Trinity Term 2021

3YP Green Hydrogen Public Mobility Project

Alexander Astor, Akshay Pal, Lingxi Tang, Antoine Thibaut

Department of Engineering Science

University of Oxford
 Table of Contents
1 Introduction 1
1.1 Project Outline and Scope 1
1.2 Project Motivation 1
1.3 Overview of Oxford’s bus network 4
1.4 Limitation of battery electric vehicles 5
1.5 Current environmental impact 9

2 Evaluation of green gases 11

2.1 Hydrogen 11
2.1.1 Hydrogen as a fuel 11
2.1.2 Fuel cell efficiency 13
2.1.3 Hydrogen Production 14
2.1.4 Hydrogen storage 16
2.2 Ammonia 19
2.2.1 Introduction to Ammonia 19
2.2.2 Ammonia as a fuel 19
2.2.3 Ammonia production 22
2.2.4 Ammonia storage and distribution 22
2.3 Verdict on green gas 24

3 Bus technology and fleet requirements 25

3.1 Diesel bus technology 25
3.2 Hybrid bus technology 26
3.3 Hydrogen bus technology 27
3.4 Oxford fleet requirements 28
3.4.1 Bus type requirements 28
3.4.2 Range requirements 29
3.4.3 Refuelling time requirements 29
3.4.4 Speed requirements 29
3.5 Hydrogen fleet advantages 30
3.5.1 Vehicle costs 30
3.5.2 Maintenance costs 31
3.5.3 Noise 32
3.5.4 Conclusion on hydrogen fleet advantages 32
3.6 Fleet retrofitting 32
3.6.1 Range verification 33
3.6.2 Refuelling time verification 33
3.6.3 Speed verification 35
3.6.4 Financing a hydrogen bus fleet 39

4 Estimating the hydrogen demand 41

4.1 Route modelling 41
4.2 Energy equivalence method 44
4.2.1 Accounting for efficiency 45
4.3 Results 46
4.3.1 Distribution demand 47
 4.3.2 Emergency demand 47
4.4 System Overview 48

5 Social and legal study 50

5.1 Social Study 50
5.2 Legal study 51
5.2.1 Hydrogen management regulations 52
5.2.2 Hydrogen fleet regulations 54
5.2.3 Infrastructure and planning regulations 54

6 Hydrogen production 56
6.1 Energy production 56
6.1.1 Selecting the energy source 56
6.1.2 Solar Energy 58
6.2 Hydrogen Production Components 61
6.2.1 Electrolyser Type 61
6.2.2 Electrolyser Model 63
6.2.3 Battery Technology 64
6.3 Smart Electrolyser-Battery System 66
6.3.1 System Operations 66
6.3.2 System Parameter Design 66
6.3.3 System Performance Results 71

7 Hydrogen storage and distribution 73

7.1 Hydrogen storage 73
7.1.1 Evaluation of different storage methods 73
7.1.2 Storage requirements 75
7.2 Hydrogen distribution 79
7.2.1 Oxford depots and general delivery strategy 80
7.2.2 Hydrogen production plant location 81
7.2.3 Transporting hydrogen to fuelling stations 83

8 Financial analysis 85
8.1 Transition to net zero 85
8.2 Expenditure Outline 85
8.3 Comparison with Business As Usual 87
8.4 Financing the transition 89
8.4.1 Loan 90
8.4.2 Increase of Passenger Fare 90
8.4.3 Grants 91

9 Conclusion 92
9.1 Project limitations 92
9.2 Next steps 93

References 94

Appendix A: Bus route data 110

Appendix B: Truck trailer distribution data 111

Appendix C: 10 Year Transition Period Expenditure 112

Appendix D: 10 Year Transition Period Cash Flow Analysis 114

 Alexander Astor

1 Introduction
An effective public transport system is one of the vital components of a thriving city. It allows for

ease of mobility for the people, both for leisure and work-related purposes. However, in order to

meet global, national and city-specific targets for reducing emissions, the public transport system in

Oxford needs swift, major adjustment. In this report, a proposed solution to this problem will be

presented. A link to a repository containing the modelling for this project is available at the top of

the list of references.

1.1 Project Outline and Scope

The main objectives of the project are as follows:

1. To design a net zero public transport system for Oxford City using green gas

2. To consider an Oxfordshire-based generation and distribution site for the green gas

production

3. To consider additional benefits, through integration with the wider energy system

The scope of this project includes buses operating within Oxford, whether they operate solely

within Oxford or connect to other cities. The bus companies included in this scope are Stagecoach

Oxfordshire (SC), the Oxford Bus Company (OBC) and Thames Travel (TT). City Sightseeing

Tours is not being included in this project due to their plans to operate a fully electric fleet, which

they have already begun to develop [1.1]. In addition to this, they only cover approximately 25,000

bus kilometres each year, whilst the three companies being considered cover over 33,000,000 bus

kilometres annually in total. The modelling by which these figures were obtained will be presented

in more detail in Section 4. Trains are not included in the scope of this project, and nor are taxis,

though this project yields some relevant results that could be applied to similar projects in those

areas.

1.2 Project Motivation

When the Paris Agreement was adopted by 196 parties in late 2015, and then brought into force in

November 2016, the signatories took upon themselves responsibility to ensure that the global rise

in temperature above pre-industrial levels remains well below 2°C, preferably below 1.5°C [1.2,
1
 Alexander Astor

1.3]. As a signatory, the UK has since been updating previous targets, and adding new measures

to reduce their carbon emissions. In 2019, the UK committed to achieving net zero harmful

emissions by 2050, improving on its original aim declared in the Climate Change Act of 2008 for an

80% reduction in carbon emissions by 2050, from a 1990 baseline [1.4]. From 2008, the

Government has been committed to setting carbon budgets [1.5]. These are legally binding caps

on carbon emissions over a period of five years, which help guide the UK toward its 2050 target.

These are put together by the Climate Change Committee more than a decade ahead of the

relevant timeframe, to provide enough time to prepare policies and investments for the most

economically beneficial route to the target [1.5]. Amongst the policy measures taken so far are

carbon pricing, low-carbon energy support, improvement of energy efficiency and increased

spending on international climate action [1.5].

