, temperature profiles along the depth below the surface becomes almost self similar. Fig. (4) shows von Mises stress along the x-axis at tree depths below the surface for the same z-axis location as shown in Fig. (1) for two heating periods while Fig. (5) shows 3D view of von Mises stress in the substrate material. von Mises stress attains low values in the surface region because of the melting; in which case, molten metal is free to expand in this region. von Mises stress increases sharply as the depth below the surface increases along the y-axis. Moreover, it remains high as the distance increases further. This sharp rise of von Mises stress is related to the sharp increase in temperature gradient at some depth below the surface. The variation of von Mises stress along the x-axis is similar at different y-axis locations, since the difference in the stress magnitude is negligibly small. This situation is associated with the similar behavior of temperature distribution at different y-axis locations along the x-axis. In the case of time period t =2 s, von Mises stress rises with increase in distance along the x-axis. This is true for all the y-axis locations. This increase is attributed to the strain developed at early periods when the laser peak intensity scans over the z-axis locations corresponding to the figure. Therefore, thermal stress developed during high temperature heating contributes to the stress field developed during the cooling period for t = 2 s. Consequently, von Mises stress for this time period does not follow exactly the temperature distribution as shown in Fig. (2). Moreover, the maximum stress level along the x-axis is below the elastic limit of the substrate material. Fig. (6) shows temperature distribution along the z-axis (laser scanning direction) at three different y-axis locations and for three time periods. It should be noted that the x-axis location is x= 0 (Fig. (1)). The location of peak temperature changes along the z-axis for different periods, which is because of the location of laser beam intensity along the scanning direction (z-axis). In this case, the laser peak intensity is at z= 1.5 Â 10 À 2 m for t =1 s, z= 2.25 Â 10 À 2 m for t =1.5 s, and z = 3 Â 10 À 2 m for t = 2 s. The small irregular behavior is observed in temperature profile at about 1600 K. This is because of the phase change process (melting), which suppresses the rate of rise of temperature in the solid due to the latent heat of fusion. However, the sharp increase in temperature occurs at a location where the laser beam intensity is the maximum. This is due to the absorption of the laser beam energy by the substrate material, which enhances significantly the internal energy gain in this region. Therefore, temperature increases sharply in the molten material. This situation is true for different y-axis locations, provided that temperature gradient around the peak temperature reduces as the y-axis location increases below the surface. The decrease in time derivative of temperature (qT/qt) is because of less energy transfer through conduction from the surface in this region where the laser energy absorption is significantly high. The rise and delay of temperature around the peak temperature is not the same. This suggests that temperature remains high at the surfaces during the heating period when the laser scans the workpiece surface. However, once the welding is over, temperature delay is sharp in the cooling cycle. This is true for all the y-axis locations and for all periods in the cooling cycle. Fig. (7) shows von Mises stress along the z-axis for three y-axis locations and three time periods. It should be noted that the locations and time periods are the same as Fig. (6). von Mises stress attains high values immediately after temperature reaches its peak value (Fig. (6)). This is associated with the temperature gradient, which is higher after the peak temperature. It should be noted that material in the liquid phase, where temperature
ARTICLE IN PRESS
B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768 765
Fig. 5. Three-dimensional view of von Mises stress distribution during welding process for two time periods. Stress is in Pa.
reaches its peak, expands freely and stress level becomes very low. However, upon decaying from the peak temperature, the solidification takes place rapidly and conduction cooling from the heated region results in attainment of high temperature gradient. This causes a sharp increase in von Mises stress in this region. von Mises stress level rises sharply and decays gradually for all time periods and the y-axis locations, despite the fact that temperature decays sharply after its peak value. Consequently, strain developed during the heating period plays an important role on the stress formation after the laser beam passes over the concerned location along the z-axis (laser scanning direction). The maximum value of von Mises stress attains almost 195 MPa, which is less than the elastic limit of the substrate material. This is true for all the y-axis locations and time periods. Moreover, the
decay in von Mises stress is observed before rising to its maximum. This may be associated with the initial solidification and temperature variation immediately after the solidification, i.e. temperature change is small in the region of solidification resulting in attainment of low temperature gradient while lowering von Mises stress in this region. Fig. (8) shows optical photographs of welding surface and different regions in the weld cross-section while Fig. (9) shows the SEM micrographs of the weld cross-section. The heat affected zone (HAZ) is evident from the SEM micrograph and the heat affected zone extends both sides of the weld section almost at a same size. This indicates the plane symmetry of the heating during the actual welding process, which is also adapted in the simulations. The grain size coarsening is evident in the HAZ due to
ARTICLE IN PRESS
766 B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768
3000 TEMPERATURE (K)
1.00E+09
x = 0 mm and y = 0 mm
1800 1200 600 0 0.0E+00 3000 t=1s t=2s 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
STRESS (Pa)
2400
1.00E+08 1.00E+07 1.00E+06 1.00E+05 0.0E+00 1.00E+09
t=1s t=2s
x = 0 mm and y = 0 mm 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
DISTANCE (m) TEMPERATURE (K) x = 0 mm and y = 0.8 mm
DISTANCE (m) t=1s t=2s
1800 1200 600 0 0.0E+00 3000 6.0E-03 t=1s t=2s 1.2E-02 1.8E-02 2.4E-02 3.0E-02
STRESS (Pa)
2400
1.00E+08 1.00E+07 1.00E+06 1.00E+05 0.0E+00 1.00E+09
x = 0 mm and y = 0.8 mm 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
DISTANCE (m) TEMPERATURE (K)
DISTANCE (m) t=1s STRESS (Pa) 1.00E+08 1.00E+07 1.00E+06 1.00E+05 0.0E+00 t=2s
2400 1800 1200 600
x = 0 mm and y = 1.6 mm
t=1s t=2s 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
x = 0 mm and y = 1.6 mm 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02
0 0.0E+00
DISTANCE (m)
Fig. 6. Temperature distribution along the z-axis at different depths below the surface and for two time periods.
