Forming micro-, nano- and macrocapsules from PCM(phase change materials) substrates with polymer materials, etc. is also a method for PCM to enhance heat transfer, it can also avoid the leakage of PCM in the solid-liquid phase transition process, and has broad application prospects in the fields of heat utilization. Taking n-octadecane as the core material and silica as the shell material, a model of n-octadecane/silica capsule PCM was established and simulated by molecular dynamics.
In the model of n-octadecane/silica capsule PCM, the inner diameter of the capsule is 4 nm, the outer diameter is 5 nm, and the interior contains 31 molecular chains of n-octadecane. The capsule shell consists of 716 silica molecules, and the total number of atoms in a capsule PCM system is 3884. In order to study the effect of the shell on the diffusion performance of the system, the capsule shells were set to two types: free and constrained, representing the soft shell and the hard soft shell, respectively, two capsule PCM systems were constructed and calculated separately. After the initial configuration of each system is constructed, the energy minimization (smart minimizer) and the annealing treatment for one cycle (280~380K) are carried out respectively, and followed the 100ps kinetic process at room temperature and pressure to perform equilibrium relaxation on the configuration of the system. After the systems were equilibrated, molecular dynamics simulations were performed for the two capsule PCM systems to be heated from 283K to 353K, respectively. The configuration at the end of each temperature was used as the initial configuration for the next temperature calculation, and the temperature gradient was 10K. The simulation time is 1000ps at each temperature with a time step of 1fs.
In order to verify the validity of the simulation time, the slope of the MSD curve of the capsule PCM system at room temperature and pressure was calculated first, which was recorded as kMSD, and the simulation time was increased from 200ps and 300ps to 1000ps. As the simulation time increases, the kMSD values of the soft-shell and hard-soft-shell capsule PCM systems tend to be constant. Therefore, it can be considered that the simulation time is 1000 ps, which is sufficient for the simulation of both systems.
When the simulated temperature was changed from 283K to 353K, the MSD curves of the two capsule PCM systems are shown in Figure 1. In order to reduce the influence of the MSD value of the first 200ps and the last 200ps on the self-diffusion coefficient, the MSD of 0~1000ps, 200~800ps and 200~1000ps were intercepted to calculate the self-diffusion coefficient.
Figure 1 - MSD curve of capsule PCM system
The self-diffusion coefficients of the two capsule PCM systems at different temperatures are shown in Figure 2. The self-diffusion coefficient of the system is also different when the time period of the intercepted MSD value is different, but the general trend remains basically unchanged. With the increase of temperature, the increasing trend of self-diffusion coefficient of each system does not change. However, when the temperature is the same, the self-diffusion coefficient when the capsule shell is a soft material is larger than that when the capsule shell is a hard material.
Figure 2 - Self-diffusion coefficients of two capsule PCM systems at different temperatures
The radial distribution functions of the two capsule PCM systems at 353K are shown in Figure 4. When r is 0.31~0.50nm, 0.54~0.68nm, 0.74~0.90nm, the RDF value of soft shell capsules is higher than that of hard shell capsules. This means that the atoms of the soft-shell capsules have a higher probability of appearing within these distance intervals.
According to the changes of the self-diffusion coefficient and radial distribution function of the two capsule PCM systems, it can be concluded that when the capsule shell is composed of soft materials, the self-diffusion performance of the capsule system is obviously better, which is more conducive to the transfer of energy, and from the perspective of macroscopic behavior, it is more conducive to the transfer of heat. The following further analysis is carried out through the change of the terminal distance of the alkane molecular chain.
Before the calculation, a certain molecular chain of the core material n-octadecane was calibrated, so as to record the change of the terminal distance of the molecular chain during the simulation process. Because the molecule can rotate freely around the carbon-carbon single bond, long-chain alkanes can show various conformations, and the molecular block conformation composed of four adjacent carbon atoms is most common in trans and cross, when each molecular segment exhibits a trans conformation (all-trans), the end-to-end distance of the alkane molecule is the largest. When the capsule shell is a soft material, the end-to-end distance of n-octadecane is mainly distributed around 19.5Å. At this time, the molecular chain of n-octadecane mainly exhibits all-trans conformation; when the capsule shell is a hard material, the terminal distance of the n-octadecane molecular chain shrinks from the main distribution around 18.9A at 283K to around 1.83Å at 353K. Since the phase transition process of alkanes is accompanied by volume changes, and the volume becomes larger after melting, when the capsule shell is a soft material, the twist and expansion of the molecular chain of the core material alkane is not affected, while when the capsule shell is a hard material, the core material alkane is bound in torsion and expansion due to the limitation of the shell material, and the hard shell will increase the gap between different capsules and increase the contact thermal resistance between different capsules. The above conclusions show that the soft shell is more conducive to the heat transfer of the capsule PCM.