On top of the aims of the UK government to reduce carbon emissions, the Oxford City Council has

targets of its own to accelerate the process of becoming a zero emissions city. In 2019, there was

an Oxford Citizens’ Assembly on Climate Change. The main relevant findings published in the

report were [1.6]:

● A majority felt that Oxford should aim to achieve net zero earlier than 2050

● Increased public transport and fewer cars were seen as a central part of achieving net zero

● The burden of change appeared to be put on individuals, so the Council needed to better

communicate a shared vision and strategy to achieving net zero

In response to this, the Council proposed to do the following among other measures [1.7]:

● Set a Climate Emergency Budget which commits a total of £19 million to address the

climate emergency, on top of £84 million of ongoing investment to tackle the climate

emergency in Oxfordshire

● Cut transport emissions and boost renewable energy installation

● Hold a Zero Carbon Oxford summit involving the organisations responsible for most of the

emissions in the city to develop a shared plan to tackle the carbon problem

● Establish a Zero Carbon Oxford Partnership and influence partners to do more

2
 Alexander Astor

● Create new carbon budgets for the city to step down to zero emissions

In December 2020, the council announced that it had achieved its target of a 40% reduction in

carbon emissions from a 2005 baseline, as well as joining the UK100’s aim of achieving net zero

harmful emissions by 2045 [1.8].

In February 2021, some of these aims were improved, whilst others were expanded upon, at the

Zero Carbon Oxford summit [1.9]. At the summit, leaders of major organisations in Oxford signed

the Zero Carbon Oxford Charter to give their support to achieving net zero by 2040, an entire

decade before the UK’s legally binding date to achieve this as a country. As well as advancing the

date by which the city will have zero emissions, the summit marked the creation of the Zero

Carbon Oxford Partnership. The partnership provides a collaborative approach to achieving the

ambitious targets for carbon reduction, enabling greater sharing of methods and insights between

partners, attracting greater financial support for innovative projects to achieve the city’s aims, as

well as engaging citizens and communities in shared action. Furthermore, the partnership will allow

for more coordinated lobbying of the UK government, which should allow the city to develop the

means to achieve their aims, whether through improved policy or greater funding.

Fig 1.1 shows a chart of the sources of Oxford’s carbon emissions. As shown, the transport sector

in Oxford is responsible for 17% of the city’s carbon emissions [1.10]. Of the remainder, the vast

majority is taken up by various building-related emissions, and a small amount (less than 2%) is

associated with waste treatment [1.10].

Although the aim is to have the city as a whole

net zero by 2040, the Council has set a separate

aspiration for all buses within the city to be zero

emissions capable by 2035 [1.11]. This lines up

with the plans to improve the current low

emissions zone to a zero emissions zone which

covers the entire city by 2035 [1.11].

3
 Alexander Astor

1.3 Overview of Oxford’s bus network

As mentioned previously, the three bus operators in the project scope are Stagecoach Oxfordshire,

the Oxford Bus Company, and Thames Travel. The latter two are both subsidiaries of the Oxford

Bus Group, and therefore share management. On top of this, Stagecoach and the Oxford Bus

Company share certain routes under a shared ticketing scheme.

The Oxford Bus Company operates out of their depot in Cowley, Thames Travel operates out of

one in Didcot, and Stagecoach Oxfordshire operates from a depot in Cowley and another one in

Witney. The Oxford Bus Company currently has a fleet of 148 buses and coaches, while Thames

Travel has 83, and Stagecoach Oxfordshire has 168, giving a total of 399 buses within the scope of

the project [1.12, 1.13, 1.14].

The Oxford Bus Company runs services including BROOKESbus, the airline, Park & Ride, and city

routes [1.15]. A number of the city routes are shared evenly between the Oxford Bus Company and

Stagecoach Oxfordshire. In addition to this, Stagecoach Oxfordshire runs services which link the

city to various locations in Oxfordshire, as well as the well-established Oxford Tube service, which

runs between Oxford and London [1.16]. Thames Travel runs services connecting Oxford to local

areas which surround it, as well as to some further locations including Henley-on-Thames and

Reading [1.17].

4
 Akshay Pal

1.4 Limitation of battery electric vehicles

An obvious solution to reducing the emissions caused by the Oxford bus network may seem to be

a transition to battery electric vehicles (BEVs). Indeed, the purchase of battery and hybrid electric

vehicles in the UK has been rapidly growing in the last five years, as shown by Fig. 1.2 below.

< Fig. 1.2: Registration of new hybrid and battery electric vehicles in the UK from 2015 - 2019
[1.18, 1.19, 1.20, 1.21, 1.22] >

City Sightseeing Tours, a member of the Oxford Bus Group has already begun to develop and

operate a fully electric fleet [1.1]. However, although BEVs are seeing a massive increase in

popularity, they still suffer from issues such as low range and long recharging time which are key

vehicle metrics for a bus network. Fuel cell (FC) buses surpass BEVs on these key vehicle

considerations. This is less of a problem for City Sightseeing Tours as their services are bound to

the city centre with very low mileage and hence they have no need for long range buses. Other

services, especially those travelling outside the city centre serving places such as Heathrow

airport, need buses that have the extra range that FC buses can provide.

Despite the growth of BEVs, they are not desirable for Oxford’s public transport system for the

following reasons. A starting point is to first examine the fuel sources for each vehicle. A summary

of the key properties is provided in Table 1.1 below.

5
 Akshay Pal

< Table 1.1: Gravimetric and volumetric energy density of various fuel types >

It is clear that whilst batteries and green gas have a similar volumetric energy density, both

hydrogen and ammonia have significantly greater gravimetric energy densities. This is important as

it contributes significantly to the overall efficiency and range of the vehicle due to the much lower

mass of fuel required for the same energy output. Range is the major advantage of FC buses over

BEVs and typical values are ~ 240 km for BEVs [1.28] and ~ 320 to 430 km for FC buses [1.29].

Another key vehicle consideration is the refuelling/recharging time. FC buses have the advantage

here as it takes only minutes to refuel an FC bus. For example, the Wrightbus H2Bus can take as

little as 7 minutes to refuel [1.29, 1.30]. This is in comparison to, for example, the Tesla Model S, a

BEV which takes 1.25 hours to fully recharge when using a supercharger, the fastest method of

charging. BEVs can take several hours to charge and this down time directly translates to financial

loss for the bus operators. Many of the public BEV chargers located across the country are

connected to the national grid and so by using these chargers there is a contribution to carbon

emissions. Green gas powered vehicles can be completely net zero.

