Tort Law Considerations for the Hydrogen Economy

Article excerpt


Hydrogen and fuel cell technologies are being championed as the way to further environmental and national security policy objectives.1 The widespread enthusiasm accompanying these technologies is apparent in recent policy initiatives.2 For example, in his 2003 State of the Union Address, President Bush proposed $1.2 billion in funding to develop "clean hydrogen-powered" automobiles.3 In January 2002, Secretary Abraham of the U.S. Department of Energy (DOE) announced the Administration's plan to develop a hydrogen fueled "Freedom CAR."4 State governments5 and a number of industrial firms6 are also involved. These programs have created a fervor where potential obstacles to widespread deployment of hydrogen technologies to the general public are rarely discussed and liability implications are seldom considered. Furthermore, none of the current studies or policy initiatives on the hydrogen economy address tort liabilities that could be associated with the unique physical properties of hydrogen.

The unique physical properties of hydrogen lead to significant differences in the use, hazard detection, and likelihood of injuries as compared to other conventional or alternative fuels. Therefore, consideration of tort law issues associated with hydrogen, including negligence, products liability, and abnormally dangerous activity could impose additional operational costs as insurance premiums or in the management of potentially adverse publicity. These additional costs have yet to be considered in the discussion of the hydrogen economy.


The heart of the hydrogen economy is the fuel cell. Fuel cells are similar to batteries in that they chemically combine two reactants to generate electricity.7 In a battery, the reactants and reaction products are permanently contained. Some batteries can be recharged with an external power supply that drives the discharge chemical reactions in reverse. A fuel cell differs from a battery because it does not permanently contain either the reactants or reaction products.8 A fuel cell operates by pumping reactants into the cell, as reaction products are vented and electricity is generated. Fuel cells are not electrically rechargeable but, in principle, can operate indefinitely as additional reactants are supplied.

The reactants used in virtually all fuel cells are oxygen, obtained from air, and hydrogen, with water as the reaction product. A few experimental systems substitute methanol for hydrogen because engineering the reactant storage system for direct methanol fuel cells is simplified as methanol is a liquid at room temperatures.9 These systems, however, have lower current densities than hydrogen fuel cells and produce carbon dioxide (CO2). Furthermore, there are serious concerns that because methanol is water soluble, leaks in storage tanks associated with the distribution infrastructure could contaminate groundwater supplies.10 Some engineers have proposed hydrogen distribution systems based on chemical compounds such as sodium hydride.11 However, at least a few of these systems involve recycling procedures that are subject to the Resource Conservation and Recovery Act (RCRA), and are burdened with compliance costs for Environmental Protection Agency (EPA) regulations associated with the handling of hazardous wastes.12

Devices called reformers, designed to convert gasoline or methanol to hydrogen (and CO2) on demand, are being developed, because they take advantage of the existing energy infrastructure.13 However, it will be a challenge for those self-contained chemical plants, combined with already expensive fuel cells, to compete with conventional or alternative fuel systems (e.g., natural gas, batteries) in terms of cost, weight, or packaging volume.

Certain characteristics of hydrogen are attractive and widely touted. Unlike petroleum or other hydrocarbon fuels, hydrogen used in vehicles or stationary fuel cell applications produces only water vapor emissions. Furthermore, hydrogen can be obtained from domestic feedstocks such as natural gas and coal or from renewable sources such as wind or solar energy.14 However, hydrogen also differs from other fuels in ways that are rarely discussed. Both hydrogen gas and its flame are invisible and odorless.15 The heat radiation from a hydrogen flame is slight,16 and the ignition energy is more than an order of magnitude lower than that of gasoline and natural gas. In addition, the flammable composition of hydrogen in air is much wider than any of the other conventional or alternative fuels in use or under consideration for future use. Figure 1 compares the ignition and combustion properties of hydrogen to methane, the primary constituent of natural gas. Table 1 lists the ignitable and combustible properties of hydrogen and most common fuels.

Hydrogen requires more extensive handling precautions than conventional or alternative fuels because its physical properties differ significantly from those fuels. For example, the Occupational Safety and Health Administration (OSHA) regulations require a more stringent "Group" rating for explosion proof equipment (e.g., light fixtures, motors, thermostats, heaters, switches, and telephones) where hydrogen is handled than for areas designed for the safe handling of other fuels.17

There have been a number of cost studies on the hydrogen economy and infrastructure, but those studies focus on engineering issues (e.g., pipe, pump, or seal compatibility) or financial incentives to attract initial consumers (e.g., tax credits).18 Safety is rarely addressed in these analyses,19 and to the extent that it is, it is the severity of injury and not the likelihood or frequency of injury that is considered.20

