Integrating hydrogen into the UK energy economy

Abstract

Hydrogen is likely to become a major new fuel for transport and district or domestic-scale combined heat and power systems in the 21st century. Depending on the specific production technology, hydrogen can displace fossil fuels and limit or completely displace the production of carbon dioxide. It may also enhance energy system security through increased storage capacity and reduction in the need for energy imports. This paper reports preliminary results from a project funded by the Tyndall Centre for Climate Change Research into the environmental impact of selected pathways for producing, storing, and distributing hydrogen in the context of a range of possible future energy economies. Historical examples of large scale technological change will be used to demonstrate how the integration of hydrogen into the energy supply system will depend on a complex interaction of government regulations, corporate strategies, institutional factors, and required technological developments. It will be shown that there is likely to be no single optimum architecture for the hydrogen economy and this will be further emphasised in exploring possible transition pathways for the integration of hydrogen into the UK energy mix.

Full text

1
INTEGRATING HYDROGEN INTO THE UK ENERGY ECONOMY
A G Dutton1, J Watson2, A Bristow3, M Page3, A Pridmore3
1 Energy Research Unit, CLRC Rutherford Appleton Laboratory, Chilton, Didcot, Oxon. OX11 0QX, UK
Tel: +44 1235 445823, Fax: +44 1235 446863, E-mail: a.g.dutton@rl.ac.uk
2 Science Policy Research Unit, University of Sussex, Falmer, East Sussex, BN1 9RF, UK
Tel: +44 1273 873539, Fax: +44 1273 685865, E-mail: w.j.watson@sussex.ac.uk
3 Institute for Transport Studies, University of Leeds, Leeds, LS2 9JT, UK
Tel: +44 113 343 5325, Fax: +44 113 343 5334, E-mail: abristow@its.leeds.ac.uk, E-mail: mpage@its.leeds.ac.uk,
E-mail: apridmor@its.leeds.ac.uk
(all affiliated to Tyndall Centre for Climate Change Research)
Keywords
Hydrogen economy, Energy supply, Transport, Hydrogen distribution
Abstract
Hydrogen is likely to become a major new fuel for transport and district or domestic-scale
combined heat and power systems in the 21st century. Depending on the specific production
technology, hydrogen can displace fossil fuels and limit or completely displace the production
of carbon dioxide. It may also enhance energy system security through increased storage
capacity and reduction in the need for energy imports. This paper reports preliminary results
from a project funded by the Tyndall Centre for Climate Change Research into the
environmental impact of selected pathways for producing, storing, and distributing hydrogen in
the context of a range of possible future energy economies.
Historical examples of large scale technological change will be used to demonstrate how the
integration of hydrogen into the energy supply system will depend on a complex interaction of
government regulations, corporate strategies, institutional factors, and required technological
developments. It will be shown that there is likely to be no single optimum architecture for the
hydrogen economy and this will be further emphasised in exploring possible transition
pathways for the integration of hydrogen into the UK energy mix.
Various hydrogen production and distribution pathways will be developed and analysed,
particularly for their carbon dioxide reduction potential. In particular, the capacity for the
integration of water electrolysis systems associated with renewable energy power generation
will be explored and compared with the conventional alternative - steam reforming of
methane.
1. Introduction
The recent UK Energy White Paper [1] has adopted the target to reduce UK carbon dioxide
emissions to 60% of current levels by 2050, in order to mitigate the effects of climate change.
The introduction of hydrogen to displace conventional fuels for heating and transport is cited
as one of the possible strategies towards achieving this target. However, although hydrogen
can be produced by multifarious processes, the principal contenders for large scale production
are still likely to result in some emissions of carbon dioxide. For example, steam methane

