As long as you can couple them covalently to a carrier protein like KLH, it's worth a try. Or you stick your lipids to a suitable test tube and screen a ScFv phage display library against them.

Consider your question. How are the lipids volatile? Are they low boiling point or are they liable to oxidation. The reason for producing antibodies is presumably so that you can identify these lipids specifically. In that case it is the unique part of the structure which one needs to prepare the antibodies to. There are 3 options. First as above you can link the lipid to a carrier, provided of course that the unique identifying structure is retained and accessible to the antibody; second determine whether it is possible to prepare a structure chemically which contains the unique feature and link that to the carrier ( an example of this is the use of caproic acid as a marker for penicillin); third you can prepare an antibody to the product of the volatile process, e.g. if oxidised then the oxidised lipid is the antigen and in establishing the assay you deliberately oxidise the lipid fraction and analyse for the oxidised product, and example of this is soya protein which in the majority of assays is analysed as the renatured protein following denaturation with urea, the antibody recognises the renatured protein specifically.