Industrial experience has shown that 22% of hydrogen accidents are caused by undetected leaks,21 despite the standard operating procedures, protective clothing, and electronic flame and gas detectors provided to the limited number of specially trained hydrogen workers. With this track record, it is difficult to imagine how the general public can manage hydrogen risks acceptably. A widescale deployment of these safety precautions would be costly and public compliance impossible to ensure.22 Hydrogen advocates rely on the development of electronic sensors to guarantee safety.23 The National Aeronautics and Space Administration (NASA) hydrogen systems are equipped with the most sophisticated (and expensive) electronic sensor technologies.24 However, NASA safety engineers know that all such sensors are subject to malfunction or improper placement. NASA workers are therefore advised to use a "dry corn straw or sage grass broom" to supplement electronic flame sensors because the fail-safe broom "easily ignites as it passes through a flame."25 Brooms, used as sacrificial hydrogen flame detectors, are one of only four categories of detection technologies that the rocket scientists at NASA have identified.26

Hydrogen technology development programs focus on safety through the use of electronic equipment such as flame and leak detectors.27 Those programs do not rely on the development of the intrinsic safety features such as the visible flame and distinctive odor that are expected and required for other fuels.28 Furthermore, these programs do not consider liability costs associated with sensor or alarm failures, or situations where those devices could be ineffective. Liability and safety are, of course, separate and distinct concepts, as any car owner with an unblemished driving record is reminded every time she pays the liability insurance premiums.

In fact, leaks from hydrogen systems are already being reported in public settings. The most recent reported incident involved one of Toyota's fuel cell vehicles that had been leased to the Japanese Ministry of the Environment.29 That leak was detected by "a strange noise in the car when [the driver] was filling up the hydrogen tank," and not by onboard sensors or alarms.30 This leak occurred despite certification from the Japanese Ministry of Land, Infrastructure, and Transport that the vehicle was market-ready.31 Fortunately, no injuries were reported. Nevertheless, Toyota recalled all of its fuel cell vehicles, which were leased to Japanese cabinet ministries and to universities in California.32


There are three long established torts that should be considered in the context of possible injuries associated with the hydrogen economy; negligence, products liability, and abnormally dangerous activity.

Negligence refers to injuries that result from a breach of a duty of care.33 This duty includes situations where "the defendant['s] conduct create[s] a foreseeable zone of risk," in addition to where the defendant could foresee the occurrence of a specific injury.34 The duty can take the form of an omission or commission,35 based on what behavior would have been expected from a reasonable professional in the field.36 Here, courts would tend to look to industry customs memorialized in four decades of compliance with published engineering practices,37 and OSHA regulations,38 as indicia of what is reasonable behavior.39 Changes to codes or regulations, like those proposed by the DOE to reduce compliance costs,40 may not reduce negligence liability costs for injuries or property damage related to hydrogen accidents.41 In fact, the National Transportation Safety Board (NTSB), in its analysis of a fatal hydrogen tanker truck crash, recommended the imposition of more stringent safety regulations "to assure the protection of hazardous materials containers, and the protection of those who work in their proximity."42 The use of hydrogen sensors and alarms, such as those being developed by the DOE or those commercially available, would not have prevented this crash or the subsequent hydrogen explosions, and was not mentioned in the NTSB's recommendations. Although hydrogen leak and flame detection was not a factor in this accident,43 the properties of the fuel are likely to make emergency response more difficult for accidents involving hydrogen than other types of fuel.44

Strict products liability attaches to everyone normally engaged in the business of selling a product having a dangerous defect whenever personal or property injuries are incurred.45 Accordingly, courts conduct neither negligence nor fault analyses. Courts have held that products liability attaches even when the product was subjected to reasonably foreseeable misuse.46 The only issue for courts to decide is whether a dangerous defect of the product caused the injury. Courts have found dangerous defects to include inadequate warnings,47 which could impose products liability for imperceptible hydrogen leaks and flames that cause injuries.48

One proposal to warn hydrogen users of leaks is to add an odorant. However, this approach is technically infeasible because many chemical compounds, such as those added to natural gas, react irreversibly to 'poison' the catalysts, which are a major component of fuel cell electrodes.49 A court could also determine that a flammable gas such as hydrogen, which is not odorized in accordance with DOT regulations,50 harbors a dangerous defect, as the Colorado Supreme Court did for unodorized propane.51 Moreover, hydrogen is the lightest, therefore most mobile molecule in existence. Even in the unlikely event that a catalyst compatible odorant could be identified, the effectiveness of such an odorant would be severely compromised, because the diffusion rate could never match that of hydrogen.52