2
reforming (SMR) will result in significant carbon dioxide emissions unless implemented
alongside a carbon dioxide sequestration strategy; electrolysis using grid electricity will result
in emissions elsewhere in the electricity supply network. The likely level of these emissions
over time and the long term prospects for the hydrogen economy to deliver sustainable
reductions in the time frame beyond even 2050 must be estimated in order to decide the
immediate priority which should be accorded to hydrogen within the overall carbon dioxide
reduction strategy.
The carbon reduction potential of introducing hydrogen into the energy supply infrastructure
depends on:
(i) the type of conventional capacity displaced,
(ii) the new plant required to supply and distribute the hydrogen,
(iii) the measures (if any) taken to limit harmful emissions associated with the hydrogen
production, and,
(iv) the end-use efficiency of hydrogen use.
The first three of these will vary between countries and even locally within any given country;
all four will vary with time.
For the analysis of longer-term developments of this kind in the energy system, conventional
forecasting techniques are severely limited. Whilst these techniques are a useful tool for
analysing trends over the short to medium term, the results of forecasts for periods longer than
a decade into the future are often unsatisfactory. As a result, there is an increasing trend
towards the use of scenario techniques to explore the uncertainties inherent in longer-term
explorations of the future (see, for example, [2]).
Scenarios are one way in which both governments and companies have sought to test the
robustness of current decisions by developing a range of possible futures. The aim is to identify
how likely it is that a particular development (in this case, the shift to a hydrogen economy)
will happen, and to identify some common trends that are important for this development
across a wide range of scenarios.
This paper will briefly review a widely used set of scenarios in the UK (section 2.1) and what
they might imply (section 2.2) for the development of a hydrogen economy up to and beyond
the year 2050. It will be shown that very different end-states (and hence transition pathways)
are likely depending on the prevailing geo-political framework.
A software model, THESIS (Tyndall Hydrogen Economy Scenario Investigation Suite), is
being developed to explore the transition pathways towards these possible end-states.
Ultimately this model will be used to assess the penetration and diffusion of hydrogen into the
transport fleet and the built environment, and the knock-on consequences for the rest of the
energy supply system (for example, build-rate and primary fuel-mix in the electricity supply
industry). As the database is still under construction, preliminary results are presented here for
2020 (section 3), based on published projections. These results clearly show the consequences
for the overall energy system of a relatively modest initial penetration of hydrogen.
2. Energy use scenarios and possible hydrogen futures in 2050
2.1. The SPRU/Foresight Contextual Futures Scenarios
A widely used set of scenarios within the UK are the contextual futures scenarios, originally
developed by the Science Policy Research Unit (SPRU) at the University of Sussex for the UK

3
Government Foresight Programme [2]. The SPRU/Foresight Scenarios have subsequently been
used in projects for other Government departments. Energy-specific applications of these
scenarios have included the Fuelling the Future study by the Foresight Energy and Natural
Environment Panel [3] and, more recently, the energy projections of the Cabinet Office
Performance and Innovation Unit (PIU) [4]. In this latter study, the PIU produced its own
results from the basic SPRU framework for the UK energy system in 2020 and 2050. Some
specific results from this study are presented later in this paper.
The starting point for the SPRU/Foresight Scenarios is the construction of logical and
consistent 'storylines' about the future. These storylines can be applied at a number of different
levels from sub-national to global. The scenario construction assumes that the whole world is
subject to the forces described within them. This is no more or less realistic than the
assumption that any one country will conform strictly to any one scenario over time.
The scenarios are defined around two main variables or dimensions of a largely qualitative
nature. The first dimension is values (individuals/consumers vs. community) and the second is
governance (autonomy vs. interdependence). This means that technology is not viewed as
autonomous and following its own independent path. It is rather viewed as dependent on the
particular combination of dominant values and governance systems being explored. This
allows the degree of success of different technologies and systems to vary according to
different scenario circumstances. When the two dimensions of the scenarios are combined,
they suggest four possible future states (see Figure 1).
WORLD
MARKETS
PROVINCIAL
ENTERPRISE
GLOBAL
SUSTAINABILITY
LOCAL
STEWARDSHIP
CONVENTIONAL
DEVELOPMENT
CONSUMERSISM
COMMUNITY
GLOBALISATION
REGIONALISATION
Figure 1 : The SPRU/Foresight Contextual Futures Scenarios
These four scenarios can be characterised as follows:
• World Markets combines an emphasis on the individual and consumer with a highly
interdependent governance structure; this world is dominated by private consumption and a
highly developed and integrated trading system.
• Provincial Enterprise has a similar emphasis on private consumption combined with a
more fragmented governance system, thus emphasising more local-level (i.e. national and
regional) and variable decision processes.

