Solving hydrogen’s catch-22

Efforts to introduce hydrogen-fulled vehicles have so far struggled. Patrick Meyer stresses the importance of distribution infrastructure.

It has recently become apparent that the fossil-fuel based transportation systems upon which the world relies are increasingly unsustainable in terms of energy, economic, environmental and social parameters. Indeed, some argue that we are now well beyond the limits of sustainable consumption of our energy resources. It has been claimed that the production of oil, upon which the transportation sector is almost 97 per cent dependent, peaked in 2005 and that we are now heavily dependent on a continually-decreasing stock (Deffeyes, 2008).

Further, although reserves of natural gas and coal will take longer to deplete, both are inherently finite and will eventually run out. Economically, we fair no better. The exponential increase in demand from China, India and other large developing nations has led to disastrous supply shortages, which are slated to become worse in the future (Tsoskounoglou et al, 2008). It is partly due to these emerging economies that we saw the price of oil skyrocket to over US$147/bbl in July 2008 (BBC, 2008). Although the price of oil has recently fallen, the current declines are probably due to overall market hardships and are likely to rise to new record levels in the next few years once the market regains its footing.

Environmentally, the resulting effects of oil consumption have been nothing less than a global catastrophe. Oil consumption and the subsequent release of greenhouse gases has many times been directly linked to climate change, which has led to observed changes in the cryosphere, hydrological systems, coastal zones, biological and agricultural systems, and has resulted in an increased frequency of natural disasters on every continent and through most oceans (IPCC, 2007).

Despite the fact that the US has traditionally had the strictest air quality standards in the world, about 30,000 Americans die annually from respiratory illness related to car exhaust (Hertsgaard, 1998). Viewing environmental effects through the economic lens can assist in gaining perspective: automobiles as a class cause an estimated US$93bn in health and environmental damage annually in the US alone (Smith, 2004).

Finally, in terms of social sustainability, oil may have the gravest negative effects as the acquisition and consumption of oil resources has led to wars that have killed hundreds of thousands of people, led to the displacement of great populations and fostered a greater level of social division amongst the rich and poor of the majority of populaces (Goodman, 2008; GPF, 2007; IBC, 2008). Researchers have mathematically demonstrated that the oil industry has strong worldwide antidemocratic properties, causing the greatest damage to economically-poor and oil-poor nations (Ross, 2001).

Clearly, the fossil fuel economy is not sustainable in terms of energy, economic, environmental, or social parameters. Fortunately, due to the aforementioned issues, it has been predicted that the oil era may be coming to an end, while many alternatives to oil have emerged in the past decades. In the transportation realm, these alternatives include biofuels, natural gas, electricity, and others. Hydrogen, however, has emerged as the most promising long-term solution.

Why hydrogen?

Hydrogen is the simplest element and most plentiful gas in the universe and is colourless, odourless and tasteless. There are many recognised benefits of hydrogen as a fuel, the most valuable of which is the fact that it has the highest energy content per weight compared to other alternative fuels: hydrogen holds 120.9kJ/g of energy whereas gasoline holds only 48.3kJ/g, ethanol 27.7kJ/g and methanol 19.9kJ/g (Larminie, 2003; Prasad, 2005). Like electricity, hydrogen is an energy carrier and must be produced from another substance. As such, it acts as a medium of storage and transmission of energy from other sources. There are a number of methods for producing and supplying hydrogen, including:
• steam reforming natural gas (or other hydrocarbons) in large centralised plants, then delivering liquid or gaseous hydrogen to filling stations
• supplying natural gas to decentralised filling stations and reforming hydrogen locally for immediate or later consumption
• creating hydrogen as a by-product in an oil-refining process and capturing that hydrogen for use
• producing hydrogen through electrolysis at a centralised plant or refuelling station (Prasad, 2005).

The fourth option is by far the most environmentally-sustainable method for producing hydrogen. This is because electrolysis needs nothing more than an electric charge, which can be obtained from totally renewable energy sources. The solar-hydrogen production pathway has gained particular attention lately as it relies on totally renewable, non-polluting solar-derived electricity to produce hydrogen in a sustainable, economical and socially-agreeable manner (Abdallah et al, 1999; Arriaga et al., 2007; Bockris, 2007; Kazim & Veziroglu, 2001).

Fuel cells are the preferred technology for utilising hydrogen because of their high efficiency (commercial demonstrations have reached up to 50 per cent) and cleanliness (water and heat are the only by-products). Further, when using hydrogen from renewable energy sources, fuel cells show advantages like low specific energy demand and low or no specific emissions in comparison to the conventional internal combustion engine (Felder & Meier, 2008; Linssen et al, 2003). There are numerous fuel cell options, but the proton exchange membrane fuel cell (PEMFC) has emerged as the most promising fuel cell for mobile applications.

