Objective To reduce the complex residual stress, deformation and coarse grains generated in the laser cladding process of thin plates, and improve the quality of cladding layer and the reliability of parts assembly. Methods A synchronous gas cooling method for laser cladding using liquid nitrogen to cool nitrogen was proposed. A double ellipsoid heat source was used, and the temperature-dependent thermophysical parameters, heat conduction, heat radiation, powder-light interaction, and the role of latent heat of phase change during solidification were comprehensively considered. A three-dimensional numerical model of laser cladding with synchronous gas cooling was established, and the transient temperature field, molten pool flow field and stress evolution process under different cooling distances (d) were numerically calculated. At the same time, the temperature field distribution was verified by thermocouples, and the microstructure and coating hardness were analyzed by scanning electron microscopy (SEM) and microhardness tester. Result The introduction of cooling nitrogen enhanced the heat exchange between the molten pool surface and the external environment, thereby improving the temperature distribution of the molten pool and accelerating the cooling rate of the molten pool. The flow field change trend of the molten pool of conventional laser cladding and synchronous gas cooling laser cladding is basically the same, showing a flow from the center to the surroundings, with a small flow velocity at the center and a large flow velocity at the edge, and a flow velocity in front of the molten pool greater than that behind the molten pool. When the cooling distance is 5 mm, the maximum lateral residual stress of the substrate decreases from 204 MPa to 181 MPa, the residual stress at the top of the cladding layer decreases from 190 MPa to 172 MPa, and the residual stress in the interface bonding area decreases from 234 MPa to 211 MPa, and the warping deformation on both sides of the substrate is also reduced by 50%. When the cooling distance is 10 mm, the grain structure of the coating is significantly refined, the disordered arrangement of the columnar crystals in the middle is significantly increased, and the microhardness of the coating is increased from 348.2HV0.2 to 375HV0.2. Conclusion This method can effectively reduce the residual stress and deformation of the substrate, and provide a new method for better controlling the grain and microstructure of the laser cladding coating.

The laser cladding process has the advantages of small heat-affected zone, low dilution rate, and good metallurgical bonding. It has been widely used in various industrial fields such as automobile industry, aerospace, shipbuilding, and petrochemical. However, in the actual forming process, due to the difference in thermophysical properties between the cladding layer material and the substrate and the high temperature gradient in the heated zone, the local thermal stress is too high and plastic deformation occurs. The subsequent cooling and shrinkage stage produces high residual stress, which easily leads to part deformation, especially in thin-walled parts. This deformation may cause the parts to fail to meet assembly requirements or fail during part maintenance. In order to solve this problem, Li Guangqi et al. studied the effects of different scanning methods on the residual stress distribution of the laser cladding layer and the deformation of the substrate, and found that the block scanning method can significantly reduce the residual stress distribution of the cladding layer and improve the degree of warping deformation of the substrate. However, this method may affect the overall cladding efficiency and cause overall inhomogeneity of the cladding layer. Du et al. prepared a Fe60 laser cladding layer on 304 stainless steel and performed heat treatment. The results showed that the heat treatment effectively reduced the residual stress and lattice deformation in the coating, but also reduced the microhardness. Therefore, developing a technology that can efficiently control residual stress and deformation during laser cladding is of great practical significance for improving the quality of laser cladding layers.