4
• Global Sustainability combines social and ecological values with a more interdependent
and collective governance structure, producing a strong world regime interested in dealing
with environmental issues.
• Local Stewardship has a fragmented governance structure, as in Provincial Enterprise, but
the predominant values are social and, to a lesser degree, ecological, rather than consumers
and individuals.
2.2. The SPRU/Foresight Scenarios and Hydrogen
This section of the report sets out the context for hydrogen production and use in the UK in
2050. The Tyndall Centre research team selected drivers and inhibitors to hydrogen in a
number of related categories including security of supply, environmental impact, quality of
life, extent of trade and competition, technical change and the policy environment. Each
category was then addressed in the context of the four SPRU/Foresight scenarios. The
elaboration of each scenario does not necessarily address each of these categories in detail.
However, they provide a guide for the kinds of issues that need to be covered to justify
subsequent quantitative judgements about the role of hydrogen in 2050. These scenarios will
be developed further and published as a Tyndall Working Paper [5].
Under the World Markets scenario, with its emphasis on low energy prices, security of supply
and the environment are regarded as secondary and there is very little government intervention
in markets. There is no incentive for a change in the fuel mix and hydrogen is only likely to
develop in specialist, niche applications, such as portable applications with fuel cells and
certain types of vehicle. Overall penetration is expected to be low, perhaps less than 5% across
all sectors by 2050, with SMR being the sole production route.
In Provincial Enterprise, governments are inward-looking and protectionist; energy and
resource policy is dominated by security concerns, so a shift to hydrogen could be expected as
UK reserves of gas and petroleum run down. However, innovation may be slow due to limited
import of foreign technology, although this may be countered by a stronger incentive to
develop a centralised hydrogen infrastructure. A penetration of 20-25% across all sectors is
expected by 2050. Coal gasification (though without carbon sequestration) and renewable
electricity are the most likely production routes.
Global Sustainability is arguably the most promising environment for the growth of a hydrogen
economy. Global economic and political systems are highly interconnected and social and
environmental goals are paramount. Energy prices reflect environmental externalities and the
goals of the Kyoto treaty have been successfully followed through with a strong international
emissions trading regime. Energy demand is considerably lower than in the previous two
scenarios, but generation diversity is low due to the adoption of a lowest cost approach.
Hydrogen emerges in local grids around areas of traffic and population density. By 2050,
hydrogen is firmly established (penetration, perhaps, 25-30%) in the Domestic, Service, and
Industry sectors as a fuel vector for fuel cell CHP systems and is starting to dominate the
Transport sector. Hydrogen is produced from renewables and nuclear power.
In Local Stewardship local and regional decision making predominate; energy and transport
decisions are driven by security, local environmental, and local employment issues. The Kyoto
framework has survived but is not so well regulated as in GS with varying degrees of
compliance. Long distance and international travel is reduced and innovation is comparatively
low. Hydrogen development is hampered by the lack of local energy sources to produce it, but

5
has established itself as the fuel of choice in the public transport sector by 2050. Renewables
are the dominant source of hydrogen.
2.3. Possible UK hydrogen futures in 2050
Any assessment of the possible impact of the hydrogen economy towards reducing carbon
dioxide emissions over a given time horizon must first estimate the likely emissions over the
time frame excluding the use of hydrogen and then evaluate the emissions assuming that
hydrogen is introduced. The SPRU/Foresight Scenarios present a useful framework for such an
assessment [6].
Figure 2 shows the historic energy consumption in the four main sectors (Domestic, Industry,
Transport, and Services) of the UK energy economy over the period 1970-2000 [7]. In the
World Markets and Provincial Enterprise scenarios, it is clear that the potential growth in the
Transport sector is likely to dominate energy consumption. By analysing recent trends in the
Transport sector only (Figure 3), it is clear that air and road travel have been and are likely to
remain the major growth sub-sectors.
Official projections of growth in the kilometrage of UK cars [8] suggest a likely increase from
around 400 billion kilometres per annum in 2000 to 550 billion kilometres per annum by 2030.
This corresponds to a growth in the car population from around 25 million to 33.5 million [9].
The problem for hydrogen forecasters is to predict the timing of penetrations of hydrogen
vehicles into the new car market and their subsequent diffusion through the population.
Kruger [10] has modelled the likely growth of the world market in hydrogen-powered road
vehicles. Assuming a baseline of 10,000 hydrogen vehicles in 2010, Kruger produces estimates
of hydrogen vehicle penetration assuming industry growth rates of between 10% and 40%.
Even in the latter case, the proportion of hydrogen vehicles in the population only becomes
significant after 2030, tending towards complete penetration by 2050. This is a high growth
rate, bearing in mind it must cover parallel growth in vehicle production lines, fuel cell
manufacture, and hydrogen production, storage, and distribution systems.
The major large scale options for producing hydrogen are currently considered to be:
(i) Electrolysis (ideally using low carbon electricity),
(ii) Steam methane reforming (SMR) (ideally with carbon dioxide sequestration).
The carbon dioxide reduction potential of any transition scenario depends on the production
route, the efficiency of the hydrogen production process, and the efficiency of the end-use of
hydrogen.
Typical state of the art conversion efficiencies are 69% for electrolysis and 81% for SMR.
It is estimated [11] that hydrogen consumption in a bivalent internal combustion engine will
require similar fuel energy input per km as for petroleum. However, the fuel cell is likely to
perform up to twice as well on typical urban driving cycles [11].
It has been claimed that hydrogen production, storage, and regeneration will facilitate the
wider penetration of renewable energy electricity by providing increased energy storage
capacity. In fact, the UK electricity grid provides sufficient buffering capacity that energy
storage is unlikely to be a general driver towards the hydrogen economy until renewable
penetrations exceed 20-30% [12]. In fact, if hydrogen for transport is produced using
electricity, hydrogen production will constitute an additional demand, not accounted for in
current industry plans.

