Abstract The ability to isolate and stabilize thylakoid membranes from photosynthetic organisms (cyanobacteria, green algae, or higher plants) facilitates their study as photocatalysts in processes for the production of hydrogen from water using solar energy (biophotolysis). The feasibility of immobilizing photosynthetic membranes with retention of their electron transport capability was explored in this study. The method of immobilizing chloroplast thylakoid membranes which was observed to best preserve their photochemical activity involved the following steps: 1. (1) mixing an aqueous suspension of thylakoid membranes (isolated from Spinacia oleracea) with an aqueous suspension of 1.5% (w/v) collagen; 2. (2) adding glutaraldehyde to a concentration of 0.01% (w/v) and cross-linking for 20 min at 0 °C; 3. (3) casting the collagen-thylakoid mixture on a flat surface; and 4. (4) freeze-drying the cast mixture to form a macroporous collagen-thylakoid film composite. Films prepared by this immobilization technique had a very open, sponge-like fibrous structure with a bulk density of 0.049 g/cm 3 of film. Photosynthethic electron transport activities of free and immobilized thylakoid membranes were determined by measuring the photoproduction of oxygen with a Clark-type polarographic probe using potassium ferricyanide as the electron acceptor (Hill reaction). A simplified kinetic model of the photosynthetic electron transport system (PETS) was formulated based on a reaction scheme which is consistent with current concepts of photosynthesis. The kinetic model relating electron transport rate (O 2 evolution) to incident light intensity is a rectangular hyperbolic function, in agreement with the observed behavior of both immobilized The photocatalytic properties of the collagen-thylakoid film were characterized with respect to electron transport kinetics. A simplified kinetic model of the photosynthetic electron transport system was formulated based on a reaction scheme which is consistent with the current concept of photosynthesis. The model accounts for the significant features of photosynthetic electron transport and yet has sufficient simplicity to be useful in the eprocess engineering of photosynbetic reactor systems. The dependence of initial reaction rates on incident light intensity predicted by the kinetic model was experimentally verified for both free and immobilized thylakoid membranes. Initial-rate data were used to evaluate the parameters of the kinetic model. Future kinetic studies might fruitfully focus on dissolved O 2 catalyst-poisoning effects, as well as photoinactivation and/or photoinhibition effects.