Synchronous gas cooling technology adjusts the laser cladding temperature field by local cooling, controls the development of plastic strain behind and on both sides of the melt channel, and produces a strong tensile effect in the cladding metal in the rapidly shrinking cold source area, compensating for the plastic deformation in the compression state of the cladding high temperature area, so that after the cladding is completed, the peak value of the residual tensile stress area is controlled, and the quenching accelerates the cooling rate of the molten pool, promotes the refinement of grains, and further improves the microstructure and performance of the cladding layer. Guan Qiao et al. proposed a dynamic low-stress and deformation-free welding method, which can effectively prevent the instability and deformation of thin plate welding. Zhang et al. proposed a dynamic cooling method of welding gas suitable for laser welding deformation control. This method uses liquid nitrogen to cool argon gas, and further uses the cooled argon gas as a cooling medium, which has a good control effect on longitudinal flexural deformation. This method can reduce deformation by 71%, longitudinal residual stress by 21%, and longitudinal plastic strain by 14%. Xie Hao et al. used solid dry ice as a quenching medium to significantly reduce the welding deformation and residual stress of aluminum alloy thin plates, and had a certain refinement effect on the microstructure of the weld area of ​​the weld joint. Liu et al. studied the effect of synchronous gas cooling on the microstructure and mechanical properties of Ti6Al4V titanium alloy laser welded joints. The introduction of cooling argon gas mainly enhanced the heat exchange between the molten pool surface and the external environment, thereby improving the temperature distribution of the molten pool and increasing the microhardness value of the weld metal from 423.7HV1.0 to 453.7HV1.0. Laser cladding has a similar principle to laser welding. At present, there are few studies on the application of synchronous gas cooling technology in laser cladding to control residual stress and deformation.

In this study, the synchronous gas cooling experiment of Q235 thin plate was carried out by using relevant equipment, and COMSOL software was used to simulate the laser cladding based on the thermoelastic-plastic finite element method, taking into full account the characteristics of the heat source and the material, and the temperature field, flow field, stress and other results were calculated. The corresponding experimental comparative analysis was carried out, and the residual stress, molten pool flow field and microstructure under different cooling distances were analyzed to optimize the synchronous gas cooling technology.

1 Experimental materials and methods

1.1 Experimental materials
This study used Q235 as the base material and 316L stainless steel as powder. Its chemical composition is shown in Table 1. Before the experiment, the surface of Q235 was polished with 400#~1000#SiC sandpaper to remove the surface oxides and impurities.

1.2 Experimental methods
The synchronous gas cooling assisted laser cladding system was built in the experiment, which consists of HWF20 laser system, liquid nitrogen assisted cooling system and protective gas system. The structure of the whole experimental system is shown in Figure 1. The temperature of nitrogen at the outlet of the liquid nitrogen device was measured using a 1310K thermocouple electronic thermometer. The results show that the special liquid nitrogen cooling device can reduce the temperature of the outlet nitrogen to -40°C. During the cladding process, the device can achieve synchronous follow-up cooling of the back of the substrate at a certain distance behind the heat source, quickly absorb most of the heat, and effectively achieve rapid cooling. The laser cladding parameters used in the experiment are: laser power 1 000 W, scanning speed 8 mm/s, and powder feeding rate 1.4 r/min. The experiment was conducted with single-pass laser cladding without cooling and with cooling distances of 5 mm, 10 mm, and 15 mm. During the laser cladding process, argon was used to protect the molten pool.
The specimens were prepared along the longitudinal and transverse directions of the cladding layer using Guangdong Datie DK350 wire cutting. After grinding and polishing, the cross section of the cladding layer was corroded by aqua regia for 10 s, cleaned with alcohol in ultrasonic waves, and then dried. The actual cross-sectional morphological parameters of the coating were measured using an ultra-depth microscope (Keyence VHX-6000), and the results were compared with the numerical simulation data to verify the reliability of the calculation model. The deformation of the substrate after laser cladding was measured using an ultra-depth microscope. With the help of the three-dimensional imaging function of the microscope, the morphology of the middle part of the substrate deformation was measured at a magnification of 100 times. Each set of parameter measurements was performed at the same position to ensure the accuracy and comparability of the data. The temperature was measured during the cladding process using a thermocouple. The microstructure of the upper, middle and lower parts of the coating was analyzed using a scanning electron microscope (Zeiss SUPRA55). A microhardness tester (TMHV-1000Z) was used to measure the hardness every 0.1 mm in the depth direction of the cladding layer (test load 200 N, holding time 15 s).