6
Figure 2 : Historic energy consumption data in the UK economy by end-use sector [7],
compared with scenario projections [6] (and linear behaviour)
Figure 3 : Transport sector - trends in energy consumption (mtoe) by technology and
major fuel-type [7]
3. Modelling the introduction of hydrogen to 2020
3.1. The UK electricity generating mix
The potential carbon savings of any change in the electricity supply system must be assessed
against the likely generating mix. Such predictions have been complicated by the liberalisation

7
of the electricity market in the UK, which has resulted in grid operations in Scotland and
Northern Ireland being controlled separately from those in England and Wales, while national
energy statistics are collected for the whole of the UK. Figure 4 shows the plant mix1 existing
in 2000 according to UK national energy statistics (DUKES) [7].
UK electricity generation plant ( 2000)
22026
28302
6236
12242
2705
4513
1976
Gas
Coal
Oil
Nuclear
Renewables
CHP
Interconnectors
Figure 4 : UK electricity generating plant mix (MW) by capacity (2000) after DUKES [7]
The 2002/03 winter peak demand for England and Wales was 54.8 GW, representing 82% of
the installed capacity of 66.5 GW [13].
The current perceived overcapacity has arisen from historically high margins allowed by the
original state-owned Central Electricity Generating Board and the so-called “dash for gas”
following electricity liberalisation, as power generators installed high efficiency Combined
Cycle Gas Turbines (CCGT) [14]. This shift from predominantly coal to gas has placed the UK
on course to meet its Kyoto obligation to reduce greenhouse gas emissions to 12.5% below
1990 levels by 2010.
The introduction of the Labour Government’s New Electricity Trading Arrangements (NETA)
in spring 2001 led to a fall in wholesale prices by 20% (adding to a similar sized fall before the
system went live). These price falls have led to financial difficulties for some generators,
notably British Energy (who operate most of the UK’s nuclear power capacity), and
discouraged further investment in plant.
However, the UK faces a steep decline in generating capacity over the next 20 years, due to:
(i) retirement and decommissioning of most of the current nuclear capacity (around 11.0
GW or 16.5 % of capacity)
(ii) retirement of a significant proportion of coal-fired generation due to age and the
implementation of the EU Large Combustion Plant Directive (LCPD), possibly up to
the entire coal-fired generating capacity of 22.8 GW.
At the same time, the Government has a target for renewable energy to supply 10% of
electricity demand by 2010 and an aspiration for this to increase to 20% by 2020.
1 Dual fuel plant distributed evenly between Coal and Oil categories; Hydro, Waste, Wind included in
Renewables

