An empirical rate expression was obtained for the photolysis of tetrachloroplatinic acid (TCPA). The data used to evaluate the parameters of the rate expression were obtained in an elliptical photoreactor operated a t high recycle rates in order to maintain a uniform reactant concentration within the photoreactor. A low-pressure mercury lamp provided light a t a wavelength of 2,537A. The light was assumed to enter the reactor radially and uniformly with respect to both the axial and angular directions.
This rate expression was incorporated in a reactant conservation equation which was solved to obtain the predicted conversion of TCPA in an elliptical photoreactor operating i n the laminar regime. Solutions of the conservation equation were obtained for photoreactors of two different diameters and for three different lengths.
Experimentally observed conversions of TCPA in the laminar regime agreed well with the theoretical solutions except a t conversions above 55%. where a solid precipitate apparently caused the lack of agreement.
F. P. Ragonese is with
Shell Chemical Company, Woodbury, New Jersey 08096. Correspondence concerning this article should be addressed to J. A. Williams.
method of Harris and Dranoff ( 9 ) is restricted to radial scale-up of annular photoreactors whose contents are well mixed.
The method of dimensional analysis was used by Dolan, Dimon, and Dranoff (7) to correlate conversion data obtained in a tubular flow photoreactor for a range of flow rates and length-to-diameter ratios. Since the correlation utilizes optical density as a parameter, this method cannot be used to predict photoreactor performance under conditions of varying optical density unless experimental data are available for every optical density encountered in the scale-up problem.
The purpose of the present work is to demonstrate an alternate approach to the problem of photoreactor scaleup. In this approach an empirical reaction rate expression, reactant continuity equation, and radiation equation are applied to the problem of predicting the performance of