2 Establishment of numerical model of laser cladding

2.1 Basic assumptions
Laser cladding is a process in which powder and substrate melt and solidify rapidly after heat input. In order to ensure the accuracy of the model calculation and improve the calculation efficiency, the following assumptions were made in the model construction [11,12]: (1) The temperature-related material thermophysical properties are calculated by JMatPro; (2) The impact of laser pulse, powder carrier gas, shielding gas and metal powder on the morphology of the cladding layer is not considered; (3) The molten pool fluid is an incompressible laminar flow, and the fluid is acted upon by gravity and buoyancy. The effects of shielding gas, powder feeding gas and powder on the molten pool interface are ignored; (4) The ambient temperature is assumed to be 293 K during the study, and only the heat convection and heat radiation between the substrate and the surrounding environment are considered during the calculation; (5) The yield of the cladding material obeys the Von Mises yield criterion.

2.2 Geometric model and meshing
A three-dimensional numerical model was established, which includes fluid heat transfer, laminar flow, dynamic mesh and solid mechanics modules. The model was modeled according to the actual size 1:1 and divided into two regions for more accurate calculation. The middle region uses a hexahedral mesh, and the two sides use a free tetrahedral mesh to mesh the substrate. The dimensions of the model are 150 mm×60 mm×2 mm and 150 mm×4 mm×2 mm, respectively, and contain a total of 197,897 domain units, as shown in Figure 2. The parameters used in the calculation are shown in Table 2. Figure 3 provides the thermophysical parameters of Q235 and 316L materials. These parameters are imported into the COMSOL software in the form of interpolation functions, and then the boundary conditions of the model are defined and solved.

2.3 Control equations of temperature field, flow field, and stress field
The laser cladding process involves multi-physics field coupling, and the mass, momentum, and energy conservation equations in the process are as follows (1)-(3): See the formula in the figure.
𝐹 is the volume force acting on the fluid (N), which can be expressed as: See formula (4) in the figure.
Where: 𝜌 is the material density (kg·m-3); 𝑢 is the velocity of the fluid flow in the molten pool (m·s-1); 𝑡 is time (s); 𝐶𝑝 is the specific heat of the material (J·kg-1·K-1); 𝑇 is the temperature (K); 𝑇𝑚 is the melting temperature (K); 𝑘 is the thermal conductivity (W·m-1·K-1); 𝑄𝑚 is the laser heat source (W·m-3); 𝑃 is the fluid pressure (Pa); 𝐼 is the unit matrix; 𝜇 is the liquid metal dynamic viscosity (Pa·s); 𝛽 is the product expansion coefficient (K-1); 𝐾0 is the mushy area coefficient, 𝐵 is a decimal (0.001) to avoid the denominator being zero; 𝑓𝑙 is the liquid phase volume fraction. See formula (5) in the figure.
Where: 𝑠 and 𝑙 represent the solid phase and liquid phase respectively. In the calculation of plastic stress field, the von Mises yield criterion is used to judge whether the material undergoes plastic transformation. The judgment expression is shown in (6).
Where: 𝜎1, 𝜎2, 𝜎3 represent the principal stresses along the x-axis, y-axis and z-axis respectively (MPa); 𝜎𝑠 represents the yield strength of the material (MPa). When the average stress 𝜎̅ is not less than the yield strength 𝜎𝑠, the material begins to undergo plastic deformation.

