Hydrogen Detection Considerations
by Karen Hall
Vice President, Technical Operations of the
National Hydrogen Association

On April 29, 2004, the NHA held a Hydrogen Safety, Codes and Standards Workshop in Hollywood, California. One topic discussed was public perception regarding hydrogen detection. In the natural gas community, odorants are used so that any gas escaping can be detected through smell. This is possible because mercaptan odorants, used in natural gas, track well with natural gas. In addition, natural gas is used in homes, where it is burned to provide heat. The addition of mercaptan odorants is not a reasonable solution for hydrogen that will be used in fuel cells. There are several reasons for this:

In fuel cells, hydrogen is not burned. It is reacted chemically in the fuel cell which contains sensitive membranes that can not tolerate high levels of impurities, such as mercaptans. It should be noted that some uses of hydrogen involve burning it in an internal combustion engine. This application would not be susceptible to poisoning. But does it make sense to have multiple forms of hydrogen, some with odorants and some without? Are odorants the best choice of hydrogen detection technologies available today for hydrogen use?

Many early uses of hydrogen fuel cells will not be in the home. They will be businesses and, of course, in vehicles. Therefore, if an odorant could be found that did not poison the fuel cell, would a nose be present to detect the escape of hydrogen in these applications? This concern is also true for vehicles using hydrogen with an internal combustion engine.

The workshop participants discussed the issues relating to odorants and more sensitive electronic detection methods, such as detectors and sensors. One concern was that the public may not accept a method of detection for hydrogen that is different than that used for natural gas. Yet the group also agreed that any odorant for hydrogen should have a unique smell, so that it was immediately recognizable as hydrogen. One participant even suggested the smell could be pleasant, and needn’t contain sulfur.

It was noted by one participant working with a utility that when a natural gas customer informs the utility that they are unable to smell the odorant in natural gas (this can be a result of damage to the sense of smell, which can happen for a number of reasons, and is also a natural part of aging), the utility will provide an electronic detector for the customer. The detector is a kind of electronic nose, and will indicate the presence of the natural gas.

It was also noted that the public is familiar with electronic detectors, as they are commonly used today to detect smoke and carbon monoxide.

So with the familiarity issue addressed, the group then addressed the potential to use a non-sulphur odorant in hydrogen for fuel cells. None of the participants were aware of any odorant being successfully used for hydrogen fueling a fuel cell. The group noted that in a fuel cell vehicle, the fuel cell does not consume the odorant. This results in the odorant collecting at the tailpipe. This could easily result in a false positive indication – smelling the odorant although there is no hydrogen escaping. In addition, as hydrogen molecules are so much lighter than air, an odorant that tracked with the hydrogen would need to be found.

Other concerns raised by the group with the possible application of odorants for hydrogen were the fact that this low-tech method of detection is not quantifiable; i.e., you may smell the odorant, but you do not know the severity of the leak. In addition, sleeping occupants in a home are unlikely to smell the odorant if the problem is occurring where early public hydrogen use is most likely, in the garage. Odorants could therefore give a false sense of security in hydrogen applications.

In today’s world of high-tech then, why base hydrogen detection on the low-tech solution used successfully for decades by the natural gas industry? Odorants make sense in natural gas applications, but really do not meet the need for hydrogen fuel cells.

Detectors and sensors provide design options for the manufacturer. Placement of these items can be based on good engineering practices. In addition, these high-tech solutions provide more information. They can include monitoring systems and alarms. They can perform leak characterization. Besides, the discerning public likes high-tech.

There is currently a concern about the cost of hydrogen sensors and detectors. The U.S. Department of Energy recently issues a Request for Proposals for the development and demonstration of hydrogen sensors. This is the latest of a set of programs to develop low-cost hydrogen sensors for the developing hydrogen energy economy. The ongoing work is likely to result in cost-reductions. In addition, several workshop participants suggested that the total system cost must be considered. Use of more sensitive and reliable equipment may reduce costs elsewhere. For example, are sensors more expensive than equipment to remove the odorant for use in the fuel cell?

This summarizes the workshop discussion. The National Hydrogen Association would like to encourage the continued discussion of hydrogen detection methods. Anyone wishing to add anything to the discussion is asked to email me at khall@ttcorp.com. Are you aware of odorants that do not poison the fuel cell, have a distinct odor, track with hydrogen, and do not collect at the tailpipe of hydrogen fuel cell vehicle? Would you like to express an opinion on the public acceptability of electronic detection systems for hydrogen? Do you have any other questions or contributions to the discussion?

Thanks to all who participated in the workshop for addressing these important issues.