Investigations of interfacial heat transfer efficiency in HFQ^® process of high strength aluminum alloy

Rolled AA7075-T6 sheet with 1.5 mm thickness was adopted in die quenching experiments and its nominal chemistry compositions are shown in table 1.

Figure 1 shows the schematic illustration of the experimental process for die quenching. At first, solution heat treatment (SHT) was conducted at 480 °C for 1 h in a furnace. Then the blank was transferred to the press machine to carry on die quenching. The die temperature and contact pressure are 25 °C, 100 °C, 200 °C, 300 °C, 400 °C and 5 MPa, 10 MPa, 20 MPa, respectively. The specimens for die quenching were prepared through cutting AA7075 alloy sheet into a disc with 100 mm diameter. The upper and lower cylindrical dies were made of H13 steel with a height of 40 mm and diameter of 100 mm, and the nominal chemistry composition was shown in table 2. A K-type thermocouple was inserted into the hole of the specimen center. Considering the symmetry, temperature collection was only performed for the lower die, and another two thermocouples were embedded into the holes at 1 mm and 2 mm below the lower die surface. Temperature change was recorded through the connecting thermocouple with a four-channel temperature collector (MIK-R9600), and the time-temperature data was used as output for computer analysis. Finally, peak-aged heat treatment was performed for AA7075.

Zoom In Zoom Out Reset image size

The microstructure evolution was characterized by transmission electron microscope (JEM2100-TEM), and the TEM samples were polished with a solution of 70% methanol and 30% nitric acid at −30 °C/12 V. Differential scanning calorimeter (DSC) with a 10 °C s⁻¹ heating rate was utilized to further study the phase transformation behavior of AA7075. Tensile tests were carried out at a strain rate of 10⁻³ s⁻¹ on a universal testing machine (SANSCMT 5000). Finite element simulation (FE) was performed by using ABAQUS software in order to verify the results of experiments. Blank and die were arranged as plastic body and discrete rigid body, respectively. Material properties, experimental parameters and calculated results were referred for model establishment and boundary condition setting. When holding the pressure, the 'surface-to-surface' contact method was utilized for blank and the die, and the 'surface film condition' and the 'surface radiation' contact method were set when the edge of the blank is in contact with the air. In the mean time, setting the temperature output variable to time points, the interval of each time point is 0.1 s, and a total of 150 time points are output.

During die quenching, heat radiation and heat convection are considered to contribute little to heat transfer, and the main mechanism of heat transfer is heat conduction [16]. Due to the cylindrical shape for specimens and forming die, the calculation of IHTC can be simplified into one-dimensional problems of heat transfer.

Biot number (B_i) can characterize the ratio of the conduction resistance to the transfer resistance, and it can be described as:

$$\begin{eqnarray}&&{{\rm{B}}}_{{\rm{i}}}=\displaystyle \frac{\delta /\lambda }{1/\alpha }=\displaystyle \frac{\alpha \delta }{\lambda }\end{eqnarray} \tag{ 1 }$$

where δ (0.75 mm) is the 1/2 thickness of the blank, α is interfacial heat transfer coefficient, and the value of which can be assumed to be 11000 W/(m² K) based on [17], λ is thermal conductivity, and the value is 183.49 (W mK⁻¹) for aluminum alloy. Finally B_i ≈ 0.045 < 0.1, indicating that there is no temperature gradient for the blank along the thickness direction. Therefore, 'lumped parameter method' can be selected to calculate the IHTC value, which can be defined as [18]:

$$\begin{eqnarray}&&{h}_{w}=\displaystyle \frac{q}{{T}_{a}-{T}_{b}}\end{eqnarray} \tag{ 2 }$$

Where h_w is the interface heat transfer coefficient, q is the heat flux density between the surface of the aluminum alloy sheet and H13 steel. T_a and T_b represent the surface temperature of the AA7075 and H13 steel, respectively. T_a can be obtained from the Time-Temperature curves (figure 2). Nevertheless, it is difficult to measure the surface temperature of H13 die directly. The heat flux density can be obtained by the temperature field of the die, thus the surface temperature of the die can be calculated, being expressed as:

$$\begin{eqnarray}&&q={C}_{a}{\rho }_{a}\nu /S(d{T}_{a}/{d}_{t})\end{eqnarray} \tag{ 3 }$$

Where C_a is the specific heat capacity of AA7075, ρ_a is the density of AA7075, S represents the contact area between the blank and die, and V is the volume of the blank. The value of all the thermal physical parameters related to AA7075 and H13 steel are listed in table 3.

