November 30, 2021, 07:55:59 PM
Forum Rules: Read This Before Posting


Topic: Quantification of Dimethicone  (Read 1007 times)

0 Members and 1 Guest are viewing this topic.

Offline Thurbs

  • New Member
  • **
  • Posts: 3
  • Mole Snacks: +0/-0
Quantification of Dimethicone
« on: August 25, 2021, 11:05:01 AM »
Management has given me the task of developing a method to quantify dimethicone in in one of our new lip balm formulations, and I think I need a little help with it. The formula is 98.5% wax/oil and 1.5% dimethicone. My lab is pretty basic, but we have an FTIR and a range of solvents.

What I understand so far about dimethicone is that there are no chromophores, so my UV/vis HPLC is of no use. It is not volatile, so a GC would be no use. My FTIR software allows you to subtract one spectra from another, however, at 1.5%, the peak remaining at 1260cm-1 is very small when you subtract the placebo spectra from the full formula. 

Does anybody have any advice regarding a method of extracting dimethicone from an oil base?

Offline Corribus

  • Chemist
  • Sr. Member
  • *
  • Posts: 3221
  • Mole Snacks: +487/-22
  • Gender: Male
  • A lover of spectroscopy and chocolate.
Re: Quantification of Dimethicone
« Reply #1 on: August 25, 2021, 01:16:09 PM »
FTIR isn't the greatest method for quantitative analysis to begin with, and it will be especially hard here because dimethicone is a polymer with variable molecular weight. I wonder if some kind of separation and/or gravimetric analysis might be a viable approach. Is the rest petroleum wax? Waxes may be separable from the siloxanes.
What men are poets who can speak of Jupiter if he were like a man, but if he is an immense spinning sphere of methane and ammonia must be silent?  - Richard P. Feynman

Offline Thurbs

  • New Member
  • **
  • Posts: 3
  • Mole Snacks: +0/-0
Re: Quantification of Dimethicone
« Reply #2 on: August 25, 2021, 01:45:41 PM »
FTIR isn't ideal, but so far it's all I have for detection. Unfortunately the formula consists of both triglycerides and petroleum waxes, and I haven't been able to find a way to chemically separate dimethicone. Gravimetric analysis using a centrifuge is something that I have been considering since dimethicone has a higher density than the other components. One method is to use a heated centrifuge to spin melted product. Another is to use a normal centrifuge to spin product that has been dissolved in acetone or similar solvent that dissolves everything but dimethicone. Both have potential I think, but I don't have a centrifuge yet, and then I will have the problem of getting the dimethicone out of the centrifuge tube without contamination.

Offline Corribus

  • Chemist
  • Sr. Member
  • *
  • Posts: 3221
  • Mole Snacks: +487/-22
  • Gender: Male
  • A lover of spectroscopy and chocolate.
Re: Quantification of Dimethicone
« Reply #3 on: August 25, 2021, 05:14:27 PM »
I was thinking more along the lines of thermogravimetric residue analysis, using the completely unsubstantiated idea that pyrolysis of polysiloanes would leave behind silica, whereas pyrolysis of everything else would leave behind... practically nothing. I did a quick search and see that it's not such a crazy idea

https://www.sciencedirect.com/science/article/pii/S0141391005004568?via%3Dihub

If you don't have a good TGA but you can get your hands on a good muffle furnace or the equivalent, and you have a lot of the test material to play around with, you could just combust a lot of it and weigh the amount of residue left behind. Use the pure polysiloxane as a control, and then back correlate how much of it is in the original formulation using a weight ratio (residue in the pure stuff versus residue in the mixture). We've done this very successfully with organically modified silicate fillers in organic polymers. As long as your additive leaves behind a consistent measurable residue and all the other stuff doesn't, you're golden.

