Effect of Laser Polishing Parameters on Residual Stress, Surface Roughness, and Microhardness of Ti6Al4V

doi:10.21203/rs.3.rs-5232352/v1

Download PDF

Research Article

Effect of Laser Polishing Parameters on Residual Stress, Surface Roughness, and Microhardness of Ti6Al4V

https://doi.org/10.21203/rs.3.rs-5232352/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The present study explores the impact of laser polishing (LP) process parameters of Ti6Al4V on the induced residual stress and surface properties. LP is a thermal process that involves melting a thin surface layer. The flows within the molten layer reduce the surface undulations caused due to initial surface roughness, resulting in a smoother surface. The material undergoes heating and cooling cycles in LP, resulting in residual stresses. This paper details the trade-offs between the residual stresses and the surface properties for various process parameters. The induced residual stresses show dependency on the cooling rate during the process, predicted using a validated finite element-based numerical model. For the set of process parameters, it was observed that the induced residual stress has no significant variation with a change in laser power. However, it increases with an increase in scan speed. A reduction of around 71% in surface roughness is observed at 100 W and 0.1 m/s, with a minimum induced residual stress of ~ 372 MPa. However, the surface hardness is maintained for all sets of process parameters. A comparative study is also conducted for the induced residual stress between pure Ti and Ti6Al4V. It is noted that for all process parameters, the induced residual stress is higher for Ti6Al4V. However, pure Ti also follows the trend of variation of laser power and scan speed on the induced residual stress.

Laser polishing

residual stress

melt pool

micro-hardness

surface roughness

Laser polishing (LP) is an advanced post-manufacturing process employed to improve the surface smoothness and properties [1][2]. LP involves the melting and subsequent solidification of the substrate's surface. The melting results in the redistribution of the molten superficial surface which fills the surface asperities to smoothen the surface. Figure 1 shows the schematic diagram of the LP process, which describes the various zones, such as the melting zone and heat-affected zone formed during the process. Since LP is a surface process, it does not affect the initial bulk properties of the material. In addition, one of the advantages of LP is that the surface polishing is accomplished without adding or removing the material. Several researchers have accomplished a significant reduction in surface roughness by employing LP. Chen et al. studied the effect of laser polishing on the microstructure and mechanical properties of SS316 [2]. They addressed the reduction of the surface roughness of laser processed sample from 4.84 µm to 0.65µm. Rosa et al. [3] and Souza et al. [4] experimentally evaluated the efficacy of laser polishing on additively manufactured SS316L and reduced the roughness by 61% and 86%, respectively. Richter et al. reported an average reduction of 87% in surface roughness through LP of additively manufactured Co-Cr [5].

LP is accompanied by high temperature and fast thermal cycles. The molten material undergoes contraction during the cooling cycle after the laser source moves away from it. This fast cooling cycle leads to the development of significant tensile stresses in the processed part [6] [7] [8] [9]. The development of residual stresses due to the thermal cycles can initiate surface cracks and lead to fractures. Formation of surface cracks is commonly observed in other laser-based processing applications, such as welding, SLM [6], surface remelting [10] and cladding [11]. A few potential technical approaches to mitigate the stress developed in laser-based processes are preheating the samples or employing optimised process parameters [12]. The current work focuses on optimising the process parameters for LP of Ti and Ti6Al4V to minimise residual stress and improve surface properties.

Many studies related to LP are conducted on Ti and its alloy due to their exceptional properties [13]. Due to the unique combination of superior strength, reduced density, and great resistance to Ti and its alloy corrosion, they have attracted attention in various industries, including aerospace, biomedical, biocompatible technology and automotive sectors [14] [15]. However, utilising these alloys often encounters challenges, such as surface defects, high tensile residual stresses, poor microhardness and surface roughness [16]. These defects can collectively undermine the overall performance and durability of parts fabricated by Ti or its alloys. In this context, techniques to improve surface finish play a crucial role in enhancing the functional attributes of the material.

In the present study, LP of Ti6Al4V is carried out using a continuous wave fibre laser. The laser-processed area is studied for the induced residual stresses, surface roughness and surface hardness. The co-relation of melt pool size and cooling rate for different process parameters with the induced residual stress is highlighted. A previously developed laser-based processing FEM model constituting heat transfer and fluid flow physics is employed to extract the thermal cycles for various process parameters. The model derived thermal history is then used to explain the induced residual stresses during LP [17].

The current work also focuses on the comparative study of the residual stresses induced in pure Ti and Ti6Al4V when processed by LP. Harda et al. studied the mechanical properties of selective laser sintering of pure titanium and titanium alloy [18]. They concluded that tensile stress in titanium alloy is more than twice that of pure titanium and has less corrosion resistance. Similar results were also observed for LP, as materials behave differently when exposed to thermal cycles due to their varying thermal properties. Since LP is a thermal process, different materials are expected to result in varying outcomes. The same concept extends to when pure Ti and Ti6Al4V undergo LP and a variation in induced residual stresses is expected.