Zoom In Zoom Out Reset image size

With regard to the lower die, the first order partial derivative of temperature to time is applied to the forward difference method, which can be expressed as:

$$\begin{eqnarray}&&\displaystyle \frac{\partial }{{\partial }_{{\rm{t}}}}T(x,t)=\displaystyle \frac{{T}_{{\rm{i}}}(t+{\rm{\Delta }}t)-{T}_{{\rm{i}}}(t)}{{\rm{\Delta }}t}\end{eqnarray} \tag{ 4 }$$

Where T(x, t) represents the temperature in a distance of x mm to the surface of the lower die at t time. Δt represents the time increment.

The second order partial derivative method can be expressed as the central difference method:

$$\begin{eqnarray}&&\displaystyle \frac{{\partial }^{2}}{\partial {x}^{2}}T(x,t)=\displaystyle \frac{{T}_{i-1}-2{T}_{i}(t)+{T}_{i+1}(t)}{{({\rm{\Delta }}x)}^{2}}\end{eqnarray} \tag{ 5 }$$

Substituting equations (4) and (5) into the heat conduction equation of the lower die (equation (3)), and equation (6) was obtained as followed:

$$\begin{eqnarray}&&{\rm{C}}\rho \displaystyle \frac{\partial }{{\partial }_{t}}T(x,t)=\lambda \displaystyle \frac{{\partial }^{2}}{\partial {x}^{2}}T(x,t)\end{eqnarray} \tag{ 6 }$$

For simplicity, T_i−1, T_i and T_i+1 are replaced by T_b, T₁ and T₂. T₁ and T₂ are the temperature 1 mm and 2 mm from the lower die surface. The die surface temperature equation can be obtained as followed:

$$\begin{eqnarray}&&{T}_{b}(t)=2{T}_{1}-{T}_{2}(t)+\displaystyle \frac{C\rho ({\rm{\Delta }}{x}^{2})}{\lambda {\rm{\Delta }}t}[{T}_{1}(t+{\rm{\Delta }}t)-{T}_{1}(t)]\end{eqnarray} \tag{ 7 }$$

Then the equation for calculating IHTC value between AA7075 and H13 steel can be obtained through combining equations (2), (3) and (7):

$$\begin{eqnarray}&&{h}_{w}=\displaystyle \frac{{C}_{a}{\rho }_{a}}{S\left(\displaystyle \frac{d{T}_{a}}{{d}_{t}}\right)}/({T}_{a}-{T}_{b})\end{eqnarray} \tag{ 8 }$$

Finally, the average IHTC value was described as:

$$\begin{eqnarray}&&\overline{{h}_{w}}=\displaystyle \sum {h}_{w}/n\end{eqnarray} \tag{ 9 }$$

Where h_w is the IHTC value corresponding to any time, n is the number of collected samples.

Figure 2 shows the comparison of temperature evolution with time between the experimental and finite element simulation results under different conditions. It can be observed that the simulation agrees well with the experimental results. In the condition of 25 °C die temperature, the cooling rate of the blank increases with the increasing contacting pressure. The slope of the curve gradually decreases, which is attributed to the decreasing temperature difference. In addition to that, die-1 and die-2 represent the temperature at 1 mm and 2 mm below the lower die surface, respectively, and both of which increased to the peak value firstly, then decreased slightly to a steady value (figure 2(a)). As can be seen in figure 2(b), at a contact pressure of 20 MPa, as the die temperature increases, the cooling rate gradually decreases.

