Signicant Contribution of Temperature on AISI D2 Tool Steel Ground Surface Characteristics and Integrities

The Controlled grinding is governed by the maximum surface temperature in the wheel-workpiece interaction. In this study, we demonstrate that temperature is the significant controller on the surface characteristic grinding, the surface integrity, the productivity and the fatigue life. Moreover, high temperature generated in abrasive processes is the main factor responsible on ground surface damage and its impact on the induced consequences in grinding of AISI D2 tool steel. The combined effects of abrasive type, cooling mode according to the cutting depth, on the temperature and residual stress distribution were highlighted by exploiting FEM. Obtained numerical results were validated with the experimental ones.

Initial residual stresses in X-and Y-directions, respectively Residual stress tensor

Introduction
Grinding operations of tool steels involve cutting forces and heat fluxes which generate gradients of properties in the surface layers [1,2]. The grinding process requires a combination grinding (abrasive typecooling mode -cutting conditions) [3,4]. This combination controls the wheelworkpiece interaction and the superficial modifications [3, 5,6]. Consequently, this results in gradient surfaces with microstructural and mechanical properties that can compromise the life integrity of the finished or machined workpiece [7][8][9]. Several studies have been devoted to the identification of these interactions and their quantitative evaluation using experimental approaches [10][11][12], analytical models [6,13,14] and numerical simulations [14][15][16][17]. The resolution of this problematic requires the development of an experimental and / or a numerical evaluation methodology. This result allows to correlate industrial needs to combinations grinding (abrasive typecooling modecutting conditions) through their adequate choice sweeping the areas of low and high productivity. However, the proportion of heat conducted into the workpiece (heat partition ratio) was estimated an experimental approach based on the temperature measurement sufficiently near the grinding zone is required. The temperature and microstructure gradients of the affected layers create plastic deformation incompatibilities generating residual stress and strain distributions. Such distributions are extensively studied in the literature [5,6,18]. The residual stress profiles were established by either experimental measurements using the following methods [9,19,20], and by modelling and numerical simulations of the grinding process [18,21,22].
Generally the distributions resulting from alloyed steel grinding relate to the grinding wheel parameters (nature of the abrasive, type of dressing, ...) [23,24], the nature of the material to be machined (behaviour law, thermophysical characteristics) [25], the lubrication mode [3, 26,27] and the cutting conditions (aw, vw, vs) [18,28,29]. The following results of the effects of grinding parameters on the property gradients (temperature and residual stresses), through literature, reveal that the surface temperatures are higher for grinding with soluble oil than for cryogenic grinding of steel EN X210CrMoV12, as well as, the temperatures reached at the surface increase with increasing depth of cut (aw) [28]. Lin shows that the temperature rises from T=100°C for aw = 10µm to T = 150°C for aw = 50µm [30]. The surface temperature of grinding alloy steel (100Cr6) reached decreases with the rise of the table speed (vw) [31]. Lefebvre shows that in grinding of AISI 1045 (C45) steel that the temperature changes from T=370°C for vw=100mm/s to T=200°C for vw=300mm/s [32], the surface temperature reached decreases as the cutting speed (vs) increases.
Hamdi [21], shows in the case of 100Cr6 steel that the temperature goes from T=200°C for vs = 30m/s to T=100°C for vs = 120 m/s [21] and the maximum temperatures at the ground surface of the steel increase with the decrease of the linear speed table (vw). Stephenson [33] attributes the increase in surface temperature to the increase in wheel-to-work friction.
It is well established in the literature that material interactions in the grinding process lead to changes in the surface layer properties of the workpiece [3,5,28,34]. Such modifications cover the microgeometrical [6,35], microstructural [12,20,36] and mechanical [5,7,37] aspects. The magnitude and extent of these modifications depend on the thermophysical and microstructural characteristics of the material to be machined, the grinding wheel [6,7], the lubrication conditions [2,3] and the cutting conditions (aw, vw, vs) [38,39].
The Controlled grinding is governed by the temperature in the wheel-workpiece interaction.
Moreover, the surface temperature is the significant controller of the surface characteristic grinding.
In this study, we developed the combination effect of the abrasive type and the cooling mode of the tool steel AISI D2 (EN X160CrMoV12) on the temperature distribution and residual stresses. In this context, that a contribution is made through finite element modeling by using the ABAQUS code with FORTRAN programmed subroutines (dflux, film and dload). Therefore, we propose to evaluate the gradients of surface layer properties resulting from the modifications generated by the various material interactions -process (surface grinding) according to the wheel typecooling mode -cutting condition combinations described above. This evaluation was established using an experimental approach based on a methodology for identifying microgeometric, microstructural and mechanical characteristic profiles in the affected layers by combining various mechanical and physical investigation techniques.  (2) 6