The current manuscript is organised as follows: Section 2 details the experimental design measurement methods and describes thermal cycle models. Section 3 discusses the experimental results, and Section 4 concludes the manuscript with the key insights.

This study used rolled Ti6Al4V alloy to investigate the effects of laser process parameters on residual stress, surface topography and surface hardness during LP. The dimensions of the sample are 10 mm X 20 mm X 2 mm. Table 1 shows the as-received chemical composition of the sample.

Table 1

Chemical composition of Ti6Al4V samples used in the current experimental work.
Element	Fe	W	Cu	V	Al	Si	Mb	Ti
Composition (%)	0.157	0.262	0.170	3.700	5.130	0.038	0.023	Bal

2.1 Experimental setup

The schematic of the laser setup used for this study is shown in Fig. 2. A continuous-wave fibre laser with a maximum power of 300W (SPI Lasers, redPOWER QUBE, model: SP-0300-C-A-020-05-PQA- 011-001-000) is used in the experiments. A laser scan head (hurrySCAN20) with a 255 mm focal length -theta objective focusing lens receives the auto-collimated laser beam and has a maximum scan speed of 1m/s. At the focal plane, the laser beam profile is close to Gaussian (~ 1.1) and has a beam radius of 19.8 µm. The software application SCANLAB laser DESK is utilised to control the predetermined path for laser scanning and the activation/deactivation of the laser. Moving the sample away from or towards the focal plane can adjust the incident beam diameter within the processing chamber's manually adjustable sample stage. The sample surface is held at a fixed distance of 10 mm from the focal plane by the stage in this investigation. The resulting beam radius for the work is 194 µm. The sample stage is kept in an argon-filled chamber to avoid oxidation or other chemical reactions.

Figure 3 details the experimental design employed for the current study. The scan speed is kept constant for the first set of process parameters, and the laser power is varied as 80 W, 90 W, 100 W, 110 W and 120 W. The scan speed is varied for the second set of process parameters as 0.1 m/s, 0.2 m/s, 0.3 m/s, 0.4 m/s and 0.5 m/s while keeping the power constant at 100 W.

Single-line and area scan experiments are performed for all the sets of process parameters, as illustrated in Fig. 3. A single line scan of 8 mm length is run for all the designed process parameters. Furthermore, the 6mm X 6mm area scan is run for the same process parameters with 50% overlap. The overlap percentage for the area scan is calculated according to the line width observed for the respective process parameters.

2.2 Measurement methods

The single-line scale is studied for melt pool dimensions. The line scan is cross-sectioned using the BEUHLER IsoMet low-speed cutting machine. The cross-section of each line scan is polished with gradually increasing grit size abrasive paper to achieve a finer surface ranging as 220, 400, 600, 800, 1000, 1200, 2500, and 5000. Finally, the cross-section is cloth polished with alumina suspension of particle size 3 µm. The polished cross-sections are etched to reveal the melt pool boundary and microstructure. For Ti6Al4V, Kroll's reagent (2% HF, 4% HNO3 and 94% distilled water in volume fraction) is used as an etchant. An optical microscope (Leica DM4P) is used to capture the melt pool, and its dimensions are measured by the ImageJ software package.

The induced residual stress in LP samples is measured using Panalytical™ X'pert PRO MRD (Material Research Diffractometer), which works on the X-ray diffraction technique. A point detector and a high-resolution Eulerian cradle with angular repeatability of 0.0001^o are combined. The device measures the inter-atomic lattice spacing, which is distorted by residual stress, using a Cu-Kα target material with a wavelength of 1.5418 Å fitted with a 2 mm collimator slit. Changing the d-spacing because of leftover stress causes the peaks to shift and broaden. According to Bragg's equation ($\:\lambda\:=2dsin\theta\:$), a change in the diffraction angle from the value of the unstressed state ($\:{d}_{o}$) represents a change in the inter-planer spacing. The amount of residual strain determines how much of a shift there is. Usually, a standard peak is found at a higher $\:2\theta\:$ for the measurement of residual stress. Additionally, residual stress is evaluated at various tilting and diffraction angles at this $\:2\theta\:$ value. The residual stress is measured using the $\:{{sin}}^{2}\varPsi\:$ approach, which computes the lattice's d-spacing at various angles between positive and negative $\:\varPsi\:$. Samples are oriented at seven distinct angles: 0º, 45º, 90º, 180º, 225º, and 270º to determine the full stress tensor using iso inclination geometry. The sample's inclination angle ($\:\varPsi\:$) is altered from $\:{0}^{\text{o}}$ and $\:{40}^{\text{o}}$. To convert the strain to stress, XEC (X-Ray Elastic Constant) values $\:{s}_{1}$ = -2.8x10-7 MPa and $\:1/2{s}_{2}$ = 11.52 x10-7 MPa are used.

