Th' Gaussling posted a question a few weeks about about polymer compatibility. Being that Th' Gaussling and I have exchanged emails in the past, I shot him back (or so I thought) a quick reply. Not hearing a simple "thanks" is uncharacteristic of the fellow, so after additional follow-up mails were unanswered, I became suspicious. It now appears that my emails have never arrived. That is certainly understandable as our IT department loves to play tricks on us (replacing networks servers over the weekend without prior notification AND supplying the new server with a new name so that all of our old shortcuts need to be edited...) so the fault is clearly on this end. Since most of what I attempted to send to Th' Gaussling is suitable for public viewing, here is what I attempted to communicate earlier.
This is one of areas of polymer science that is not as well known as it should be. There has been quite a bit of work done in the area, but mostly by paint chemists so it just isn’t very broadly known. As you would expect, it is a horribly complex area, full of endless exceptions-to-the-rules, so here are what I see as general guidelines.
The concerns you have all closely related: as a polymer dissolves, swelling increases, migration increases… Avoid one, and you avoid them all.
Most of the focus is on polarity of the polymer and the solvent, but it’s more complicated than in organic chemistry. Start with the Hildebrand cohesive energy density which is the heat of evaporation divided by the molar volume. The square root of this is the solubility parameter. You get weird units, such as (cal/cc)1/2. This number will range from about 6 to 24 or so, with the higher numbers indicating greater polarity. You can run into problems if the solubility parameter of the polymer and the chemical are too close, as they are then too compatible.
It has been found that this one-dimensional approach falls short in many cases, and that a two-dimensional approach is needed. Introducing the hydrogen bonding parameter, which is measured by looking at shifts in the IR spectra following deuteration.Using two parameters allows you to differentiate between chlorinated solvents for example, which have high polarity but negligible hydrogen bonding.
There are a couple of other bonding parameters as well to carry this to further dimensions, but they are much more difficult to measure and I’ve never found that they greatly increase the usefulness.
Working with these two parameters, people have prepared 2-D maps showing the regions where solubility occurs. The regions are big amorphous blobs often with one or more arms. Like something from a low budget sci-fi movie. (“The Blob that Ate Boise.”) Given this, you can then imagine situations where a polymer dissolves in some solvents with a given solubility parameter, but not all of them, since the hydrogen bonding also need to be given heed.
A big wrench in the workings is crystallinity of the polymers. Based on the arguments above, xylene is a good solvent for polyethylene, but for it to solvate, you need to more or less melt the resin. The thermodynamic well for the crystallization is that deep. Too bad, as a RT solvent for PE would be a godsend.
So that’s it in a nutshell. Going forward, there are a number of options for getting more detailed information. Alan F. M Barton has a couple of authoritative books in the area (“CRC Handbook of Solubility Parameters and Other Cohesion Parameters”, and “CRC Handbook of Polymer-Liquid Interaction Parameters and Solubility Parameters) that have tabulations of the parameters and some solubility maps. Unfortunately, they are published by CRC, which to me has always meant “Costly, Really Costly”. The coverage is not as broad as I would like, but measurement and tabulation of basic physical properties has never been sexy. Google Books also has some sections online too. Lastly, now that you know the terminology, you might be able to locate appropriate journal articles as well.