X-ray powder di®raction (XRPD) is an excellent tool for characterising the bulk structure of crystalline materials. Along with the growing interest in exploiting materials with decreasing particle sizes and increasing number of defects, factors that complicate the traditional interpretation of the experi- mental XRPD patterns, the need for new interpretation methods has arisen. The method described in the present thesis is by no means new, in fact it was developed by Debye in 1915. However, the Debye method it is rather computationally heavy, so in practise it is only applicable to the X-ray char- acterisation of nanostructured materials because of modern computers. The Debye equation was implemented into a general GUI based program which is not custom-made to characterise a speci¯c type of material as op- posed to many earlier implementations. The Debye program is able to read a Crystallographic Information File (CIF), simulate the XRPD pattern given information about the nature of the sample and the experimental setup, and ¯nally ¯t the simulated di®ractogram to experimental XRPD data. Three very di®erent materials were studied using the Debye approach: 1) Cellulose, an organic polymer with a nano¯brous structure. The study was initiated based on the need for a reliable crystallinity determination from XRPD. It was shown that for future crystallinity determinations a new method based on Rietveld re¯nements should be preferred. If additional in- formation about particle shape, size or size distribution is required, this can be obtained from Debye simulations. 2) Nitrogen expanded austenite, a highly defective material. Debye simulations con¯rmed that this material contains deformation stacking faults and that screw dislocations are abun- dant. A combined XRPD/EXAFS characterisation of nitrogen expanded austenite produced using a novel method showed that CrN formed even at temperatures below 450± where the mobility of Cr is very low. 3) Carbon nanotubes, a non-crystalline material with a periodic structure. It was shown that the mean bulk structural properties of the nanotubes can be obtained from XRPD using a combination of Debye simulations and param- eterised Principal Component Analysis (PCA).