The surface topography of the LP-processed surface is measured using a three-dimensional optical measurement system based on white-light interferometry (ALICONA infinite focus). The 3D data is further subjected to a least square polynomial fitting to remove the form and a Gaussian filter using a MATLAB algorithm to determine the average surface roughness ($\:{S}_{a}$) without waviness. The Gaussian filter is implemented with a cutoff wavelength of 0.25 mm.

The Vickers hardness testing machine, with a diamond-shaped indenter, is used to measure the hardness of the laser-polished surface. A load of 500 g is employed for a dwell time of 15 s. These parameters are chosen to maintain indentation depth less than melt pool depth. The hardness number for the processed sample is calculated as an average of 15 indentations.

2.3 Numerical model description

A finite element (FE) based 3D symmetrical model for moving laser source is employed to predict the thermal cycles for various parameters during LP. The description of the developed model, including mesh details and grid impendence study, can be accessed through Hijam et al. [17]. The model is developed for Ti6Al4V alloy with a computational dimension of 2 mm X 0.7 mm X 0.4 mm with coupled heat transfer and fluid flow physics. The laser source runs only for a length of 0.5 mm till the time $\:{t}_{p}$, sufficient to capture the thermal and flow history during the process, to reduce the computational cost. The laser source term is then equated to zero for all time $\:t>{t}_{p}$ and the simulation is run for total time of $\:17{t}_{p}$. The simulation of $\:t>{t}_{p}$ allows the study of cooling behaviour during the process. A brief model description is given in the following sections.

2.3.1 Assumptions

To simplify the model, the top surface of the model is considered to be flat, and surface evolution is not concluded. The flow within the molten pool is Newtonian incompressible laminar flow. Also, the material properties are considered homogeneous and isotropic. Any chemical reactions are not included, as LP experiments are carried out in an inert shielding gas environment.

2.3.2 Governing equations and boundary conditions

The fluid flow is governed by the Navier-Stokes equation given as

$$\:\rho\:\frac{\partial\:\overrightarrow{u}}{\partial\:t}+\rho\:\left(\overrightarrow{u}.\nabla\:\right)\overrightarrow{u}=\nabla\:.\left[-pI+\mu\:\left(\nabla\:\overrightarrow{u}+{\left(\nabla\:\overrightarrow{u}\right)}^{T}\right)\right]+\rho\:\left(1-\beta\:\left(T-{T}_{m}\right)\right)\overrightarrow{g}$$

where $\:\rho\:$ is the density of the material (kg/m³), $\:\overrightarrow{u}$ is the flow velocity field within the molten metal pool (m/s), $\:t$ is the time (s), $\:p$ is the pressure (Pa), $\:\mu\:$ is the dynamic viscosity (Pa.s), $\:\beta\:$ is the thermal expansion coefficient (1/K), $\:T$ is the temperature field (K), $\:{T}_{m}$ is the melting temperature (K), and $\:I$ is the identity matrix. Considering the incompressibility assumption, the continuity equation is given as

$$\:\nabla\:.\overrightarrow{u}=0$$

The fluid flow dynamics are coupled with the energy equations to model through the convection. The heat transfer equation is given as

$$\:\rho\:{C}_{p}^{eff}\frac{\partial\:T}{\partial\:t}+\rho\:{C}_{p}^{eff}\left(\overrightarrow{u}.\nabla\:T\right)=\nabla\:.(k\nabla\:T)$$

where $\:{C}_{p}^{eff}$ is the effective specific heat of the material in the mushy zone (J/kgK), and $\:k$ is the thermal conductivity of the material (W/mK). The governing equations are coupled and solved simultaneously. As the process is symmetric about the plane passing through the centre of the laser beam along the direction of its movement, only half of the process is simulated.

For the current model, the laser heat source is introduced as a boundary heat flux on the top surface. The laser source boundary condition is given as

$$\:-k\nabla\:T=\alpha\:q$$

where $\:\alpha\:$ is the absorptivity of the material, and $\:q$ the Gaussian heat flux representing the laser heat source. It is mathematically expressed as

$$\:q=\frac{2P}{\pi\:{r}_{b}^{2}}exp\left(\frac{2({\left(x-vt\right)}^{2}-{y}^{2})}{{r}_{b}^{2}}\right)$$

where, $\:P$ is the laser power (W), $\:{r}_{b}$ is the effective laser beam radius (m), $\:v$ is the laser scanning speed (m/s), and $\:x,y$ are the coordinates along the $\:x$- and $\:y$-directions. All the faces other than the symmetric face are also exposed to the convective and radiative heat losses, which are given by,

$$\:-k\nabla\:T=h\left(T-{T}_{a}\right)+\epsilon\:\sigma\:({T}^{4}-{T}_{a}^{4})$$

where $\:h$ is the convective heat transfer coefficient (W/mK), $\:\epsilon\:$ is the surface emissivity, $\:\sigma\:$ is the Stefan-Boltzmann constant, and $\:{T}_{a}$ is the ambient temperature. The top face, which indicates the liquid and gas interface, is also subjected to tangential viscous stress as the laser-impinged interface is prone to deformation. In the free surface, the viscous stress is modelled as a counterbalance to the surface tension gradient, which is mathematically given as

$$\:{\sigma\:}_{t}=\frac{\partial\:\gamma\:}{\partial\:T}\nabla\:T.\overrightarrow{t}$$

where $\:\overrightarrow{t}$ is the unit vector in the tangential direction, $\:{\sigma\:}_{t}$ is the tangential stress, and $\:\partial\:\gamma\:/\partial\:T$ is the surface tension coefficient. All the faces except the top and symmetric faces are also subjected to no-slip conditions. The symmetric boundary condition for fluid flow and heat transfer is also applied to the symmetric face, which is expressed as

