The construction and use of a thin layer flow cell test system employing a TiO2 working electrode, a platinum quasi-reference electrode and the ruthenium dye (H-2-dcbpy)Ru(NCS)(2) (H-2-dcbpy = 2,2'-bipyridine-4,4'-dicarboxylic acid) is described. The efficient design enables significant advantages to be gained over presently available procedures for the measurement of photocurrents of dye-sensitized solar cells. The widely used iodide/triiodide redox shuttle system has been investigated over a wide range of conditions. A linear dependence of photocurrent on cation radius was revealed. Under certain conditions, the photocurrent measured in the presence of the Li+ cation is five times larger than when the (C4H9)(4)N+ cation is used. Additionally, the addition of low concentrations of cations with small diameters has a significant catalytic enhancement effect on the photocurrent. Other redox shuttles, based on ferrocene, thiocyanate, triiodide and bromide, were tested for their performance in the flow cell and compared to iodide. However, despite some apparent thermodynamic advantages, the photocurrents obtained with these redox shuttles were more than two orders of magnitude lower than those measured with iodide. This finding implies that the efficiency of redox shuttles is limited by kinetic restraints rather than their thermodynamic properties and confirms that the iodide/triiodide system is the dominant redox shuttle.