Cooling rate determines the supersaturation of solid solution of carbon in metal. The higher cooling rate the higher growth rate. pls see details in my papers at RG.
Dear Amin, please first check the reproducibility of your experiments and corresponding Raman measurements. Normally, I can not see any reasons for change of CNT crystallinity in dependence on cooling rate from the usual temperature of CVD (500-1000 C). Apparently, such effects could be observed only at significantly higher temperatures (>1500 C), at which carbon could change its crystallinity. Otherwise, contact of sample with air at 400 C could also lead to burning some contaminations leading to removal of nanostructures with low crystallinity and survival of structures with high one.
Actually i tried to characterize the sample by FESEM and also I found some improvements in the uniformity as well.I thought that removing the sample at high temperature can be identify one of the parameters to control the activity of the vaporized carbon nanotubes. It means that we can control the deposition of CNT by removing the sample in high temperature while annealing because the deposition is still continue in the annealing process,
FESEM is not a suitable method for checking crystallinity of the sample because it gives (1) only results of local investigations and (2) could not give insight into the arrangement of graphene layers inside of CNTs. Furthermore, it is not usual to perform CVD at low temperature (400 C). As I know, this temperature is not enough for the thermal CVD. Therefore, I can not comment your idea.