You should prepare standard samples containing a known quantity of Pt and plot calibration graph (Intensity vs. concentration). Best of all, if standards will also contain both Pt and carbon.
alternately you can use x-ray transmission spectrometry by looking at the K-edge jump of the Pt K-edge (if the sample is suited for transmission experiments). From known values of the Pt attenuation coefficient (see for example the NIST tables) you can easily deduce the Pt mass coverage of your sample (without any calibration using well known systems).
But Sergei is right. Calibration in XRF is essential in order to kick off all experimental issues concering solid angles, distances, self absorption, detector properties, scatter, etc.
Which (Pt) sample/layer thicknesses do you work with or expect ?