8
Over the same period, the expansion of conventional demand is expected to be from around
343-345 TWh in 2000 to 387-408 TWh in 2020 [15]. Any electrical power plant capacity to
supply the Transport sector with hydrogen must be considered as an additional demand to this
conventional prediction. Since the UK’s current targets for renewable energy expansion are
generally considered ambitious (at least within Government departments), it is unlikely that
additional renewable capacity would be available to meet the additional demand from
transport, which would therefore most likely be met by CCGT generation (or directly as
hydrogen produced by SMR). This would not realise the expected environmental benefit from
using hydrogen. In fact, on a national aggregate basis, it is not clear when there would be
sufficient low carbon electricity to meet this demand, although high regional renewable energy
penetration levels might create local markets for hydrogen production.
The immediate options for an accelerated introduction of low carbon electricity are:
(i) accelerated renewable energy programme,
(ii) replacement of existing fossil fuel based (coal/oil/OCGT) plant with modern, high
efficiency CCGT capacity,
(iii) increased penetration of combined heat and power (CHP) systems,
(iv) staged programme of new nuclear construction,
(v) large scale clean coal or steam methane reforming, with large scale carbon/carbon
dioxide sequestration.
Strong objections exist to the last two options: there is likely to be widespread public
opposition to a revitalised nuclear programme (on grounds of nuclear waste and cost), while
technology for large scale carbon dioxide sequestration remains unproven (and the possibly
catastrophic implications of leakage are not well understood). However, the ability to
accelerate the installation of renewable energy capacity must also be considered doubtful given
the widespread difficulty in obtaining planning permission for renewable energy plant,
particularly wind farms, in the UK; the installation of 0.5 GW of wind capacity (which took 10
years) would have to be expanded at least one hundredfold. Offshore installations are likely to
improve the installation rate, but, from the point of view of carbon dioxide reduction, all this
electricity will best be used to displace existing fossil fuel based capacity.
The National Grid Co. publishes a 7 year plan, listing likely changes to capacity available to
the network in England and Wales to 2010 [13]. Projections of the UK electricity generating
mix to 2020 have also been made in terms of primary fuel in Energy Paper 68 [15] published in
2000. More recently, the Energy White Paper [1] has listed some additional targets and
aspirations. In the absence of any decision to re-instate the nuclear power programme and
following the existing strategy, the balance of demand in 2020 seems likely to be supplied by
natural gas. Figure 5 shows the authors’ best estimate2 of the likely UK electricity generating
mix in 2020.
The potential to develop a low carbon electricity economy in time to meet significant demands
for hydrogen fuel even in the time frame beyond 2020 therefore seem limited.
2 Figure 4 assumes that: the Government target for 10 GW of CHP and “aspiration” for 20% of electricity
demand to be supplied by renewables (shown as declared net capacity) are met; nuclear capacity is phased
out, as scheduled, to leave only Torness, Heysham, and Sizewell B; the coal mix declines to 13% of
production (Energy Paper 68 [15], projection CH); oil-fired generation is completely phased out; and
CCGT’s take up the balance of demand. Total capacity has been scaled from the capacity in 2000 according
to the predicted change in total electricity demand [15] over the twenty years (i.e. identical plant margin).

9
UK electricity generation plant (projection 2020)
42163
11732
3700
18049
10000 4600 Gas
Coal
Oil
Nuclear
Renewables
CHP
Interconnectors
Figure 5 : Projected UK electricity generating plant mix (MW) by capacity (2020), after
[15] and [1]
If hydrogen is generated from electricity therefore, it must be considered to have a carbon
dioxide penalty in the same way as it must if reformed directly from fossil fuels.
3.2. The carbon implications of various hydrogen production pathways to 2020
A software package, the Tyndall Hydrogen Economy Scenario Investigation Suite (THESIS),
has been developed to assess the carbon emissions implications of future energy and hydrogen
scenarios. THESIS accepts as input the energy demand for various sectors of society, accounts
for energy conversion and distribution losses (including those associated with electrical power
generation), and outputs the total primary energy required and the associated carbon emissions.
Sub-models have been developed for the introduction of hydrogen into the Transport sector.
Further sub-models are being developed to include CHP systems and hydrogen storage and
regeneration into the electrical grid.
Table 1 shows the projections of total primary energy supply and associated carbon emissions
for the year 2020 with a range of hydrogen production scenarios, assessed using THESIS. The
baseline year (H1) is taken as 1990 for comparison with Kyoto climate change targets. The
energy projections are based on the Central GDP Growth / High Energy Prices (CH) scenario
in the UK Government’s Energy Paper 68. The historic data shows a dip in carbon emissions
due to the “dash for gas” in the 1990s, as already discussed. The projection (F1), published in
2000, underestimated the primary energy demand and emissions for that year (H2), largely due
to an unexpected rise in natural gas prices, which led to an increased proportion of electricity
production from less efficient coal-fired power stations.
For purposes of comparison, it is assumed that 5% of Transport energy demand in the UK is
met by hydrogen in 2020 (since approximately half of the UK’s transport emissions come from
motor cars, this corresponds to a 10% penetration of hydrogen vehicles into the car population
by 2020, representing around 2.5 million vehicles). This may be considered an ambitious
target, but the conclusions regarding the relative merits of different supply routes remain valid
for a wide range of vehicle penetrations at this time horizon. Half the hydrogen vehicle
population is assumed to have internal combustion engines (at equivalent efficiency to petrol