$\:u.\overrightarrow{n}=0$	(8)
$\:\overrightarrow{n}.(k\nabla\:T)=0$	(9)

2.3.3 Material properties

The study investigates the melting and cooling during laser polishing of Ti6Al4V. The entire domain is considered fluid, however, the viscosity is varied according to the effective viscosity method [19]. The implementation of an effective viscosity method allows to restrict of the flow in the region where the temperature is below the solidus temperature $\:\left({T}_{s}\right)$, which indicates the solid region. The viscosity below the temperature of $\:{T}_{s}$ is taken almost 2000 times that of the viscosity at temperatures higher than the liquidus temperature $\:\left({T}_{l}\right)$, the liquid region. The value of the viscosity is dropped smoothly in the mushy zone from $\:{T}_{s}$ to $\:{T}_{l}$ ensuring the convergence of the numerical model. The material properties for Ti6Al4V such as thermal conductivity, density and specific heat, are chosen to be temperature dependent and are taken from the experimental work reported in [20]. Within the mushy zone, the effective specific heat capacity is introduced to implement the physics of phase transformation and to incorporate the latent heat of fusion $\:\left({L}_{f}\right)$. The implementation is carried out using the equation given as

$$\:{C}_{p}^{eff}={C}_{p}+{L}_{f}\frac{\partial\:{f}_{l}}{\partial\:T}\:$$

where $\:{C}_{p}$ is the temperature-dependent specific heat capacity and $\:{f}_{l}$ is the liquid mass fraction, which is calculated using the COMSOL's phase change material module [21].

2.3.4 Model validation

It is to be noted that the developed numerical model has already been validated and is available in previous work from the authors. A detailed description of the model, including the computational domain, mesh sensitivity analysis, simulation environment and constant thermophysical properties of Ti6Al4V, can be found in [22]. An extension of the validation of the numerical model is carried out for the set of parameters relevant to the current work. Figure 4 compares experimental and simulation variations of melt pool depth with laser power and scan speed, respectively. The errors between the experimental and simulated melt pool depth for all the process parameters lie within 15%, and the developed numerical model is acceptable for the current study.

The current section discusses the experimental results obtained for the laser-polished Ti6Al4V surfaces for various process parameters. The residual stress induced in the samples is correlated to the melt pool dimensions and the cooling rate. The variation of surface topography and surface hardness is also discussed. A comparative discussion on the residual stresses induced on pure Ti and Ti6Al4V by LP is also included.

3.1 Temperature profile and melt pool depth

The validated numerical model is run for all the process parameters, and the surface transient temperature is exported. The variation of surface temperature against time for the case of constant scan speed and constant laser power is plotted in Fig. 5(a-b). The cooling rate is calculated using the predicted transient temperature profile, from the peak temperature for each process parameter to the recrystallization temperature of Ti6Al4V (1223 K). The lower bound of temperature for cooling rate calculation is chosen as recrystallization temperature as there are no changes in microstructure below the recrystallization temperature[13]. Hence, the cooling rate will not affect residual stress below the recrystallization temperature. Figure 6(a-b) shows the calculated cooling rates from the numerical model data for various process parameters. It is observed that the cooling rate does not have any significant change when there is a variation in laser power at constant scan speed (Fig. 6a).

Moreover, the cooling rate seems to have been significantly affected by the changes in the scan speed when laser power is kept constant (Fig. 6b). It is noted that higher scan speed results in a higher cooling rate and vice-versa. A high scan speed results in lower material interaction time and less energy impingement on the surface, which results in lower peak temperature, as shown in Fig. 5(a-b). A lower peak temperature solidifies faster, giving rise to higher cooling rates.

However, while the scan speed is kept constant, the material interaction time remains the same. The change in laser power affects the peak temperature but does not contribute to cooling rate changes. The linear energy density (laser power/scan speed) is introduced to determine the cumulative effect of laser power and scan speed rather than focusing on their individual effects. When the laser power is varied and scan speed is kept constant, the minimum and maximum linear energy density is calculated as 267 J/m and 400 J/m, respectively. Whereas, when the laser power is kept constant and scan speed is varied, the minimum and maximum linear energy density is calculated as 200 J/m and 1000 J/m, respectively. The changes in scan speed exhibit larger variations in the linear energy density. It is concluded that for the current set of process parameters, the process and cooling rate is much more sensitive to the changes in the scan speed than to the changes in the laser power.

