The feasibility of distributed hydrogen production from renewable energy sources and the financial contribution from UK motorists on environmental grounds

Accepted Manuscript Title: The feasibility of distributed hydrogen production from renewable energy sources and the financial contribution from UK motorists on environmental grounds Author: Geoffrey D. Southall Anshuman Khare PII: DOI: Reference:

S2210-6707(16)30088-9 http://dx.doi.org/doi:10.1016/j.scs.2016.05.009 SCS 424

To appear in: Received date: Revised date: Accepted date:

29-3-2016 18-5-2016 19-5-2016

Please cite this article as: Southall, Geoffrey D., & Khare, Anshuman., The feasibility of distributed hydrogen production from renewable energy sources and the financial contribution from UK motorists on environmental grounds.Sustainable Cities and Society http://dx.doi.org/10.1016/j.scs.2016.05.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The feasibility of distributed hydrogen production from renewable energy sources and the financial contribution from UK motorists on environmental grounds

Geoffrey D [email protected], Anshuman Khareb*[email protected]

a

University of Liverpool

b

Athabasca University

*

Corresponding author. 201 - 13220 St. Albert Trail, Edmonton AB T5L 4W1, CANADA. Tel.: +1 780 5097533; fax: +1 780 4882832

1

Highlights Decarbonisation of transport sector would serve to meet climate change obligation. Examines case when hydrogen is produced from renewable energy sources for HFCVs. Business case depends on production of hydrogen at an acceptable price to public. Paper presents the infrastructure costs and attractiveness to the public. Certain infrastructure configurations offer acceptable hydrogen production costs.

2

Abstract Decarbonisation of the transport sector in the UK would serve to meet climate change obligations through greenhouse gas reductions, reduced pollution levels and an improvement of energy security through a reduced dependency on imported crude oil and refined fuels. Hydrogen fuel cell vehicles have the potential to fulfil these aims, more so when the hydrogen used to power them is produced from renewable energy sources. The electrolytic production of hydrogen at the point of sale eliminates the need for hydrogen distribution costs and promotes sustainability but the business case is dependent upon the production of hydrogen at an acceptable cost to the public. This study calculates the hydrogen production cost using varying capacity hybrid wind/solar PV systems and associated hydrogen generation, storage and dispensing technologies in supplying a scenario population exposed to average UK weather conditions. It examines if the environmental benefits are evident through the UK car-owning public’s willingness to pay for a cleaner transportation fuel. Whilst the respondents’ concerns for the environment are strong, a high degree of sensitivity over fuel pricing remains. In spite of this, the projected demand for hydrogen fuel cell vehicles through to 2030 and the use of renewable energy tariffs show that certain configurations of hydrogen production infrastructure are still financially viable. The possibility exists for the application of fuel taxation whilst still maintaining price parity with conventional hydrocarbon fuels. Keywords: Transport sector; Hydrogen fuel cell; Renewable energy; Hydrogen production infrastructure; Clean transportation fuel

3

1. Introduction Hydrogen is the simplest and most abundant chemical element in existence. It is the power source of stars and here on our Earth it forms a component of chemical compounds which humankind has been able to harness as fuels. Organic compounds make up traditional fossil fuels and one hydrogen-containing inorganic compound, water, can be utilised as a source of hydrogen which in itself offers huge potential for use as a transportation fuel. A transition to hydrogen fuel offers a number of societal benefits which include a greater degree of energy security, a reduction in pollution levels and a drop in greenhouse gas emissions (Hoffmann, 2012). Addressing each issue in turn, the successful propagation of hydrogen as a transportation fuel would result in a reduced dependency on crude oil for the production of petrol and diesel fuels, a large proportion of which is located in politically sensitive regions of the world, subjecting it to price volatility and resulting in economic and political pressures at home. Used in its cleanest form the only by-product of hydrogen when used in transportation is water, following its reaction with oxygen and resulting energy release. This means exhaust pipe pollutants CO2, NOx and SOx, typically associated with combustion engine (CE) vehicles have the potential to be eliminated, leading to a reduction of inner-city pollution levels. The degree of pollution is however dependent upon the means by which hydrogen fuel is consumed. Direct combustion in a manner similar to petrol and diesel engines still produces NOx compounds due to the high temperatures involved (Sørensen, 2012). Hydrogen fuel cell vehicles (HFCVs), using hydrogen to produce electricity and power a vehicle battery, emit only water as the byproduct thereby reducing overall inner-city NOx levels. Thirdly, on a global scale, a lower level of greenhouse gas (GHG) emissions is possible when compared to using conventional hydrocarbon fuels in transportation. The degree of reduction is dependent upon the method of hydrogen generation employed as some production processes generate GHGs as by-products thereby offsetting some of the gains through its use. The UK Government, through enacting the Climate Change Act, has committed to reducing the level of CO2 emissions to 80% lower than the 1990 levels by 2050 (Great Britain, Climate Change Act

4

2008) and hydrogen could play a key role in achieving this target. In this context, Spataru et al. (2015) utilised 48% of non-commercial vehicles in the UK being hydrogen fuelled as a means of achieving this target. Like gasoline, hydrogen is not a primary energy source i.e. it must be created before it can be used, unlike other sources of energy such as coal and natural gas. Hydrogen and gasoline are therefore commonly referred to as energy carriers. Described by Bartels, Pate and Olson (2010) the current main industrialscale methods of producing hydrogen are either by the steam reforming of methane (SMR) or the gasification of coal, both of which emit CO2. A third method, the electrolysis of water, accounts for 4% of hydrogen production (Ngoh and Njomo, 2012). The electrolytic generation method involves using electricity to split water into hydrogen and oxygen in an electrolysis cell. Reactions occur at both the positive and negative electrodes when a potential difference is applied. At the positive electrode oxygen is liberated and at the negative electrode hydrogen is formed (Sørensen, 2012). Whilst the actual production process of hydrogen through electrolysis produces no GHG emissions, the means by which the generation of the required electricity occurred will have an influence on the overall environmental impact. Electricity generation from fossil fuels produces GHGs whereas generation from renewable energy sources (RES) offers perhaps the best means of environmentally sustainable hydrogen production with no GHG emissions from direct operation. The widespread use of hydrogen as a vehicle fuel currently requires significant supply chain development due to the virtual absence of HFCVs from the UK’s road network. Whilst commercial scale hydrogen production already exists, the development of the distribution network to the point of sale is dependent upon the sale of hydrogen vehicles leading to what Kriston, Szabó and Inzelt (2010) term a “chicken and egg problem”. Addressing this issue, the UK industry and government formed a coalition to produce a roadmap for the introduction of HFCVs on the UK’s roads. It proposed siting the initial filling stations in major cities and connecting roads in the early years through to full population coverage by 2030

5

(UKH2Mobility, 2013). One means of assisting this in the short-term until the larger scale infrastructure is developed could involve producing hydrogen from locally distributed electrolysers within the towns and cities. The principle materials and utilities, water and electricity, already being widely distributed. Such an approach would contribute to regional sustainability. From a business perspective, the competitive strategy of distributed hydrogen production using RES could be argued as providing a source of hydrogen fuel with the lowest direct carbon footprint possible (albeit at possibly a higher cost than other means of hydrogen production) and from a supply chain design perspective, this research would seek to answer the question of whether the distributed production of hydrogen by RES has the potential to be a viable future supply chain configuration for the UK, subject to local variations in RES availability and whether the environmental benefits offered are acceptable to the UK public by means of a willingness to pay (WTP) a price premium for hydrogen fuel over conventional petrol and diesel fuels. Insights into the viability of such a supply configuration could be gained by examining environmental and government fiscal policy towards the taxation of hydrogen fuel and whether a subsidy would be required for hydrogen produced from RES to be competitive with conventional hydrocarbon fuels. Specifically: 1) Is the importance of environmental concerns reflected in the WTP amount? 2) The determination of a possible taxation level on hydrogen fuel. The objective of the research is therefore to investigate the extent to which the UK public is prepared to pay for a product that offers greater environmental sustainability.

2. Literature Review Despite the hydrogen supply chain lacking in current physical infrastructure much academic research has already been performed pertaining to future supply chain design in order to meet the market demand for HFCVs. A key requirement for the successful transition to hydrogen as a large-scale vehicle fuel is being able to successfully match the emerging hydrogen demand with emerging infrastructure (Melendez and

6

Milbrandt, 2008). This is commonly achieved through the use of mathematical modelling techniques to determine optimal supply chain configurations to meet service requirements at the lowest cost performed on different spatial scales of hydrogen infrastructure. Winskel et al. (2009) in utilising the Markal linear programming model to show the prospects for hydrogen as a future vehicle fuel for buses and cars in the UK running through to the year 2050 and basing the modelling on vehicle lifetime, energy efficiency and capital and maintenance costs with assumptions made for how these costs may change in future. Vehicle penetration levels form the output from the model and so the likely requirements of an aggregated national vehicle hydrogen demand can be made. Modelling performed at a regional level has been used to determine the optimal layout of hydrogen production, storage and transportation infrastructure with which to meet the national hydrogen requirement at the lowest cost. Almansoori and Shah (2006) have performed a number of studies on hydrogen infrastructure applied against Great Britain. Their initial model being further developed and subsequently refined and expanded in later papers. Almansoori and Shah (2006) first presented a single framework for the production, storage and distribution of hydrogen and developed a supply chain based on optimised infrastructure and operational costs. Being a snapshot model of an established hydrogen market it does not consider the development of the infrastructure over time as the number of HFCVs increases. This initial model was refined further in Almansoori and Shah (2009) to include the availability of hydrogen feedstock materials and the scale of the hydrogen production and storage facilities brought about by modelling the infrastructure over five time periods from 2005-2034 ranging from 5% market share in the first period to 100% in the fifth. Almansoori and Shah (2012) in their latest research introduced a further spatial scale to their model giving consideration to filling stations within a particular geographic area in terms of number, operating cost and the logistics used to replenish them with. Hydrogen demand uncertainties relating to HFCV uptake were also included in the optimisation model with each iteration becoming more comprehensive. Infrastructure models that focus on lowest cost such as those previously discussed, the source of hydrogen in virtually all the scenarios tended to be from SMR which as previously demonstrated is not the most

