As others have noted, for a properly-operating detector, the FID responce for liquid HCs above carbon number 6 is purely a function of the relative C and H content.
From memory, Benzene usually has the highest response, but still only about 120% of the n-alkane, and Toluene about 107%, and the aromatic response difference decreases with increasing carbon number.
Your problem is most likely related to the different standard solvents selected, and it's important to use the same solvent for your samples and standards.
The use of different solvents for different analytes would require internal standards to provide correction factors, which is not very efficient.
You should be following one of the many standard methods that exist for analysng TPH from contaminated sites, especially if the data is to be reported to a client or regulator.
From memory, the US EPA has several, as do some US states, some countries, and there is the European Eugris www site as well. Many newer ones will use MS detection, but there are still FID methods around, which a WWW search should identify.
Those methods specify the standard and sample solvents, which may be DCM, or a CS2/DCM mixture - if solubility is an issue.
Please keep having fun,
Bruce Hamilton
