When high thermal conductivity particles are added to PCM(phase change materials) to enhance heat transfer, it is referred to as particle-enhanced PCM. After adding high thermal conductivity particles of different particle sizes, if there is no chemical reaction between the particles and the PCM, the microstructure of the PCM will not change. In order to simplify the calculation in the molecular dynamics simulation process, in this paper, aluminum nanoparticles are used as high thermal conductivity particles, and linear alkanes are used to represent paraffin, and the mixed PCM models of n-nonadecane and aluminum nanoparticles are established respectively. Firstly, an amorphous structure model of n-nonadecane was established in a cube, and the energy was minimized by the Smart Minimizer method, and equilibrium relaxation was performed at normal temperature and pressure. After the system reached equilibrium, aluminum nanoparticles with diameters of 1 nm, 2 nm, 3 nm and 4 nm were added to construct four different PCM systems composed of n-nonadecane and aluminum nanoparticles. Each mixed PCM system contains 80 n-nonadecane molecular chains. Due to the increase in the particle size of aluminum nanoparticles in the system, the corresponding number of aluminum atoms in the four mixed PCM systems are 43, 249, 887, and 1985, respectively, and the total number of atoms corresponding to each system is 4763, 4969, 5607 and 6705, respectively. The initial molecular velocity is still sampled according to the Maxwell distribution, and is solved according to the Verlet velocity algorithm. The Atom and Ewald methods were used for van der Waals and electrostatic interactions, the COMPASS force field was used for the force field, the periodic boundary conditions were used for the boundary conditions, and the Nosé-Hoover A and Berendsen methods were used to control temperature and pressure, respectively.
Before analyzing the n-nonadecane and aluminum nanoparticle hybrid PCM systems, energy minimization was performed for each system separately, and relaxation was performed for 1000 ps in the NPT ensemble, and then for another 500 ps under the NVT ensemble. The energy changes of one of the hybrid PCM systems during energy minimization and equilibrium relaxation are shown in Figure 1. With the increase of simulation time, the energy of the system basically fluctuates up and down in a constant range. At this time, it can be considered that the PCM systems have tended to balance.
Figure 1 - Energy changes during energy minimization and equilibrium relaxation in a hybrid PCM system
After each mixed PCM system of n-nonadecane and aluminum nanoparticles reached equilibrium, molecular dynamics simulations of 1000 ps were performed at temperatures of 283 K and 353 K, respectively, and the simulations were performed in the NPT ensemble. The structural parameters and density of the system are shown in Table 1. The temperature environment in which the simulation is performed is the temperature before and after the phase transition of n-nonadecane. It can be seen from the density change that after the phase transition of n-nonadecane, the density of the system decreases, which is consistent with the macroscopic property that the density of liquid paraffin is smaller than that of solid paraffin.
Table 1 - Simulation parameters and densities of hybrid PCM systems
Figure 2 shows the mean square displacement (MSD) and self-diffusion coefficient of the n-nonadecane/aluminum nanoparticle hybrid PCM system at simulated temperatures of 283 K and 353 K, respectively. When the particle sizes of the added aluminum nanoparticles are 1nm, 2nm, 3nm and 4nm, the corresponding self-diffusion coefficients of the mixed PCM system at 283K are 4.743×10-11m2·s-1 and 1.993×10-11m2·s-1, 0.832×10-11m2·s-1 and 0.307×10-11m2·s-1, respectively. When the temperature is 353K, the corresponding self-diffusion coefficients of the mixed PCM system are 23.870×10-11m2·s-1, 11.708×10-11m2·s-1, 2.448×10-11m2·s-1 and 1.153×10-11m2·s-1. As the temperature increases, the self-diffusion coefficient also increases. The self-diffusion coefficient of the hybrid PCM system decreases with the increase of the particle size of Al nanoparticles, but when the particle size of aluminum nanoparticles increases to a certain extent, the self-diffusion coefficient of the mixed PCM system is basically the same before and after the melting of n-nonadecane. The reason is that the content of alkanes in the mixed PCM system is constant, and with the increase of the particle size of aluminum nanoparticles, the mass fraction and volume fraction of aluminum in the mixture gradually increase. After the particle size of aluminum particles increases to a certain value, the phenomenon of precipitation is easy to occur due to the different density of aluminum and alkanes and the mass effect. In addition, the increase of the mass fraction and volume fraction of the high thermal conductivity particles will also lead to their excessive filling, resulting in the weakening of the fluidity of the PCM substrate in the system. At the same time, limited by the total latent heat of the system, high thermal conductivity particles should not be added too much as a heat transfer enhancement material. When the high thermal conductivity particles are filled too much, the volume ratio is too large, and when the PCM is mixed with it in the molten state, due to the weak fluidity of the PCM, the mixing is likely to be uneven or even impossible to mix, or the mixture will appear loose and porous, which is not conducive to heat transfer. Therefore, when the heat transfer of paraffin-based PCM is enhanced by adding high thermal conductivity particles, it is not that the larger the particle size, the better.
Figure 2 - MSD and self-diffusion coefficient of n-nonadecane/aluminum nanoparticles hybrid PCM system