10
engines) and half fuel cell drives (assumed to require 60% of fuel required by petrol engine due
to improved performance over a typical driving cycle). The production routes considered are:
(i) electrolysis at 69% efficiency (additional required electricity capacity provided by
CCGTs) [F3a],
(ii) electrolysis at 69% efficiency (additional required electricity capacity provided by
renewable or nuclear power generation) [F3b],
(iii) steam methane reforming (SMR) at 81% efficiency (without carbon sequestration)
[F3c],
(iv) steam methane reforming (SMR) at 81% efficiency (with carbon sequestration) [F3d].
Detailed scenario Primary hydrogen
source Total primary
energy supply
(1990 = 100)
Relative C
emissions3
(1990 = 100)
H1 Historic (1990) 100 100
H2 Historic (2000) 107 95.9
F1 EP68 projection CH
(2000) [15] 103 91.2
F2 EP68 projection CH
(2010) [15] 110 98.8
F3 Baseline projection
(2020) No hydrogen 114 100.8
F3a
5% transport
demand (2020) Electrolysis (increased
CCGT capacity) 116 102.3
F3b
5% transport
demand (2020) Electrolysis (increased
(renewables/nuclear
capacity)
118 99.0
F3c
5% transport
demand (2020) SMR (without C
sequestration) 114 100.5
F3d
5% transport
demand (2020) SMR (with C
sequestration) 114 99.2
F3e
Baseline projection
(2020) + enhanced
renewables capacity
No hydrogen 114 93.5
Table 1 : Projected energy and emissions scenarios for 2020 with various hydrogen
production pathways
Electricity is arguably the most convenient energy delivery vector. However, current
electrolysis systems are only around 69% efficient and any additional electricity generating
3 THESIS model

11
capacity beyond current plans is likely to be the cheapest and easiest to install, which is
expected to be CCGT technology. Under this scenario (F3a), carbon emissions actually
increase with the change in fuel type. Clearly it would be better to opt for direct production of
hydrogen from natural gas by SMR rather than utilise electricity generated in CCGT’s, but the
consumer will make their decision based on production cost (which would probably favour
SMR), availability and utility at the point of use (which might favour electrolysis), and
additional storage and distribution considerations. Only if the additional electricity for
electrolysis is generated using renewable (or nuclear) electricity (F3b) can substantial carbon
emission savings be made.
A very small carbon emission reduction (F3c) from the baseline (F3) could apparently be
achieved by using hydrogen derived directly from natural gas without carbon sequestration, if
the promised performance improvements from fuel cell vehicles can be realised. However, this
reduction is likely to be more than negated by improvements to the fuel efficiency of
conventional engines proposed by vehicle manufacturers for 2008 [16]. If the carbon from the
SMR process can be sequestrated (F3d) then carbon savings of around 1.6% can be made for
every 5% of vehicle population (note that no energy penalty has been added for the carbon
sequestration). The emissions savings are slightly less than with renewable electricity due to
distribution losses in the gas network. The technology for carbon sequestration remains largely
unproven and the concept is not yet widely accepted.
While the most beneficial hydrogen production route would seem to be by electrolysis from
renewable electricity, the final scenario (F3e) in Table 1 shows that the same additional
renewables power capacity required to produce the 5% penetration of hydrogen into the
Transport sector would result in a greater carbon emissions benefit if used to displace coal- and
then gas-fired generation and the Transport sector remained petroleum-based.
4. Hydrogen pathways to 2050
Work is proceeding to develop full hydrogen transition pathways to 2050 from the scenarios.
One of the crucial questions currently facing policy-makers is whether a hydrogen
infrastructure must be centrally planned and implemented, or whether the policy framework
should rather seek to encourage a more piecemeal development. Current developments at a
European level [17] appear to favour the first approach. However, as has been shown above,
the environmental drivers for the hydrogen economy rely on the supply of carbon-free
hydrogen and this cannot be guaranteed a priori. Rather, the hydrogen supplied is likely to be
derived from fossil fuel sources or as an additional demand on the existing electricity system
(thereby causing the retention for longer of carbon-based generation capacity); in either case,
the environmental benefit may be small or even negative. There is a danger that the gulf
between current energy supply and the very long term vision of a renewables-supported
hydrogen energy economy will remain so large that it becomes difficult to sustain the
continued investment required to install a centralised distribution system if other technologies
are seen to deliver earlier and larger carbon emission benefits.
But is the centralised approach to the development of a hydrogen infrastructure the best way to
proceed, in any case? Large technical systems, such as the electricity supply infrastructure,
tend to consist of a complex network of new and old technologies, bespoke equipment and
organisational relationships. Thomas Hughes, a pioneer in the analysis of large technical
systems, has identified three distinctive features of such systems [18]:
• they combine sets of technical (e.g. power stations, transmissions lines) and non-technical
(distribution companies, environmental laws) components,