 The initial and boundary conditions
The initial temperature is set through the structure at T0 = 20°C.
Δt: The time interval corresponding to a displacement of the flow from Δx.
This displacement is carried out digitally using a DFLUX Subroutine written in Fortran and called by Abaqus standard when the calculation is performed.

 Spatial discretization:
The mesh size of the area to be affected by the process is about 1 mm in thickness and has been refined to a minimum size of 94 µm to improve computational accuracy ( Figure 4). The geometry of the sample was discretized in space by 3400 elements of the CPE4RT type (4-node quadrilateral elements, in plane deformation, with reduced integration thermal coupling and bilinear displacement) adapted to this type of calculation and available in the element library of the calculation code used.

 Numerical simulation results
The results reveal characteristics such as surface temperature and the width of the affected areas. The characteristics are as shown in the figure. Surface temperature values between 20°C < T < 430°C (aw=15 µm) and 20°C < T < 817°C (aw=30 µm) are found.
The thicknesses of the thermally affected layers varied from (ea >50µm) to (ea=150µm). The distributions of residual stresses induced by grinding.

Effect of combination (A-C-Cd) on the temperature evolution on grinding surface:
numerical approach The thermomechanical interaction resulting from grinding simulation using the standard abaqus code corresponding to the combined effects of abrasive type, cooling mode and cutting depth (A-C-Cd) were reported in Table 5. The thermal fields, on the affected layers, resulting from different combinations revealed characteristics such as maximum grinding temperature and residual stress profile as shown in Figures 5-8.
The numerical result of température gradient evolution, obtained with the Abaqus standard code, are presented in Table 5. They are shown, the effect of depth of cut on the temperature of the grinding AISI D2 tool steel. It was revealed the evolution of temperature versus time at various surface locations as shown in Figure 5 Table 5. Large spectra of stress variations ranging from residual stress compression  R =-500MPa (aw=15 µm) to high tensile residual stress  R = 1150MPa (aw = 40µm) are shown. The widths of the layers under residual stress vary from (ec=50µm) to (ec=300µm).

Experimental validation
This part is aimed to validate, experimentally, the effects of combinations (abrasive type, cooling mode and cutting depth) on temperature and residual stresses in surface and sub-surface in plunge grinding. A good correlation is found between the simulated and measured profiles, as framed by their error bars. The validation of the temperature distributions was carried out by using the experimental database for the Al2O3 abrasive grain using all combinations in conventional grinding of AISI D2 steel.

Experemental approach : Temperature evolution
The quasi-stationary state is well reached upstream of the junction but the maximum temperature increases before the sensor and decreases afterwards. This is due to the very low thermal conductivity of the mica which constitutes a thermal barrier. As a result, the cooling curves are almost merged following the heat source flow.
The assembly of the calibrated RC filter circuit is prepared as a whole (Room, Oscilloscope, RC circuit) taking into account the assembly of the thermocouple for surface temperature measurement ( Figure 8, 9 and 10). Grinding experiments using the effect of depth of cut (aw=15µm and aw=50µm) on the température surface of the AISI D2 using the conventional cooling mode combined with Sol gel abrasive type (SG CC) with conventional conditions (vw=9m/min and vs=22m/s). 9 The validation of numerical residual stress gradient simulations is based on the results of X-ray diffraction measurements performed in this study for different combinations (A-C-Cc) (Figure 12).