Finally, the costs of these vehicles can also be compared. Both BEVs and FC buses can use

exactly the same electric drivetrain system and will only differ on the methods used to power the

drivetrain. For a FC bus the main capital costs will be the fuel cell and the hydrogen storage,

whereas in a BEV the capital cost will be for large-capacity lithium-ion batteries. Due to the high

cost of modern, high-density lithium ion batteries and the fact that quite a large number of batteries

are required to give the BEV a decent range (again this is due to the very low gravimetric energy

density of lithium-ion batteries) BEVs lose out to FC buses in this criteria. The cost of a FC bus

capable of the ranges mentioned previously is ~ £356,000 [1.31] whereas a BEV with a lower

range will cost ~ £554,000 [1.28].

6
 Akshay Pal

The major vehicle metrics have been summarised in Table 1.2 below.

< Table 1.2: Summary of the major vehicle metrics between BEVs and FC Buses >

It is also important to look at the wider environmental impact caused by both vehicle technologies.

While BEVs may seem to be carbon free, there are many other secondary sources of emissions to

consider such as transport, manufacture and mining. Many of these secondary emissions, namely

manufacture and transport, will be mirrored in a green gas based system. The major source of

damage to the environment, caused primarily by BEVs, is the impact on wildlife and local

ecosystems due to the mining and processing of metal ores. While a green gas based system is

certainly not battery free as batteries are used onboard FC buses, it does contain fewer batteries

than a BEV system. Batteries onboard FC buses are also much smaller (lower capacity and

therefore containing less lithium) than those used on BEVs as their main purpose is to ensure that

there is a constant current through the fuel cell, not to be the primary power source [1.32, 1.33].

The idea of replacing the batteries in an FC bus with ultracapacitors is something that is already

being explored, specifically, Wrightbus has been working with a company called Skeleton

Technologies to replace the lithium based batteries currently used in the H2Bus with

ultracapacitors which are much more environmentally friendly [1.34].

There are two primary metals used in lithium-ion batteries, they are lithium and cobalt which is

found in the cathode of the batteries. The mining of these two metals is extremely harmful to the

environment and has had damaging effects on the landscape and inhabitants of the countries in

which they are found.

The process of lithium extraction uses a large amount of water, approximately 500,000 gallons per

tonne of lithium extracted [1.35] and produces many toxic chemicals as by-products, including

hydrochloric acid (HCl) [1.35]. These damage the soil and leak into nearby rivers, killing fish and

any land animals that drink from the river. Many environmental incidents have occured due to the

7
 Akshay Pal

mining and processing of lithium such as, in May 2016, when environmental protestors threw dead

fish onto the streets of Tagong, Tibet in opposition to China’s mining operations at the Ganzizhou

Rongda lithium mine which had leaked toxic chemicals into the nearby Liqi river, which is from

where the dead fish had been taken [1.36]. In Chile, lithium mining operations have left pools of

toxic liquids which are visible from the sky. The American photographer David Maisel has

documented the devastating impact lithium mining has had on the once beautiful Chilean

landscape in his work Desolation Desert [1.37].

The majority of the world’s cobalt ores are found in the Democratic Republic of Congo (DRC).

Kinshasa, the country’s capital, is a leading source of cobalt, supplying approximately 70% of the

cobalt produced worldwide in 2020 [1.38]. The leading consumer of cobalt in 2020 was China, with

80% of its consumption being used by the rechargeable battery industry [1.38]. The mining of

cobalt in the DRC has had a severe human toll and there have been several reports of child-labour,

exploitation, dangerous working conditions and serious health problems due to the mining of cobalt

[1.39]. These include thyroid and breathing problems and come as a result of workers not being

provided with adequate machinery, often having to resort to mining with hand tools which results in

exposure to dangerous chemicals [1.39]. The bigger concern, however, is the link to birth defects in

children whose parents worked in a cobalt mine. Children in the DRC are being born with a

condition known as holoprosencephaly which is usually fatal. This is being investigated by doctors

at the University of Lubumbashi, a large centre for cobalt mining in the DRC [1.39].

Overall, the FC bus outperforms the BEV in the most important vehicle considerations of range,

refuelling time and cost. This point is reinforced later by the results of the route modelling (Section

4.1) which shows that buses with a significant range are required to maintain coverage that the

current bus network requires. The environmental impact of battery technology is much greater than

first meets the eye and hence, it is vital that the system design minimises the need for battery

technology and is able to take advantage of other forms of energy storage such as ultracapacitors.

Hence, it is for these reasons that a green gas based system is the way forward and most likely to

achieve a successful net zero transport system.

8
 Akshay Pal

1.5 Current environmental impact

The current environmental impact of the transport network in Oxford is considerable and it is worth

estimating the emissions caused in order to quantifiably measure the change that switching to a

net zero system will have. In order to estimate the emissions it was necessary to gain a thorough

understanding of the current bus network. This was achieved by creating a model of the current

network using Google Maps [1.40] and the online timetables of the network operators [1.15, 1.16,

1.17]. The modelling is discussed in much detail in Section 4.1, here only the results of the

emissions aspect of the route modelling will be presented.

The results of the route modelling came to a total of 35,867 tonnes of CO2 released annually by the

bus network in Oxfordshire, which contributes to the 17% of the city’s emissions caused by the

transport sector [1.10]. This value can be down into the individual contributions of the three

networks operators identified in the project scope. These are Stagecoach Oxfordshire, the Oxford

Bus Company and Thames Travel. Fig 1.3 shows the breakdown of weekly CO2 emissions by the

network operators.

< Fig 1.3: Estimate of the weekly CO2 emissions by each network operator >

Stagecoach contributes most to the emissions of the bus network, around 330 tonnes weekly.

It is important to note the impact of the COVID-19 pandemic on the timetables of the networks.

