Dear Mr. Sebastian

Thank you so much for your support indeed

Might I suggest using volume fraction analysis through Archimedes principle (for a binary composite). This can be easily accomplished if the density of the pure matrix and reinforcing fibres are already known.  

For fiber-reinforced composites mass ratio concentration (i.e. mass of fiber : mass of composite), or the corresponding percentage ratio, can be conveniently used. Let us call the mass of fiber, Mf, that of matrix phase, Mx, and that of composite, Mc. The mass ratio concentration will be given by: R = Mf / Mc. Let the volume of fiber, composite, and matrix phase, be Vf, Vc, Vx; and the corresponding densities ρf = Mf /Vf, ρc = Mc /Vc, ρx = Mx /Vx. Mass balance for mixing the two phases reads Mf + Mx = Mc, and we will have: Mf = Mx / (1/R – 1).(*). Accepting volume additivity, we may write Vx = Vc - Vf, so that Mx = (Vc - Mf/ρf)·ρx. Substitution at (*) yields Mf/Vc = [ρx/(1 - R·(1 - ρx/ρf))]·R. Partial substitution of R = Mf / Mc at last equation now conducts to ρc = ρx/(1 - R·(1 - ρx/ρf)), and after some transformations one obtains: R = (ρx/ρc - 1)/(ρx/ρf - 1). Last equation can be used to calculate the mass ratio concentration, knowing the density of the composite, and those of the fiber and matrix phases.

You have also the image analsys method for determine  fiber(glass), Matrix (resin) and porosity content in your composite. I suggested this method because is an non destructive test (contrary of thermogravimetry). You could realise a picture frame of  your sample by SEM and after that  used the software as imageJ or Aphelion™

Imaging Software for determine each constituent content in your composite. Technique is simple: in each frame picture the contrast of Matrix, fiber and porosity is different. The soft ware find the pixel number of each phase based on the contrast and calculate the fraction.  Good luck man ...

For a correct detection of fiber content the burn-off technique is highly recommended:

ASTM D2584 Standard Test Method for Ignition Loss of Cured Reinforced Resins

If only glass fabric or filament is used as the reinforcement of an organic resin that is completely decomposed to volatile materials under the conditions of this test and the small amount of volatiles (water, residual solvent) that are potentially present are ignored, the ignition loss can be considered to be the resin content of the sample.

Density measurements as described in preceeding answers may suffer some errors coming from void content.

Measurements from image analysis are quite costly.

If the resin is organic based, I highly recommend the burn-off method, "Standard Test Method for Ignition Loss of Cured Reinforced Resins ASTM D2584"

its is important to give the sample good time to rest in the furnace to let the resin burns off completely.

Thank you for your useful advises.

Dear All,

Thank you for your kind support. I prefer burn-off  & TGA method, but still worried about the accuracy of this method, because if sample content calcium carbonate (CaCO3), so such methods can not be compatible.

Hi

use rule of mixture to find

use the rule of mixture