12
• there are various horizontal and vertical interconnections between the components, which
means that changes in one component often lead to changes in others, and,
• they have a control component that sets out the way in which the economic and wider
social performance of the system is regulated; this control is exercised by management and
economic systems (e.g. wholesale power markets), technical systems (e.g. control
technologies) and regulatory systems (e.g. through regulators such as OFGEM).
These features have far reaching consequences for the operation and development of the
system. These consequences are particularly important for those wishing to make radical
changes to current technical systems, such as attempting to shift the UK energy system towards
the use of hydrogen as its primary energy carrier. Beyond the technical challenge, Hughes also
points out the powerful vested interests inherent in the existing system, which the new
technology is seeking to supplant. The hydrogen energy economy as it is usually conceived
represents a direct challenge to the current energy system. It calls for new technologies and
infrastructures, and also new relationships between energy suppliers and consumers (e.g.
through the expected deployment of fuel cell heat and power systems), new firms to supply
equipment, new modes of energy service delivery and new challenges for government
regulation. Faced with this wide ranging list of potential barriers, a transition to the mass use of
hydrogen seems to be an enormous challenge.
An important case study is the early development of the electricity industry [19]. Whilst
modern electricity industries are highly integrated with large numbers of co-ordinated
components and organisations, this state is the end result of decades of system development –
some carefully planned and some extremely chaotic.
The development of the electricity supply industry has shown that:
(i) Developing technical systems are characterised by complexity and, at times, an
apparent lack of rationality. They are not just a collection of new and old technologies
linked together, but incorporate new regulatory arrangements, new corporations,
entrepreneurs and financiers, where success does not depend solely on attractive
economics. The first electricity systems were not cost competitive, but were developed
anyway due to reasons of novelty, prestige and the preparedness to take risks, a
situation not unlike the current circumstances of the fuel cell industry.
(ii) New system growth is often full of uncertainty. The national scenario exercises, carried
out by experienced and influential energy experts from industry and academia, show
how much uncertainty there is about the future role of hydrogen – i.e. about whether or
not it has a future, and how it might be introduced. The case of the electricity industry
has demonstrated that the role of the State has its limits. Whilst governments can set up
new regulatory frameworks and market rules to shape developments, the transition path
to a new energy system cannot be predetermined or centrally planned. This transition is
likely to be accompanied by many unsuccessful technical experiments as well as
corporate and policy failures [20]. The key challenge for government is to be able to set
a framework that makes space for these failures, and openly acknowledges the role of
the unknown and unforeseen.
(iii) The growth of new systems is a combination of evolution and revolution. A new
system might build on the old (e.g. by transporting hydrogen through gas pipelines) but
may contain revolutionary elements (e.g. the concept of the fuel cell as a mobile,
distributed power generation system capable of being connected to the electricity
network at a time of need [21]). The more revolutionary the new system, the more it

