What type of light source are you using in your reactor? Usually cut off filters are installed/supplied on/with the light source that can be tuned or changed e.g solar simulator.
Too general question. It depends on the photocatalyst. If it absorbs e.g. in blue, substrates absorb in UV, cut only UV (not much in xenon lamp, but consider the emission spectrum declared by the supplier of the lamp). When using the optical bench for some analysis/mechanistic study, xenon lamp and filters are good option for its tunability. If you use it in preparative set-up, use LEDs, they are more efficient and monochromatic. First, measure the absorption spectrum of the catalyst, then use monochromatic LEDs with peak wavelength in the lowest energy band. Actinometric measurement of light intensity is a must for each set-up.
If you would like to test your photocatalyst if absorbs Visible light or not, you can use simply a cut off filter >400 nm or >420 nm to eliminate UV light.