A study of p-xylylene polymerization kinetics using in situ vacuum differential scanning calorimetry

摘要

Vapor deposition polymerization (VDP) of p-xylylenes, also known as parylenes, is a unique vacuum deposition process enabling preparation of thin uniform pinhole-free coatings of controlled thickness in the range from several nanometers to several hundreds of micrometers, which is carried out without solvents or catalysts. This method is also promising for preparation of thin hybrid polymer-metal (semiconductor) nanocomposite films with a high filler content (up to several tens of volume percent) and homogeneous distribution of inorganic nanoparticles in polymer matrix. We developed a differential scanning heat flux calorimeter integrated into a VDP deposition chamber as well as its control electronics and software for a study of p-xylylene polymerization kinetics in condensed state. The temperature and peak area calibrations were performed over a wide temperature range with indium and mercury standards. Time constants and apparatus function (Green's function) of the calorimeter were estimated, and desmearing procedure of DSC signal employing Bayesian deconvolution (Richardson-Lucy method) was implemented. The temperature range (-110 - -80℃) and the total heat effect (Q = 86 ± 6 kJ/mol) of polymerization reaction were found to be identical with the results measured ex situ with a commercial calorimeter Perkin Elmer DSC7 [2]. The desmeared DSC curves were treated employing the differential isoconversional method by Friedman in order to determine the variation of effective activation energy (E_a) with the degree of conversion and therefore to gain insights in the p-xylylene polymerization mechanism in condensed phase. The initial part (? < 0.2) of the E_a vs. ? dependence reveals a decrease of E_a from about 50 kJ/mol to 40 kJ/mol probably indicating a larger contribution of the initiation step at low conversion degrees. In the range from ? > 0.2 to ? < 0.7 E_a = 40 ± 2 kJ/mol which is practically identical with activation energy of the chain propagation reaction E_a = 36 kJ/mol reported for polymerization of p-xylylene in low-temperature hexane and toluene solutions [1]. At ? > 0.7 E_a decreases from 40 to 25 kJ/mol revealing a transition from kinetic to diffusion regime of polymerization in viscous reaction system. We suppose that some discrepancy between E_a vs. ? dependence at high ? calculated from the data measured by in situ DSC and the same dependence reported in [2] can be explained by effect of transfer of samples from vacuum deposition chamber to a DSC cell, resulting in initiation of additional radicals frozen in monomer matrix, faster polymerization at the initial stage and vitrification of the reaction system at lower conversion degrees. To estimate reaction model f(?) we have used a model-free method proposed by Sbirrazzuoli [3] based on the use of the compensation effect. The calculated f(?) values are adequately fitted to the Avrami-Erofeev A2 model in the conversion degree interval of 20-70%. The discrepancy observed at the end of the reaction is probably due to diffusion control. The work was supported by Russian Foundation for Basic Research (grant no. 15-03-08502) and the Council of the President of the Russian Federation for Support of Young Scientists (grant no. MK-5590.2016.3).