Experimental approach : Residual stress evolution
A good correlation is found between the simulated and measured profiles, as framed by their error bars. This procedure allows the effects of the different combination elements (A-C-Cd) on the residual stress distributions induced by grinding ( Figure 12 a and b). The validation of the temperature distributions was carried out by using the experimental database for the Al2O3 abrasive grain using all combinations in conventional grinding of AISI D2 steel studied by S. Paul [4]. The numerical simulation data, obtained by the procedure developed in this study, are compared with S.
Paul's experimental results (Table 6). A satisfactory prediction is obtained within the experimental dispersion range provided by the author.

Discussion:
Analysis of the results obtained in this study highlight the effect of combinations (A-C-Cd): abrasive type -cooling mode -cutting depth, on the microgeometric, microstructural and mechanical characteristics of AISI D2 tool steel surfaces generated by the grinding process. The combinations studied involves the conventional grinding with the various modes of material tool interaction (cutting forces, friction and temperatures) and their consequences on roughness, surface softening of the microstructure, residual stresses and characteristics of the thermal crack network. The results of the numerical simulation showed that mechanical exchanges and in particular the contact pressure between grinding wheel and part, characterized by normal cutting forces, do not seem to have a significant effect on surface characteristics, due to their low intensity. Many studies related to the grinding of steels confirm this finding [6,21,46]. Therefore, the temperatures generated at the wheel / workpiece interaction is of increasing significance in the sense that they govern most of the changes in the surface layers, whether microstructural, microgeometric or mechanical.

Effects of the grinding temperature on the Microstructural surface characteristic
The superficial softening of the ground surface layers of AISI D2 tool steel shows a microstructural evolution phenomenon generated by the temperature increases at the grinding wheel -workpiece interaction independently of the combinations (A-C-Cc) applied ( figure 13). This result is in accordance with the literature and particularly with Murthey's [39] and Zhejun's [47] work on micro alloy steel 18MnNi2 and low-alloy steel 100Cr6 respectively. It can be used to translate the effect of combinations by the temperatures generated at the grinding wheel -workpiece interaction. This evolution can vary from the simple tempering of the basic martensite (a slightly decreased hardness) to a partial austenitization (a significantly decreased hardness). Surface austenitization, induced by grinding the martensite of AISI D2 tool steel, seems to occur at lower temperatures (around 800°C) than predicted by the equilibrium charts (around 1080°C -pseudo-binary diagram fe-C). This suggests that the high deformations accelerate the decomposition kinetics of martensite, which appears to occur at relatively lower temperatures [3]. This point deserves further investigation in relation to the microstructure of the white layer, if any, highlighted by Shaji [48], in the case of the grinding of low-alloy steels.

The effects of the grinding temperature on the Microgeometrical surface characteristic
The elementary mechanisms of material removal by single abrasive grains is based to referring this result. We can understand that the roughness, traces of large deformations of the surface layers, is essentially controlled by the contact pressure (at each grain) and by the behaviour of the material under the extreme loading conditions. This control is imposed by the process (speed, temperature and deformation mode) such as :  the contact pressure of each grain depends on the stability of its geometry during grinding (wear, flattening, regeneration and others).  Under soft grinding conditions aw < 40µm, the temperatures generated at the grinding wheel -workpiece interface remains less than or equal to 800°C, corresponding to a tempered to a tempered martensite but brittle ( Figure 14) which favours a chip formation mechanism by cleavage for both types of grinding wheels and both lubrication modes. As a result, the low effect of wheel type and lubrication mode on the arithmetic roughness Ra and total Rt. These evolve moderately as the running depth increases for all types of grinding wheel and lubricant as a result of the increase in temperature at the grinding wheel -workpiece interface and the contact pressure (Figure 14 (Figure 14 (a) and (b)). 11