The melt pool depths are calculated for the laser polished samples and Fig. 7(a-b) shows the variation of melt pool depth for the case of constant scan speed and laser power. It can be observed that the melt pool depth increases with an increase in power when the scan speed is kept constant at 0.3 m/s. The calculated melt depth for a minimum power of 80 W is 22.8$\:\:{\mu\:}\text{m},$ and for a maximum power of 120 W is 41.3 $\:{\mu\:}\text{m}$. For the case of varying scan speed with constant laser power of 100 W, the melt depth decreases with the increase in scan speed. The calculated melt depth for minimum velocity of 0.1 m/s is 66.9 $\:{\mu\:}\text{m},$ and for maximum scan speed is 20.1 $\:{\mu\:}\text{m}$. Higher laser power is associated with higher energy density impinged on the surface, resulting in a deeper melt pool. Higher scan speed is associated with lower material interaction time, leading to a shallow melt pool. Also, when laser power is varied, the range of the variation of the melt pool depth is lower when compared to the range of melt pool depth when scan speed is varied. The range of the variation of melt depth further supports the statement of scan speed being more sensitive than laser power for the current set of process parameters.

Melt pool depth increases with the power as more energy density has impinged on the sample surface, which melts more material, leading to a deeper melt pool. Furthermore, melt pool depth decreases with the increase in scan speed, as shown in Fig. 7(b). At high scan speed, laser interaction time with the sample surface decreases, due to which a shallow and wider melt pool is formed. Melt pool depth varies from 66.9 $\:{\mu\:}\text{m}$ to 20.1 $\:{\mu\:}\text{m}$ with respect to the scan speed from 0.1 m/s to 0.5 m/s, respectively.

3.2 Residual stress

The induced residual stress of laser-polished samples for all the process parameters is measured using an XRD machine. The XRD data is processed through the software package STRESS PLUS synced with the equipment. The sample is placed in the XRD machine in such a way that $\:{\sigma\:}_{xx}$ is the stress along the direction of the laser scan and $\:{\sigma\:}_{yy}$ is the stress perpendicular to the direction of the laser scan. The stress in the z-direction is considered to be zero, as the penetration depth of the X-ray is very low. The magnitude of shear stress is of order 1, whereas the magnitude of lateral and longitudinal stress is of order 2. Therefore, the plane in which the stress is calculated is considered as principal plane, and $\:{\sigma\:}_{xx}\:$become $\:{\sigma\:}_{11}$ or $\:{\sigma\:}_{1}$ and $\:{\sigma\:}_{yy}\:$become $\:{\sigma\:}_{22}$ or $\:{\sigma\:}_{2}$. However, $\:{\sigma\:}_{1}$ and $\:{\sigma\:}_{2}$ independently cannot give the essence of the effect of process parameters. Therefore, von Mises stress ($\:{\sigma\:}_{v}$) is introduced to get a cumulative effect of $\:{\sigma\:}_{1}$ and $\:{\sigma\:}_{2}$ [23] [24] [25] [26]. The von Mises stress is calculated by using the equation,

$$\:{\sigma\:}_{v}=\sqrt{{\sigma\:}_{1}^{2}+{\sigma\:}_{2}^{2}-{\sigma\:}_{1}{\sigma\:}_{2}}$$

The standard deviation of von Mises stress, $\:{\sigma\:}_{v}^{SD}$, is given by [27],

$\:{\sigma\:}_{v}^{SD}=\sqrt{{\left(\frac{\partial\:{\sigma\:}_{v}}{\partial\:{\sigma\:}_{1}}{\sigma\:}_{1}^{SD}\right)}^{2}+{\left(\frac{\partial\:{\sigma\:}_{v}}{\partial\:{\sigma\:}_{2}}{\sigma\:}_{2}^{SD}\right)}^{2}}$

(12)

where $\:{\sigma\:}_{1}^{SD},\:{\sigma\:}_{2}^{SD}$ is the standard deviation of stress in the$\:\:x$ and y direction, respectively.

The as-received sample of Ti6Al4V has a compressive residual stress of 321 $\:\pm\:$ 13.8 MPa. It is known that high thermal gradients and non-uniform cooling during the laser polishing of Ti6Al4V result in tensile residual stresses. Similar observations are also made in the current experiments where laser polishing has induced tensile stresses. For the case of constant speed of 0.3 m/s, the residual stress measured for the laser power of 80 W, 90 W, 100 W, 110 W and 120 W is 511$\:\pm\:$60.9 MPa, 556$\:\pm\:$62.2 MPa, 541$\:\pm\:$54.7 MPa, 537$\:\pm\:$58.5 MPa, and 553$\:\pm\:$61.1 MPa, respectively. The measurements are also plotted in Fig. 8(a). It is observed that there is no significant change in the residual stress with the variation of laser power. It has been noted from Fig. 6(a) that the cooling rate remains constant with the variation of laser power. It is concluded that the residual stress extensively depends on the cooling rate, increasing with the cooling rate [28] [29]. Hence, no significant changes in induced residual stress are observed due to insignificant variations in cooling rates.