A catch-22

Despite hydrogen vehicles’ emergence as a possible technology for addressing the long-term challenges associated with oil consumption, the mainstream implementation of a global hydrogen energy system, or a “hydrogen economy”, is faced with tremendous far-reaching challenges (Rifkin, 2002). The greatest of these is the creation of an efficient and affordable fuel production and distribution infrastructure while simultaneously developing the proper vehicle technology to complement it. This dilemma has received so much attention of late, that it has been dubbed the “chicken-and-egg” phenomenon in the alternative fuels literature.

In 2007, my co-researcher, James J Winebrake, PhD, and I set out to create a logical and scientific method for assisting policy makers seeking to tackle the vehicle-infrastructure phenomenon. The now completed work has been published in the academic journal Technovation and is currently accessible online, while it will be available in print in the coming months (see Meyer & Winebrake, 2008).

After conducting a detailed literature review, it was found that the vast majority of public policies aimed at spurring the development of the hydrogen economy have been heavily oriented towards vehicle technology and often neglect the necessary infrastructure to support such vehicles. Based on real-world experiences, such as those with the natural gas vehicle market in Canada reported by Flynn (2002), we hypothesised that public policies aimed solely at vehicles or solely at infrastructure would lead to severely-underdeveloped hydrogen economies and that overcoming such underdevelopment would require coordinated policy approaches aimed at vehicles, fuel, infrastructure and other industry components.

Simulating the solution

To explore our hypothesis, we created a systems dynamics model to better understand and evaluate the diffusion of hydrogen vehicles and infrastructure through a lens which assumes that the vehicles and infrastructure are complementary goods. The complementarity of the two implies that the infrastructure cannot exist without the vehicle and vice versa. We sought to prove this notion through scenario analysis.

Dubbed H2VISION (Hydrogen Vehicle and Infrastructure Simulator for Integrated and Operational Transportation Networks), the model makes use of system dynamics (SD) techniques to simulate the diffusion paradigm associated with hydrogen vehicles and refuelling infrastructure. The model explores:
• the fundamental dynamics of the vehicle-infrastructure, complementary goods phenomenon and long-term mainstream hydrogen technology diffusion
• consumer preferences regarding hydrogen fuel cell vehicles; the impact of convenience costs associated with the availability or lack of availability of refuelling infrastructure
• the role that policy makers, fleet operators, governments, and investors can play in spurring the early adoption of hydrogen technologies.

H2VISION is primarily a discrete choice analysis model which focuses on consumer choice and vehicle market share simulation. There are three sub-sections of the model, each dealing with particular areas:
1. vehicle purchases, life, and retirement
2. refuelling station market development
3. conventional and hydrogen vehicle markets.
The calculations for the most important component of the model, the vehicle markets, make use of linear utility functions originally developed by Greene (1994). In the model, whether or not a consumer becomes an adopter of a hydrogen vehicle is dependent on the cost per mile to drive the vehicle (determined by fuel price), the vehicle purchase price, and the refuelling station density (called “fuel availability” in Greene’s work). We call these terms “attractiveness factors”, each of which has a detailed definition provided in the Technovation article. H2VISION is constructed in a manner that any demographic area can be simulated. We chose to base our simulations on the approximate equivalent of greater Washington, DC. We assume an area of approximately 60 square miles, an initial population of approximately 565,000 people growing at about 0.6 per cent annually, and a vehicle ownership rate of approximately 0.8 vehicles per person.

Using the model, we explored four scenarios to illustrate the role of complementary goods (vehicles and infrastructure) in the hypothetical hydrogen economy. Through the scenarios we investigated the effects of early adoption of vehicles, early development of infrastructure, and fuel and vehicle costs on overall market development.

The first scenario focused almost entirely on infrastructure. It assumed that 20 hydrogen refuelling stations were constructed in a city-level market in which hydrogen vehicles were economically unfavourable. This scenario was meant to illustrate the influence of infrastructure-oriented legislation which may inadvertently overlook the impact that vehicle promotion can have on overall market growth. The selection of 20 hydrogen refuelling stations in our simulation area was done to represent a situation in which there would be far greater hydrogen stations than the current situation, but less than the current conventional vehicle-to-station ratio. Based on the demographics incorporated in the model, the construction of 20 hydrogen stations would represent a situation where hydrogen stations would be 13 times less accessible than conventional (gasoline) stations (based on the notion that there have been, nationwide on average, 0.92 refuelling stations per 1000 vehicles over a 10-year average). Thus, this scenario is really not very aggressive at all, but is reflective of propositions in the current policy arena.

It is shown through the scenario analysis that when hydrogen infrastructure incentives are pursued in a market in which hydrogen vehicles are considerably more expensive than conventional vehicles, the station availability is not enough to reduce the negative convenience costs associated with accessing those stations. That is, although the infrastructure is available, the average consumer does not view the fuel’s availability as reason enough to purchase a vehicle which complements that fuel. Thus, the scenario results in a situation where there is limited vehicle market penetration (only a small handful of vehicles are purchased) and the vehicle population collapses entirely in the long run.