Figure 3 shows the simulation temperature distribution in the cross-section of the blank in three cases. It can be seen that, under all conditions, the temperature in the center of the blank is the highest, and the temperature gradually decreases from the center to the edge, showing a gradient distribution. Because a bit heat radiation and heat convection occurs between the blank and air during die quenching process besides the heat conduction, the closer to the edge, the larger the volume of the blank in contact with air, the more obvious the temperature difference is. However, the temperature difference between the center and the edge position is not large, and the ΔT_max was calculated for three cases. When the die quenching time is 1 s, 5 s and 15 s at a contact pressure of 5 MPa and 25 °C die temperature, the ΔT_max is 17 °C, 9 °C and 4 °C, respectively (figure 3(a)). The similar ΔT_max values were obtained under the other two conditions (figures 3(b) and (c)). Therefore, the temperature distribution in the radius direction can be considered as relatively uniform, indicating that the effect of heat radiation and heat convection is much weaker when compared with heat conduction during die quenching. Apart from that, the assumption of one-dimensional heat conduction model related to the calculation of IHTC value was also verified.

Zoom In Zoom Out Reset image size

Table 4 summarized the calculated value of average IHTC for AA7075. It can be seen that the increasing contact pressure and the decreasing die temperature are both beneficial to the increment of IHTC value. The maximal IHTC was achieved under 20 MPa contact pressure and 25 °C die temperature, and the value is 3560 W·m² k. In contrast, the minimum IHTC value with 1563 W·m² k appears at 5 MPa contact pressure and 400 °C die temperature.

Figure 4 shows microstructures in specimens conditioned by different die temperature and contact pressure. It can be observed that a number of fine η' phases are uniformly distributed on the matrix at 5 MPa contact pressure. Apart from that, only a small amount of η phase can be discovered (figure 4(a)). With the increasing contact pressure (10 MPa and 20 MPa), the microstructure change is not obvious (figures 4(b) and (c)). The TEM morphologies of the specimens under 200 °C die and 10 MPa/20 MPa contact pressure are similar with those under 25 °C and 5–20 MPa conditions (figures 4(d) and (e)). Nevertheless, when the contact pressure decreases to 5 MPa, the fine η' phase is almost invisible, and replaced by a large number of large η phases (figure 4(f)). Figures 4(g)–(i) shows TEM images at a contact pressure of 20 MPa and a die temperature of 100 °C, 300 °C and 400 °C, respectively. Combining figures 4(c) and (f), it can be found that, a large amount of η appear when the die temperature reaches 300 °C under condition of 20 MPa (figure 4(h)), and this phenomenon developed further under 400 °C die temperature (figure 4(i)).

Zoom In Zoom Out Reset image size

Figure 5 shows the differential scanning calorimetry (DSC) curves of AA7075 under different die temperatures at a contact pressure of 20 MPa. The peaks marked by yellow dash lines indicated the transformation from η' to η, and the area is proportional to the volume fraction of η [20]. Therefore, the area of exothermic peak enables to be utilized to make a qualitative analysis of the volume fraction of precipitates in all cases. It is thus evident that, the area of the samples under 300 °C/400 °C die are larger than those ranging from 25 °C–200 °C, indicating the appearance of more η precipitates under the former condition. Simultaneously, the temperature of exothermic peak moves left with the increasing die temperature, demonstrating the η precipitates are easier to form under a higher die temperature. The above DSC results are well consistent with TEM observation.

Zoom In Zoom Out Reset image size

Figure 6 shows the mechanical properties in specimens conditioned by different die temperature and contact pressure. It can be seen from figure 6(a) that, the ultimate tensile strength (UTS) and yield strength (YS) decreased with the increasing die temperature under 5 MPa contact pressure, and the significant reduction in strength occurs when the die temperature is 200 °C. Apart from that, the total elongation (TE) increased slightly. With regard to 10 MPa and 20 MPa conditions (figures 6(b) and (c)), the trends related to the strength with the increasing die temperature keep the same as the case of 5 MPa contact pressure. With the increasing contact pressure at the same die temperature, the UTS and YS of the specimens increase. The evident difference is the significant decrease in strength at 300 °C die temperature.

