Driving Commercial Advancements in PEM Electrolysis – from Lab to Plant Floor

Abstract

Proton exchange membrane (PEM) electrolysis has advanced significantly in the last 10 years, scaling from units in the sub 100 kW range to installed systems at 10 megawatts and greater. This technology, among others, is likely to play a major role in decarbonizing the industry, transportation, and energy sectors, as both the cost of electrolyzer technology and the renewable energy to power it have decreased. Still, the capital cost of electrolysis needs to be further reduced to economically replace hydrogen from steam methane reforming in very large applications, especially as capacity factor goes down to take advantage of the lowest cost electricity. There is also significant potential to decrease cost, as electrolyzer manufacturing is still relatively undeveloped.PEM electrolysis has been developed for many decades and is proven as a reliable, scalable technology. However, electrodes are overdesigned due to the manual methods used to produce them, as well as the aerospace legacy of these systems. Many advancements such as reductions in catalyst loadings, use of thinner membranes, novel porous transport layers, and alternate cell configurations have been shown to be feasible in the lab, but have been slow to transition to commercial products.There are many factors in this lag, including the need to often change manufacturing methods when moving from the lab to a product environment. While this effort is sometimes viewed as “just engineering”, there is significant fundamental understanding required to translate a slow, manual process where finished parts can be individually scrutinized, to a fast, automated process that has to include automated inspection as well. For example, a catalyst ink that is hand painted onto a substrate will need different properties and formulation to deposit the catalyst by spray printing, or yet different properties for other methods such as slot die or gravure printing. Determining acceptability of the resulting parts can no longer rely on human judgment but must be well enough understood that the most important properties can be quickly measured and analyzed in real time. In addition, the scale over which high quality needs to be achieved increases by orders of magnitude when moving from lab scale to product scale. This talk will discuss these types of challenges and approaches to solve them.