DISTANCE (m)
Fig. 7. von Mises stress distribution along the z-axis at different depths below the surface and for two time periods.
the slow cooling rates in this region. However, some fine grains are observed in the region away from the weld interface. The grain refining is formed because of high cooling rates due to high temperature gradient. In general, the fine grains consist of circular ferrite and fine pearlite structures (Fig. (9)), in which case, the light region is ferrite while dark region is pearlite. Moreover, the nucleation of pearlite results in partial formation of cementite lamellae structure. The pearlitic modulus grows from the nuclei along the austenite boundaries and the scattered colonies of alternating ferrite and cementite lamellae are formed within the pearlitic structure. In the weld core, the austenite phase allowing a considerable grain growth is observed. In this case, grain boundary ferrite and ferrite/carbide aggregate appear to be fine pearlite. In addition, in the fusion zone, very fine grains, as well as locally scattered large grains, where crystallization takes place, are evident. Fig. (10) shows microhardness distribution across the welding cross-section along the line parallel to the free surface (transverse to the weld section). The hardness decays sharply with increase in distance across the weld. The peak hardness is almost 40% higher than the base material hardness. It should be noted that the increase in hardness is one of the indications of formation of fine structures and the residual stress developed during the welding process. The residual stress measured using the XRD technique
and observed from the simulations are 105 MPa and 90 MPa, respectively. It should be noted that the residual stress is measured 2.5 mm away from the welding line and at the midway of the welded sample along the welding direction. Moreover, the prediction of the residual stress is compared at the same location of the measurement. It can be observed that both results are in agreement (in the order of 100 MPa), provided that a small discrepancy occurs, which is because of the assumptions made in the simulations such as uniform structure and mechanical properties are considered.
7. Conclusions Laser welding of mild steel sheets is carried out and stress field developed in the welding zone is modeled using the finite element method (FEM). The residual stress developed in the weld region is measured using the XRD technique and compared with the FEM predictions. The morphological and metallurgical changes in the welding region are examined using the optical microscopy and the SEM. Microhardness distribution across the welding zone is measured. It is found that temperature decay rate in the molten zone is lower than in the solid. This is because of the absorption and dissipation of the laser energy in the molten zone,
ARTICLE IN PRESS
B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768 767
2 mm Top view of laser welded section
Fusion Zone Fusion Zone Heat Affected Zone Base Material Partially formed pearlite
1 mm Cross-section of laser welded section. Heat Affected Zone
Fig. 9. SEM micrographs of welding cross-section.
200 180 HARDNESS (HV) 160 140 120 100 0.0E+00
0.2 mm Fusion Zone and Heat Affected Zone Interface.
Fig. 8. Optical photographs of welding cross-section.
6.0E-04
1.2E-03
1.8E-03
2.4E-03
3.0E-03
which is generated in the surface region. This, in turn, increases the internal energy gain in this region. In the solid region, temperature variation along the x-axis becomes similar for all the y-axis locations. This results in similar von Mises stress distribution along the x-axis provided that at large y-axis locations (at some depth below the surface); the strain developed in this region modifies this behavior slightly due to temperature increase, which reaches its peak at the z-axis location when the laser beam intensity is the maximum. Moreover, once the laser beam scans over this region, temperature decays sharply. This causes sharp increase in von Mises stress due to the attainment of high temperature gradient in this region. This situation is true for all the y-axis locations and time periods considered in this study. However, the maximum stress level is less than the elastic limit of the substrate material. The residual stress predicted from the FEM agrees well with the XRD results, which is in the order of 100 MPa. The grain coarsening occurs in the HAZ and grain refining and recrystallization takes place in the fusion zone as observed from the optical and the SEM micrographs. The closed examination of welding section reveals that the weld section is free from cracks and voids.