7

environmentally friendly production method available. Sabio et al. (2012) modelled the optimum future Spanish hydrogen supply chain from both a cost perspective and also considering multiple categories of environmental impact both relating to the ecosystem and human health caused as a result of the supply chain’s operation. In a shift away from the work of Almansoori and Shah, a greater inclusion of electrolytic hydrogen from RES was made in order to lower GHG emissions. Consideration of using mixed RES to produce hydrogen for vehicle refuelling stations has been made by Dagdougui, Ouammi and Sacile (2012) through modelling the hydrogen requirements at various market penetration rates using a cluster of actual petrol refuelling stations in Italy with a local hydrogen production and storage facility serving their requirements. In this case, assumptions were made relating to the selling back of excess electricity to the grid along with an industrial outlet for excess hydrogen produced and furthermore last mile distribution costs were not considered in the model. As the use of existing gasoline refuelling infrastructure can be used as a basis for the future supply of hydrogen a logical question therefore could be whether the existing gasoline infrastructure could be utilised to supply hydrogen? Katikaneni et al. (2014) carry out a techno-economic assessment relating to the on-site production of hydrogen by steam reforming existing gasoline station fuels compared to on-site electrolysis and pipeline hydrogen from centralised SMR production. The resulting production costs based on a hydrogen production volume of 1000kg per day were almost $12.5 USD/kg for on-site electrolysis and between $6-7 USD/kg for the hydrocarbon-based production methods. The total GHG emissions however were comparable to or (in the case of diesel) actually exceeded those of the CE equivalents. Decentralised hydrogen production by electrolysis has been studied in a techno-economic assessment by Prince-Richard, Whale and Djilali (2005), using technological aspects along with local electricity prices and fuel taxation levels in the cities of Vancouver, Los Angeles and Paris. It was found that the technical aspect was not as significant in affecting the competitive position of hydrogen as fuel taxes and electricity prices. In Vancouver, the low electricity price makes electrolytic hydrogen competitive with gasoline and in Paris the high levels of fuel tax also promotes the competitiveness of hydrogen. Los Angeles with its low fuel taxation level and high electricity cost makes hydrogen uncompetitive against gasoline. This research complements this existing 8

work and applies it to the UK with the further inclusion of RES in a decentralised hydrogen production configuration. Infrastructure models across these different spatial scales, whilst serving to effectively match supply and demand, are subject to approximations and fail to account for other factors that would affect the actual supply chain configuration deployed. Agnolucci and McDowall (2013) posit for national scale hydrogen infrastructure models human behaviour is inadequately considered with the assumption that people are both informed about the technology and also little or no risk aversion to new technology exists, resulting in an overstated vehicle penetration rate. For regional scale models a general assumption is made that demand is uniform across regions which in reality is unlikely to reflect the reality of hydrogen demand and therefore its resulting infrastructure. Local scale models would benefit from factoring in low utilisation rates in the early stages of hydrogen deployment to provide comfort to early adopters not having to drive an excessive distance to the nearest filling station. In line with the reduction in GHG emissions that accompanies using renewable sources of electricity, Balta-Ozkan and Baldwin (2013) optimised the hydrogen supply chain for the UK’s total transport requirements i.e. automobiles, goods vehicles, rail, bus and air travel to achieve the transportation sector’s contribution to the overall 80% reduction in the 1990 level of CO2 emissions by 2050. This model considers the import of hydrogen with which to partially meet these requirements alongside the domestic production routes of RES /electrolysis and SMR/carbon capture and sequestration methods. In scenario analyses the effects on the supply chain configuration caused by the rising costs of fossil fuels along with a doubling or halving of hydrogen pricing due to high demand or innovations in production technology respectively are modelled. In terms of hydrogen exports, Iceland is a country with the potential for exporting hydrogen due to its abundance of untapped hydro and geothermal power as explored by Ingason, Ingolfsson and Jensson (2008). Production locations and energy sources were modelled in order to produce hydrogen at the lowest cost at increasing demand levels to represent the timely uptake of hydrogen as a fuel with the southwest of the country becoming the main hydrogen producing region. The 9

relatively short distance from Iceland to the UK and mainland Europe demand centres adds interest to the proposition and highlights that hydrogen generation does not have to be the domestically-produced product as is the focus of many studies. Biomass has the potential for being used as a sustainable means of hydrogen production as various types of agricultural waste can be utilised through different processes to produce hydrogen. Balat and Balat (2009) in a study of the political, economic and environmental impacts of the use of biomass in hydrogen production describe biomass sources as crop straw, grain residues, black liquor by-product from paper making, nut shells and agricultural manure which in turn can be treated by a number of processes such as fermentation, high temperature decomposition (pyrolysis), steam gasification and supercritical extraction to yield hydrogen. Solar energy can be utilised for hydrogen production both through concentration of solar radiation to perform high temperature splitting (thermolysis) of chemicals into hydrogen and other constituents along with using the sun’s energy to produce electricity via a solar photovoltaic (PV) cell. With the former method, Bhosale et al. (2015) study the two-step thermochemical splitting of water by firstly reacting iron (II) oxide with water and sulphur dioxide to produce hydrogen. The second stage regenerates the iron (II) oxide from the iron (II) sulphate by-product allowing it to be reused in the first stage. Overall, temperatures of up to 1000˚C are required for the reactions to proceed effectively. The disadvantages stated by the authors of a single stage water thermolysis process (using heat to directly dissociate water) include higher reaction temperatures in excess of 2450˚C to get a reasonable degree of dissociation and the formation of a potentially explosive mixture of hydrogen and oxygen. An overview of the scale required for solar thermolysis is given by Roeb et al. (2011) who, in using an alternative iron-based thermochemical process, targeted a production yield of 3kg of hydrogen per day but the use of 93 reflective mirrors (heliostats) limits its placement to large, open areas of land with the heliostat and tower array occupying around 29,000m2. Hydrogen has also been created from other chemicals via thermolysis. Robat, Abanades and Flamant (2011) utilised a small-scale (10kW) reactor to split methane into carbon and hydrogen operating at temperatures in the range of 1460-1800°C. This method alleviates GHG emissions when compared to SMR and the solid carbon black by-product produced could be sold into other markets. Process analysis of a 55MW reactor 10

suggests that if the carbon black is sold at typical market prices then the hydrogen production cost is similar to that of SMR and with it potential CO2 equivalent savings over SMR is the region of 11.9kg per kg of hydrogen and 5.7kg CO2 equivalent per kg of carbon black over conventional production methods. However as none of the methods discussed so far are at a widely available, commercially viable stage of development, RES-produced hydrogen needs to be from established technologies. Solar PV is a commercialised technology and currently provides electricity in many applications and can be connected to an electrolyser to produce hydrogen. In a commercial application, Boudries et al. (2014) performed a techno-economic study into hydrogen production for use in a glass production plant in Algeria using a solar PV array. It was deemed commercially viable with an electricity production cost of $0.445/kWh in producing a typical 150m3/h of hydrogen (assuming room temperature and pressure and hydrogen behaving as an ideal gas this is approximately 12.5kg/h of hydrogen). Ngoh et al. (2014) apply hybrid solar thermal and PV to simulate hydrogen production via the electrolysis of steam using the climate of Cameroon and from the available sunlight and equipment performance derive a hydrogen production cost of $5.2/kg. Whilst a relatively large hydrogen production volume of 1843kg/day is predicted, a correspondingly large amount of land (53,400m2) is required in line with Roeb et al. (2011) researching solar thermal hydrogen production. Whilst solar energy shows the potential for hydrogen production, the necessary surface area to achieve it could prove problematic in an urban setting for distributed fuel production. Wind energy is a mature RES that has also been studied for use in electrolytic hydrogen production. A feasibility study by Genç, Çelic and Karasu (2012) assesses Turkey’s regional viability for hydrogen production with different configurations of wind turbines and electrolyser capacities employed in different regions of the country which have different average wind velocities. Using a levelised cost of electricity (LCE) method to determine the electricity and hydrogen production costs for each configuration and region the hydrogen costs were found to vary from a competitive $3.1/kg to an unfeasible $56.3/kg. Turkey is in the position of being the country with the greatest potential for wind energy within European OECD nations. Notably the UK lies in second place (Kenisarin, Karslı and

11

Çağlar, 2006). Wind-solar PV hybrid systems to produce hydrogen are feasible from a technical perspective as demonstrated by Sopian et al. (2009) in a small-scale trial using a 10kW wind turbine and 1kW PV cells coupled to an electrolyser in Malaysia. The resulting hydrogen flow was measured around 135mL/min at ambient pressure (equivalent to 0.7g/h of hydrogen). With the hybrid technology proven, a significant shift in scale is needed for its feasibility as an automotive fuel source to be determined and thus far there has been no research uncovered in the application of hybrid RES to produce hydrogen in the quantities required for automotive use in the UK. A hybrid system is considered necessary in this case due to the intermittent nature of individual RES, the choice of which is influenced by the climate of a particular geographic area. Whilst hydrogen has been safely manufactured and handled industrially for many years, its use as a vehicle fuel requires safety assessments to be performed in this new application. The hazard potential of a hydrogen release in a refuelling area involving actual hydrogen combustion experiments have been conducted by Royle and Willoughby (2011) by simulating pipework failure allowing the release of high pressure hydrogen and using the ignition data to predict safe distances for people from high pressure components. A further study by Shirvill et al. (2012) involved igniting hydrogen releases into a full-scale model of a refuelling area complete with vehicle to simulate leaks from refuelling equipment. The resulting explosion characteristics in terms of overpressure could be used to predict the damage potential of the blasts. A key safety indicator of the viability of an alternative fuel is to assess that the risk (frequency of occurrence and the consequence) associated with its use is no higher than an existing option. Rosyid, Jablonski and Hauptmanns (2007) assess the likelihood of component failure or human error resulting in a release of hydrogen from a storage tank installation fed from an electrolyser and the consequences of its ignition. The overall result of the assessment being that the risk of storing hydrogen as a fuel is no higher than using LPG. Public attitudes towards hydrogen safety have also been discussed and whilst associations of hydrogen fuel with the hydrogen bomb and the Hindenburg disaster have been made in the literature (Cherryman et al., 2008; O’Garra, Mourato and Pearson, 2005), opposition to the