For the case of constant laser power of 100 W and varying scan speed of 0.1 m/s, 0.2 m/s, 0.3 m/s, 0.4 m/s and 0.5 m/s, the measured residual stresses are 372$\:\pm\:$39.7 MPa, 473$\:\pm\:$33.7 MPa, 541$\:\pm\:$54.7MPa, 523$\:\pm\:$57 MPa, and 533$\:\pm\:$55.5 MPa, respectively. It is observed that the induced residual stress increases with the increase in the scan speed, and the trend is plotted in Fig. 8 (b). As the cooling rate increases with the increase in the scan speed (Fig. 6b), the induced residual stress shows similar behavior.

Further, the comparative residual stress analysis for pure Ti and Ti6Al4V can be seen in Fig. 8(a-b). For pure Ti, similar to Ti6Al4V, induced the residual stress has no significant variation with change in power when scan speed is kept constant. When laser power is kept constant, the induced residual stress increases with increased scan speed. The trend of induced residual stress in pure Ti is similar to the one observed in Ti6Al4V. However, for all the process parameters, the tensile residual stress induced in pure Ti is lower than the Ti6Al4V. The phase transformation of pure Ti is at a constant temperature; however, for Ti6Al4V, the transformation happens over a range of temperatures. The range of temperature for the solidification in Ti6Al4V arises due to alloying elements, all of which have different thermal properties. During solidification, different alloying elements solidify at different temperatures. The elements solidifying earlier hinder grain growth, which is still molten. The hindrance forms larger grain boundaries, which act as a source of stress concentration. Hence, for Ti6Al4V, the induced residual stress is higher than those induced in pure Ti.

3.2 Surface roughness and hardness

Figure 9 shows the variation in surface topography and $\:{S}_{a}$ of laser processed sample with laser power ranging from 80W to 120W by keeping the scan speed constant at 0.3m/s and with scan speed ranging from 0.1m/s to 0.5m/s by keeping the power constant at 100W. The magnitude of surface roughness at different power 80 W, 90 W, 100 W, 110 W and 120 W are 0.08$\:\:{\mu\:}$m, 0.09$\:\:{\mu\:}$m, 0.09$\:\:{\mu\:}$m, 0.10$\:\:{\mu\:}$m, and 0.11 $\:{\mu\:}$m, respectively. Similarly, variations of surface roughness with different scan speeds of 0.1 m/s, 0.2 m/s, 0.3 m/s, 0.4 m/s and 0.5 m/s are 0.04$\:\:{\mu\:}$m, 0.06$\:\:{\mu\:}$m, 0.09$\:\:{\mu\:}$m, 0.11$\:\:{\mu\:}$m, and 0.12 $\:{\mu\:}$m, respectively. The average surface roughness of the as-received Ti6Al4V sample is 0.14 $\:{\mu\:}$m. After LP, the surface roughness is reduced significantly compared to the bare sample. The optimised parameter for laser polishing is 0.1 m/s scan speed with 100 W laser power, resulting in $\:{S}_{a}$ of 0.04$\:\:{\mu\:}\text{m}$ which is a reduction of ~ 71%. The process parameter also reflects the lowest induced residual stress among all the process parameters. Hence, the induced residual stress is minimised while resulting in the lowest $\:{S}_{a}$ value. The optimised parameter has a linear energy density of 1000 J/m, the highest of the process parameters for the current work. Higher energy density results in a larger melting zone and hydrodynamics flows, which enhances the redistribution of the surface material and improves the surface finish.

The hardness of the as-received Ti6Al4V sample is measured as ~ 427 VHN. Figure 10 shows the variation of the surface hardness of laser polished Ti6Al4V with process parameters at constant scan speed and laser power, respectively. It is observed that the micro-hardness of the laser processed sample is less than the bare sample but is maintained with the variation of process parameters.

In this study, the effect of process parameters on residual stress, surface roughness and surface hardness are studied during laser polishing. A validated FEM-based numerical model predicts the transient temperature and cooling rates for various process parameters. Some of the key observations are discussed as follows:

The induced residual stress and cooling rate have no significant variation with changes in laser power for the process parameters used in this study. The changes in power shift the energy density to a narrow range, which does not affect the induced residual stress. However, the variation in scan speed gives a larger range of energy densities. Hence, the cooling rate increases significantly with increased scan speed, resulting in higher residual stress.
A higher magnitude of residual stress is induced in Ti6Al4V compared to pure Ti for all the process parameters. The presence of alloying elements in Ti6Al4V results in a non-uniform solidification rate, giving rise to larger grain boundaries, which are the source of stress concentration.
Around 71% reduction in the surface roughness after laser polishing is obtained with the process parameter of 100 W and 0.1 m/s. The same process parameter is responsible for the lowest induced residual stress of ~372 MPa.
The hardness of laser processed sample is maintained for sets of process parameters.