The effects of the grinding temperature on the Mechanical surface characteristic
The effect of combinations (A-C-Cd) on the residual stress distributions of grinding can be understood as the predominant contribution of thermal stresses. These are all the more important as the higher temperatures reached at the grinding wheel -workpiece intersection ( Figure 15 (a) and   (b)). Therefore, the residual surface stresses remain compressive for all combinations (A-C-Cc) that generate temperatures at the grinding wheel -workpiece interface less than or equal to 350°C. These combinations correspond to the soft grinding conditions (aw<30µm) with a Sol-Gel wheel and cryogenic lubrication by higher temperatures there is a linear relationship between the residual tensile stresses and the temperatures generated at the wheel -part interface of the shape: Likewise, the layer thickness undergoing the remaining grinding stresses is quite comparable to the depth of the layers thermally affected by the grinding process, regardless of the nature of the combinations applied (Figure 15 (a) and (b)). These results are in line with the findings of the literature presented in [3,7] referring to the specific effects of the different types of wheels [47,49], lubrication modes [29,50] and cutting depth [5,27].

The effects of the grinding temperature on the surface integrities
The effect of combinations (A-C-Cd) on thermal crack distributions in the case of AISI D2 tool steel can be assessed by referring to the temperature generated at the grinding wheel -workpiece interface ( Figure 17 (a) and (b)). Indeed, the crack density length is all the more important as the temperature generated on the surface of the part is higher. These two values, which are characteristic of the integrity of ground steel surfaces AISI D2, can be written as a function of the temperature at the grinding wheel -workpiece interface for all combinations (A-C-Cd) in the following manner: These results reveal qualitatively and quantitatively the achievements of the literature presented in [3] regarding the effects of wheel type [47], lubrication mode [3] and cutting depth [4] on the grinding burn network.

The effects of the grinding temperature on productivity
The favorable combinations study of high productivity combined with better quality involves controlling the temperatures generated at the grinding wheel -workpiece interface. This combination is attributed to the low roughness and low level of residual tensile stresses and integrity, low density and length of thermal cracks of ground surfaces in AISI D2 tool steel ( Figure 18). This approach shows that steel grinding productivity doubles when changing from an Al2O3 wheel with soluble oil lubrication to a Sol-gel wheel and cryogenic lubrication, also.

Conclusion
The temperature generated at the grinding wheel -workpiece interface reveals the

Declarations
Ethics approval and consent to participate: Not applicable.

Consent for publication:
The copyright permissions have been taken and consent to submit has been received explicitly from all co-authors.

Competing interests:
The authors declare no competing interests.  Table 1 The constants of the Johnson & Cook equation of AISI D2 steel established by [43][44][45]. Table 2 Mechanical and thermo physical characteristics of AISI D2 tool steel [42,43].      Numerical approach of the predictive accuracy of surface grinding temperature and the residual stress.

Fig. 2
Typical experimental stress-strain curves obtained from the high strain rate tests: shear stress to plastic strain of AISI D2 tool steel as a function of initial test temperature [43]

Fig. 4
Spatial discretisation of the specimen geometry.       Wheel/piece contact time.     Numerical approach of the predictive accuracy of surface grinding temperature and the residual stress.

Figure 2
Typical experimental stress-strain curves obtained from the high strain rate tests: shear stress to plastic strain of AISI D2 tool steel as a function of initial test temperature [43] Figure 3 Grinding modelisation workpiece of AISI D2 tool steel. Spatial discretisation of the specimen geometry. Temperature gradient: effect of depth of cut on the AISI D2 tool steel (vw=0.15m/s and vs=22m/s. (a) aw=15µm and (b) aw=40µm.

Figure 11
Thermocouple test mounting and un ltered tension at the ends of the thermocouple. Wheel/piece contact time.

Figure 12
Effect of depth of cut on temperature evolution (vw = 0.15m/s -lc=2mm Al2O3 abrasive type and conventional cooling mode)    Effect of temperature on surface productivity of AISI D2 tool