When doing the research for the route modelling it was noticed that the value obtained for the

distance travelled yearly using 2020 timetables was significantly lower (around half) than the value

9
 Akshay Pal

published by the bus companies for the total distance travelled in 2019 [1.41, 1.42]. It was obvious

that this was due to a downsizing of operations caused by the national restrictions put in place to

combat the spread of the coronavirus. As such, the results were adjusted to better reflect the

operation in a normal year. This modelling was also used to estimate the total annual hydrogen

demand, which is a key metric of the system, hence it was doubly important that the figures were

reflective of what is expected in a normal year.

10
 Akshay Pal

2 Evaluation of green gases

In order to choose an appropriate source of power for the system, it was necessary to review the

options and compare them with each other. This section of the report aims to explore the technical

aspects of two different potential fuels for the public transport network; these are hydrogen and

ammonia. This section will aim to evaluate and compare the performance, production and storage

of these fuels, providing technical information where necessary, in order to make an informed

choice for the system.

2.1 Hydrogen

This subsection will give a brief overview on the performance of hydrogen as a fuel, its different

methods of production, the environmental impact of these processes and finally, the different

methods of hydrogen storage.

2.1.1 Hydrogen as a fuel

Hydrogen is a highly flammable, colourless gas that exists as a diatomic H2 molecule. It has an

atomic weight of 1.008 kg kmol-1 [2.1], making it the lightest element on the periodic table. It is also

the universe’s most abundant element. It has a mass density of 0.090 kg m-3 (at 0°C, 1 atm) [2.1]

and a gravimetric energy density of 120 MJ kg-1 [1.23], making it much more energy dense than

modern lithium ion batteries (0.36 - 0.95 MJ kg-1 [1.26]).

Electricity is produced from hydrogen by an electrochemical device known as a fuel cell. This

converts the chemical energy stored within hydrogen directly into electricity via electrochemical

reactions known as reduction and oxidation (redox) reactions which take place at the electrodes.

There are many different types of fuel cells which mainly differ by type of electrolyte they use and

their size. The most common of these are:

1. Alkaline fuel cells (AFCs) in which the mobile ion is OH- and the electrolyte used is

potassium hydroxide (KOH). They operate at temperatures between 50 and 250°C and

have aerospace applications, having been used on both Apollo and Space Shuttle

spacecraft [2.2, 2.3].

11
 Akshay Pal

2. Polymer Electrolyte Membrane/Proton Exchange Membrane fuel cells (PEMFCs) conduct

protons (H+ ions) through a membrane which forms the electrolyte. They have an operating

temperature between 30 and 100°C [2.2, 2.3].

3. Solid Oxide Fuel Cells (SOFCs) conduct oxygen O2- ions via an electrolyte made of a solid,

nonporous metal oxide. An example of this is yttrium stabilised zirconium (YSZ) which

consists of zirconium dioxide (ZrO2) stabilised with yttrium oxide (Y2O3). They have an

operating temperature range of 500 - 1000°C and are used in large-scale commercial

applications [2.2, 2.3].

Fuel cells are appropriate for transport applications as they have no moving parts, operate quietly

and have a modular design [2.4]. Being modular means that the fuel cell can be scaled up or down

depending on the application, giving them more flexibility to operate within different parameters

and use cases. PEMFCs, having the lowest operating temperature, is the favoured fuel cell type for

transport applications and hence, the rest of this section will focus on them. Fig. 2.1 shows the

basic operation of a fuel cell and the arrangement of the two electrodes (anode and cathode) and

the electrolyte.

< Fig. 2.1: Basic operation of a fuel cell [2.3] >

12
 Akshay Pal

The first PEMFC was developed in the 1960s by General Electric for use by NASA on their Gemini

spacecraft as an auxiliary power source [2.2, 2.5], there have obviously been significant advances

in the technology since then but the basic principles of operation remain the same. At the anode,

hydrogen is continuously fed in and is oxidised to form H+ ions and electrons as shown in equation

2.1 below [2.2],

H2 → 2H+ + 2e- (2.1)

At the cathode, oxygen is fed in and reduced as shown below [2.2],

½O2 + 2H+ + 2e- → H2O (2.2)

Combining these two half equations gives the following overall equation for the fuel cell,

H2 + ½O2 → H2O (2.3)

Note that this equation is identical to the equation for the combustion of hydrogen in air and so it

must be an exothermic reaction, meaning it will have a negative change in enthalpy ∆𝐻 and will

release heat.

2.1.2 Fuel cell efficiency

An important metric to understand is the efficiency of fuel cells as this will play a key role in

designing the system requirements later on, specifically the hydrogen demand of the system.

Hence, it is important to look at fuel cell efficiency and compare this with the efficiency of a modern

internal combustion engine.

The maximum theoretical efficiency of a PEMFC is ~ 83% [2.2] which is much greater than the

efficiency of a modern internal combustion engine. This is expected as fuel cells have no moving

parts and subsequently have no mechanical friction losses which is a major source of loss in an

internal combustion engine. The irreversible energy losses in a fuel cell can be summarised as

follows:

1. Activation losses: These are due to the speed of the reactions taking place at each

electrode. These reactions are not instantaneous, in fact they are rather slow, and so a

proportion of the voltage is lost in order to drive the reactions to completion [2.3].

13
 Akshay Pal

2. Concentration losses: As the fuel is used, the concentration of reactants on the surface of

the electrodes will reduce and this will affect the cell potential [2.3].

3. Ohmic losses: This is perhaps the most obvious form of voltage drop. There is a resistance

to the flow of electrons through the electrodes and there is a resistance to the flow of ions

through the electrolyte [2.3].

Modern PEMFCs are manufactured in a way that reduces these losses to increase the fuel cell

efficiency. This involves increasing the roughness of the electrodes to give them a higher surface

area, using electrodes with high electrical conductivity 𝜎, such as platinum, although this may

increase cost and finally by trying to make the electrolyte layer as thin as possible to reduce the

ionic resistance [2.3]. Fuel cell efficiencies are ever increasing with new advances in materials

science and chemistry, the current standard value is around 60% [2.6].

The fuel cell process produces only water and heat as by-products and is completely carbon free if

pure, green hydrogen is used [2.7]. This is an important distinction to make as not all hydrogen is

classed as green hydrogen and thus would not adhere to the project aim of being net zero, this will

be discussed further in Section 2.1.3.

