474790 A Comparative Techno-Economic Analysis of Renewable Hydrogen Production Using Solar Energy

Monday, November 14, 2016: 2:40 PM
Mason (Hilton San Francisco Union Square)
Matthew Shaner, Carnegie Institution for Science, Stanford, CA, Harry A. Atwater, Applied Physics, California Institute of Technology, Pasadena, CA, Nathan S. Lewis, Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA and Eric W. McFarland, Chemical Engineering, University of California, Santa Barbara, Santa Barbara, CA

A technoeconomic analysis of photoelectrochemical (PEC) and photovoltaic-electrolytic (PV-E) solar-hydrogen production of 10,000 kg H2 day-1 (3.65 kilotons per year) was performed to assess the economics of each technology. The systems analyzed include: (i) two PEC systems that are defined by a single unit operation with inputs of solar power and water and outputs of
hydrogen and oxygen, and that are differentiated primarily by the extent of solar concentration (unconcentrated and 10x concentrated), (ii) two PV-E systems that are defined by two unit
operations, solar panels that convert solar power to electricity and proton exchange membrane (PEM) electrolyzers that convert electricity to hydrogen, and that are differentiated by the
degree of grid connectivity (unconnected and grid supplemented), and (iii) PEM electrolyzers supplied by grid electricity to serve as a commercially practiced benchmark.

For each system, a base-case design and bill of materials was developed and used to determine the untaxed, plant-gate levelized cost of hydrogen (LCH). Based solely on laboratory scale
demonstrations and thus likely optimistic projections for system costs, the unconcentrated and 10x concentrated PEC systems base-case capital expense and LCH values were $205 MM ($293
per m2 of solar collection area (mS -2), $14.7 W-1 H2) and $11.5 kg-1, and $160 MM ($428 mS -2, $11.5 W-1 H2) and $9.2 kg-1, respectively. Consisting of significantly more mature technologies, the non-grid connected and grid connected PV-E systems base-case capital expense and LCH values were $260 MM ($371 mS -2, $18.8 W-1 H2) and $12.1 kg-1, and $66 MM ($441 mS -2, $4.9 W- 1 H2) and $6.1 kg-1. The benchmark PEM grid electrolysis systems capital expense and LCH were $34 MM ($2.4 W-1 H2) and $5.5 kg-1, respectively. A sensitivity analysis indicated that, relative to the base-case, increases in system efficiency could effect the greatest cost reductions for all systems, due to the areal dependencies of many of the components. The balance-of-systems (BoS) costs were the largest factor in differentiating the PEC and PV-E systems.

Successful research and development, measured solely by achieving a system efficiency of greater than 20% within the current embodiments of solar H2 generators, is not sufficient to produce systems with LCH values comparable to that of fossil-fuel derived hydrogen. Panel mounting materials, labor, and other balance of systems costs, irrespective of the active materials, amount to LCH values in excess of current hydrogen and energy prices. For basecase PEC systems, a carbon tax of greater than $800 (ton CO2)-1 would be required in an unconstrained CO2 energy market for fossil-fuel derived hydrogen to reach price parity with these solar technologies.

Expected electricity prices from CO2-neutral nuclear and low-CO2 fossil fuel with CCS technologies are currently an order of magnitude lower than expected electricity prices from solar or wind systems with battery or fuel storage solutions that provide reliability similar to that of current base-load or dispatchable generation technologies. Given these low electricity prices, electrolytic hydrogen prices are also projected to be significantly lower than the solar hydrogen routes analyzed, requiring disruptive approaches to solar hydrogen generation relative to the present embodiments of the technology. Efforts to directly reduce CO2 photoelectrochemically or electrochemically could potentially produce products with higher
value than hydrogen, but many as yet unmet challenges include catalytic efficiency and selectivity, and CO2 mass transport rates and feedstock cost.

As compared to other, commercial, energy production and storage technologies, the solardriven technologies analyzed herein are limited by an intrinsically diffuse solar power density of
<1 kW m-2, as well as by a low capacity factor, of under 25%, for fixed-tilt panels in the optimal locations in the U.S., and by a conversion efficiency of under 20%. The capacity factor of
presently known solar-based energy systems is their fundamental limitation; any capital item used only 25% or less of the day will be at a disadvantage to capital used more effectively.
Efforts to increase the low capacity factor of terrestrial solar power, as well as re-designed solar installation schemes that significantly reduce the balance of system costs and increase the
efficiency to near the Shockley-Queisser limit without increasing the capital cost, will have the largest impacts on the economic competitiveness of the resulting technology implementations.
Radically new materials and system designs that achieve fully installed costs similar to simple material installations such as artificial grass are required to achieve the equally dramatic cost
reductions needed for solar power to compete with current generation technologies.

Extended Abstract: File Not Uploaded