12

introduction of hydrogen vehicles due to safety concerns are not widespread (O’Garra, Mourato and Pearson, 2005; O’Garra et al., 2007). For HFCVs to be commercially successful there must be an incentive for existing transport users to switch from CE vehicles. The previously discussed environmental benefits are one such incentive that could be gained through the uptake of HFCVs. Hardman et al. (2016) find that the UK public perceive far superior environmental attributes of HFCVs over CE vehicles with slightly superior perceptions of performance and fuel economy in a study conducted at a low carbon vehicle event in the UK. Concerns over high vehicle purchase price and a lack of refuelling infrastructure put HFCVs at a significant disadvantage over CE vehicles. The attendees at this event were classed as early technology adopters and only 10% of those surveyed said they would adopt a HFCV. For later technology adopters these barriers to entry would appear even greater. The personal inconvenience for benefitting the environment is researched in a financial manner by Mair (2011) investigating the purchase of carbon offsets by air travellers and those attending conferences. Allocation of the survey respondents into ecocentric, midgroup and anthropocentric categories through framing criteria determined that a greater percentage of carbon offsetters belonged to the ecocentric group. This indicates underlying environmental values are capable of invoking a financial outlay from an individual. From a more psychological perspective, Steg et al. (2014) propose a framework to encourage pro-environmental behaviour. The authors view behaviour as a conflict of hedonic (feel good) and gain goals (improving status) versus normative (the way a person ought to behave) goals. The most dominant of these behaviours will steer thought processes and decision making on environmental issues. The key to strengthening pro-environmental behaviour thereby lies in manipulating the goals, either through elevation of hedonic and gain goals e.g. incentivising economically or reinforcing legally which the authors describe as obtaining “cheap morals”, in effect buying the desired behaviour. The promotion of normative goals is related to increasing the values relating to concern for the environment. With normative goals influenced by concern for the environment, awareness of environmental consequences relating to GHG emissions would help in their

13

shaping. Stern (2006) in a comprehensive review prepared for the UK Government of the impacts of climate change highlighted the impact on the environment from current GHG emissions and predicted the future consequences of inaction of GHG reduction from both environmental and economic perspectives. The conclusion being that a global response to tackle climate change now would be much less costly than doing nothing and suffering the consequences. Application of the model by Steg et al. (2014) to the above study of Hardman et al. (2016) suggests that the success of HFCVs on the characteristics of pro-environmental attributes would signify the dominance of normative goals over hedonic (in this case improved performance) and gain goals (significantly more expensive, a lack of refuelling infrastructure and improved fuel economy). The strength of hedonic and gain goals is further emphasised by Schramme (2011) through taking a liberalist approach to environmental attitudes explains that individual liberties are seemingly placed at the forefront over environmental concerns as people carry out actions they know are harmful to themselves, others and the environment e.g. smoking and automobile use. People drive the cars they do, whether polluting or not, because they represent an expression of choice that means something to the owner. This behaviour is not easily modified through increased taxes and so lends support to Steg et al. (2014) and the description of “cheap morals”. This trade-off between hedonic, gain and normative goals can be applied through this research in the context of RES-produced hydrogen sales in the UK to indicate the relative strength of normative i.e. environmentally positive goals and the personally satisfying hedonic and gain goals represented by a higher WTP for a more environmentally friendly fuel. There is a general need for governments of developed nations to raise revenues through fuel taxation, the levels of which vary between countries and geographic regions. Cheon, Urpelainen and Lackner (2013) in a comprehensive study of gasoline prices across 137 countries from 2002 to 2009 substantiated a link between the levels of subsidies (or taxation) and the country’s status as an OPEC exporter and the level of institutional capacity e.g. the sophistication of the taxation system. Those exporting nations with an abundance of crude oil were more likely to subsidise the cost of fuel for domestic consumption evident by 14

Venezuela’s average cost of fuel being $0.04USD/litre and the nations of Libya, Saudi Arabia, Iraq, Qatar and Kuwait pricing between $0.12-0.23USD/litre with these nations also having a comparatively lower degree of bureaucratic sophistication. Conversely, in developed nations, particularly in western European countries, gasoline was priced between $1.76-1.93USD/litre with a corresponding higher degree of institutional capacity. Whilst the results pertaining to the UK were not directly mentioned in the study, it could be inferred that a high degree of taxation would be applicable to the UK with it being a net crude oil importer with developed government institutions. The actual situation in the UK is in line with predictions in that fuel duty is levied at £0.58/L of petrol, diesel, biodiesel and bioethanol (HM Revenue & Customs, 2014). On top of this is a value added tax charged at 20% of the fuel and duty total (Office of Fair Trading, 2013) resulting in average retail prices during January-May 2015 of £1.10/L ($1.72USD) for petrol and £1.17/L ($1.83USD) for diesel (UKPIA, 2015). Such relatively high pricing serves to work in the favour of hydrogen fuel as it means that depending on an adjustment to taxation levels the price to the consumer could be normalised with that of conventional fuels more easily than heavily subsidised countries. The question of whether hydrogen could be sold as a fuel with the current tax regime employed was explored by Hansen (2010) researching this question in the European context and compares the costs of petrol/diesel compared to SMR-produced hydrogen and non-fossil fuel hydrogen (hydro, nuclear and wind). It was found that the price of crude oil is critical to the competitiveness of hydrogen. The price of crude oil influences the price of petrol/diesel, and also influences the price of natural gas through its substitution as a fuel and so linking the costs of SMR-produced hydrogen and petrol/diesel whereas nonfossil fuel produced hydrogen is decoupled from this. The application of fuel taxation levels in the EU suggests that non-fossil hydrogen could be financially viable in addition to it not being subject to preferential subsidies or reduced taxation levels. The research however does not incorporate the infrastructure costs of hydrogen production. Earlier studies such as in Ogden, Williams and Larson (2004), assume lower fuel prices in their evaluation of life cycle costs of alternative fuel vehicles against a

15

CE standard case. In their case gasoline has a price of $0.95/gallon (before tax averaging $0.42/gallon). The US-focused study incorporates the purchase costs, present value lifetime costs and also costs assigned to externalities such as air pollution, GHG emissions and oil security concerns to determine which advanced vehicle type, if any, has the potential to become successful in future. HFCVs are singled out as the vehicle choice of the future largely due to the externality values assigned to more polluting vehicle types making HFCVs more cost effective. In reality these charges are not fully reflected in vehicle costs so the HFCV’s strength is possibly overstated. As with Hansen (2010), higher prices of crude oil evident in higher gasoline prices in the US than used in the study by Ogden, Williams and Larson (2004) will serve to strengthen the hand of HFCVs. Whilst a high price of crude oil will serve to favour the competitiveness of HFCVs so will the production costs of hydrogen itself. The key to these aims is reflected in the price the UK public would be prepared to pay for hydrogen produced by RES. WTP studies involving HFCVs have been previously conducted. O’Garra et al. (2007) describe the overall positive response to the introduction of hydrogen-fuelled buses trialled in London, Berlin, Luxembourg along with Perth in Western Australia. Using interval regression the authors determined the factors influencing WTP amounts for users in each city with environmental attitudes and behaviours positively influencing results. However the authors note this is not the usual trend and point to the fact public transport was being studied as opposed to private vehicles. These findings highlight a WTP an increased fare for passengers to cover the higher technology costs and increased taxation by non-users who benefit from inner-city air quality improvements amounting to an average of €0.32 (£0.25) per single fare and in London an annual tax of €24 (£19) by residents to support the more environmentally friendly technology. The means of hydrogen generation is omitted from the study so the possibility of GHG releases during its production e.g. SMR, cannot be discounted and so the environmental attitudes and behaviours could be explored further with RES-produced hydrogen. A study of taxi drivers in London by Mourato, Saynor and Hart (2004) explored the WTP for using a HFCV as a taxi as both an 18-month trial and also as a long-term purchase using least-squares regression to highlight the independent variables

16

influencing the WTP behaviour in each case. The influences of the trial taxi use appear to be financially driven with the WTP amount being effectively cancelled out by incentives offered as part of the scheme such as free maintenance, servicing and insurance. For the purchase of HFCV taxis, environmental considerations were more important as was an understanding of the technology. The wide range of literature examined failed to describe the complete picture for the use of distributed hydrogen production from RES in the context of the UK. Whilst the technology has been proven and production costs determined and in some cases the UK public’s attitudes towards HFCVs have been researched, this has not fully engaged people’s environmental concerns by explicitly focusing on RES to produce hydrogen. It is this gap in the literature that this research aims to address. This research examines the distributed production avenue of the overall mission of determining the most effective deployment of resources to align with the strategy of producing hydrogen fuel from RES in order to seek a competitive advantage. This is achieved by quantitatively examining environmental factors discussed in the literature exploring the public’s attitudes towards HFCVs which in turn may influence the WTP amount for RESproduced hydrogen. The findings of the WTP analysis can then be applied to determine the financial feasibility of the distributed hydrogen production infrastructure configuration. Such a notion fits within the resource based view of strategic management theory.