2.1.3 Hydrogen Production

Not all hydrogen that is produced is suitable for this project and so, it is important to understand the

different methods of production and their environmental impact. The current most common method

of hydrogen production is a process known as steam methane reformation. This is a chemical

reaction between the simplest hydrocarbon, methane (CH4) and steam. It is a highly endothermic

(∆𝐻 = +206 kJ/mol) and reversible reaction with the following chemical equation [2.8],

CH4 + H2O ⇌ CO + 3H2 (2.4)

Additional hydrogen can be produced by the water-gas shift reaction which is an exothermic (∆𝐻 =

-40.6 kJ/mol), reversible reaction between carbon monoxide and water [2.9], equation 2.5 shows

the water-gas shift reaction,

CO + H2O ⇌ CO2 + H2 (2.5)

14
 Akshay Pal

Therefore the overall equation for this process can be written as follows,

CH4 + 2H2O ⇌ CO2 + 4H2 (2.6)

The reaction has an overall enthalpy change of ∆𝐻 = +165.4 kJ mol-1, which means that it is an

endothermic process and takes in heat. Hydrogen produced in this way is known as grey hydrogen

which is not net zero and therefore, unsuitable for the project. The carbon emissions caused by

this process can be quantified from the overall chemical equation. Taking the molecular weights

(𝑀𝑟) of hydrogen (H2) and carbon dioxide (CO2) to be 2.016 kg kmol-1 and 44.009 kg kmol-1

𝑚𝑎𝑠𝑠
respectively [2.1] and using the equation 𝑚𝑜𝑙𝑒𝑠 = 𝑀𝑟
it is possible to calculate that there will be

5.46 kg of carbon dioxide released per kg of hydrogen produced. This is a “best case” scenario

and due to real-world inefficiencies, there will likely be a greater emission of carbon dioxide than

this calculation suggests. In actuality grey hydrogen emits roughly 9.3 kg of CO2 per kg of

hydrogen produced, which translates to approximately 0.28 kg of CO2 per kWh of energy [2.10].

Hydrogen is also produced through the gasification of coal, hydrogen produced this way is known

as brown hydrogen. Brown hydrogen is the least clean form of hydrogen production.

If a system is implemented to capture and store the carbon dioxide waste, then the hydrogen

produced is known as blue hydrogen. Blue hydrogen is much cleaner than grey hydrogen but due

to the limitations with current carbon capture technology and the fact that such a system could

never capture 100% of the carbon emissions, it is incorrect to call blue hydrogen a net zero fuel. It

is therefore unsuitable for this project. However, it can be viewed as a potential transition fuel as it

makes use of the most common hydrogen production method which is the steam methane

reformation process described above.

Green hydrogen is hydrogen that is produced by electrolysis of water powered by renewable

energy [2.11] and is the only form of hydrogen production capable of achieving the net zero goal of

this project. Electrolysis is the reverse of a fuel cell reaction and hence, it is a non-spontaneous

reaction that requires energy to take place thereby consuming electricity. Just like fuel cells,

electrolysers are electrochemical devices and are classified by the type of electrolyte they use and

their size, the most common are:

15
 Akshay Pal

1. Alkaline electrolysers which use a solution of KOH or NaOH as the electrolyte and have a

typical operating temperature range of 80 - 200°C [2.12].

2. PEM electrolysers which use an acidic electrolyte allowing for the transfer of protons (H+

ions) from the anode to the cathode. Their operating temperature range is 25 - 80°C [2.12].

3. Solid oxide electrolysers use a solid metal oxide electrolyte which conducts oxygen ions.

The most commonly used metal oxide is zirconium dioxide (ZrO2) which is doped with

around 8% yttrium trioxide (Y2O3) or scandium trioxide (Sc2O3). The doping produces

crystallographic defects, allowing oxygen ions to pass through [2.12]. Solid oxide

electrolysers operate at extremely high temperatures in the range 900 - 1000°C [2.13].

It is important to understand the different types of electrolyser and their operating parameters when

designing the system so that an informed choice could be made about what type of electrolyser is

needed. Further comparison, the choice of electrolyser and modelling of its operation is detailed in

Section 6.2 and 6.3 respectively.

The overall water splitting reaction that occurs in an electrolyser is,

H2O → H2 + ½O2 (2.7)

The exact half equations that take place at the anode and cathode differ depending on the type of

electrolyser, specifically they depend on what ion the electrolyte is able to conduct. This overall

equation is the reverse of the fuel cell equation and so input energy is required to make this

reaction happen. This must be provided by a green, renewable electricity source in order to meet

the project goal of being net zero.

2.1.4 Hydrogen storage

Hydrogen can be stored in a variety of ways and this section will give a brief, qualitative overview

of the most common storage methods currently in use. Fig 2.2 breaks down the different

classifications of hydrogen storage methods.

16
 Akshay Pal

< Fig. 2.2: Classification of hydrogen storage methods [1.23] >

Hydrogen can be stored in elemental form in two ways, either as a compressed gas or as a liquid.

Storage as a compressed gas is the most common of the two being utilised by several existing

hydrogen buses such as the Wrightbus H2Bus [1.29] and the Mercedes-Benz EvoBus [2.14]. The

pressure used in onboard storage for buses is 350 bar whereas 700 bar is commonly used for

passenger cars such as the Toyota Mirai [2.15]. Storage as a gas is the simplest storage method.

However, it has a low volumetric energy density and is inefficient due to the energy consumed by

the compression process itself [2.6]. To combat the low volumetric energy density, hydrogen can be

stored as a liquid. This, however, comes with other problems including having to maintain the

hydrogen at a temperature of −253°C at ambient pressure [2.6], which requires specialist tanks.

The liquefaction process is also very energy intensive and hence this is an inefficient storage

method [2.6].

Hydrogen can also be stored in the form of another compound. The two main methods being metal

and chemical/organic hydrides. These storage methods are more useful as long term, back-up

storage in case of a system failure, such as an electrolyser breaking down. Metal hydrides are by

far the safest form of hydrogen storage as they are stable solids which will not immediately

combust upon ignition. However, they do have issues which need to be considered, mainly their

weight and the fact that there are multiple conversion steps required throughout the process which

reduces the overall efficiency [2.6, 2.16]. The most popular metals being considered for this type of

storage are aluminium (Al) and magnesium (Mg) due to their low weight and cost [2.16].