2.4 Boundary conditions of temperature field, flow field and stress field

2.4.1 Heat source model
The double ellipsoid heat source model comprehensively considers the characteristics of large temperature gradient at the front end of the molten pool and small temperature gradient at the tail end in the actual temperature distribution. In this model, the heat source 𝑞𝑓 (W·m-3) in the front part and the heat source 𝑞𝑟 (W·m-3) in the back part are expressed by formulas (7) and (8), respectively.
Where: 𝑝 is the laser power (W); 𝛼 is the absorption rate of the material to the laser energy, which is set to 0.5 here; 𝑉 is the moving speed during the laser cladding process; the geometric parameters a, 𝑏𝑓, 𝑏𝑟, and c of the double ellipsoid heat source model
respectively represent the semi-axis lengths in different directions. According to Figure 4, the values ​​of these parameters are 1 mm, 1 mm, 1.5 mm, and 0.4 mm, respectively; 𝑓1 and 𝑓2 represent the energy distribution coefficients of the front and back of the double ellipsoid heat source, which are 0.6 and 0.4, respectively; 𝑡 is the duration of the laser action (s).
During the laser cladding process, the heat dissipation between the molten pool and the environment is mainly convection 𝑞𝑐 (W·m-2) and radiation 𝑞𝑓 (W·m-2) between the molten pool and the environment. The equations are as follows: See formulas (9) and (10) in the figure.
Where: ℎ𝑐 is the convection heat transfer coefficient (W·m-2·K-1); 𝑇𝑎𝑚𝑏 is the set ambient temperature (K); 𝜎𝑏 is the Boltzmann constant (W·m-2·K-4); 𝜀 is the surface emissivity of the material.
In the laser cladding flow field, the capillary force is presented in the form of surface tension. The surface tension acts on the surface of the molten pool, restricting the flow of surface fluid and producing the Marangoni effect. The equations are as follows: See formula (11) in the figure.
Where: 𝛾1 is the surface tension of pure metal (N·m-1); 𝛾0 is the surface tension at reference temperature (N·m-1·K-1); 𝑇𝑀 is the melting temperature (K). The shape of the melt channel is determined by the interaction between the laser energy distribution and the powder density distribution. Coaxial powder feeding is adopted, and the powder is Gaussian distribution. The equation of 𝑉𝑝(m·s-1) is: See formula (12) in the figure.
Where: 𝑚𝑓 is the powder feeding speed (g·s-1); 𝜂𝑝 is the utilization efficiency of the powder; 𝜌𝑝 is the powder material density (kg·m-3); 𝑟𝑝 is the mass flow radius (mm); 𝑧⃗ is the unit vector in the z direction.
2.4.2 Cold source model
The nitrogen gas cooled by liquid nitrogen is blown onto the substrate surface to cool the local area at a certain distance behind the heat source. Forced convection is used in COMSOL analysis, and the expression of the convection heat transfer coefficient is as follows: see formula (13) in the figure.
The convection heat transfer coefficient ℎ𝑐(W·m-2·K-1)[19] at the center of the cold source can be written as follows: see formula (14) in the figure.
Where: the effective radius of the cold source 𝑟0 is 2 mm; the room temperature 𝑇0 is 293 K; the outlet temperature of the cooling nitrogen 𝑇𝐶 is 233 K; since the cooling effect is applied to the Q235 steel substrate in the experiment and simulation, the physical parameters of Q235 are used to estimate the convection heat transfer coefficient. 𝛿 is the thickness of the substrate (mm); 𝜌 is the density of the material (kg·m-3); 𝑐𝑝 is the specific heat capacity (J·kg-1·K-1); 𝜕𝑇⁄𝜕𝑡 is the cooling rate (K·s-1).

2.5 Model Verification
In order to verify the validity of the numerical model, experimental verification was carried out. The melt height, melt depth, melt width and heat affected zone (HAZ) of the coating were measured by electron microscopy. Figure 5 shows the comparison between the experimental results and the simulation results of the cross-sectional morphology of the coating. Through comparative analysis, it was found that the maximum deviations between the simulation values ​​and the experimental measurements in terms of melt height, melt depth, melt width and heat affected zone did not exceed 1%, 4%, 2% and 19%, respectively, thus verifying the validity of the model.