3. Methodology In this study personal vehicle drivers responsible for the running costs of their own vehicles in the UK were chosen to take part in a study. By definition this excluded people who may have driven a companyfinanced vehicle and also people who were not resident in the UK. The use of an online survey introduced some degree of bias due to the fact not all UK households were Internet users. As of 2013 this figure stood at 73% of adults and 83% of households (Office for National Statistics, 2014a). The survey was 17

conducted in English which may have precluded some foreign nationals living and working in the UK e.g. Eastern European nations as a result of the expanded European Union who operated their own vehicles but may not have been fully conversant with the English language. Collecting demographic information from the survey respondents helped define population representativeness within the sample such as age ranges, geographic location etc. and in total 500 qualifying responses were received. In addition to the survey from which the WTP amount and any factors influencing it were derived, the feasibility of the study was ultimately determined by the indicated acceptable RES hydrogen selling price compared to the price of electricity production to produce the hydrogen compared to a business rate grid tariff. This involved data collection on the pricing of renewable electricity sources such as a wind turbine and solar PV array from which a levelised cost of electricity (LCE) calculation was performed based on the equations (1-3) of Genç, Çelic, and Karasu, (2012) in which the annualised cost of the system components was divided by the electricity generated but in this case modified to incorporate two sources of renewable electricity, wind and solar PV. The LCE calculation was based on the following equation: -

LCE(£ / kWh) =

C wt CRFwt + C pv CRFpv + Cci CRFci + Cin CRFin + C misc CRFmisc + Com Ep

(1)

where Cwt, Cpv, Cci, Cin, Cmisc, Com and Ep were the costs of the wind turbine, solar PV array, civil costs, inverter, miscellaneous equipment, operation and maintenance and electricity generated respectively. CRF, the capital recovery factor, was expressed as: -

CRF =

(1 + r )n r (2) (1 + r ) n r − 1

where r being the discount rate and n the useful system life. The electricity costing was then applied to give a hydrogen production cost using the following equation: -

CH 2 (£ / kgH 2 ) =

C EACC CRFEACC + C EE + COM − EE − C EXP CRFEXP (3) mH 2 18

where CEACC, CRFEACC, CEE, COM-EE, CEXP and CRFEXP and mH2 were the annual capital cost of the electrolyser, the annual cost of the energy used by the electrolyser, operation and maintenance costs, the received cost of any electricity exported to the grid and annual hydrogen production respectively. By utilising the CEE term with the cost of grid electricity and subsequently for RES electricity the difference in hydrogen production costs was obtained. This was then used as a basis for determining if the public WTP amount was sufficient to cover the assumed higher costs of RES-produced hydrogen. The independent variable of the WTP amount in this study was the consideration for environmental issues and the influence of government incentives and the relationship of these with the WTP amount was explored by statistical analysis of the survey results. 4. Results & Analysis The data gathering and analysis process can be divided into two distinct sections, each addressing a different aspect of the study. Firstly, the determination of the infrastructure costs relating to renewable electricity production need to be evaluated along with the associated costs of hydrogen generation, compression, storage and dispensing to refuel the HFCV. These costs can then be evaluated against the survey results exploring the opinions of the UK car-owning public to highlight the degree of support for RES-hydrogen technology and the environmental motivations behind this behaviour. In order to apply a scale by which the costs of the hydrogen infrastructure can be calculated, a scenario was developed by which an existing refuelling station was modified to produce and dispense hydrogen on-site to provide fuel to a town’s population of 100,000 people. A number of assumptions are made in this scenario: 1) Only one hydrogen filling station is present for the town’s fleet of HFCVs. 2) Wealth distribution, environmental attitudes and hydrogen technology acceptance within the town’s population were assumed equal to that of the nation as a whole thereby reflecting the national distribution of HFCVs within the town. 19

3) The English national average annual driving distance of 7,900 miles p.a. per car (GOV.UK, 2014a) applies to this study. 4) The fuel efficiency of HFCVs is assumed to be 60 miles per kilogram of hydrogen rising by 2% p.a. to 2020 and 1% p.a. to 2030 in line with Brand, Anable and Tran (2013) to reflect technology improvements. 5) The projected HFCV uptake rates by the UK population as given by UKH2Mobility (2013) were adopted for this study starting from the year 2015 through to 2030. 6) All HFCVs remain on the road for the duration of the study years 2015-2030 and are introduced onto the roads evenly during the year i.e. not all on January 1st. Using the above assumptions, the projected HFCV uptake is summarised (Table 2) for the UK as a whole and the representative scenario town of 100,000 inhabitants. (Insert Table 2 here) By basing the HFCV fuel consumption of 60 miles per kilogram of hydrogen (Toyota, 2014) and the incremental improvement in fuel efficiency and average driving distance, the hydrogen demand can be calculated. Furthermore, with an electricity requirement of 50kWh per kilogram of hydrogen (Siemens, 2011), the electricity required to meet the hydrogen demand can also be determined (Table 3). (Insert Table 3 here) Specifically, the growth in requirements for hydrogen and electricity to meet the growing demand for HFCVs through to 2030 is summarised in Fig. 1. (Insert Figure 1 here)

20

With the predicted demand for both hydrogen and electricity now established the necessary equipment needed to meet these requirements requires evaluation. The selection of a hybrid wind/solar PV system offers the potential of mitigating the intermittency issue often associated with RES as when the prevailing weather conditions for one form of electricity generation are unsuitable then the second means may still be capable of electricity production. In considering a solar PV array, the available space on an existing filling station lies on the canopy roof situated above the pumps and the roof of the retail store. Further ground level space may be available on some stations but is not considered in this study. A typical filling station canopy roof (covering sufficient area to refuel twelve vehicles simultaneously) was measured at 548m2 and was assumed to be structurally capable of supporting the additional weight of a solar array. The solar panels, each measuring 1.65m x 1m were arranged in rows a landscape configuration facing southwards at an incline of 25°, held in place by a flat roof fixing bracket. A space of 2m was left between rows of panels to minimise the effect of shade on PV panel performance. Each panel was rated to 250W output at peak performance (Wp). This configuration allowed a total of 66 solar panels to be fitted to the canopy roof. Furthermore, the pitched roof of the retail store area allowed a total of 54 panels to be fitted on all four sides of the roof giving an array total of 120 panels, equal to 30kWp of installed capacity. Due to the intermittency of solar radiation in the UK the annual generation per kWp is reduced to around 850kWh per year (Energy Saving Trust, 2011). Seasonal variation caused by lower light intensity and reduced daylight hours in the winter months will adversely affect the monthly output from solar PV. Based on the solar PV performance data for a 2.2kWp system in the UK (Energy Saving Trust, 2011), the electricity generation and subsequent hydrogen production capability could be calculated. A quote obtained from a UK-based solar PV firm gave installed costs of £1,400 per kWp for a flat roofmounted array and £1,200 per kWp for a pitched roof-mounted array, giving a total installation cost in this scenario of £39,300.

21

For the generation of electricity by wind power, the turbines are available in a wide variety of sizes, both in terms of operating height (hub height) and power output. In this study five different turbines were selected in order to assess the impact of turbine capacity on the production cost of hydrogen. The peak power ratings selected were 55kW (24m hub height), 225kW (30m), 800kW (60m), 2MW (80m) and 3.3MW (84m). To calculate each turbine’s performance in the UK climate it was necessary to determine the average wind speeds at the hub heights of each turbine for each month of the year to account for seasonal variation in average wind speed. The UK wind speed is typically measured at ten metres above ground level (GOV.UK, 2014b) and because of the wind gradient affect the wind speed increases at greater hub heights. Using equation (4) described by Kaltschmitt, Streicher and Wiese (2007), the wind speeds at the different hub heights were obtained and summarised (Fig. 2).

v w (h) = v10 (h / h10 ) a (4) where vw (h) is the wind speed at height (h), v10 is the wind speed at 10m height, h10 is a height of 10 metres and a is the Hellmann exponent which in this case was taken as 0.34, corresponding to neutral air over human inhabited areas. (Insert Figure 2 here) By comparing the average monthly wind speeds with the power curve of each turbine (the graph of the output of the turbine at different wind speeds) it was possible to obtain a figure for the electricity generation for each turbine during each month of the year, from which the corresponding amount of hydrogen could be calculated (Fig. 3a-e). (Insert Figure 3a-e here) With the performance established, it is necessary to assign costs to the electricity generation infrastructure in order to obtain levelised costs to be used for hydrogen pricing calculations. A supplier-quoted installed cost of a 55kW turbine was £330,000. The remaining turbine installed costs were calculated using actual 22

turbine installation projects where costs of projects according to turbine size were stated (Wiser and Bolinger, 2014). For turbines between 100-1,000kW a 2013 $USD cost of $3,000USD/kW is used, 23MW projects are costed at $1,900USD/kW whereas projects greater than 3MW are $1,850/kW. Using the 2013 average exchange rate of $1.5643 to £1GBP (X-rates, 2014a) gives installed costs of £1,917.81/kW for 100-1,000kW turbines, £1,214.61/kW for 2-3MW turbines and £1,182.65/kW for turbines larger than 3MW, summarised in Table 4. (Insert Table 4 here) The Capital Recovery Factor (CRF) associated with the turbines assuming a discount rate of 10% and a system life of 20 years was calculated at 0.11746. Wind turbine operations and maintenance (O&M) costs were taken as $0.025/kWh (IRENA, 2012) in 2010 $USD, equivalent of £0.0162/kWh (X-rates, 2014b) and was added to the calculated electricity unit price for each turbine. Furthermore, the incentive offered in the UK in the form of the Feed-in Tariff (FiT) payable to generators of renewable electricity is also considered for both wind (Ofgem, 2015a) and solar PV (Ofgem, 2015b) with the index-linked payment guaranteed for twenty years and assumed to rise at 2% p.a. For the purpose of this study an average FiT in pence per kWh was calculated over the period 2015-2030. The LCE for each turbine type is summarised (Table 5, Fig. 4). (Insert Table 5 here) (Insert Figure 4 here) The next stage of infrastructure selection concerns the generation and storage of hydrogen with the differing outputs of each turbine and solar PV configuration able to match demand to the hydrogen and electricity demand profile allowing Fig. 1 to be modified (Fig. 5). (Insert Figure 5 here)

