Overview | Summary | Volume I | Volume II | Volume III | Volume IV
This study examines how a hydrogen transition could plausibly unfold in four metropolitan areas: Boston, Denver, Houston, and Seattle, as well as in the USA as a whole. The study takes a full fuel cycle approach and seeks to answer three main questions for a transition to hydrogen in each of the five regions, namely, what are the energy savings? What are the carbon dioxide reductions, and what are the costs? The report is divided into four Volumes, as follows:
Hydrogen, an energy carrier that can be produced from domestic zero-carbon sources and can be consumed in zero-pollution devices, offers the hope that it can effectively deal with climate, environmental, and energy security concerns. Some have questioned, however, whether a large-scale transition to a hydrogen economy is more hype than hope. To them, a national transition to a hydrogen economy poses serious challenges, foremost among them being the highly inertial investments over the past century or more in conventional energy technologies and infrastructure based on petroleum, natural gas, coal, and electricity. Moreover, considerable research and development is still needed throughout the hydrogen supply chain - from production to final conversion – before one could be sanguine about its long-term sustainability prospects.
This study takes these perspectives as a point of departure in an analysis of a transition to hydrogen in the USA and in the cities of Boston, Denver, Houston, and Seattle. Is hydrogen the sustainable energy alternative portrayed by some, warranting aggressive action by decision makers to foster a smooth transition? Or, is it an ill-fated distraction that highjacks the energy agenda away from more promising strategies such as biofuels, renewable electricity, and large-scale equipment efficiency improvements? The study follows the full fuel cycle modeling approach to explore the energy and carbon implications of fuel choices and addresses three questions: what are the energy savings? What are the carbon dioxide reductions, and what are the costs?
A national transition to hydrogen would have to occur through the year 2050
to allow for a gradual stock turnover in vehicles, plants and equipment. We
explicitly address the notorious “chicken-and-egg” problem – i.e., there must be
a widespread hydrogen-refueling infrastructure before anyone will buy hydrogen
equipment, conversely there must be a demand for hydrogen before anyone will
invest in a hydrogen-refueling infrastructure - through seeding experiments,
demonstration projects and fleet risk-reducing incentives in the early to
intermediate years. Lastly, we explore hydrogen transitions in two contrasting
policy contexts: a “business-as-usual” (BAU) context where there is a continued
reliance on fossil fuels, and a “GHG-Constrained” (GHG) context where the
reduction of greenhouse gas emissions is a strong policy objective. We built a
new analytical tool, called H2M to model integrated hydrogen transition
scenarios relative to these policy contexts and calibrated the model to fuel
cycle emissions, hydrogen production costs & performance, and electric and
natural gas system characteristics embodied in the Energy Information
Administration’s National Energy Modeling System (NEMS).
Underlying our analysis are several major premises. First, since hydrogen
production requires the use of energy feedstocks, we tracked depletion of
reserves to ensure that sufficient levels are annually available without
exceeding available supplies and without triggering price shocks stemming from
approaching resource scarcity. Second, in response to the low and relatively
dispersed initial hydrogen demand in early years, on-site hydrogen production
units (electrolysis or natural gas reforming) form the initial basis for
hydrogen production, with a shift to large centralized facilities with pipeline
delivery once demand has increased considerably. Third, we assume that hydrogen
is mostly produced by fossil fuels (i.e., coal and natural gas) in the BAU
context,, consistent with future trends in the national energy sector; while in
the GHG context, it is mainly produced from renewable sources (biomass,
municipal waste, and remote non-grid-connected renewables), consistent with the
trends toward greater reliance on renewable energy.
In addition, we assume that even after a large-scale transition in which equipment stock is nearly fully replaced with hydrogen-consuming equipment, rural areas will continue to rely on more costly on-site production options due to unacceptably high costs associated with pipeline delivery of hydrogen. Even within relatively dense metropolitan areas, there are areas that are suburban, or even rural, in character, with spatially dispersed demand.
We model the cost-effectiveness of pipeline delivery by assuming a hydrogen demand density threshold of 240 kg H2 per day per square mile. Above this demand level, the costlier investments in central production facilities and transmission and distribution pipeline networks are offset by scale economies and the resulting delivered cost of hydrogen is lower than the on-site alternative. Finally, we assume future improvements occur in carbon capture and storage technology such that greenhouse gas emissions from central reforming units can be sequestered at reasonable carbon capture levels - 91% for coal and biomass, 85% for natural gas.
For both the BAU+H2 and GHG+H2 scenarios, overcoming the inertia inherent in
current energy infrastructure based on petroleum, natural gas, coal, and
electricity requires substantial investments in new infrastructure and
technology. Total national hydrogen demand by 2050 reaches between 120 in the
BAU+H2 and 190 million tonnes in the GHG+H2 scenario., most of this demand –
about 70% - is taken up by light duty fuel cell vehicles in the BAU+H2 scenario,
while cogeneration in the commercial and industrial sectors has a roughly equal
share as hydrogen demand in transport in the GHG+H2 scenario. Meeting these
levels of hydrogen demand requires substantial investments in new production and
delivery infrastructure, as summarized below.