Acknowledgements

The authors are thankful for the support of IIT Gandhinagar in conducting the present study. The authors acknowledge Mr. Jatin Khankhal, Indian Institute of Technology Gandhinagar for helping in experiments.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work the author(s) used Grammarly in order to improve language and readability. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

Conflict of Interest statement: The authors have no competing interests as defined by Springer, or other interests that might be perceived to influence the results and/or discussion reported in this paper.

Ethical Approval: Not applicable

Funding: The authors thank the financial support of IIT Gandhinagar in carrying out the work presented in this manuscript. The authors also thank the fellowship received by Ms Rama Balhara and Mr Justin Hijam from the Ministry of Education, India.

Availability of data and materials: Not applicable

R. Avilés, J. Albizuri, A. Lamikiz, E. Ukar, A. Avilés, Influence of laser polishing on the high cycle fatigue strength of medium carbon AISI 1045 steel, International Journal of Fatigue 33 (2011) 1477–1489. https://doi.org/10.1016/j.ijfatigue.2011.06.004.
L. Chen, B. Richter, X. Zhang, K.B. Bertsch, D.J. Thoma, F.E. Pfefferkorn, Effect of laser polishing on the microstructure and mechanical properties of stainless steel 316L fabricated by laser powder bed fusion, Materials Science and Engineering: A 802 (2021) 140579.
B. Rosa, P. Mognol, J.-Y. Hascoët, Laser polishing of additive laser manufacturing surfaces, Journal of Laser Applications 27 (2015) S29102.
A.M. Souza, R. Ferreira, G. Barragán, J.G. Nuñez, F.E. Mariani, E.J. da Silva, R.T. Coelho, Effects of Laser Polishing on Surface Characteristics and Wettability of Directed Energy-Deposited 316L Stainless Steel, J. of Materi Eng and Perform 30 (2021) 6752–6765. https://doi.org/10.1007/s11665-021-05991-y.
B. Richter, T. Radel, F.E. Pfefferkorn, Sensitivity of surface roughness to laser parameters used for polishing additively manufactured Co-Cr alloy, Surface and Coatings Technology 451 (2022) 128872. https://doi.org/10.1016/j.surfcoat.2022.128872.
A. Bacciochini, N. Glandut, P. Lefort, Surface densification of porous ZrC by a laser process, Journal of the European Ceramic Society 29 (2009) 1507–1511. https://doi.org/10.1016/j.jeurceramsoc.2008.09.002.
R. Comesaña, F. Lusquiños, J. del Val, T. Malot, M. López-Álvarez, A. Riveiro, F. Quintero, M. Boutinguiza, P. Aubry, A. De Carlos, J. Pou, Calcium phosphate grafts produced by rapid prototyping based on laser cladding, Journal of the European Ceramic Society 31 (2011) 29–41. https://doi.org/10.1016/j.jeurceramsoc.2010.08.011.
H. Yves-Christian, W. Jan, M. Wilhelm, W. Konrad, P. Reinhart, Net shaped high performance oxide ceramic parts by selective laser melting, Physics Procedia 5 (2010) 587–594. https://doi.org/10.1016/j.phpro.2010.08.086.
A. Lodh, T.N. Tak, A. Prakash, P.J. Guruprasad, C. Hutchinson, I. Samajdar, Relating Residual Stress and Substructural Evolution During Tensile Deformation of an Aluminum-Manganese Alloy, Metall Mater Trans A 48 (2017) 5317–5331. https://doi.org/10.1007/s11661-017-4280-x.
Q. Han, Y. Jiao, Effect of heat treatment and laser surface remelting on AlSi10Mg alloy fabricated by selective laser melting, Int J Adv Manuf Technol 102 (2019) 3315–3324. https://doi.org/10.1007/s00170-018-03272-y.
Laser cladding | Journal of Laser Applications | AIP Publishing, (n.d.). https://pubs.aip.org/lia/jla/article-abstract/11/2/64/374959/Laser-cladding (accessed October 28, 2023).
K. Shah, A.J. Pinkerton, A. Salman, L. Li, Effects of Melt Pool Variables and Process Parameters in Laser Direct Metal Deposition of Aerospace Alloys, Materials and Manufacturing Processes 25 (2010) 1372–1380. https://doi.org/10.1080/10426914.2010.480999.
Froes, F. (ed), Titanium: physical metallury, processing, and applications, ASM International, 2015.
Optimization of the mechanical properties of Ti-6Al-4V alloy fabricated by selective laser melting using thermohydrogen processes - ScienceDirect, (n.d.). https://www.sciencedirect.com/science/article/abs/pii/S0921509317307785 (accessed September 30, 2023).