23

The pricing for hydrogen production, storage and dispensing were obtained from a UK equipment manufacturer supplying an integrated unit comprising of a water purifier, proton exchange membrane (PEM) electrolyser, a small compressed hydrogen storage tank, compressor and a dispenser from which HFCVs can be filled with hydrogen at a pressure of 350 bar. The unit merely requires connection to water and electricity supplies. Various sized systems are available producing hydrogen at 1.4kg/hour, 4.6kg/hour, 9.2kg/hour and 25kg/hour priced at £500,000, £1,100,000, £1,800,000 and £3,500,000 respectively. Due to the intermittent nature of RES electricity and equipment downtime a nominal hydrogen storage capacity of five days was included in order to decouple hydrogen production from customer demand. The appropriate number of storage cylinders, each of 350kg capacity were included in the equipment cost. A value of €500/kg as used by Menanteau et al. (2011) in a study of wind-generated hydrogen was used. Using the 2010 average exchange rate of €1.168 to £1GBP (X-rates, 2014c) gives a cost of £428/kg storage capacity. Annual O&M costs were taken at 5% of the capital cost of the hydrogen generating units and storage cylinders. The CRF for the hydrogen production units (based on a 16-year life and discount rate of 10%) was 0.12782 and for the storage cylinders a CRF of 0.11746 (assuming a 20-year life and 10% discount rate). In line with the FiT that serves to lower the LCE, there is an export tariff paid by utility companies for the sale of excess electricity by RES to the grid. This is currently 4.85p/kWh for both solar PV and wind electricity (Ofgem, 2015a; Ofgem 2015b) and in line with the FiT is index-linked and guaranteed for twenty years. The increased demand for hydrogen over time means less electricity will be exported and instead used to produce hydrogen. In the case of each turbine, sufficent hydrogen production and storage capacity was installed initially until the hydrogen production quantity became limited by the turbine size. The mean average hydrogen production from 2015-2030 could then be calculated and used in the levelised hydrogen production cost calculation. A summary of which is shown in Table 6 and Fig. 6 both with and without FiT (as with the LCE calculations to highlight its contribution to the hydrogen production cost). By means of comparison, also displayed is the hydrogen cost using grid electricity using the average unit price of £0.08772/kWh. For illustrative purposes, the hydrogen production costs using the peak charge of £0.09914/kWh for 17 hours per day and an off-peak 24

charge of £0.06000/kWh are displayed. The values used were obtained from the tariff details of electricity supplied to a UK manufacturing company. (Insert Table 6 here) (Insert Figure 6 here) Whilst the above hydrogen production figures are based on turbine size being the limiting factor on production quantity, a downsizing of hydrogen production capacity resulting in higher electrolyser utilisation and lower capital costs thus making the hydrogen production equipment the limiting stage can serve to lower the hydrogen production costs. Increased revenue from exported electricity also serves to further lower the hydrogen cost (Table 7 and Fig. 7). Note the 55kW+30kWp PV system was not downsized as this was already at the lowest equipment configuration and revised grid electricity prices were included to reflect the lower infrastructure costs for each configuration. (Insert Table 7 here) (Insert Figure 7 here) The high capital costs of the hydrogen production infrastructure to meet long-term hydrogen demand installed initially (Table 6 and Fig. 6) result in a high hydrogen production price which is higher in all cases than the equivalent cost of petrol or diesel in the UK (around £5.45 (or 4.8L of fuel) to drive the same distance as a one kilogram of hydrogen based on an average new car fuel consumption of 5L/100km (BP, 2015)). The incorporation of the FiT acts as a hydrogen production subsidy and lowers the cost quite significantly but only with the larger 3.3MW turbine can it be considered competitive with hydrocarbon fuels. The lower capital costs and higher FiT revenues from installing a lower hydrogen generating capacity (Table 7 and Fig. 7) reduces the hydrogen production cost on the 800kW, 2MW and 3.3MW turbines to a level competitive with hydrocarbon fuels. The option for further hydrogen production capacity installation remains open due to a surplus of electricity and future advancement along the

25

technology learning curve may offer reduced capital and O&M costs and improved performance from next generation electrolysers. In assessing the public sentiment to HFCVs and RES hydrogen, a survey was conducted to obtain the views of 500 UK car owners who are at least partially responsible for the ownership costs of their vehicles. This qualifying requirement was to ensure that the respondents answered the questions with the personal cost of motoring to themselves borne in mind as opposed to having a company-financed vehicle where it was felt that consideration for running costs may not have been considered as important. The online survey was conducted from 24th-31st October 2014. The survey was distributed to receive 500 qualifying responses and be nationally representative in both age and gender. Rapid responders were automatically disqualified and their responses do not contribute to the 500 responses obtained. The results were compared against the equivalent respondent profile determined from published national statistics (Office for National Statistics, 2014b) with results from chi-square tests (Table 8) showing a low degree of difference between the data sets (Age p=0.981, Gender p=0.850). (Insert Table 8 here)

Research question 1 - Importance of environmental concerns reflected in the WTP amount. The survey obtained the respondents’ attitudes to two environment-related questions. The first was to place a level of importance towards environmental wellbeing on a seven point scale ranging from not at all important to extremely important and the second question to gauge the frequency that respondents donate to environment welfare causes, again on a seven point scale, with each point relating to different frequencies ranging from never to at least once per month. It was only after providing responses to these questions the respondents were informed of the environmental benefits of using RES-produced hydrogen in HFCVs. Linear regression analysis was performed on the responses for each of the two questions (independent variable X) against the WTP amount (dependent variable Y) for each respondent. The null 26

hypothesis in both cases being there is no relationship between the degree of environmental concern expressed and the WTP amount indicated. In this case, the coefficient of determination (R2) is extremely low (0.2% for environmental wellbeing and 6.2% for donation level) meaning very little of the variation is explained by the analysis. The histograms of residuals (Fig. 8 a-b) are not of normal distribution with a high frequency around the zero mark suggesting that although respondents may indicate concern for the environment, a corresponding degree of financial outlay in response to those concerns is not evident. In this case the null hypothesis cannot be rejected as there is insufficient evidence to suggest the WTP amount is influenced by concerns for environmental wellbeing. Research question 2 - Determination of a possible taxation level on hydrogen fuel. The current fuel duty (£0.58/L) and value added taxation (VAT) measures (20% of fuel and duty price) amount to around £0.77/L of fuel. As previously stated, 4.8L of petrol or diesel will travel approximately the same distance as 1kg of hydrogen meaning a taxation amount of £3.70/kg of hydrogen needs to be levied to remain revenue neutral for the government. The distribution of WTP amounts collected in the survey showed the largest number of respondents (68/500) were only prepared to pay a price equivalent to petrol or diesel (0p premium) which equates to £5.45/kg of hydrogen. The distribution of WTP amount from all survey respondents was combined into 10p bins (Fig. 9). (Insert Figure 9 here) In determining which hydrogen generation infrastructure configurations can produce hydrogen at an acceptable price to allow a taxation mechanism to be applied the production costs determined in Tables 6 and 7 were further refined with any production price below the £5.45/kg threshold highlighted (Table 9). (Insert Table 9 here)

27

The cost of hydrogen and the taxation level that can be applied to maintain price parity with petrol and diesel is related to the size of the wind turbine and the capacity of the hydrogen production equipment installed. Installing all the hydrogen production and storage capacity initially to meet demand through to 2030 and beyond requires the largest scale wind turbine of those evaluated, the 3.3MW unit, to achieve a feasible hydrogen price. Installing smaller scale hydrogen production and storage equipment, whilst reducing the length of time that demand can be met, lowers the hydrogen production price significantly to such a level that the smaller capacity 2MW wind turbine can be employed whilst still maintaining a full taxation level. Reduction to an 800kW turbine requires an 87% reduction in tax level to be a feasible option. It should be noted however that according to UK taxation laws, the FiT and export tariffs are classed as an income and may be subject to corporation and income tax liabilities (HM Revenue & Customs, n.d). A flow diagram depicting the steps performed in the feasibility study is shown in Figure 10. (Insert Figure 10 here) 5. Conclusions & Recommendations Calculations were performed on the costs of electricity produced from different sizes of wind turbine coupled with a fixed 30kWp solar array. In turn these costs were used to obtain hydrogen production costs utilising different configurations of hydrogen generating infrastructure, with the hydrogen production quantity being ultimately constrained by turbine capacity or electrolyser capacity. The survey responses were analysed to provide answers to the research questions and to ultimately address the research aim of stating the feasibility of producing hydrogen from RES at the point of sale. In both of the environmental questions relating to the stated amount of importance placed on environmental wellbeing (i.e. saying) and the donation frequency to environmental causes (i.e. doing), the results failed to find a statistically significant relationship in either question between the degree of environmental concern and the WTP amount. What was notable was the number of respondents who, 28

despite having been informed of the environmental benefits of RES hydrogen in the survey after responding to these two questions, still went on to state they would like to see a reduction in pricing over conventional fuels. This would suggest that other factors are influencing the respondents’ decision over fuel pricing over and above concern for the environment. The findings of Mair (2011) in linking proenvironmental attitudes and increased financial outlay could not be determined in this study. Comparing the study’s findings to the hedonic, gain and normative goals proposed by Steg et al. (2014) suggests in this case that hedonic and gain goals (i.e. the feel good and status improvement goals) are more dominant than the normative goal (acting for the good of the environment). The inclusion in the survey of a question gauging the sentiment towards current fuel pricing would have been useful in giving an insight into the reasons behind the hedonic and gain goal dominance i.e. if there is a strong feeling that fuel is currently overpriced then the willingness to pay more for hydrogen, even with the environmental benefits it brings, may be reduced. The cost of produced hydrogen has been shown to be heavily influenced by the economies of scale of the electricity generating equipment installed, the size of the Feed-in and export tariff subsidies and the capacity of the electrolyser and hydrogen storage equipment. For any level of taxation to be applied at all to the scenario of a filling station supplying a population of 100,000 people the higher capacity turbines (800kW+) are required and in all but one case (3.3MW turbine) a smaller hydrogen electrolyser than the turbine capacity could supply over the demand period is needed (which increases revenues from the Feedin and export tariffs). The 800kW turbine running a smaller electrolyser requires an 87% reduction in overall taxation level to remain cost neutral to petrol and diesel. No other configurations allow the taxation levels equivalent to petrol and diesel of £0.58/L (HM Revenue & Customs, 2014) and the 20% value added tax level (Office of Fair Trading, 2013) to be applied. The study has determined that it is financially and technologically viable to produce hydrogen at the point of sale for areas in the UK that encounter national average wind speeds. Solar PV in isolation cannot be used as a form of electricity generation in the distributed production configuration to meet the fuel 29

