Does (borosilicate) glass adsorb methanol to an exceptional extent? We are doing vapor density measurements of organic solvents using a microbalance and a sealed glass sphere. Other polar molecules such as acetone present no problems, but we are getting anomalous results for methanol for some reason. [Alternatively, does it absorb less than other solvents?]

  • $\begingroup$ What does 'exceptional' mean in this case? $\endgroup$ – Lighthart May 10 '16 at 13:31
  • $\begingroup$ @Lighthart Compared to other alcohols and ketones $\endgroup$ – Dirk Bruere May 10 '16 at 13:38
  • $\begingroup$ Again, this cannot really be defined until you dictate what you mean by 'exceptional'. I understand you are comparing it to other molecules, but how? $\endgroup$ – Lighthart May 10 '16 at 14:40
  • $\begingroup$ @Lighthart I want to know whether more molecules of methanol will stick to a bit of glass at a given temperature than (say) ethanol or acetone. Not sure how much simpler I can describe it. $\endgroup$ – Dirk Bruere May 10 '16 at 15:06

Exceptional is subjective term, and can't be responded to sensibly. But more/less can be discussed. This is somewhat complex. The isotherm is an equilibrium between adsorption and desorption.

Methanol is small, and when adsorbing to glass, presumably through hydrogen bonding, a single molecule will do very little to block other adsorption sites. Other, larger, molecules might 'lay down' on the surface or have branched structures that may inhibit their peers from adsorbing, giving smaller surface coverage. However, long linear molecules will tend to orient, and there is an energetic favorability to forming a monolayer for long linear alcohols (and thiols, carboxylates, amines, silanes, etc).

In these highly ordered linear systems there is an attractive force of roughly 0.2 kcal/mol per methylene unit. Since methanol has the fewest number possible, if a methanol molecule desorbs it can (almost) freely leave the molecular layer, as opposed to longer chain molecules. When you begin to discuss molecules that are more topologically diverse than simple linear chains, this attraction is significantly weakened, and as such those molecules may also (almost) freely desorb.

In essence the methanol isotherm has maximum kinetic lability, and as such the coverage will happen reasonably quickly, and be fairly complete. At equilibrium I would expect methanol to have apparently higher coverage than any branched molecule, and before equilibrium, I would expect methanol to have more coverage than long-chain analogs.

Please note this is on a per molecular basis, and the mass equivalents of methanol will be smaller than any other molecule to be compared with (save water).

| improve this answer | |
  • $\begingroup$ Thanks - sounds plausible. I will pass it on to the PhD student who discovered the anomalous results we got and see whether it might be testable. $\endgroup$ – Dirk Bruere May 10 '16 at 19:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.