Development of Novel Sampling Devices for Analysis of Impurities in Hydrogen at Refueling Stations for Fuel Cell Vehicles

Hydrogen refueling stations for fuel cell vehicles are being demonstrated by a number of energy and industrial gas supply companies. The hydrogen dispensed at these stations must meet stringent quality standards that limit the allowable concentrations of contaminant species to extremely low levels. For example, SAE guidelines state that carbon monoxide levels must not exceed 0.2 parts per million (ppm). To meet requirements such as these, the hydrogen suppliers will have to certify the quality of the gas on a regular basis. Certification of impurities at such low levels is currently done in specialized analytical laboratories, requiring considerable time and expense.  Preliminary estimates suggest that the cost of analysis may add up to 10 cents to the cost of hydrogen (per kg). It is highly desirable to conduct these analyses at the hydrogen fueling stations using inexpensive analytical equipment that is accurate for measurements at higher concentrations.

One approach to achieving the needed analysis in a cost-effective manner is to enrich the impurities in hydrogen by taking advantage of the high pressure of the hydrogen gas available at the nozzle. This paper will report on the proof-of-concept and development of two enriching methodologies. The first process, based on the selective removal of hydrogen, enriches the impurities by removing the hydrogen from the sample through a hydrogen-transport membrane. The second method traps (by adsorption on a suitable sorbent) the impurities from a high pressure flowing stream, and then analyzes the impurities that are subsequently released (desorbed) at a low pressure.

We are using an experimental and modeling approach to prove the viability of these concepts. Laboratory units have been set up and tests have been conducted to enrich the impurities – N₂, CO, CH₄, H₂S, and CO₂ – in hydrogen. Both methods have been repeatably observed to achieve enrichment factors as high as 30X. It is projected that higher enrichment factors can be achieved by adjusting the operating parameters. The experimental results have been found to be consistent with the theoretical predictions.

The hydrogen membrane approach is theoretically simple and offers the potential for very high enrichment factors. The hydrogen-transport membrane is based on a palladium-alloy and the hydrogen transport occurs at elevated temperatures.

The adsorption approach uses inexpensive sorbents, such as activated carbon and zeolites, which have been shown to be durable over thousands of adsorption-desorption cycles in industrial settings. This process is simple, inexpensive, and does not require any temperature cycling. Compared with the hydrogen membrane method, this approach may be more limited in the enrichment factors achievable. This paper will describe the two concepts, experimental data showing the close correspondence with theoretical predictions, reproducibility of the data, and compare the relative merits of the two systems and their applicability for the desired analysis at the refueling stations.