I can't use different type of column as the sample contains non polar compounds.
I don't understand why this is an issue.
If you must stay with an HP-5 column: of course ethylene glycol peak shape will be bad, as it's very polar and your column is essentially non-polar. So if you are wedded to HP-5, dissolve your samples in DMF and make trimethylilyl derivatives of the polar compounds using BSTFA. The derivatized ethylene glycol will elute just after the derivatizing agents, so be sure to run a blank. See Journal of Cosmetic Science, May/June 2010, pp. 1-10.