화학공학소재연구정보센터
Solar Energy, Vol.86, No.1, 99-111, 2012
Development of a system model for a hydrogen production process on a solar tower
An attractive path to the production of hydrogen from water is a two-step therm chemical cycle powered by concentrated sunlight from a solar tower system. In the first process step the redox system, a ferrite coated on a monolithic honeycomb absorber, is present in its reduced form while the concentrated solar energy hits the ceramic absorber. When water vapour is fed to the honeycomb at 800 degrees C, oxygen is abstracted from the water molecules, bond in the redox system and hydrogen is produced. When the metal oxide system is completely oxidised it is heated up for regeneration at 1100-1200 degrees C in an oxygen-lean atmosphere. Under those conditions and in the second process step, oxygen is set free from the redox system, so the metal oxide is being reduced and after completion of the reaction again capable for water splitting. Since the overall process consists of two core reaction steps, which need to be carried out sequentially in a reactor unit at two different temperature steps, a special process and plant concept had to be developed enabling the continuous supply of product regardless of the alternating nature of the solar reactor operation. The challenge of the process control is to keep the two core reaction temperatures constant and to ensure regular temperature switches after completion of the individual process steps, independent of the weather conditions, like DNI fluctuation, clouds and wind speed. Also start-up, the fast switching after completion of half-cycles and the shutdown must be controlled. State of the art is the manual switching of heliostats to fulfil those control tasks. This paper describes the development and use of a system model of this process. The model consists of three main parts: the simulation of the solar flux distribution at the receiver aperture, the simulation of the temperatures in the reactor modules and the simulation of the hydrogen generation. It can be used for the analysis of the operational behaviour. The model is intended to be used in the future for the control of the whole process. (C) 2011 Elsevier Ltd. All rights reserved.