requirements for HFCVs in anything above the short term due to surface area constraints, whilst its contribution to a wind/solar PV hybrid system becomes largely insignificant over time as demand increases. Wind technology has demonstrated the ability to meet all demand from distributed production until HFCVs have become established in the UK transport sector although the size of the wind turbine (in terms of both height and generating capacity) needed is high. The large electricity requirements for hydrogen production and the revenues needed from the Feed-in and export tariffs mean that small microgenerating turbines are not economically viable. The required scale of infrastructure came as a surprise when carrying out the research, as were the capital costs with combined turbine, electrolyser and storage costs (excluding O&M) to produce a levelised cost of hydrogen below the acceptable selling price of £5.45/kg ranging from £2,800,000 to £8,900,000. With the primary focus of the study on the environmental benefits of HFCVs it was surprising to observe that whilst environmental concerns featured positively in the opinions of the survey respondents, it did not translate into a willingness to undertake a financial inconvenience (in the form of the WTP) to purchase hydrogen fuel produced from RES which would have benefitted the environment. The WTP figures in general showed high support for cost neutrality or even reduced pricing over petrol and diesel which, unless it is successfully achieved, may be liable to slow or limit the uptake of HFCVs. With the scope of the study aiming to arrive at a feasibility decision for the UK a number of generalisations were made. This included the scenario of a filling station meeting the requirements of 100,000 people whereas this would be set against actual towns and cities of varying populations. Additionally, average wind speeds for the whole of the UK were used as the basis for the calculations. In reality, regional variation is evident with coastal and northern areas generally experiencing higher average wind speeds than southern and inland areas (Department of Energy & Climate Change, 2014), meaning the performance of wind turbines would vary depending on location. The electricity generating performance of a turbine was calculated using the national average wind speed but the output from a turbine is not linear to wind speed due to the power in the wind being proportional to the cube of the 30

velocity (Beggs, 2009). With the varying average wind speeds across the UK, incorporating this would have likely led to the need for multiple scenarios in different regions as regional wind speed probability data would need to be included in the calculations. It is felt that exploring this additional level of detail would be useful in future studies in which this general feasibility study for the whole of the UK could be modified and applied to particular geographic regions. Lastly, the scope of this study was limited to the evaluation of environmental attitudes of the UK car-owning public and it was shown that environmental concerns are not evident in an increased WTP for the use of HFCV technology which would benefit the environment. Further studies could seek to evaluate what other factors could play a role in promoting a WTP for hydrogen fuel produced from RES.

References Agnolucci, P. and McDowall, W. (2013) ‘Designing future hydrogen infrastructure: Insights from analysis at different spatial scales’, International Journal of Hydrogen Energy, 38(13), pp. 5181-5191. [Online]. DOI: 10.1016/j.ijhydene.2013.02.042 Almansoori, A. and Shah, N. (2006) ‘Design and operation of a future hydrogen supply chain: snapshot model’, Chemical Engineering Research and Design, 84(A6), pp. 423-438. [Online]. DOI: 10.1205/cherd.05193 Almansoori, A. and Shah, N. (2009) ‘Design and operation of a future hydrogen supply chain: multiperiod model’, International Journal of Hydrogen Energy, 34(19), pp. 7883-7897. [Online]. DOI: 10.1016/j.ijhydene.2009.07.109 Almansoori, A. and Shah, N. (2012) ‘Design and operation of a stochastic hydrogen supply chain network under demand uncertainty’, International Journal of Hydrogen Energy, 37(5), pp. 3965-3977. [Online]. DOI: 10.1016/j.ijhydene.2011.11.091

31

Balta-Ozkan, N. and Baldwin, E. (2013) ‘Spatial development of hydrogen economy in a low-carbon UK energy system’, International Journal of Hydrogen Energy, 38(3), pp. 1209-1224. [Online]. DOI: 10.1016/j.ijhydene.2012.11.049 Balat, M. and Balat, M. (2009) ‘Political, economic and environmental impacts of biomass-based hydrogen’, International Journal of Hydrogen Energy, 34(9), pp. 3589-3603. [Online]. DOI: 10.1016/j.ijhydene.2009.02.067 Bartels, J. R., Pate, M. B. and Olson, N. K. (2010) ‘An economic survey of hydrogen production from conventional and alternative energy sources’, International Journal of Hydrogen Energy, 35(16), pp. 8371-8384. [Online]. DOI: 10.1016/j.ijhydene.2010.04.035 Beggs, C. (2009) Energy: management, supply and conservation. 2nd edn. Abingdon: Taylor & Francis. Bhosari, R. R., Kumar, A., van den Broeke, L, J, P., Shahd, G., Dardor, D., Jilani, M., Folady, J., AlFakia, M. S. and Tarsad, M. A. (2015) ‘Solar hydrogen production via thermochemical iron oxide-iron sulfate water splitting cycle’, International Journal of Hydrogen Energy, 40, pp. 1639-1650. [Online]. DOI: 10.1016/j.ijhydene.2014.11.118 Boudries, R., Khellaf, A., Aliane, A., Ihaddaden, L. and Khida, F. (2014) ‘PV system design for powering an industrial unit for hydrogen production’, International Journal of Hydrogen Energy, 39(27), pp. 15188-15195. [Online]. DOI: 10.1016/j.ijhydene.2014.04.166 BP (2015) Energy outlook 2035 [Online]. Available from: http://www.bp.com/content/dam/bp/pdf/Energy-economics/energy-outlook2015/Energy_Outlook_2035_booklet.pdf (Accessed: 06 July 2015). Brand, C., Anable, J. and Tran, M. (2013) ‘Accelerating the transformation to a low carbon passenger transport system: the role of car purchase taxes, feebates, road taxes and scrappage incentives in the UK’, Transportation Research Part A, 49, pp. 132-148. [Online]. DOI: 10.1016/j.tra.2013.01.010 32

Cheon, A., Urpelainen, J. and Lackner, M. (2013) ‘Why do governments subsidize gasoline consumption? An empirical analysis of global gasoline prices, 2002-2009, Energy Policy, 56(4-5), pp. 382-390. [Online]. DOI: 10.1016/j.enpol.2012.12.075 Cherryman, S. J., King, S., Hawkes, F. R., Dinsdale, R. and Hawkes, D. L. (2008) ‘An exploratory study of public opinions on the use of hydrogen energy in Wales’, Public Understanding of Science, 17, pp. 397-410. [Online]. DOI: 10.1177/0963662506068053 Dagdougui, H., Ouammi, A. and Sacile, R. (2012) ‘Modelling and control of hydrogen and energy flows in a network of green hydrogen refuelling stations powered by mixed renewable energy systems’, International Journal of Hydrogen Energy, 37(6), pp. 5360-5371. [Online]. DOI: 10.1016/j.ijhydene.2011.07.096 Department of Energy & Climate Change (2014) Annual mean wind speed map [Online]. Available from: http://webarchive.nationalarchives.gov.uk/20140711173142/https://restats.decc.gov.uk/cms/annual-meanwind-speed-map (Accessed: 11 May 2015). Endurance Wind Power (2014a) Endurance E series [Online]. Available from: http://www.endurancewindpower.co.uk/wp-content/uploads/2014/06/E3120-Tuv-Nel-Brochure-July2014.pdf (Accessed: 11 May 2015). Endurance Wind Power (2014b) Endurance X35 [Online]. Available from: http://www.endurancewindpower.co.uk/wp-content/uploads/2014/06/Endurance-X35-FINAL-Hi-ResA4.pdf (Accessed: 11 May 2015). Enercon (2014) Enercon product overview [Online]. Available from: http://www.enercon.de/p/downloads/ENERCON_Produkt_en_web_072013.pdf (Accessed: 11 May 2015).

33

Energy Saving Trust (2011) A buyer’s guide to solar electricity panels [Online]. Available from: http://www.wenhastongreen.org/documents/buyers_guide_solar_pv_panels.pdf (Accessed: 27 April 2015). Genç, M. S., Çelic, M. and Karasu, İ. (2012) ‘A review on wind energy and wind-hydrogen production in Turkey: A case study of hydrogen production via electrolysis system supplied by wind energy conversion system in Central Anatolian Turkey’, Renewable & Sustainable Energy Reviews, 16(9), pp. 6631-6646. [Online]. DOI: 10.1016/j.rser.2012.08.011 GOV.UK (2014a) National travel survey: England 2013 [Online]. Available from: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/342160/nts2013-01.pdf (Accessed: 24 April 2015). GOV.UK (2014b) Average wind speed and deviations from the long term mean. [Online]. Available from: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/284989/et7_2.xls (Accessed: 27 April 2015). Great Britain. Climate Change Act 2008: Elizabeth II. Chapter 27 (2008) [Online]. Available from: http://www.legislation.gov.uk/ukpga/2008/27/section/1 (Accessed: 23 April 2015). Hansen, A. C. (2010) ‘Will hydrogen be competitive in Europe without tax favours?’, Energy Policy, 38(10), pp. 5346-5358. [Online]. DOI: 10.1016/j.enpol.2009.03.035 Hardman, S., Chandan, A., Shiu, E. and Steinberger-Wilckens, R. (2016) ‘Consumer attitudes to fuel cell vehicles post trial in the United Kingdom’, International Journal of Hydrogen Energy, 41(2016), pp. 6171-6179 [Online]. DOI: 10.1016/j.ijhydene.2016.02.067 HM Revenue and Customs (n.d) BIM40510: specific receipts: domestic microgeneration: renewables obligation and feed-in tariff. [Online]. Available from: http://www.hmrc.gov.uk/manuals/bimmanual/bim40510.htm (Accessed: 11 May 2015).