3 Results and analysis

3.1 Temperature field calculation results
Based on the established temperature field model, the temperature field distribution cloud diagrams of conventional laser cladding and synchronous gas cooling laser cladding (d=5 mm) are shown in Figure 6. It can be observed that the temperature field distribution range of synchronous gas cooling is much smaller than that of conventional laser cladding. After establishing the isotherms of the heat affected zone (800 K) and the melting zone (1 800 K), it can be found that the 800 K isotherm of synchronous gas cooling cladding is closer to a circle due to the cooling at the rear, while the isotherm of conventional laser cladding is closer to an ellipse. For the 1 800 K isotherm, the shape does not change significantly. This shows that the influence of synchronous gas cooling on temperature is mainly in the heat affected zone. Moreover, under the same cladding conditions, the molten pool of synchronous gas cooling laser cladding reaches a steady state earlier than that of conventional laser cladding.
Figure 7 shows the temperature distribution on the surface of the conventional laser cladding and synchronous gas cooling laser cladding substrates when the cladding reaches x=120 mm. The data in Figure 7 show that in front of the molten pool, the temperature field distribution is basically similar, the temperature gradient is large, and the isotherms are dense. However, behind the molten pool, the temperature field of synchronous gas cooling has denser isotherms than conventional laser cladding. Synchronous gas cooling enhances the heat exchange between the molten pool surface and the external environment by introducing cooling nitrogen, forming a local low-temperature area. This improves the temperature distribution of the molten pool and accelerates the cooling rate of the molten pool.
Figure 8 shows the temperature acquisition curves at 3 mm (point A), 6 mm (point B), and 9 mm (point C) from the center of the melt. Due to the irradiation of the laser, the material in the laser beam irradiation area melts rapidly, and the temperature of the material in this area rises sharply. As the heat source of the laser beam moves to different positions, the material in the heated area gradually cools down. The simulated thermal cycle curve is consistent with the experimental thermal cycle curve, which lays the foundation for the subsequent molten pool flow rate and stress analysis.

3.2 Velocity flow field calculation results
The Marangoni force is the main driving force for fluid flow in the laser molten pool. The Marangoni force in the molten pool will have an important impact on the temperature distribution and the solidification behavior of the microstructure, thereby changing the growth process of the molten pool, causing the generation of internal stress in the material, and leading to deformation and cracks. Therefore, it is of great significance to study the movement of the molten pool. In order to obtain the law of change of the molten pool flow field, the distribution of the molten pool flow field of conventional laser cladding and synchronous gas cooling laser cladding is selected, as shown in Figure 9. The arrows in the figure indicate the flow direction of the fluid in the molten pool. Since the surface tension temperature coefficient of 316L stainless steel is negative, the liquid metal on the upper surface flows from the center to the surroundings, with a low flow velocity in the center and a high flow velocity at the edge, and the flow velocity in front of the molten pool is greater than the flow velocity in the rear of the molten pool. This phenomenon is due to the small temperature difference in the center of the molten pool and the large temperature gradient 𝑑𝑇⁄𝑑𝑥 ​​in front of the molten pool, which in turn causes the tension gradient 𝑑𝛾⁄𝑑𝑥 in front of the molten pool to gradually increase. The large surface tension difference will further strengthen the Marangoni convection, resulting in a phenomenon in which the flow velocity in front of the molten pool is higher than the flow velocity in the rear. It can be seen that the flow field change trends of conventional cladding and synchronous gas cooling cladding are basically similar. Since synchronous gas cooling mainly acts on the rear of the molten pool, its effect on the overall temperature of the molten pool is limited. Therefore, the flow inside the molten pool will not be significantly enhanced, which avoids the uneven temperature distribution caused by the intensified flow inside the molten pool and reduces the possibility of residual stress in the cladding layer during the solidification process.
In order to further study the relationship between the flow velocity and temperature in the molten pool, the temperature, flow velocity and temperature gradient along the X direction of the molten pool were calculated in a steady state, and the specific relationship was obtained as shown in Figure 10. According to the figure, it can be observed that the temperature of the molten pool in the center area of ​​the laser action is higher, while the corresponding flow velocity is lower. As the distance from the laser center point increases, the flow velocity on both sides of the molten pool begins to gradually increase, especially in the periphery of the high-temperature area of ​​the molten pool, where the flow velocity reaches a peak. The main reason for this phenomenon is that the temperature gradient in the center of the molten pool is relatively small, while the temperature gradient on both sides increases significantly. The trend of the temperature gradient and the flow velocity is consistent, which also shows that the surface tension gradient caused by the temperature gradient in the molten pool is the key factor driving the flow of metal liquid.