17
 Akshay Pal

Hydrogen can also be stored as a liquid organic chemical such as ammonia (NH3), methanol

(CH3OH) or methanoic acid (HCOOH). These chemicals are widely used in other markets and the

infrastructure for their transport and distribution is already in place making them an attractive option

[2.16]. They are all light organic liquids with a high weight percentage of hydrogen, especially when

compared to metal hydrides [2.16]. However, they also have drawbacks which include a loss of

efficiency due to the extra conversion steps and safety issues as these are all flammable organic

compounds. Of note, however, is ammonia which has a relatively high hydrogen weight

percentage. The synthesis and transportation infrastructure for ammonia is already extremely

mature and widespread [2.17].

Finally, the last form of hydrogen storage which is being explored by researchers is the sorbent

method. This method involves the adsorption of hydrogen onto a framework surface and exploits

physical, rather than chemical, bonds such as the Van Der Waals bonding between hydrogen

molecules and specific materials known as metal-organic frameworks (MOFs) [1.23, 2.16]. This

technology is largely still in development and furthermore, both a low temperature and high

pressure are required in order to obtain a similar hydrogen storage density as existing technologies

[2.16]. This could be a fantastic technology in the future. However, currently it has not reached a

level of scientific maturity where it could become commercially viable.

18
 Lingxi Tang

2.2 Ammonia

2.2.1 Introduction to Ammonia

Ammonia is the second of the two emission-free fuels taken into consideration to be used as an

energy source for Oxford’s public transport system. This section will explain how ammonia is used

as vehicle fuel, the different ways to produce ammonia and the costs of ammonia storage and

distribution. Throughout Section 2.2, comparisons between ammonia and hydrogen will be made

and by the end of the section, readers will understand why hydrogen is chosen as the ideal green

gas.

2.2.2 Ammonia as a fuel

Ammonia Combustion

Ammonia is a colourless and pungent gas at room temperature, with each molecule consisting of a

nitrogen atom and 3 hydrogen atoms. It can be used as a fuel for vehicles in two main ways: direct

combustion to power an internal combustion engine (ICE) or to power fuel cells to generate

electricity.

To fuel an ICE, ammonia combusts with air via the following equation:

NH3 + 3O2 → N2 + 6H2O (2.8)

and moves pistons which provide power to the wheels. As such, ammonia-fuelled vehicles operate

similarly to existing petrol vehicles, except they only emit nitrogen and water vapour, making

ammonia a zero-emission fuel. However, there are concerns with regards to NOx pollution, which

can be rectified using a selective catalyst reduction (SCR) system in NH3 fuelled vehicles [2.18].

Ammonia has similar physical properties to propane [2.18]. Most notably, their condensation

pressures at room temperature are similar at 9.90 atm or 10 bar for ammonia and 9.40 atm or 9.5

bar for propane [2.19]. Thus, with just straightforward modifications, a conventional petrol vehicle

can be retrofitted to run on liquid ammonia [2.18]. The ease of storing ammonia onboard a vehicle

is a strong advantage over hydrogen, which requires a storage pressure of 350 to 700 bar [2.20].

While ammonia excels in storage ease compared to hydrogen, it suffers in terms of energy density:

losing to diesel and gasoline in both volumetric and gravimetric density, and losing to hydrogen in

19
 Lingxi Tang

gravimetric energy density. This means that for the same desired range, a vehicle would have to

carry roughly 7 times the weight of ammonia compared to hydrogen.

There are additional challenges with ammonia combustion, including high ignition temperature, low

flame velocity and slow chemical kinetics, all of which will affect engine performance.

<Fig. 2.3: Gravimetric and volumetric energy density of combustible materials and batteries [2.19]>

These challenges are usually tackled by combining ammonia with traditional fuels such as diesel,

gasoline or even hydrogen to make ammonia/fuel blends. However, this inevitably will result in

carbon emissions for most traditional fossil fuels, and increased toxic NOx emissions for all

ammonia fuel blends (see Fig. 2.4).

Ammonia Fuel Cells

Ammonia fuel cells generally work on the principle of converting the chemical energy of ammonia

into electricity through redox reactions. They are split into direct and indirect fuel cells, with the

latter involving thermal decomposition of ammonia into hydrogen before utilising said hydrogen for

20
 Lingxi Tang

energy, and the former directly using ammonia as an energy source. This report will focus on direct

ammonia fuel cells since this eliminates the need for onboard storage of hydrogen and

decomposition process of ammonia, thereby reducing costs and increasing efficiency [2.22].

< Fig. 2.4: (A) CO2 and (B) NOx emissions for different ammonia fuel blends [2.21] >

Direct ammonia fuel cells can be categorised into solid oxide fuel cells (SOFC), alkaline fuel cells

or microbial ammonia fuel cells [2.22]. SOFCs are generally deemed to be the most ideal type of

direct ammonia fuel cells due to its high efficiency and it is the most studied ammonia-fed fuel cell

technology [2.22, 2.23]. Thus, this section will focus on ammonia SOFCs.

In ammonia SOFCs, ammonia is cracked into hydrogen at high temperatures,

2 NH3 → N2 + 3 H2 (2.9)

and the hydrogen is then immediately utilised to generate electricity [2.22]. The conversion of

hydrogen into electricity was discussed in Section 2.1.1. At high operating temperatures, the

decomposition process and electricity generation process is effectively merged, negating the need

for on-board storage of hydrogen [2.22].

21
 Lingxi Tang

On-board vehicular use of ammonia, however, pose significant challenges, including high operating

temperature, purification requirements (of hydrogen), and high cost of on-board “cracking” of

hydrogen [2.24].

2.2.3 Ammonia production

N2 + 3 H2 → 2 NH3 (2.10)

The predominant method of ammonia production today is the Haber-Bosch Process, where

hydrogen is combined with nitrogen to form ammonia, using an iron-based catalyst, under high

temperature ranging from 325–525°C and high pressure ranging from approximately 150–350 bar

[2.22].

