A Novel Solar-Fossil Hybrid Power Plant

The Pacific Northwest National Laboratory (PNNL) is developing a promising solar-fossil hybrid power system for integration with a conventional combined-cycle power plant. The hybrid system uses concentrated solar power (CSP) from a parabolic dish to drive the endothermic steam-methane reforming (SMR) reaction (see Equation 1) in a reactor mounted at the focal point of the dish (Figure 5). The mixture of carbon monoxide (CO) and hydrogen (H2) resulting from the reaction, called syngas, has 27% more chemical energy than the entering methane. The syngas is then routed to the combustion turbine instead of using methane (natural gas) directly.

Equation 1: CH4 + H2O = CO + 3H2

The hybrid power system has several advantages that suggest promise for developing a cost-effective system. The parabolic dish concentrator is always pointed directly at the sun, which results in 50% more solar energy reaching the solar receiver on an annual basis, per unit of concentrator area, compared to fields of heliostats pointing to a receiver on top of a tower or parabolic troughs, the most commonly implemented CSP technology to date. This performance advantage is particularly important because the concentrator, with its mirrored surface, is typically the most costly component in a CSP system.

The parabolic dish provides the solar concentration necessary to reach the temperature (an estimated 800C) required to drive the reforming reaction forward. High heat and mass transfer rates made possible via small channels in the reactor allow a compact design to minimize cost and thermal losses. After recovering most of the thermal energy in the product stream by preheating the reactants, thermal losses are minimal compared to that experienced in other CSP piping systems. Best of all, capture of solar energy in syngas allows its conversion to electricity via a combined-cycle power plant. Collectively, these advantages translate into solar-generated chemical energy (i.e., the solar fraction of syngas chemical energy) that is projected to cost no more than forecast natural gas prices in the U.S. over the lifetime of the power plant, assuming a site with solar insolation typical of the Southwestern U.S.

PNNL has completed the first of three research and development phases planned in work funded by SolarThermoChemical LLC and the U.S. Department of Energy through its SunShot Initiative. Much of the first phase was spent fabricating and assembling the single dish system (shown in Figure 5). The system uses an early version of a parabolic dish concentrator developed by Infinia Corp. of Ogden, Utah. Infinia developed the concentrator to mate with its Stirling engine generators, but it works equally well providing concentrated solar energy to the PNNL steam-methane reformer.

5. SolarSMR test system in late afternoon. Source: PNNL

Phase 1 testing achieved solar insolation to chemical energy conversion efficiency as high as 69% for periods of several hours, which is believed to be a world record. One objective of Phase 2 will be to achieve similar or better performance for more extended operations. Developing control methods for automatic operation is another. Phase 1 activities also included computer simulations of the reactor, manufacturing cost studies of the reactor and recuperator (for preheating reactor feed stream with reactor product stream), and laboratory investigations of reaction catalyst durability and regeneration. The results of these activities have identified several design changes for improving performance and reducing cost that will be implemented in Phase 2.

Phase 2 will use the same concentrator as Phase 1, which is a quarter of the size of Infinia’s commercial-scale unit, but an improved reactor design. During Phase 3, the reactor and recuperator will be scaled up to match a commercial-scale concentrator. If all goes well, the system will be ready for commercial demonstration in 2016.

—Contributed by Daryl Brown (daryl.brown@pnnl.gov), a senior staff engineer at the Pacific Northwest National Laboratory.