Zoom In Zoom Out Reset image size

Heat conduction is the main heat transfer mechanism during die quenching, and in this process, contact pressure play an important role on interfacial heat transfer efficiency. As shown in the previous Time-Temperature curves (figure 2(a)), the cooling rate increases with the increasing contact pressure, which is due to the increased IHTC value, which is related to the increasing contact area between the blank and die. Figure 7 shows the contact schematic diagram of the blank and the die surface. When the blank is in contact with the die, micro-convexes on the both surfaces will collide each other. Because the hardness of the 7000 series aluminum alloy is much softer than that of H13 die steel [13], the micro-convexes of AA7075 can be crashed by those of H13 die steel, inducing a reduction of the micro-convexes height. Then the micro-convexes of the H13 die steel embed into those of AA7075. It can be inferred that, with the increasing contact pressure, the depth of H13 micro-convexes inserting into those of AA7075 gradually increases, leading to an increased real contact area, finally resulting in an increased IHTC value. When warm dies were applied, the IHTC value decreased from 3112 W·m² k to 1563 W·m² k with the increasing die temperature ranging from 25 °C to 400 °C, as shown in table 4. This is because the blank temperature prepared for die quenching is 430 °C, the increasing die temperature leading to a decrease in interface temperature gradient, finally resulting the decrease of the cooling rate (figure 2(b)) and interfacial heat transfer efficiency.

Zoom In Zoom Out Reset image size

In order to have a deep understanding of the correlation between interficial heat transfer efficiency and microstructures, a schematics (figure 8) was constructed based on the previous experimental results, where TTP curves of AA7075 in T6 temper was also included by using the data in [9]. The TTP curves demonstrated the sensitive temperature range of AA7075 is 185 °C–415 °C, in which the phase transformation occurs more easily. Because when the holding temperature is higher than 415 °C, the driving force became small due to the low supersaturation, inducing the reduced nucleation rate of precipitates, resulting in a low rate of phase transformation. In the case of lower than 185 °C, the similar result was obtained due to the low solute diffusion rate. Only in the temperature range of 185 °C–415 °C, the solute diffusion rate and the supersaturation can satisfy the rapid transformation conditions of the precipitated phase.

Zoom In Zoom Out Reset image size

Therefore in the present study, microstructure is strongly affected by interfacial heat transfer efficiency in the sensitive temperature range. Table 5 summarized the average cooling rate under all conditions in this range. Cooling rate is proportional to the interfacial heat transfer efficiency. Q_C in figure 8 is defined as the critical quenching rate, inducing the massive phase transformation. Q_a and Q_b represent the quenching rate under 10 MPa/200 °C and 5 MPa/200 °C conditions, respectively, as marked by the dash line circles. Combined the evolution of microstructure with IHTC value for analysis, it can be inferred that, the value of Q_C is between 14.3 °C s⁻¹ (Q_a) and 18.5 °C s⁻¹ (Q_b), and η' and η dominate area A (Q > Q_C) and B (Q < Q_C) respectively. The reason is that, when the quenching rate is less than Q_C, the solid solution decomposes and the equilibrium phase η would precipitate in grains. The formation of large number of η enables to cause the loss of solute atoms, such as Mg and Zn atoms, finally resulting in a low volume fraction of η'.

η' is the main strengthening phase for AA7075, and η nearly has no strengthening effect [21, 22]. The strengthening model of the precipitation can be described as [23]:

$$\begin{eqnarray}&&{\sigma }_{p}=\displaystyle \frac{M}{b}\sqrt{\displaystyle \frac{3f}{2\pi }}(2\beta G{b}^{2})\displaystyle \frac{1}{r}\end{eqnarray} \tag{ 10 }$$

where b denotes the Burgers vector, and the value of which is 2.84 Å, G denotes the shear modulus (27 GPa for aluminum); f denotes the volume fraction of η', r denotes the size of η'. Based the equation (10), it can be speculated that, larger f and smaller r are beneficial for the improvement of the strength. As evidenced in TEM observation (figure 4) and mechanical properties (figure 6), the specimens with high fraction volume of η' induced by high interfacial heat transfer efficiency exhibited higher strength than those with low fraction volume of η'.