The equilibrium between evaporation and condensation - the vapor pressure of p-xlylene and its temperature dependence are of vital interest in attempting to understand the Parylene VDP process.  Monomer is delivered to the growing coating via condensation.

Measuring vapor pressure by conventional methods of course is impossible, due to its chemical reactivity.  Accepting the hypothesis that p-xylylene (C₈H₈) is unique only in its chemical reactivity, we expect that its vapor pressure might be close to a compound similar in size, weight, shape and polarity, such as p-xylene (C₈H₁₀), differing only by two hydrogens.  The vapor pressure data for p-xylene in today's handbooks originated in a very extensive 1947 compilation due to Stull.   In that pre-computer era, the tools used for evaluating existing data and interpolation were large sheets of graph paper, thumb tacks and string.  Hence we can be assured there are no computational biases in the numbers, only that which might exist in the personal judgment of the evaluators.

An estimate of the vapor pressure of p-xylylene is reported by Ganguli, using the Lee-Kesler relation and the Ambrose group contribution method for estimating the critical parameters:

Here, T_r  is reduced temperature, equal to T(K)/626.  This is a good match for the p-xylene handbook data, as seen in the plot below. Unfortunately, he gives no details on the estimation process, and my attempts to reproduce the estimate have failed to give as good a fit.  

The T ⁶ term in Ganguli's estimation contributes little at pressures below one atmosphere, and our interests are pressures below about 1 torr.  A least squares fit to the handbook data on p-xylene without the T⁶ term gives the following result:

As can be seen in the plot, it gives a somewhat better fit to the data points at the low pressure end.

One problem with the handbook data for p-xylene is that p-xylene melts at 13°C, and the lowest temperature point (-8.1°C) alone should be on the solid vapor pressure curve, which would be steeper by the heat of fusion, 16.81 kJ/mole, meeting the liquid curve at 13°C.  Instead, the data seem to line up on one continuous line.  It seems likely that this -8.1°C point is the vapor pressure of the supercooled liquid rather than the sublimation pressure of crystalline p-xylene.  Unfortunately, Stull is silent on this question.

The reason for the slight bending of the vapor pressure curve is that the heat of vaporization is a weak function of temperature, growing a little larger at lower temperatures. Here is the temperature dependence of the heat of vaporization in the temperature range of interest in VDP, obtained by differentiation of the least squares fit to the handbook data for p-xylene.

BACK