Our second scenario focused on infrastructure and vehicles to explore the impact of a simultaneous coordinated approach towards both complementary goods. In the scenario, 20 hydrogen refuelling stations were constructed in a market in which hydrogen vehicles and their fuel were also economically incentivised. It is shown that the combined promotional effort towards both infrastructure and vehicles allowed vehicle adoption and infrastructure development to overcome market tipping points and, relatively quickly, reach a sustainable level. That is, this scenario resulted in the development of a hydrogen economy.

The third scenario focused on vehicles only, providing investments for hydrogen vehicles and fuel as well as an initial hydrogen vehicle procurement for fleets. In this scenario, where only one initial hydrogen station is constructed, we do see the emergence of the hydrogen economy – ie the emergence of both complementary goods – but only after about three decades (whereas the second scenario resulted in a near-instantaneous market development). The delay in economic development is due to the lack of initial infrastructure investment. One refuelling station is simply not enough to yield a rapid growth, but given the highly favourable economics of the vehicles, the station does manage to serve as the foundation for an eventual hydrogen economy.

The fourth and final scenario explored a situation in which there was a large bulk hydrogen vehicle procurement, but there were no other monetary incentives provided to hydrogen fuel or vehicles. Further, only one refuelling station was constructed in the first year of the simulation. This scenario resulted in a stagnant hydrogen market. That is, the small procurement purchase and singular station construction did not provide enough incentive to spur the early adopter population, resulting in an entirely crashed market by year 25 of the simulation.

There are many important implications of these findings. First, when attempting to promote hydrogen vehicles and their supporting refuelling infrastructure, here viewed as complementary goods, policies must be applied to both vehicles and infrastructure in order to produce the most optimal results. Our two scenarios which applied incentives to both vehicles and infrastructure brought about complete hydrogen technology saturation in the long-run. On the other hand, the two scenarios which applied incentives to only one of the complementary goods, resulted in a stagnant or underdeveloped hydrogen economy. Moreover, based on the second and third scenarios, it can be concluded that policies which lean more heavily towards infrastructure will have a more substantive effect on market growth compared to those that lean more heavily towards vehicles – but again, only if vehicles and fuel are also economically incentivised. The absolute bottom line is that investing solely in infrastructure or solely in vehicles will not yield adoption.

The road ahead

Some policy makers have begun to understand the complementary nature of hydrogen vehicles and infrastructure. For example, consider the New York State Hydrogen Energy Roadmap (Badin et al, 2005) which provides a route for New York to implement a hydrogen economy by 2020. The three-phase plan is well-balanced and includes research and development, demonstrations, expansion of fuel production facilities, development of numerous refuelling stations, and incentives for the development and purchase of hydrogen vehicles. According to the roadmap, incentives will be applied simultaneously to vehicles and infrastructure. We view this as well-written policy and look forward to the roadmap’s implementation. Elsewhere, attempts to introduce large-scale hydrogen transportation schemes have encountered significant complications. In California, numerous attempts to establish hydrogen refuelling stations have failed due to the high cost of hydrogen production and the overall lack of demand for hydrogen fuel (Kindy, 2008).

In one example, Pacific Gas & Electric Co was allocated US$1.5m in state funds to build a large-scale, retail hydrogen station in San Carlos, California, but then cancelled plans to do so. The site was intended to serve as a hub for hundreds of Mercedes-Benz fuel cell vehicles that were going to be leased in 2009, but now potential vehicle leasers have been left without a place to refuel (Kindy, 2008). The complete lack of coordination and failure to understand the complementarity of infrastructure and vehicles evident in this example is exactly what must be avoided if the hydrogen economy is to ever come to fruition.

Rifkin (2002) maintained that the development of the hydrogen economy will “finally end the long and barbaric reign of geopolitics and begin a new pilgrimage to create a lasting biosphere politics.” Indeed, he believes that the hydrogen economy serves as a route to solve the majority of the world’s energy-, economic-, environmental-, and social-related tribulations. Rifkin may be correct, but without well-designed public policies in place today which successfully build the foundations of tomorrow’s hydrogen economy, the hydrogen industry may crash well before it has a chance to get off the ground. Our work confirms that the multi-focus methodology of policies such as the New York State Hydrogen Energy Roadmap is not in vain.

The aforementioned examples of real-world hydrogen development, in combination with academic work such as that completed under this project, must be used as a guide for future hydrogen policies so that all strategies aimed at spurring the development of the hydrogen economy are well-rounded and incorporate both vehicle and infrastructure dynamics. Most importantly, through systems dynamics techniques, we have shown that incentivising only vehicles or only infrastructure will not serve to create a thriving hydrogen market. Only when vehicles and infrastructure are simultaneously promoted will policies have enough bearing to push humanity into the next era of energy.

Patrick E Meyer is a doctoral candidate at the Center for Energy and Environmental Policy at the University of Delaware, Newark, DE. He holds a BS in public policy and an MS in science, technology, and public policy from the Rochester Institute of Technology, Rochester, NY. Mr Meyer can be contacted at:
patrickmeyer@gmail.com