Expanded graphite has been applied widely in thermal insulation, adsorption, vibration damping, gasketing, electromagnetic interference shielding etc. It is made by intercalation of natural flake graphite followed by thermal expansion. Intercalation is a process whereby an intercalant material is inserted between the graphene layers of a graphite crystal. Exfoliation, a huge unidirectional expansion of the starting intercalated flakes, occurs when the graphene layers are forced apart by the sudden decomposition and vaporization of the intercalated species by thermal shock. Along with production methodologies, such as the intercalation process and heat treatment, the raw material characteristics, especially particle size, strongly influence the properties of the final product. This report evaluates the influence of the particle size of the raw material on the intercalation and expansion processes and consequently the properties of the exfoliated graphite. Natural crystalline flake graphite with wide particle diameter distribution (between dp = 80 and 425 µm) was divided into four size-range portions by sieving. Graphite was intercalated via perchloric acid, glacial acetic acid and potassium dichromate oxidation and intercalation procedure. 5.0 g of graphite, 7.0 g of perchloric acid, 4.0 g of glacial acetic acid and 2.0 g of potassium dichromate were placed in glass reactor. The mixture was stirred with n = 200 min–1 at temperature of 45 °C during 60 min. Then it was filtered and washed with distilled water until pH~6 and dried at 60 °C during 24 h. Expansion was accomplished by thermal shock at 1000 °C for 1 min. The prepared samples were characterized by means of exfoliation volume measurements, simultaneous differential thermal analysis and thermo-gravimetry (DTA/TGA), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), BET measurements and scanning electron microscopy (SEM). X-ray diffraction indicated a change of distance between the graphene layers due to oxidation and repulsion of positively charged layers. The increased width and reduced height of diffraction peaks were a consequence of small-sized ordered domains. The intercalation is partial, intercalated layers are divided by a considerable number of non-intercalated layers. FTIR spectra revealed that dominant intercalating species is perchloric acid. Thermo-gravimetric analysis revealed that deintercalation occurs in the temperature interval between 150 and 300 °C and that a mass loss in this temperature interval is dependent on particle diameter, i. e. the intercalation is more intensive for greater particles. The fact that deintercalation proceeds as a one-stage process indicates the existence of only one intercalating specie. Additional mass loss at higher temperatures is a consequence of graphite oxidation. The particles with a higher amount of interlcalant showed greater expansion volumes as well as specific surface area. The fraction with greatest particle diameter (315–425 µm) showed expansion specific volume of v=86 cm3 g–1. Weaker expansion of smaller particles is a consequence of intercalant thermal degradation gaseous products loss at the layer edges, as well as of lesser amount of intercalants due to their removal during washing. Adsorption-desorption isotherms of expanded graphite could be classified as type III, according to BDDT/IUPAC classification, characteristic for macro porous materials. Small variations in adsorption and desorption pressure for the same amount of adsorbed gas indicate that the macro pores are open. Specific surface area was calculated using BET equation and for sample 315–425 yields s = 36 m2 g–1. SEM micrographs revealed typical worm-like microstructure generated by exfoliation of graphene sheets. The areas of intense exfoliation forming typical pores, as well as less exfoliated sheets can be observed.