3.3 Stress field calculation results
In order to study the effect of cooling distance on residual stress, residual stress simulation was performed in the transverse and middle sections of the cladding layer along the thickness direction, as shown in Figure 11. As can be seen from Figure 11(a), before and after the substrate is cooled, the maximum points of the residual stress curve in the transverse direction appear near the two ends of the path, that is, close to the edges of the two sides of the molten path. From the center of the melt to the side of the substrate, the residual stress shows a trend of increasing first and then decreasing, which is consistent with the simulation results of Huang Guoshun et al. [24]. This is mainly due to the fact that the double ellipsoid laser heat source provides higher energy in the central area, while the energy in the edge area is lower, resulting in a faster cooling rate of the molten pool at the edge of the cladding layer. This uneven cooling rate causes the area near the center of the heat source to bear greater stress. However, after the implementation of synchronous gas cooling, it was observed that the transverse residual stress was significantly reduced, and the maximum transverse residual stress of the substrate decreased from 204 MPa to 181 MPa, a decrease of 11%. Figure 11(b) shows that in the layer thickness direction, the maximum residual stress appears in the bonding area between the cladding layer and the substrate. This is mainly due to the difference in mechanical properties between the 316L cladding material and the Q235 substrate, especially the significant difference in volume shrinkage between the two, which leads to stress concentration. As the distance from the interface bonding area with the substrate increases, the residual stress value gradually decreases. It is worth noting that synchronous gas cooling does not change the distribution position of the stress extreme point in the layer thickness direction, but reduces the residual stress value. The residual stress at the top of the cladding layer decreases from 190 MPa to 172 MPa, and the residual stress in the interface bonding area decreases from 234 MPa to 211 MPa. This is mainly because the nitrogen gas cooled by liquid nitrogen is sprayed onto the high-temperature metal behind the heat source, which effectively reduces the area of ​​the high-temperature area and causes a significant tensile effect on the metal surface between the heat source and the cold source, which can largely offset the compressive plastic deformation that has occurred, so it can effectively regulate and control the residual stress. By comparing the results, when the cooling distance is 5 mm, the control effect of residual stress is the best.

3.4 Substrate deformation
The deformation results of the laser cladding substrate are shown in Figure 12. It can be observed that both conventional laser cladding and synchronous gas cooling laser cladding show warping deformation, in which the melt channel is at the lowest point and the maximum deformation occurs on both sides of the melt channel. Compared with conventional laser cladding, when the cooling distance is 5 mm, it is found that the maximum deformation of the cladding substrate is reduced by 50%. This is because the effect of the cold source gas at this position reduces the range of the high-temperature area, accelerates the cooling rate, and exerts a tensile effect on the high-temperature area of ​​the melt channel that is cooling and shrinking, significantly reducing the residual compressive stress, and effectively reducing the bending deformation caused by the residual compressive stress exceeding the critical instability stress.

