I used a product on my family containing 90 percent Polyethylene glycol. Now I’m reading how Polyethylene glycol can be contaminated with ethylene oxide and also the lower the weight the higher the chance. The Polyethylene glycol in the product I used has no number following and it was a soft cream used for skin care. I was wondering if someone could explain to me how Polyethylene glycol is safer than ethylene oxide and how exactly is Polyethylene made? I’m very worried that I put a toxic cream on my children one of them who had eczema at the time. Thank you

  • 2
    $\begingroup$ Do not trust random websites. I suspect that widespread misinformation is made for the interest of "bio products" companies. Polyethylene glycol can be synthetized by reaction with ethylene oxide. But ethylene oxide, being far from healthy, is carefully removed from polymers suitable for human use (and it's not hard to do, being it a gas). I would say not to worry $\endgroup$ – The_Vinz Jun 17 '19 at 15:54
  • $\begingroup$ @The_Vinz: I like your comment. Also, any remaining ethylene oxide may be converted to ethylene glycol and would be okay to external applications. Lynn has nothing to worry about. $\endgroup$ – Mathew Mahindaratne Jun 17 '19 at 16:20
  • $\begingroup$ Agree with @The_Vinz, also, polyethylene oxide (or polyethylene glycol) has nothing to do with polyethylene and is only formally related to ethylene glycol as well. The name polyethylene glycol comes from the formal polycondensation of ethylene glycol (which does not work in practice), while the name polyethylene oxide comes from the anionic polymerisation of ethylene oxide (this is the synthesis route used in practice). $\endgroup$ – szentsas Jun 17 '19 at 16:37
  • 1
    $\begingroup$ While polyethylene glycol [PEG] is safe for human consumption, and the worst part of a colonoscopy, ethylene glycol is not. It is metabolized to oxalic acid which is toxic. $\endgroup$ – user55119 Jun 17 '19 at 18:57
  • $\begingroup$ @user55119 Oxalic acid is not toxic in such trace amounts. It is present in a lot of vegetables which are safely edible. Rhubarb, chocolate, spinach, ... $\endgroup$ – Karl Jun 17 '19 at 20:01

Poly(ethylene glycol), commonly known as PEG is non-ionic, hydrophilic, non-toxic, and the most used polymer in wide variety of applications. For example, PEG is the gold standard for stealth polymers in the emerging field of polymer‐based drug delivery (Ref.1). You may find mostly unbiased information about PEG here, which is generally considered biologically inert and safe according to the clinical studies. However, drawback is these studies of clinical safety of PEG are generally based on adults, not children. Nonetheless, at least one study using children have proven it is safe for children as well (Ref.2).

So far, I addressed the suitability of PEG for human consumption. Note that these PEG molecules suitable for human intake have specific lengths such as PEG 3350, which has an average molecular weight of $\pu{3350 g/mol}$ (The structure of PEG is commonly expressed as $\ce{H−(O−CH2−CH2)_n−OH}$). Yet, my argument is if PEG is clear to human intakes, then it should be safer for external applications such as skin creams. To my understanding, among available PEG molecules ($n=1, 2, 3, ..., n$), only toxic one to human is its monomer, $\ce{H−O−CH2−CH2−OH}$ (when $n=1$), commonly known as ethylene glycol or antifreeze. The apperent toxicity of that molecule is mainly due to its enzymatic oxygenation inside the body to oxalic acid, which produce insoluable calcium oxalate crystals.

As you concern, PEG is the basis of many skin creams. An example of PEG derivative used in skin creams is cetomacrogol 1000, which is the trade name for polyethylene glycol hexadecyl ether ($\ce{-O-C16H33}$) with the general formula $\ce{HO(CH2CH2O)_nC16H33}$ (where $n= 2-20$). When $n=10$, it is named Brij56, and when $n=20$, it is Brij58, etc. Another polyethylene glycol ether commonly used in cosmatic industry is Isoceteth-20, which has the same general formula $\ce{HO(CH2CH2O)_nC16H33}$ where $n$ has an average value of 20. The difference between Isoceteth-20 and cetomacrogol 1000 is use of iso-cetyl alcohol instead of cetyl alcohol. Few examples for cosmetics containing Isoceteth-20 are: Nivea, Aveeno, and Nu Skin products, among many others. So, I'd be comfortable with using these products, just based on their reputation.

Now, how PEG is synthesized?

The production of polyethylene glycol was first reported in 1859. Both A. V. Lourenço and Charles Adolphe Wurtz independently isolated products that were polyethylene glycols. Polyethylene glycol is produced by the interaction of ethylene oxide with water, ethylene glycol, or ethylene glycol oligomers. The reaction is catalyzed by acidic or basic catalysts. Ethylene glycol and its oligomers are preferable as a starting material instead of water, because they allow the creation of polymers with a low polydispersity (narrow molecular weight distribution). Polymer chain length depends on the ratio of reactants (Wikipedia).

For example, branched PEG haas been recently prepared by Copolymerization of ethylene Oxide and glycidol (Ref.3):

Copolymerization of Ethylene Oxide and Glycidol

$\mathrm{^{13}C \; NMR}$ of the product mixture shows no presence of unreacted ethylene oxide and glycidol:

13C-NMR of Products


  1. Katrin Knop, Richard Hoogenboom, Dagmar Fischer, Ulrich S. Schubert, “Poly(ethylene glycol) in Drug Delivery: Pros and Cons as Well as Potential Alternatives,” Angew. Chem. Int. Ed. Engl. 2010, 49(36), 6288–6308 (https://doi.org/10.1002/anie.200902672).
  2. Dinesh S. Pashankar, Vera Loening-Baucke, Warren P. Bishop, “Safety of Polyethylene Glycol 3350 for the Treatment of Chronic Constipation in Children,” Arch Pediatr. Adolesc. Med. 2003, 157(7), 661–664 (https://doi:10.1001/archpedi.157.7.661).
  3. Daniel Wilms, Martina Schömer, Frederik Wurm, M. Iris Hermanns, C. James Kirkpatrick, Holger Frey, “Hyperbranched PEG by Random Copolymerization of Ethylene Oxide and Glycidol,” Macromol. Rapid Commun. 2010, 31(20), 1811–1815 (https://doi.org/10.1002/marc.201000329).
| improve this answer | |

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