1.5% should be doable using an approach like that, as long as the polysiloxane leaves behind a sizeable fraction of its mass as silicate residue.
What men are poets who can speak of Jupiter if he were like a man, but if he is an immense spinning sphere of methane and ammonia must be silent?  - Richard P. Feynman

Offline Thurbs

  • New Member
  • **
  • Posts: 3
  • Mole Snacks: +0/-0
Re: Quantification of Dimethicone
« Reply #4 on: August 27, 2021, 11:45:51 AM »
It makes complete sense to me that silica is all that would remain after the pyrolysis of the sample material. We also have a muffle furnace that we use in our zinc oxide titrations.

I added ~100mg of dimethicone to a few crucibles and let them bake at 800C for an hour after a ramp rate of 10C/minute. Weight retention should have been 59.4% based on the weight percent of the methyl groups. However, actual retention was 43.8 and 46.7%, and there were faint white streaks up the inside of the crucible. Silica that remained in the crucible was such a fine powder that I think some of it simply floated away on convention currents. I wouldn't mind having a correction factor in whatever calculation I end up with, but I would need a consistent percentage of silica residue first. Considering that a typical sample size of 500mg would only leave me with 7.5mg of dimethicone to work with prior to pyrolysis, I would expect that variation to only increase.

Offline Corribus

  • Chemist
  • Sr. Member
  • *
  • Posts: 3221
  • Mole Snacks: +487/-22
  • Gender: Male
  • A lover of spectroscopy and chocolate.
Re: Quantification of Dimethicone
« Reply #5 on: August 27, 2021, 12:29:29 PM »
0.5%-1% residue pushes against the limit of even a good TGA with standard sized sample pan, so this definitely is not going to be an easy experiment. Not sure what kind of crucible you are using, but a lot of TGA vendors cell narrow ceramic pans with high perpendicular walls that allow for analysis of larger sample sizes and help to prevent the kind of spatter loss you might be seeing. They're not cheap but could be an option. You might also try playing around with the heating conditions (e.g., ramp rate) to reduce spatter loss.
What men are poets who can speak of Jupiter if he were like a man, but if he is an immense spinning sphere of methane and ammonia must be silent?  - Richard P. Feynman

Offline wildfyr

  • Global Moderator
  • Sr. Member
  • ***
  • Posts: 1662
  • Mole Snacks: +178/-10
Re: Quantification of Dimethicone
« Reply #6 on: August 28, 2021, 05:44:47 PM »
You could try to hydrolyze it with strong aqueous acid or base and see if the monomers are visible on GC.

Offline marquis

  • Full Member
  • ****
  • Posts: 357
  • Mole Snacks: +32/-3
Re: Quantification of Dimethicone
« Reply #7 on: September 02, 2021, 04:12:24 PM »
There are several  methods for polydimethylsiloxane( PDMS or silicone).  If you have access to an AA, that might work.  It did for us.   It requires the use of the nitrous oxide/ acetylene flame with an organic solvent.  Which can be an education in itself.  Then there is FTIR. We used to wash silicone off stoppers in pet ether.  Then evaporate the pet ether.  Then dissolve the residual silicone in toluene.  Use a.liquid cell and run an FTIR.  Reference the silicone peak to a toluene peak for the calculation.  That should give you a start.  Good luck.

Offline marquis

  • Full Member
  • ****
  • Posts: 357
  • Mole Snacks: +32/-3
Re: Quantification of Dimethicone
« Reply #8 on: September 02, 2021, 04:27:00 PM »
Another option would be to add an RI (refractive index) detector to your hplc.  You might be able to detect the silicone with your hplc, then.  The problem is, you will probably need a whole new set of columns and probably need to switch columns (probably different solvents, too).  Not always something that is easy to do in a qa/qc environment.

Offline marquis

  • Full Member
  • ****
  • Posts: 357
  • Mole Snacks: +32/-3
Re: Quantification of Dimethicone
« Reply #9 on: September 03, 2021, 01:36:20 PM »
Re - Corribus's remark about the variable molecular weight affecting the calculation.   For the AA procedure, this was handle by basing the calculation on the base polymer unit.  That way, polymer length didn't come into the calculation. 

Sponsored Links