34

HM Revenue and Customs (2014) Excise duty –hydrocarbon oil rates [Online]. Available from: https://www.gov.uk/government/publications/rates-and-allowances-excise-duty-hydrocarbon-oils/exciseduty-hydrocarbon-oils-rates (Accessed: 23 April 2015). Hoffmann, P. (2012) Tomorrow’s energy: hydrogen, fuel cells, and the prospects for a cleaner planet. Cambridge: MIT Press. Ingason, H. T., Ingolfsson, H. J. and Jensson, P. (2008) ‘Optimizing site selection for hydrogen production in Iceland’, International Journal of Hydrogen Energy, 33(14), pp. 3632-3643. [Online]. DOI: 10.1016/j.ijhydene.2008.04.046 IRENA (2012) Renewable energy technologies: cost analysis series – wind power, [Online]. Available from: http://www.irena.org/menu/index.aspx?mnu=Subcat&PriMenuID=36&CatID=141&SubcatID=230 (Accessed: 08 June 2015). Kaltschmitt, M., Streicher, W. and Wiese, A. (2007) Renewable energy: technology, economics and environment. 2007 edn. Heidelberg: Springer. Katikaneni, S. P., Al-Muhaish, F., Harale, A. and Pham, T. V. (2014) ‘On-site hydrogen production from transportation fuels: An overview and techno-economic assessment’, International Journal of Hydrogen Energy, 39(9), pp. 4331-4350. [Online]. DOI: 10.1016/j.ijhydene.2013.12.172 Kenisarin, M., Karslı, V. M and Çağlar, M. (2006) ‘Wind power engineering in the world and perspectives of its development in Turkey’, Renewable & Sustainable Energy Reviews, 10(4), pp. 341369. [Online]. DOI: 10.1016/j.rser.2004.09.011 Kriston, A. Szabó, T. and Inzelt, G. (2010) ‘The marriage of car sharing and hydrogen economy: a possible solution to the problems of urban living’, International Journal of Hydrogen Energy, 35(23), pp. 12697-12708. [Online]. DOI: 10.1016/j.ijhydene.2010.08.110

35

Mair, J. (2011) ‘Exploring air travellers’ voluntary carbon-offsetting behaviour’, Journal of Sustainable Tourism, 19(2), pp. 215-230. [Online]. DOI: 10.1080/09669582.2010.517317 Melendez, M. and Milbrandt, A. (2008) Regional consumer hydrogen demand and optimal hydrogen refueling station siting, National Renewable Energy Laboratory Report No. TP-540-4224. [Online]. Available from: http://www.nrel.gov/docs/fy08osti/42224.pdf (Accessed: 23 April 2015). Menanteau, P., Quéméré, M. M., Le Duigou, A. and Le Bastard, S. ‘An economic analysis of the production of hydrogen from wind-generated electricity for use in transport applications’, Energy Policy, 39(5), pp. 2957-2965. [Online]. DOI: 10.1016/j.enpol.2011.03.005 Mourato, S., Saynor, B. and Hart, D. (2004) ‘Greening London’s black cabs: a study of driver’s preferences for fuel cell taxis’, Energy Policy, 32(5), pp. 685-695. [Online]. DOI: 10.1016/S03014215(02)00335-X Ngoh, S. K. and Njomo, D. (2012) ‘An overview of hydrogen gas production from solar energy’, Renewable and Sustainable Energy Reviews, 16(9), pp. 6782-6792. [Online]. DOI: 10.1016/j.rser.2012.07.027 Ngoh, S. K., Ohandja, L. M. A., Kemajou, A and Monkam, L. (2014) ‘Design and simulation of hybrid solar high-temperature hydrogen production system using both solar photovoltaic and thermal energy’, Sustainable Energy Technologies and Assessments, 7, pp. 279-293. [Online]. DOI: 10.1016/j.seta.2014.05.002 Office for National Statistics (2014a) Statistical bulletin: internet access – households and individuals 2013. [Online]. Available from: http://www.ons.gov.uk/ons/rel/rdit2/internet-access---households-andindividuals/2013/stb-ia-2013.html (Accessed: 24 April 2015).

36

Office for National Statistics (2014b) Population estimates for UK, England and Wales, Scotland and Northern Ireland, mid-2013. [Online]. Available from: http://www.ons.gov.uk/ons/publications/rereference-tables.html?edition=tcm%3A77-322718 (Accessed: 11 May 2015). Office of Fair Trading (2013) UK petrol and diesel sector: An OFT call for information. [Online]. Available from: http://webarchive.nationalarchives.gov.uk/20140402142426/http://www.oft.gov.uk/shared_oft/marketswork/oft1475.pdf (Accessed: 23 April 2015). Ofgem (2015a) Feed-in tariff generation and export payment rate table for non-photovoltaic installations-FIT year 6 (1April 2015-31 March 2016) [Online]. Available from: https://www.ofgem.gov.uk/ofgem-publications/92753/fitnonpvtarifftablefor1april2015.pdf (Accessed: 08 June 2015). Ofgem (2015b) Feed-in tariff payment rate table for photovoltaic eligible installations for FIT (1April 2015-30 September 2015 [Online]. Available from: https://www.ofgem.gov.uk/ofgempublications/94658/fitpaymentratetableforpublication1july2015pvtariffs-pdf (Accessed 08 June 2015). O’Garra, T., Mourato, S., Garrity, L., Schmidt, P., Beerenwinkel, A., Altmann, M., Hart, D., Graesel, C. and Whitehouse, S. (2007) ‘Is the public willing to pay for hydrogen buses? A comparative study of preferences in four cities’, Energy Policy, 35(7), pp. 3630-3642. [Online]. DOI: 10.1016/j.enpol.2006.12.0319 O’Garra, T., Mourato, S. and Pearson, P. (2005) ‘Analysing awareness and acceptability of hydrogen vehicles: A London case study’, International Journal of Hydrogen Energy, 30, pp. 649-659. [Online]. DOI: 10.1016/j.ijhydene.2004.10.008 Ogden, J. M., Williams, R. H. and Larson, E. D. (2004) ‘Societal lifecycle costs of cars with alternative fuels/engines’, Energy Policy, 32(1), pp. 7-27. [Online]. DOI: 10.1016/S0301-4215(02)00246-X

37

Prince-Richard, S., Whale, M. and Djilali, N. (2005) ‘A techno-economic analysis of decentralized electrolytic hydrogen production for fuel cell vehicles’, International Journal of Hydrogen Energy, 30(11), pp. 1159-1179. [Online]. DOI: 10.1016/j.ijhydene.2005.04.055 Rodat, S., Abanades, S. and Flamant, G. (2011) ‘Co-production of hydrogen and carbon black from solar thermal methane splitting in a tubular reactor prototype’, Solar Energy, 85(4), pp. 645-652. [Online]. DOI: 10.1016/j.solener.2010.02.016 Roeb, M., Säck, J.-P., Rietbrock, P., Prahl, C., Schreiber, H., Neises, M., de Oliveira, L., Graf, D., Ebert, M., Reinalter, W., Meyer-Grünefeldt, M., Sattler, C., Lopez, A., Vidal, A., Elsberg, A., Stobbe, P., Jones, D., Steele, A., Lorentzou, S., Pagkoura, C., Zygogianni, A., Agrafiotis, C. and Konstandopoulos, A. G. (2011) ‘Test operation of a 100kW pilot plant for solar hydrogen production from water on a solar tower’, Solar Energy, 85(4), pp. 634-644. [Online]. DOI: 10.1016/j.solener.2010.04.014 Rosyid, O. A., Jablonski, D. and Hauptmanns, U. (2007) ‘Risk analysis for the infrastructure of a hydrogen economy’, International Journal of Hydrogen Energy, 32, pp. 3194-3200. [Online]. DOI: 10.1016/j.ijhydene.2007.02.012 Royle, M. and Willoughby, D. B. (2011) ‘Consequences of catastrophic releases of ignited and unignited hydrogen jet releases’, International Journal of Hydrogen Energy, 36, pp. 2688-2692. [Online]. DOI: 10.1016/j.ijhydene.2010.03.141 Sabio, N., Kostin, A., Guillén-Gosálbez, G. and Jiménez, L. (2012) ‘Holistic minimisation of the life cycle environmental impact of hydrogen infrastructures using multi-objective optimization and principle component analysis’, International Journal of Hydrogen Energy, 37(6), pp. 5385-5405. [Online]. DOI: 10.1016/j.ijhydene.2012.09.039 Schramme. T. (2011) ‘When consumers make environmentally unfriendly choices’, Environmental Politics, 20(3), pp. 340-355. [Online]. DOI: 10.1080/09644016.2011.573356 38

Shirvill, L. C., Roberts, T. A., Royle, M., Willoughby, D. B. and Gautier, T. (2012) ‘Safety studies on high-pressure hydrogen vehicle refuelling stations: releases into a simulated high-pressure dispensing area’, International Journal of Hydrogen Energy, 37, pp. 6949-6964. [Online]. DOI: 10.1016/j.ijhydene.2012.01.030 Siemens (2011) Second wind for hydrogen [Online]. Available from: http://www.siemens.com/innovation/apps/pof_microsite/_pof-spring-2011/_html_en/electrolysis.html (Accessed: 27 April 2015). Sopian, K., Ibrahim, M. Z., Daud, W. R. W., Othman, M. Y., Yatim, B. and Amin, N. (2009) ‘Performance of a PV-wind hybrid system for hydrogen production’, Renewable Energy, 34(8), pp. 19731978. [Online]. DOI: 10.1016/j.renene.2008.12.010 Sørensen, B. (2012) Hydrogen and fuel cells: emerging technologies and applications. 2nd edn. Oxford: Elsevier Spataru, C., Drummond, P., Zafeiratou, E. and Barrett, M. (2015) ‘Long-term scenarios for reaching climate targets and energy security in UK’, Sustainable Cities and Society, 17(2015), pp. 95-109. [Online]. DOI: 10.1016/j.scs.2015.03.010 Steg, L., Bolderdijk, J. W., Keizer, K. and Perlaviciute, G. (2014) ‘An integrated framework for encouraging pro-environmental behaviour: the role of values, situational factors and goals’, Journal of Environmental Psychology, 38, pp. 104-115. [Online]. DOI: 10.1016/j.jenvp.2014.01.002 Stern, N. (2006) Stern Review: the economics of climate change. [Online]. Available from: http://webarchive.nationalarchives.gov.uk/+/http:/www.hmtreasury.gov.uk/independent_reviews/stern_review_economics_climate_change/stern_review_report.cfm (Accessed: 23 April 2015).

39

Toyota (2014) 2016 Toyota Mirai fuel cell sedan production information [Online]. Available from: http://pressroom.toyota.com/releases/2016+toyota+mirai+fuel+cell+product.htm (Accessed: 03 July 2015). UKH2Mobility (2013) Phase 1 results [Online]. Available from: http://www.itm-power.com/wpcontent/uploads/2013/04/UK-H2Mobility-Phase-1-Results-Apr-2013.pdf (Accessed: 24 April 2015). UKPIA (2014) EU pump price comparisons [Online]. Available from: http://www.ukpia.com/docs/default-source/default-document-library/eu-major-brand-petrol-and-dieselaverage-prices-january---april-2015.pdf?sfvrsn=0 (Accessed: 06 July 2015). Vestas (2014a) 2MW platform [Online]. Available from: http://nozebra.ipapercms.dk/Vestas/Communication/Productbrochure/2MWbrochure/2MWProductBroch ure/ (Accessed: 11 May 2015). Vestas (2014b) 3MW platform [Online]. Available from: http://nozebra.ipapercms.dk/Vestas/Communication/Productbrochure/3MWbrochure/3MWProductBroch ure/ (Accessed: 11 May 2015).