J.S. Jesus, L.P. Borrego, J.A.M. Ferreira, J.D. Costa, C. Capela, Fatigue crack growth behaviour in Ti6Al4V alloy specimens produced by selective laser melting, Int J Fract 223 (2020) 123–133. https://doi.org/10.1007/s10704-019-00417-2.
T. DebRoy, H.L. Wei, J.S. Zuback, T. Mukherjee, J.W. Elmer, J.O. Milewski, A.M. Beese, A. Wilson-Heid, A. De, W. Zhang, Additive manufacturing of metallic components – Process, structure and properties, Progress in Materials Science 92 (2018) 112–224. https://doi.org/10.1016/j.pmatsci.2017.10.001.
J. Hijam, R. Balhara, M. Vadali, Investigating Double-Scan Strategies for Reducing Heat-Affected Zone in Laser Surface Melting, (2023). https://doi.org/10.2139/ssrn.4531337.
Y. Harada, Y. Ishida, D. Miura, S. Watanabe, H. Aoki, T. Miyasaka, A. Shinya, Mechanical Properties of Selective Laser Sintering Pure Titanium and Ti-6Al-4V, and Its Anisotropy, Materials 13 (2020) 5081. https://doi.org/10.3390/ma13225081.
S. Kou, D.K. Sun, Fluid flow and weld penetration in stationary arc welds, MTA 16 (1985) 203–213. https://doi.org/10.1007/BF02815302.
K.C. Mills, Recommended values of thermophysical properties for selected commercial alloys, Woodhead Publishing, 2002.
"Phase Change Material". https://doc.comsol.com/5.5/doc/com.comsol.help.heat/heat_ug_ht_features.09.025.html, (n.d.).
J. Hijam, R. Balhara, M. Vadali, Investigating double-scan strategies for reducing heat-affected zone in laser surface melting, Optics & Laser Technology 170 (2024) 110289. https://doi.org/10.1016/j.optlastec.2023.110289.
B.S. Yilbas, S.S. Akhtar, C. Karatas, Laser trepanning of a small diameter hole in titanium alloy: Temperature and stress fields, Journal of Materials Processing Technology 211 (2011) 1296–1304. https://doi.org/10.1016/j.jmatprotec.2011.02.012.
I. Serrano-Munoz, A. Ulbricht, T. Fritsch, T. Mishurova, A. Kromm, M. Hofmann, R.C. Wimpory, A. Evans, G. Bruno, Scanning Manufacturing Parameters Determining the Residual Stress State in LPBF IN718 Small Parts, Adv Eng Mater 23 (2021) 2100158. https://doi.org/10.1002/adem.202100158.
Scanning Manufacturing Parameters Determining the Residual Stress State in LPBF IN718 Small Parts - Serrano-Munoz - 2021 - Advanced Engineering Materials - Wiley Online Library, (n.d.). https://onlinelibrary.wiley.com/doi/full/10.1002/adem.202100158 (accessed September 26, 2023).
B.S. Yilbas, A.F.M. Arif, C. Karatas, K. Raza, Laser treatment of aluminum surface: Analysis of thermal stress field in the irradiated region, Journal of Materials Processing Technology 209 (2009) 77–88.
J. Tellinghuisen, Statistical Error Propagation, J. Phys. Chem. A 105 (2001) 3917–3921. https://doi.org/10.1021/jp003484u.
Z. Zhang, W. Wang, H. Fu, J. Xie, Effect of quench cooling rate on residual stress, microstructure and mechanical property of an Fe–6.5Si alloy, Materials Science and Engineering: A 530 (2011) 519–524. https://doi.org/10.1016/j.msea.2011.10.013.
Y. Yi, Q. Li, J. Xing, H. Fu, D. Yi, Y. Liu, B. Zheng, Effects of cooling rate on microstructure, mechanical properties, and residual stress of Fe-2.1B (wt%) alloy, Materials Science and Engineering: A 754 (2019) 129–139. https://doi.org/10.1016/j.msea.2019.03.061.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
18 Nov, 2024
Reviews received at journal
18 Nov, 2024
Reviews received at journal
13 Nov, 2024
Reviews received at journal
07 Nov, 2024
Reviewers agreed at journal
17 Oct, 2024
Reviewers agreed at journal
17 Oct, 2024
Reviewers agreed at journal
15 Oct, 2024
Reviewers invited by journal
14 Oct, 2024
Editor assigned by journal
14 Oct, 2024
Submission checks completed at journal
14 Oct, 2024
First submitted to journal
09 Oct, 2024

You are reading this latest preprint version

Effect of Laser Polishing Parameters on Residual Stress, Surface Roughness, and Microhardness of Ti6Al4V

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods and materials

2.1 Experimental setup

2.2 Measurement methods

2.3 Numerical model description

2.3.1 Assumptions

2.3.2 Governing equations and boundary conditions

2.3.3 Material properties

2.3.4 Model validation

3. Results and discussion

3.1 Temperature profile and melt pool depth

3.2 Residual stress

3.2 Surface roughness and hardness

4. Conclusions

Declarations

References

Additional Declarations

Status:

Version 1