Dear Dolla, i think that you can use XPS to determine 

TGA（thermogravimetric analysis）

All methods cited above are useful. The XPS is the best, but it is a very expensive and difficult to work. EDX not. I've been using potentiometric titration and TG-MS. Due to the heterogeneity of the carbon material surface we use a non-linear regression progam to fit the experimental data.

See: Characterization of the surface chemistry of potentiometric titrations and carbonmaterials by temperature-programmed desorption. Carbon, 2008.

I suggest XPS for quantification. In addition if you couple High-resolution XPS with FTIR you can have also an idea of what kind of surface groups you have. TGA can be useful but generally CNT decompose over 550-600 °C and this can be overlapped with the decomposition of some oxygenated groups. 

Thank you all for your answers. But, I want to compare the different oxidation methods and take the one with a high oxygen functionalities for coating with mesoporous metal oxides.