Hydrogen infrastructure in 2050, USA
Significant reductions of greenhouse gas emissions can be achieved when using carbon sequestration from centralized reforming processes, especially when coordinated with national energy policy that seeks to contain economy-wide growth in greenhouse gas emissions for non-hydrogen consuming end uses, as shown in the figure below. Transitioning to hydrogen in the BAU+H2 scenario avoids about 400 million tonnes of CO2-equivalent across the full fuel cycle in 2050. A large-scale switch to hydrogen in the GHG+H2 scenario avoids nearly 800 million tonnes in 2050 – and over 2 billion tonnes relative to BAU conditions in that year.
Carbon Emission trajectories, USA
These carbon emission trajectories highlight the importance of the energy policy context. Against the BAU context, with its continued dependence on fossil fuels and notable absence of aggressive energy efficiency and renewable energy strategies, switching to hydrogen still leads to about 1 billion more tonnes of carbon released in 2050 relative to year 2000 levels. Against the GHG context, with its underlying commitment to large-scale improvements in energy efficiency and widespread use of renewable energy, switching to hydrogen leads to over 600 million less tonnes of carbon released in 2050 relative to year 2000 levels. Hydrogen transitions without carbon sequestration only highlight this point as the carbon reduction benefits would be roughly halved by 2050. If the uncertainties regarding the long-term security of sequestered carbon dioxide can not be adequately resolved, investments in carbon capture and sequestration technologies may not be warranted and in such a case, the energy policy context takes on even more prominence.
Annual delivered costs of hydrogen are shown in Figure 3. Embedded in these costs are the annualized capital costs (on-site and centralized facilities, and pipelines), fixed and variable operating and maintenance costs associated with operating hydrogen production facilities and pipelines, fuel costs associated with the hydrogen production fuel cycle, as well as incremental costs associated with electric sector expansion to meet new hydrogen loads. The national gasoline price in the BAU scenario is shown for contrast. By 2050, delivered costs of hydrogen converge to between $20 and $21 per mmbtu (or $2.3 to $2.4/kg of H2), roughly 1.5 times the assumed price of gasoline (see online materials for sensitivity analyses assuming gasoline prices that are two and three times higher). Delivered costs for hydrogen are volatile during the early years of the transition associated with new electric generating capacity, pipeline installation, and central hydrogen production facilities.
Delivered hydrogen cost trajectory, USA
We conclude from this analysis that hydrogen could be an important part of a national climate policy. But, hydrogen alone cannot fully address the problems of climate change – or air pollution and energy insecurity for that matter. Introducing hydrogen against a BAU context does not even match the benefits of the GHG context. Indeed, hydrogen is only as compelling as its feedstock supply: The benefits of hydrogen do not derive from its greater life-cycle efficiency, but rather from the prospect of using hydrogen to exploit clean, zero-carbon energy supply options. These conclusions hold equally true for the transitions we modeled at the city level in Boston, Denver, Houston, and Seattle.
In fact, in both the BAU+H2 and GHG+H2 scenarios, the use of hydrogen represents a step backward in the early years: Near-term hydrogen supply options have negative benefits because they rely on on-site production (electrolysis and natural gas reforming) where sequestration is feedstocks are mostly fossil fuel based. Long-term hydrogen supply options, which are able to yield the GHG benefits that make hydrogen attractive, are all centralized options.
Hydrogen also has stiff competition for the benefits it provides. There are other options such as electricity from renewable energy and biofuels that are strong contenders as well. Battery technologies for electric vehicles continue to advance. Ethanol, methanol and biodiesel, in particular, have the potential for making a significant contribution to total fuel supply.
Without a coherent national strategy whose objective is to foster an orderly
shift away from conventional energy carriers and toward hydrogen, it is
implausible to consider that markets will spontaneously transform and that a
hydrogen transition will happen. A coherent transition strategy would have to
include sustained investments in R&D, identification of niche applications,
targeted incentives, development of several transitional technologies,
incentivizing hydrogen demand, and ensuring policy consistency across sectors.
Finally, even with a concerted effort and a coordinated strategy to shift toward
hydrogen, an orderly transition that avoids serious economic disruption would
likely require a long period of transition. In the scenarios that we modeled,
the transition gradually unfolds over the course of roughly 50 years. A complete
transition could require at least this amount of time, and perhaps more to
overcome the many thorny technical questions (e.g., on-board storage of hydrogen
in vehicles), the shear inertia and scale of the existing energy capital stock,
and the chicken-and-egg problems inherent in the transformation of the energy
system. A staged approach is needed, whereby initial niche markets are exploited
before the broad market is accessible to hydrogen technologies, and hydrogen
supply grows in stride with hydrogen demand – all built off an underlying
national commitment to energy efficiency and renewable energy.