13
will have to confront the entrenched position of existing systems [22], triggering the
defenders of the old system to fight back, innovate and reassert its dominance.
It could be argued from this particular case study that the hydrogen economy would be
stronger for having established itself through local, niche development than from being
imposed centrally in a misguided attempt to direct the pace of change. The most appropriate
role of Government might be to provide the appropriate policy framework and to ensure a level
playing field of standards and certification. However, it is difficult to make definite judgements
about the likely balance between ‘top-down’ government direction of hydrogen development
and more ‘bottom-up’ niche developments from just one historical case. The implications of a
range of possibilities which include both types of development will be analysed within the
ongoing Tyndall Centre project using the scenarios outlined in section 2 of this paper.
Conclusions
If the hydrogen economy is to be realised, it will need to establish itself against the background
of existing energy infrastructures and entrenched interests, which vary from country to country.
The ultimate development route will depend as much on the strength of international
agreements to reduce carbon emissions and the local implementation of those agreements as on
the technical merits of different technologies.
Since hydrogen is only a fuel vector, the carbon-reduction potential of any given transition
scenario depends heavily on the hydrogen production route. The leading contenders are usually
cited as steam methane reforming (SMR) and electrolysis. Although the cheapest production
route at current natural gas prices, SMR will provide only slight environmental improvements
unless associated with carbon sequestration technology, which is yet to be proven and the use
of which is ethically uncertain. However, the major alternative, electrolysis of water, can lead
to increased overall emissions unless the source of electricity is low or zero carbon. In the UK,
there will be insufficient low carbon electricity until all the current coal and natural gas
generation (67% of current capacity) is displaced and the status of nuclear power (15%) is
resolved.
A software package, THESIS, has been developed to assess the relative carbon reduction
potential of different hydrogen pathways in the context of the overall energy system.
It is clear, even from the preliminary results presented here, that the implementation of the
hydrogen economy is likely to require increased investment in the deployment of renewable
energy technologies, especially where uncertainties remain about the merits of carbon dioxide
sequestration and public opinion dictates that nuclear power is not an option.
References
1. Department of Trade and Industry (DTI), Energy White Paper, Our energy future -
creating a low carbon economy, DTI, 2003
2. Office of Science and Technology, Environmental Futures, Foresight Programme
(March 1999)
3. Energy Futures Task Force, Fuelling the Future, Foresight (2000)
4. Performance and Innovation Unit, The Energy Review, Cabinet Office (February 2002)
5. UK Hydrogen Futures to 2050, Tyndall Centre Working Paper, to be published at:
http://www.tyndall.ac.uk/publications/working_papers/working_papers.shtml

14
6. Performance and Innovation Unit, Energy Systems in 2050, Energy Review Working
Paper, Cabinet Office (2001), p. 22
7. Digest of United Kingdom Energy Statistics 2000, DUKES, DTI, 2001
8. Department for Transport, Focus on Public Transport Great Britain: 1999 Edition, see
http://www.transtat.dft.gov.uk/tables/1999/fpublic/fpublic.htm
9. National Road Traffic Forecast (1997), Working Paper No. 1: Car ownership:
modelling and forecasting
10. Kruger, P., ‘Electric power requirement for large-scale production of hydrogen fuel for
the world vehicle fleet’, Int. J. Hydrogen Energy 26 (2001) p. 1137-47
11. Wurster, R., Zittel, W., Hydrogen energy, published at the workshop on Energy
Technologies to Reduce CO2 Emissions in Europe, ECN, Petten, April 1994, available
at http://www.hyweb.de/gazette-e/.
12. Milborrow, D., Penalties for intermittent sources of energy, Working Paper submitted
to UK Cabinet Office Performance & Innovation Unit (PIU) Energy Review process,
2001 (see http://www.cabinet-
office.gov.uk/innovation/2002/energy/workingpapers.shtml, last accessed 22/05/2003)
13. National Grid Company PLC, Seven year statement, March 2003, available from:
http://www.nationalgrid.com (last accessed 23/06/2003)
14. Watson, J., The technology that drove the dash for gas, IEE Power Engineering Journal
11:1, p. 11-19, 1997
15. Department of Trade and Industry, Energy Paper 68, December 2000
16. Association des Constructeurs Europeens d'Automobiles/European Community, CO2
emissions from cars: The EU Implementing the Kyoto Protocol, 1998
17. High Level Group, Hydrogen energy and fuel cells – a vision of our future, HLG draft
report, European Commission, April 2003
18. Hughes, Thomas P., Networks of Power: Electrification in Western Society, 1880–
1930, Johns Hopkins University Press, 1983, further elaborated in Geyer, A. and
Davies, A., ‘Managing project–system interfaces: case studies of railway projects in
restructured UK and German markets’, Research Policy 29 (2000) p. 991-1013
19. Watson, J. The development of large technical systems: implications for hydrogen,
Tyndall Working Paper No. 18, available from the Tyndall Centre Internet site at:
http://www.tyndall.ac.uk/publications/working_papers/working_papers.shtml
20. Schumpeter, Joseph, Capitalism, Socialism and Democracy, Harper (1975) p. 82-85
21. Lovins, A., A strategy for the hydrogen transition, 10th Annual US Hydrogen Meeting,
National Hydrogen Association, Virginia, 7-9 April 1999
22. See, for example, Walker, W., ‘Entrapment in Large Technology Systems: Institutional
Commitment and Power Relations’, Research Policy 29 (2000) p. 833-846
Acknowledgements
The research reported in this paper was funded by the Tyndall Centre for Climate Change
Research in the UK.