DISTANCE (m)
Fig. 10. Microhardness across the weld cross-section and parallel to the free surface.
Acknowledgements The authors acknowledge the support of King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia, for the funded project, Project #SB080003.
References
[1] Mackwood. AP, Crafer RC. Thermal modeling of laser welding and related processes: a literature review. Optics and Laser Technology, 2005;37:99–115. [2] Vedani M. Microstyrctural evolution of tool steels after Nd-YAG laser repair welding. Journal of Materials Science 2004;39:241–9. [3] Li LQ, Chen YB, Wang WY, Lin SY. FEM simulation for laser forming processing. Acta Metallurgica Sinica (English Letters) 2004;17:317–22. [4] Papadakis L, Tobias H. Interaction between laser beam, process effects, and structural properties during welding using models based on the finite element analysis. Journal of Laser Applications 2007;19(3):189–96.
ARTICLE IN PRESS
768 B.S. Yilbas et al. / Optics & Laser Technology 42 (2010) 760–768
[5] R. Holtz, M. Jokiel,, D.Lavoie, U. Duerr, High-speed single-pulse welding with Nd:YAG lasers. Proceedings of the SPIE—the international society for optical engineering, 5713, photon processing in microelectronics and photonics IV, January 24–27, 2005, pp. 471–78. [6] Ehlen G, Ludwig. A, Sahm PR. Simulation of time-dependent pool shape during laser spot welding: transient effects. Metallurgical and Materials Transactions A (Physical Metallurgy and Materials Science) 2003;34A(12): 2947–2961. [7] M.P. Coelho, A.M. Joao, F.Abreu, C. Carvalho, Influence of thermal stresses in modeling the lap welding of thermoplastic films. Conference on lasers and electro-optics Europe, Munich, Germany, June 12–7, 2005. [8] H. Engstrom, K. Nilsson, J. Flinkfeldt, T. Nilsson, A. Skirffors, M. Miller, High power laser welding of construction steels. Proceedings of 18th international congress on applications of lasers and electro-optics (ICALEO’99), 15–18 November, San Diego, CA, USA, 1999. [9] Zambon A, Ferro P, Bonollo F. Microstructural, compositional and residual stress evaluation of CO2 laser welded superaustenitic AISI 904L stainless steel. Materials Science and Engineering A 2006;424:117–27. [10] Olabi AG, Benounis KY, Hashmi MSJ. Application of response surface methodology in describing the residual stress distribution in CO2 laser welding of AISI 304. Strain 2007;43(1):37–46. [11] Zeng-rong H, Jian-zhong Z, Hua-feng G, Jian-jun D. Simulation of temperature field of laser welding by ABAQUS. Laser Technology 2007;31(3): 326–329.
[12] Shin BH, Yoo YT, Shin HJ, Kim JH. laser welding characteristics of cold rolled carbon steel utilize continuous wave Nd:YAG laser. Key Engineering Materials 2007;345-346(pt. 2):1461–4. [13] Zhang X, Ashida E, Shono S, Matsuda F. Effect of shielding conditions of local dry cavity on weld quality in underwater Nd:YAG laser welding. Journal of Materials Processing Technology 2006;174(1-3):34–41. [14] Zhou J, Tsai H. ‘‘Porosity formation and prevention in pulsed laser welding’’, Transactions of the ASME. Journal of Heat Transfer 2007;128(9):1014–24. [15] Mahrle A, Beyer E. Hybrid laser beam welding-classification, characteristics, and applications. Journal of Laser Applications 2006;18(3):169–80. [16] M. Ihor, M. Schmidt, Laser micro welding of copper and aluminum. Proceedings of the SPIE—the international society for optical engineering, laser based micropackaing, San Jose, CA, USA, 6107, January 25–26, 2006, 610703. [17] Chen YB, Lei ZL, Li LQ, Wu L. Experimental study on welding characteristics of CO2 laser TIG hybrid welding process. Science and Technology of Welding and Joining 2007;11(4):403–11. [18] Bayraktar E, Kaplan D, Yilbas BS. Comparative study: mechanical and metallurgical aspects of tailored welded blanks (TWBs). Journal of Materials Processing Technology 2007. [19] /www.engineeringtoolbox.com/emissivity-coefficients-d_447.htmlS. [20] Totemeier TC, Wright RN. Residual stress determination in thermally sprayed coatings—a comparison of curvature models and X-ray techniques. Surface and Coatings Technology 2006;vol. 200(12–13):3962–95.
