Dear Jerald,

I hope you can find more useful information from these articles about decomposition properties of PEG

http://onlinelibrary.wiley.com/doi/10.1002/pat.276/pdf

http://www.conservationphysics.org/ppubs/pegdeg.pdf

At first , you should be sure what molecular weight PEG are you using in your experiment.

Poly (ethylene glycol) (PEG) has been widely used in studies of polymer–clay nanocomposites because it readily intercalates in smectite clays. Nanocomposites were formed from PEG with molecular weights (Mw) ranging from 300 to 20,000, as evidenced by expansion of the basal planar spacing of the clay (d001) in X-ray diffraction. However PEG with high molecular weight (≥ 10,000) readily underwent degradation during preparation of composites when heated at low temperature (60°C) due to oxidative attack. Molecular weight distribution determined by gel permeation chromatography showed that this degradation always happened with or without the presence of clay and it became more serious when the molecular weight was higher. The reduction in pH of aqueous PEG solutions after degradation increased with molecular weight. Since d001 was independent of molecular weight over a wide range, such degradation cannot be detected by this method. Precautions against oxidative attack are therefore recommended to avoid decomposition when preparing PEG–clay nanocomposites. © 2004 Wiley Periodicals, Inc. J Appl Polym Sci 94: 548–552,

Poly (ethylene glycol) (PEG) has been widely used in studies of polymer–clay nanocomposites because it readily intercalates in smectite clays. Nanocomposites were formed from PEG with molecular weights (Mw) ranging from 300 to 20,000, as evidenced by expansion of the basal planar spacing of the clay (d001) in X-ray diffraction. However PEG with high molecular weight (≥ 10,000) readily underwent degradation during preparation of composites when heated at low temperature (60°C) due to oxidative attack. Molecular weight distribution determined by gel permeation chromatography showed that this degradation always happened with or without the presence of clay and it became more serious when the molecular weight was higher. The reduction in pH of aqueous PEG solutions after degradation increased with molecular weight. Since d001 was independent of molecular weight over a wide range, such degradation cannot be detected by this method. Precautions against oxidative attack are therefore recommended to avoid decomposition when preparing PEG–clay nanocomposites. © 2004 Wiley Periodicals, Inc. J Appl Polym Sci 94: 548–552,