3.5 Microstructure
Figure 13 shows the microstructure of the top, middle and bottom bonding areas of the sample under no cooling and different cooling distances. It can be seen that the cladding layer and the substrate show good metallurgical bonding, without defects such as cracks or inclusions. The top of the cladding layer is mainly composed of equiaxed crystals, the middle part presents an equiaxed or columnar crystal structure with inclined growth, and the bottom area contains planar crystals. According to the solidification theory, the temperature gradient (𝐺 ) and the solidification rate (𝑅 ) have an important influence on the size and morphology of the coating grains. During the laser cladding process, the powder on the surface of the substrate is rapidly melted and diffused under the action of the laser, and then rapidly cooled by the heat dissipation of the substrate. Due to the effect of the laser, the powder is heated to an extremely high temperature during the whole process, but the substrate is not affected by the laser, so the 𝐺 value is very large, resulting in a very high 𝐺⁄𝑅 value. The excessively large 𝐺⁄𝑅 ratio determines that the nucleation rate of the crystals in the coating is faster than the growth rate, resulting in the formation of planar crystals or honeycomb crystals on the interface in the heat dissipation direction. As the solid-liquid interface advances, 𝐺 gradually decreases and 𝑅 increases relatively, resulting in a decrease in the 𝐺⁄𝑅 ratio. At this time, the planar crystals are transformed into columnar crystal structures. As the 𝐺 value decreases, heat begins to accumulate on both sides, and the crystals in the central area have relatively sufficient time to grow along the heat flow direction, which may cause the columnar crystals that originally grew in the vertical direction to grow obliquely or form oblique equiaxed crystals. As the solid-liquid interface continues to advance, the top of the molten pool is in direct contact with the outside world, which accelerates the heat dissipation rate. This process promotes the increase in the crystal nucleation and solidification rates, resulting in the formation of fine equiaxed grains on the surface of the cladding layer.
In addition, by comparing with and without cooling, it is found that the crystal type of the cladding layer is the same, and the crystal size of the cladding layer after cooling is smaller than that of the crystal size without cooling. Especially when the cooling distance is 10 mm, the size of the top equiaxed crystal reaches the minimum value, because the rapid cooling rate of the molten pool hinders the complete growth of the equiaxed crystal. In the middle area, the size of the columnar crystal becomes smaller after cooling, and the disordered arrangement of the columnar crystal in the middle increases significantly under cooling conditions. The cooling condition changes the temperature gradient perpendicular to the growth direction of the columnar crystal, so that their growth direction changes. In general, the sample with cooling has a more disordered and refined microstructure than the conventional laser cladding sample, which helps to improve the performance of the coating.

3.6 Hardness
Figure 14(a) shows the microhardness of the coating at different cooling distances. It can be observed that the microhardness value decreases from the top of the cladding layer to the substrate position, and Figure 14(b) shows the average microhardness comparison of the 316L cladding layer and the Q235 substrate at different cooling distances. The results show that after cooling, the hardness is improved. The average hardness of the cladding layer with cooling is 5-29HV0.2 higher than that of the sample without cooling. Under the cooling condition of 5 mm behind the cladding layer, the cooling effect of the coating is not significant due to the proximity to the upper high-temperature zone, and the hardness of the heat-affected zone is greatly affected. Under the cooling condition of 10 mm behind, the hardness of the cladding layer is more significantly affected as the cooling proceeds. At a higher cooling rate, the grain size decreases and the number of grain boundaries increases, which effectively hinders the displacement of dislocations and significantly improves the hardness of the cladding layer by about 40HV0.2. However, when the cooling distance is increased to 15 mm, due to the long distance, the heat dissipation rate of the substrate is too fast, and the effect of cooling on the temperature drop is small, resulting in a less significant improvement in hardness. In summary, the degree of grain refinement is the main reason for the change in hardness. A fine microstructure is formed at a higher cooling rate, resulting in an increase in the hardness of the coating.

4 Conclusions

1) During the synchronous gas cooling laser cladding process, the cooling nitrogen causes local low temperature, which has little effect on the temperature of the melting zone, but has a greater effect on the temperature of the heat affected zone, especially behind the molten pool. This effect causes the isotherms behind the molten pool to become more dense.
2) It can be seen that the flow field change trends of the molten pool in conventional laser cladding and synchronous gas cooling laser cladding are basically the same, showing a flow from the center to the surroundings, with a small flow velocity at the center and a large flow velocity at the edge, and the flow velocity in front of the molten pool is greater than that behind the molten pool. The change trends of the temperature gradient and the flow velocity are consistent, which also shows that the surface tension gradient caused by the temperature gradient in the molten pool is the key factor driving the flow of metal liquid.
3) When the cooling distance is 5 mm, the maximum lateral residual stress of the substrate decreases from 204 MPa to 181 MPa, the residual stress at the top of the cladding layer decreases from 190 MPa to 172 MPa, the residual stress in the interface bonding zone decreases from 234 MPa to 211 MPa, and the warping deformation on both sides of the substrate is also reduced by 50%. When the cooling distance is 10 mm, the grain structure of the coating is significantly refined, the disordered arrangement of the columnar crystals in the middle is significantly increased, and the microhardness of the coating is increased from 348.2HV0.2 to 375HV0.2.