The apparent first-order rate constant, k(app), of the Acid Blue 113 textile azo dye photocatalytic disappearance in aqueous suspensions of 4 commercialized TiO(2) samples (50 m(2) g(-1)-Degussa P25; 54 m(2) g(-1)-Millennium Chemicals PC50; 250-270 m(2) g(-1)-Sachtleben Chemie Hombikat UV100; 320 m(2) g(-1)-Ishihara ST01) was determined from the decrease in the calibrated 566 nm-absorbance of the supernatant. Under the conditions used (1 liter-batch reactor; water-filtered irradiation from a Philips HPK 125-W high-pressure mercury lamp; [TiO(2)] = 1 g/L; [dye] = 70 mg/L), the k(app) ranking was: UV100 > PC50 > P25 > ST01. The corresponding maximal difference in k(app) was roughly equivalent to that caused by adding an optimal dose of H(2)O(2) to the P25 suspension. Taking into account the dye adsorbed amount, q(ads), in the dark, it is deduced that the recombination rate k(r) of photogenerated charges was higher for poorly crystallized TiO(2) samples, as expected; additionally, a low content of impurities is suggested to be at the origin of the low k(r) of UV100 relative to ST01. It is also argued that differences in the values of q(ads) divided by the surface area, S, was less due to pores inaccessible to the dye than to thicker layers of surface tightly bound water for the samples with high S. A FTIR study showed that the dye weakly interacted with TiO(2) and, however, remained the dominant adsorbed species during the degradation. This latter conclusion was in accord with the fact that the dye initial disappearance rates -- directly measured or calculated from k(app) -- were almost equal, indicating that the intermediate products of degradation did not significantly compete with the remaining dye for active species.