Due to the energy demands of maintaining high temperature, ammonia production via the

Haber-Bosch process currently accounts for 1% of the world’s energy consumption and 1% of

global carbon dioxide emissions [2.25]. This makes ammonia production the highest emitter of

greenhouse gases in the chemical industry [2.26]. In addition to emissions from high energy

demands, most commercial production of ammonia uses brown or grey hydrogen [2.19], which

emits carbon dioxide in their production, as mentioned in Section 2.1.3. Despite its current

environmental impacts, the Haber-Bosch process can be made clean by powering it with

renewable energy and to use renewable energy to produce the green hydrogen needed for

ammonia production.

2.2.4 Ammonia storage and distribution

Ever since the first commercial ammonia plant started production in Germany in 1913 [2.27],

demand for ammonia has increased steadily to meet demands from various industries, including

agricultural, industrial and chemical.

Historically, ammonia is most commonly used as a fertiliser in the agricultural industry. It is also

used as raw material for industrial purposes and as refrigerants for industrial cooling. Additionally,

ammonia is a crucial ingredient for NOx emissions removal [2.19]. As such, the infrastructure

technology for ammonia storage and production is already mature and widespread, with liquid

22
 Lingxi Tang

ammonia being transported around the world via various means including ship, trucks and

pipelines. This compares with hydrogen, which will require new investments in infrastructure.

< Fig. 2.5: Global ammonia production [2.27] >

Due to the maturity of the industry and the physical properties of ammonia, it is much cheaper to

store and transport ammonia compared to hydrogen. As can be seen from Fig. 2.6, not only is it

more expensive to transport hydrogen at £27.83/MWh (compared to £4.72/MWh for ammonia), it is

also more expensive to store hydrogen for transport (£21.61/MWh) compared to stationary

hydrogen storage (£6.18/MWh). As such, the case of ammonia is much stronger when

transportation of fuel over long distances is required. The costs indicated in Fig. 2.6 are for

transportation distances of over 100 km [2.28].

< Fig. 2.6: Comparison between centralised and decentralised models of ammonia decomposition
[2.28] >

23
 Lingxi Tang

2.3 Verdict on green gas

At first consideration, the obvious choice for the fuel of choice is hydrogen since there will be clear

inefficiencies to produce ammonia via hydrogen, when hydrogen can be used directly to power the

public transport system. The undesirability of ammonia is further worsened by the challenges of

on-board ammonia storage for ammonia fuel cells, as well as the energy intensive process to

manufacture ammonia.

Nevertheless, It must be considered that ammonia has two main advantages over hydrogen: first,

ammonia is a suitable fuel for conventional ICE vehicles, indicating an easy transition process from

majority-fossil fuel public transport system to a majority-ammonia public transport system.

Secondly, ammonia storage and distribution technology is more mature and much cheaper than

that of hydrogen. The cost savings are amplified for long distance distribution.

Unfortunately, these advantages are largely rendered irrelevant due to the goal of net zero

emissions and local generation and distribution. For the first advantage, ammonia combustion is

only effective when combined with traditional fossil fuel such as gasoline or diesel. This will

inevitably create carbon emissions, which undermines the goal of net zero emissions. Additionally,

NOx emissions from ammonia combustion will be undesirable for the population of Oxford. Thus,

fuelling combustion vehicles with ammonia will not be considered for this project. For the second

advantage, the cost savings from ammonia storage and distribution will not be realised due to the

goal of having a Oxford-based generation and distribution site, which means minimal transportation

of fuel.

Amidst the criticisms of ammonia discussed so far, perhaps the most prominent flaw of

ammonia-fuelled public buses is the lack of commercial technology and ongoing projects. This

compares with hydrogen bus technology, with existing hydrogen-powered double decker buses by

Wrightbus [2.29] already operating on the streets of Aberdeen [2.30] and soon-to-be on the streets

of Birmingham [2.31] and Liverpool [2.32]. As such, this project will focus on powering Oxford’s

public transport with green hydrogen.

24
 Antoine Thibaut

3 Bus technology and fleet requirements

It is useful to explore the bus technologies which are currently used by bus companies in Oxford.

This will help define the performance targets that will be required by a net zero fleet, in order to meet

or exceed the current standards. Firstly, current diesel and hybrid electric vehicles will be discussed.

Secondly, fuel cell vehicle capabilities will be investigated. The reasons as to why battery electric

vehicles are not considered for a net zero solution have been discussed in Section 1.4. Once vehicle

technologies have been explored, the details of a future bus fleet for Oxford will be discussed. This

section involves financial calculations. To convert euros or US dollars into British pounds, currency

exchange rates as of February 2021 have been used. This results in a factor of 0.87 applied to

convert euros to pounds [3.1] and a factor of 0.72 to convert US dollars to pounds [3.2].

3.1 Diesel bus technology

The technical aspects of modern diesel powertrains will be analysed to understand their benefits

and limitations when operating Oxford bus routes.

One of the most common diesel buses used in Oxford is the Wrightbus Streetdeck [1.12]. A benefit

of diesel buses is their range. An average diesel bus has a range of 690 miles [1.28] or 1,100 km

[2.1]. This large range implies that bus operators do not need to schedule refuelling frequently. This

is useful when operating a busy service. The range requirements for buses to operate Oxford

routes will be described in further detail in Section 3.4.2.

Another benefit of diesel buses is the ease of refuelling. Diesel can be quickly transferred in liquid

form from a large storage tank to the on-board fuel tank. The fuel does not need to be compressed,

refrigerated or transformed into a different state. It is then essential that a net zero solution in

Oxford also relies on a fast refuelling service for bus companies to keep their current service times,

reduce idle time, maximise equipment availability and optimise the return on their investment. The

refuelling requirements for a hydrogen fleet will be discussed in Section 3.4.3.

One drawback to diesel powertrains is how they must rely on pollution reducing systems to meet

environmental targets [3.3]. These systems add to the total cost of the bus. Diesel engines typically

produce soot as a by-product of the combustion process [3.4]. In order to avoid the formation of

25
 Antoine Thibaut

soot particles, current generation diesel engines run on a lean fuel mixture [3.3]. In addition to

running the engine lean, diesel powertrains also feature particulate exhaust filters, which trap the

soot particles which could still be produced by the engine [3.5]. The excess oxygen used to

achieve the lean fuel mixture leads to the formation of a significant amount of NOx emissions [3.3].