Abnormally dangerous activity liability,53 as with strict products liability, does not include negligence as an element. One factor a court must consider in its analysis of abnormally dangerous activity liability is an actor's "inability to eliminate the risk by exercise of reasonable care."54 The general public is not routinely exposed to abnormally dangerous activities, such as pile driving, storage of toxic gases, blasting with explosives, crop dusting with airplanes,55 the shipping of gasoline as cargo,56 or the piping of gasoline under a residential neighborhood.57 In the context of a claim based upon abnormally dangerous liability, courts recognize that everyday life involves all sorts of dangers. Therefore, courts consider several factors in determining whether the injuries were caused by a danger that was normal or abnormal. For hydrogen related injuries, this determination could include examination of the physical properties of hydrogen, such as imperceptible leaks and flames, extremely low ignition energy, and wide flammable composition in air. A finding of a normal danger requires the additional negligence determination of whether a sufficient standard of care was met,58 while a finding of an abnormal danger imposes strict liability without the need for further analysis. However, it is difficult to imagine how hydrogen use could be assured a normally dangerous classification through the relaxation of the existing regulations for hydrogen storage facilities (specifying tank placement, capacity, features, and the size of buffer zones), or alternatively, how additional regulations could lead to cost savings.59


Large quantities of hydrogen are currently handled by a limited number of specially trained and equipped workers in the food, glass, metallurgy, and petroleum refining industries. Engineering research on storage, production, and use technologies is ongoing to support these industries and the prospects for the hydrogen economy.6 Research to facilitate the development of the infrastructure should include rigorous economic analyses for deployment to the general public, including the relationship between insurance premiums, the potential for substantial liability costs, and tort law generally. Tort liability laws do not prohibit the use of hydrogen, but they do allocate the responsibility for injuries. These laws make it clear that firms should be prepared to financially defend claims for that responsibility through the purchase of liability insurance.61 This cost of doing business, however, is not included in any of the existing cost studies and would likely be higher for full-scale public deployment, as compared to a purely industrial setting. Consequently, an economic analysis of the potential liability costs of the hydrogen economy is relevant. Such an analysis would serve to fill a gap in the current public discourse and policymaking in this area.

Only one reported case directly addresses liability issues associated with hydrogen storage and transportation.62 However, other cases have addressed injuries sustained when hydrogen was produced as a byproduct of chemical reactions,63 as well as in batteries64 and balloons,65 and in a military setting.66 Nevertheless, there are a number of hydrogen related incidents and injuries that have been documented and are shown in Tables 2 and 3. One particularly troubling incident occurred in New Mexico in 2001. There, firemen were attempting to fill hydrogen balloons as a part of a Fourth of July celebration. One of the balloons exploded during filling, injuring a fireman and severely damaging a fire truck. If this is how safety professionals handle hydrogen, what can be expected from the general public?

Any liability analysis would likely consider potential benefits associated with the use of hydrogen as a fuel. However, recent studies suggest that hydrogen leaks, associated with full-scale deployment of the hydrogen economy, could adversely affect the environment. One analysis indicates that hydrogen leakage could lead to a "deeper, larger . . . , and more persistent . . ." atmospheric ozone hole. This could offset improvements expected from the regulation of chlorofluorocarbons67 and, as another study suggests, hydrogen may be a more potent greenhouse gas than CO2.68 A recent engineering report from the ABB, the Swedish-Swiss energy engineering conglomerate, indicates that approximately "1.2 to 1.4 energy units of valuable electricity, natural gas, gasoline etc. have to be invested to obtain one energy unit of hydrogen. . . . [M]ost of these source energies could be used directly by the consumer at comparable or even higher source-to-service efficiency and lower overall CO2 emission."69

Since the invention of the fuel cell by Sir William Grove in 1839, many performance and cost improvements have been made.70 What future performance and cost improvements are possible remain to be seen. One thing, however, is certain: with his experience as a practicing lawyer and judge,71 Sir William would not have ignored liability issues associated with the hydrogen economy. There is no reason why policymakers and technologists should ignore these issues today.

[Author Affiliation]

Russell May*

[Author Affiliation]

* Mr. Moy holds a Juris Doctor from Wayne State University Law School, as well as a Ph.D. in Chemical Engineering from the University of Michigan. The author is also a member of the Bars of Michigan and the District of Columbia and is a licensed Professional Engineer. He was formerly Group Leader for Energy Storage Programs and Project Manager for the design and construction of hydrogen storage and vehicle refueling facilities at the Ford Motor Company. This article is based on the author's work at the Ford Motor Company and at the Georgetown University Law Center. The views expressed in this article are those of the author and do not necessarily represent the position of his current or former employers.


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.