Winskel, M., Markusson, N.,Moran, B., Jeffrey, H., Anandarajah, G., Hughes, N., Candelise, C., Clarke, D., Taylor, G., Chalmers, H., Dutton, G., Howarth, P., Jablonski, S., Kalyvas, C. and Ward, D. (2009) Decarbonising the UK energy system: accelerated development of low carbon energy supply technologies. UKERC energy 2050 research report no. 2. [Online]. Available from: http://www.ukerc.ac.uk/asset/12106DFD-9785-44F6-BEE1F9E5854427F2/ (Accessed: 23 April 2015). Wiser, R. and Bolinger, M. (2014) 2013 wind technologies market report [Online]. Available from: http://emp.lbl.gov/sites/all/files/2013_Wind_Technologies_Market_Report_Final3.pdf (Accessed: 5 May 2015).

40

X-rates (2014a) Monthly average 2013. [Online]. Available from: http://www.xrates.com/average/?from=GBP&to=USD&amount=1.00&year=2013 (Accessed: 11 May 2015). X-rates (2014b) Monthly average 2010. [Online]. Available from: http://www.xrates.com/average/?from=GBP&to=USD&amount=1.00&year=2010 (Accessed: 11 May 2015). X-rates (2014c) Monthly average 2010. [Online]. Available from: http://www.xrates.com/average/?from=EUR&to=GBP&amount=1.00&year=2010 (Accessed: 11 May 2015).

FIGURES

Fig. 1. Growth in hydrogen and corresponding electricity demand from 2015-2030.

41

Fig. 2. Calculated monthly average wind speeds at different hub heights (based on GOV.UK, 2014b).

42

Fig. 3a. 55kW turbine

Fig. 3b. 225kW turbine

Fig. 3c. 80kW turbine

Fig. 3d. 2MW turbine

Fig. 3e. 3.3MW turbine Fig. 3 (a-e). Electricity and corresponding hydrogen production possible from wind turbines in the capacity range 55kW to 3.3MW (based on Endurance Wind Power, 2014a and 2014b; Enercon, 2014; Vestas, 2014a and 2014b).

43

Fig. 4. Levelised cost of electricity for wind and solar PV hybrid systems.

Fig. 5. Degree of hydrogen demand fulfilled by solar PV and wind turbine configurations.

44

Fig. 6. Summary of hydrogen production costs configured with excess hydrogen production capacity.

Fig. 7. Summary of hydrogen production costs configured with excess electricity generation capacity.

45

Fig. 8a. Environmental wellbeing

Fig. 8b. Donation frequency

Fig. 8 (a-b). Residual histograms for WTP amount against the degree of concern for environmental wellbeing and donation frequency.

Fig. 9. Distribution of WTP amount amongst all survey respondents.

46

47

Fig. 10. Flow diagram depicting the steps in determining the viability of hydrogen from RES.

Table 1 Abbreviations. Abbreviations CE

Combustion Engine

CRF

Capital Recovery Factor

FiT

Feed-in Tariff

GHG

Greenhouse Gases

HFCV

Hydrogen Fuel Cell Vehicle

LCE

Levelised Cost of Electricity

LPG

Liquefied Petroleum Gas

O&M

Operations & Maintenance

PV

Photovoltaic

RES

Renewable Energy Sources

SMR

Steam Reforming of Methane

VAT

Value Added Tax

WTP

Willingness To Pay

Table 2 The annual and cumulative uptake of HFCVs as a UK total and per 100,000 population (based on UKH2Mobility, 2013). Year

New HFCV p.a.

National cumulative onroad HFCV

New HFCV per 100k population

Cumulative HFCV per 100k population

2015

2,600

2,600

4

4

48

2016

2,600

5,200

4

8

2017

2,600

7,800

4

12

2018

2,600

10,400

4

16

2019

2,600

13,000

4

20

2020

10,000

23,000

15

35

2021

12,000

35,000

19

54

2022

18,000

53,000

28

82

2023

30,000

83,000

47

129

2024

67,000

150,000

105

234

2025

127,000

277,000

198

432

2026

172,000

449,000

268

700

2027

235,000

684,000

367

1,067

2028

263,000

947,000

410

1,477

2029

293,000

1,240,000

457

1,934

2030

320,000

1,560,000

499

2,433

Table 3 Hydrogen and electricity demand 2015-2030 (based on UKH2Mobility, 2013; GOV.UK, 2014a; Brand, Anable & Tran, 2013). Year

Cumulative HFCV per 100k population

Hydrogen efficiency (miles/kg)

Average annual mileage

Total annual hydrogen requirement (kg)

Growth in annual hydrogen requirement (kg)

Annual electricity demand (kWh/year) per 100k population

2015

4

60.0

7,900

330

330

16,500

2016

8

61.2

7,900

851

323

42,550

2017

12

62.4

7,900

1,361

317

68,050

2018

16

63.7

7,900

1,860

310

93,000

49

2019

20

64.9

7,900

2,350

304

117,500

2020

35

66.2

7,900

3,458

925

172,900

2021

54

66.9

7,900

5,484

1,161

274,200

2022

82

67.6

7,900

8,300

1,734

415,000

2023

129

68.3

7,900

12,731

2,892

636,500

2024

234

68.9

7,900

21,755

6,479

1,087,750

2025

432

69.6

7,900

39,462

12,146

1,973,100

2026

700

70.3

7,900

66,021

16,230

3,301,050

2027

1,067

71.0

7,900

101,895

22,116

5,094,750

2028

1,477

71.7

7,900

144,993

24,426

7,249,650

2029

1,934

72.5

7,900

192,729

27,018

9,636,450

2030

2,433

73.2

7,900

244,848

29,349

12,242,400

Table 4 Installed costs of five differint wind turbines (based on Wiser & Bolinger, 2014; X-rates, 2014a). Turbine size 55kW

2013 USD/kW

Installed cost (£GBP)

N/A

£330,000

225kW

$3,000

£431,506

800kW

$3,000

£1,534,248

2MW

$1,900

£2,429,220

3.3MW

$1,850

£3,902,745

Exchange rate $1.5643USD/£1GBP (X-rates, 2014a)

Table 5 Levelised cost of electricity for wind and solar PV hybrid systems. Equipment selection

LCE (p/kWh) After FiT 50

(p/kWh) 55kW + 30kWp PV

29.02

13.71

225kW + 30kWp PV

10.36

-3.41

800kW + 30kWp PV

6.79

-0.82

2MW + 30kWp PV

4.13

1.35

3.3MW + 30kWp PV

4.00

0.76

Table 6 Hydrogen production costs from RES and grid electricity with excess hydrogen production capacity (based on Ofgem, 2015a; Ofgem, 2015b).

Electricity generation Hydrogen generation and Hydrogen production cost Hydrogen production cost method storage equipment without FiT (£/kg) with FiT (£/kg)

Hydrogen pro elec Average

55kW+30kWp PV (1.4kg/hr H2)

1.4kg/hour, no additional storage

£48.16

£40.50

£38.58

225kw+30kWp PV (4.6kg/hr H2)

4.6kg/hour, 1x350kg cylinder

£29.54

£22.66

£28.74

800kW+30kWp PV (9.2kg/hr H2)

9.2kg/hour, 3x350kg cylinders

£13.86

£10.06

£19.16

2MW+30kWp PV (25kg/hr H2)

25kg/hour, 10x350kg cylinders

£9.44

£8.05

£21.12

3.3MW+30kWp PV (25kg/hr H2)

25kg/hour, 10x350kg cylinders

£0.84

-£0.77

£20.85

Table 7 Hydrogen production costs from RES and grid electricity with excess electricity generation capacity (based on Ofgem, 2015a; Ofgem, 2015b).

51

Electricity generation Hydrogen generation and Hydrogen production cost Hydrogen production cost method storage equipment (£/kg) with FiT (£/kg)

Hydrogen p el

Avera 55kW+30kWp PV (1.4kg/hr H2)

1.4kg/hour, no additional storage

£48.00

£38.46

£38.5

225kW+30kWp PV (1.4kg/hr H2)

1.4kg/hour, 1x350kg cylinder

£18.15

£11.27

£19.3

800kW+30kWp PV (4.6kg/hr H2)

4.6kg/hour, 2x350kg cylinders

£8.78

£4.98

£17.5

2MW+30kWp PV (9.2kg/hr H2)

9.2kg/hour, 4x350kg cylinders

-£2.47

-£3.86

£18.2

3.3MW+30kWp PV (9.2kg/hr H2)

9.2kg/hour, 4x350kg cylinders

-£17.30

-£18.28

£18.2

Table 8 Comparison of actual survey respondents with published statistics (Office for National Statistics, 2014b). Age Data source 18-29 Count Actual survey respondents

60-69

>70

All

90

81

71

82

98.50

78.00

89.00

79.00

69.00

86.50

0.00254

0.01282

0.01124

0.05063

0.05797

0.23410

98

79

88

77

67

91

98.50

78.00

89.00

79.00

69.00

86.50

0.00254

0.01282

0.01124

0.05063

0.05797

0.23410

Expected count Contribution to Chi-square

50-59

77

Count ONS (2014) respondent profile

40-49

99

Expected count Contribution to Chi-square

30-39

Age Pearson chi-square = 0.739, DF = 5, p-value = 0.981 Likelihood ratio chi-square = 0.739, DF = 5, p-value = 0.981 Gender 52

500

500

Pearson chi-square = 0.036, DF = 1, p-value = 0.850 Likelihood ratio chi-square = 0.036, DF = 1, p-value = 0.850 Fisher’s exact test: p-value = 0.899350

Table 9 Possibility of tax application to hydrogen fuel to maintain a selling price equivalent to petrol or diesel. Electricity generation method

Turbine constrained hydrogen Electrolyser constrained Hydrogen demand Hydrogen de production cost with FiT hydrogen production cost with met until during met until du (£/kg) FiT (£/kg)

55kW+30kWp PV

£40.50

2020

£38.46

225kW+30kWp PV

£22.66

2023

£11.27

800kW+30kWp PV

£10.06

2027

£4.98 (partial taxation)

2MW+30kWp PV

£8.05

2030

-£3.86 (full taxation)

-£0.77 (full taxation)

Est. 2033

-£18.28 (full taxation)

3.3MW+30kWp PV

53