In the present work the existence of mass transfer limitations in slurry, photocatalytic reactors is studied. Experimental validation is made in a flat reactor that is part of a recycling system. The reactor is described with a mathematical model previously developed [Ballari et al., 2008a. Chemical Engineering Journal 136, 50], considering a transient, two-dimentional mass balance (TDM). The complete reactor model was developed to show the existence of these effects, which result from the occurrence of concentration gradients in reaction space. They develop when these reactors are operated under some operating conditions whose effects should be always analyzed before assuming the validity of existence of perfect mixing in reaction space. Dichloroacetic acid (DCA) was the adopted model compound. To solve TDM, a kinetic expression for DCA acid was determined before under well mixed conditions [Ballari et al., 2009. Industrial and Engineering Chemistry Research 48(4), 1847]. The studied variables are flow rate, catalyst loading, and irradiation rates. The experimental data agree quite well when they are interpreted in terms of the two-dimensional model (TDM) regardless of the operating mode. The perfect mixing model (PMM), normally employed to describe this and other types of slurry photoreactors, does not have the same level of universal application; i.e. it is restricted to perfect mixing, but in many cases far simpler to use. However, it can be concluded that when the photocatalytic reaction is not fast, employing catalyst loadings below 1 g L-1, irradiation rates at the reactor wall below 1 x 10(-6) Einstein cm(-2) s(-1) and good mixing operation (Re > 1700) it will be always safe to assume that mass transport limitations in the bulk of the fluid are nonexistent. In a typical batch reactor the above flow conditions are equivalent to very intense mixing. If the catalyst concentration is increased, the mixing conditions should be improved in the same proportion. Within limits, higher solid loadings can be compensated with lower irradiation rates [Ballari et al., 2008a. Chemical Engineering Journal 136, 50]. In addition, with the validated model, additional simulations are shown, operating the reactor under different virtual reactor thicknesses to widen amplitude of the reached conclusions. These findings will be useful in kinetic studies to prevent incursion in certain ranges of experimental conditions that could lead to erroneous interpretation of the obtained kinetic data. (C) 2010 Elsevier Ltd. All rights reserved.