As I read that the experiment is limited to aqueous extracts, this suggests that the compounds will be fairly polar. I would run a series of columns which work well with polar compounds. Using different columns allow different selectivity in case one compound of interest is "hidden" behind another on a particular column. I would run LC-MS with the following columns:

• C18, an AQ type column designed to resolve polar compounds. Water/methanol gradient, acid modifier optional.
• Silica HILIC using acetonitrile or methanol/water gradient; if the peaks elute at the front, try silica with dichloromethane/methanol gradient
• Alumina HILIC, run the same as silica HILIC
• Diol HILIC, run the same as silica HILIC

I also suggest that you do a literature search on the plant species from which the extract came from. This will let you know compounds that you can expect to see. When looking at the mass spectra, keep in mind that some compounds may fragment, or form solvent adducts. For example, capsaicins in my lab always seem to appear as the [M+Na]+ ion. Use both positive and negative ionization if using ESI or APCI.

Although this refers to flash chromatography, the links below show what can be done with a column screen:

http://www.isco.com/WebProductFiles/Applications/101/Poster_and_Paper_Reprints/New_Techniques_In_Extraction_Column_Screening.pdf

http://www.isco.com/WebProductFiles/Applications/101/Poster_and_Paper_Reprints/New_Techniques_in_Extract_Column_Screening_II.pdf

Dear Bachir, You may follow the process suggested by Dear Jack. In addition you may also consult the following threads.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3218439/

https://www.env.go.jp/en/chemi/pops/Appendix/04-GuideLine/04Chapter3.pdf