Supplementary cementitious materials (SCMs) like fly ash (FA) and blast furnace slag (BFS) are normally used to replace parts of Ordinary Portland cement (OPC) to reduce the cost and CO 2 emission. Some consequences are the reduction of portlandite (CH) content and the formation of C-S-H with low Ca/Si ratio, due to pozzolanic reactions. It is known that carbonation of portlandite leads to a reduction in the porosity which is ascribed to the positive difference of molar volumes between CH and CaCO 3 . However, the influence on the porosity caused by the carbonation of C-S-H is still controversial. The molar volume change due to the carbonation of C-S-H depends on the properties of C-S-H (like Ca/Si ratio, water content) and the water remained in silica gel. Moreover, the decalcification of C-S-H with the Ca/Si ratio lower than 1.2 can cause more structure changes and shrinkage of C-S-H. During the carbonation of cement paste blended with SCMs, less portlandite but a relatively high amount of C-S-H with low Ca/Si ratio will be carbonated. The pore structure will evolve in a different way, comparing with Portland cement paste. Therefore, it's very important to figure out the pore structure development of cement paste blended with SCMs under carbonation.In this paper, the binary cement pastes (B70, blended with blast furnace slag, and F30, blended with fly ash) and ternary cement pastes (F10B54 and F30B30, blended with blast furnace slag and fly ash) are studied and compared with Portland cement paste. Mercury Intrusion Porosimetry (MIP) and nitrogen adsorption isotherm are used to determine the pore volume and size distribution of capillary pores and gel pores (2-37 nm), respectively. Thermogravimetric analysis (TGA) is used to determine the amounts of portlandite and CaCO 3 . The results show that the amount profiles of portlandite and CaCO 3 can be used as a more accurate method to study the carbonation in blended cement paste, comparing with the phenolphthalein test. Carbonation of most of the species of C-S-H results in the increase of the porosity of cement paste. CaCO 3 contributed by the carbonation of low Ca C-S-H is dominant in blended cement paste B70, F10B54 and F30B30. Both the total and effective capillary porosity increases in the above-mentioned paste after the carbonation. Moreover, total porosities of B70 and F10B54 increase with the increasing amount of C-S-H involving in carbonation. However, the increment of the total porosity of F30B30 decreases with the increasing amount of C-S-H being carbonated. Carbonation of C-S-H increases the volume and size of the small gel pore, and creates more capillary pores. This peculiar phenomenon is more evident for the mixture with a higher proportion of SCMs, like B70 and F30B30. The results reveal that the carbonation of C-S-H formed in pozzolanic reactions cause the increase of the total and capillary porosity in cement paste blended with SCMs, which will bring adverse effects on the durability of blended cement concrete exposed to the carbonation...