To reduce these NOx emissions, a diesel exhaust fluid is injected into the exhaust gas [3.5]. Carbon

emissions have been discussed in Section 1.5.

In order to meet future regulations, diesel buses will have to feature more complex pollution

reducing systems, further adding to cost. Hybrid technologies are an alternative for bus companies.

3.2 Hybrid bus technology

A common short term goal for Oxford bus companies is to replace conventional diesel buses by

hybrid buses [1.41, 3.6]. It would be useful to analyse what features of a hybrid bus can be carried

over to a net zero solution.

Hybrid buses feature both a diesel engine and an electric motor [3.7]. They are equipped with

battery packs which can be charged by the diesel engine or through regenerative braking [3.7].

There are three main types of hybrid vehicles illustrated in Fig. 3.1 below, adapted from [3.7]. The

arrows represent directions of energy transfer either mechanically or electrically.

< Fig. 3.1: Various hybrid systems used on buses, adapted from [3.7] >

26
 Antoine Thibaut

The series hybrid is the most popular configuration for low speed city buses [3.7]. The drivetrain is

entirely driven by an electric motor. This allows the engine to operate efficiently at an optimised

rpm setting regardless of the vehicle speed [3.7].

Hybrid buses provide ways to reduce emission levels by optimising the engine operating point and

to improve fuel efficiency through regenerative braking. Hydrogen buses can also benefit from using

electrical energy storage to optimise the power plant (fuel cell) operating point [3.8]. Regenerative

braking systems are also incorporated in the design as will be discussed in the next section [3.8].

3.3 Hydrogen bus technology

A hydrogen bus uses hydrogen to generate electricity through a fuel cell [3.9]. The process of

generating electricity from a fuel cell is discussed in Section 2.1.1. Hydrogen buses only emit water

vapour [3.9]. Fig. 3.2 below outlines the main features of a hydrogen bus.

< Fig. 3.2: Hydrogen bus outline adapted from [3.9] >

There are many subsystems in a hydrogen bus which ensure the fuel cell runs as efficiently as

possible. For example, the fuel cell system consists of a power electronics controller and a thermal

system [3.8]. The power electronics controller distributes electrical power between the fuel cell, the

motor and on-board electrical storage [3.8]. The fuel cell’s performance is closely related to its

operating temperature [3.10], hence a thermal management system controls the fuel cell

temperature [3.8]. Additional subsystems also ensure the fuel cell performs well during the winter

under sub-zero temperatures [3.11]. If the bus is held at sub-zero temperatures for long periods of

time, a control system will purge water from all lines and fuel cell components [3.11]. This allows

the bus to be stored at sub-zero temperatures without damaging the fuel cell unit. This ensures

that bus companies in Oxford will be able to park their buses outside throughout the year, avoiding

the need to store buses in temperature controlled depots.

27
 Antoine Thibaut

Fuel cell vehicles have been developed for more than a decade [3.12]. Practical applications have

demonstrated their robustness for automotive applications, making hydrogen a viable option for a

net zero emission fleet. These applications include hydrogen buses currently used in France [3.13].

3.4 Oxford fleet requirements

This section will outline the performance targets and specific requirements which must be met by a

hydrogen bus fleet in Oxford.

3.4.1 Bus type requirements

To determine the types of hydrogen vehicles required for Oxford, the current number of single and

double deck buses in Oxford will be tallied. As mentioned in Section 1.3, 148 buses are currently

owned by the Oxford Bus Company (OBC) [1.12], 83 buses are owned by Thames Travel [1.13] and

168 buses are owned by Stagecoach [1.14, 3.14]. This represents around 399 total buses in Oxford.

The amounts of double deck and single deck buses operated by each company are shown below in

Table 3.1. Table 3.1 also illustrates how many diesel and hybrid buses are owned by each company.

< Table 3.1: Types of buses operated by Oxford bus companies >

In total, 150 single deck buses and 249 double deck buses are used in Oxford.

The hydrogen storage tanks are situated on the roof of typical single and double deck hydrogen

buses [3.17]. Since the passenger compartment space is not sacrificed, it can be suggested that

hydrogen buses have similar passenger capacities compared to diesel buses. It will be assumed

that the number of hydrogen buses required is equal to the current number of buses in Oxford bus

fleets.

28
 Antoine Thibaut

3.4.2 Range requirements

It is useful for bus companies to only refuel their buses once a day. This maximises the vehicle

usage. Oxford buses must then have a range larger than the distance they are expected to travel

per day. Using bus timetables [3.18], road distances [1.40, 2.1] and a Python model to compile the

data, the total distance travelled by all OBC Park and Ride routes on a single weekday is

approximately 5,812 km. These are some of the longest routes in Oxford which span outside of the

city [1.40] (see Fig. 4.1). There are 19 buses operating these routes [1.12]. This results in around

306 km required by each bus to travel in a single day on average. This range requirement will be

analysed in Section 3.6.1.

3.4.3 Refuelling time requirements

Most buses in Oxford typically operate between 4am and 8pm [3.18]. This leaves an 8 hour period

in which refuelling can take place. Refuelling can either be conducted in the early morning or late

evening. There are 148 buses [1.12] assigned to the Oxford Bus Company depot in Cowley [1.40].

This is the largest number of buses assigned to a single depot (see Table 7.1). It is then a

requirement for a single depot to be able to refill 148 buses in an 8 hour period. This refuelling

requirement will be analysed in Section 3.6.2.

3.4.4 Speed requirements

Hydrogen buses will have to sustain the same speeds used by current diesel buses. There are

A-type roads in Oxford where buses are required to reach a speed of 60 mph [1.40, 3.19]. For

instance, the S5 Park and Ride route covers 9.2 km on the A34 road [1.40, 2.1]. This speed

requirement will be analysed in Section 3.6.3.

29
 Antoine Thibaut

3.5 Hydrogen fleet advantages

This section will discuss the impact of moving from a diesel/hybrid fleet to a hydrogen fleet in terms

of vehicle cost, maintenance and environmental concerns. Fuel costs will be discussed in Section 4.

3.5.1 Vehicle costs

The prices of diesel and hybrid buses are outlined below in Table 3.2. This table shows the price

evolution as a function of time. The table also extrapolates these prices to 2040, when the net zero

target is expected to be achieved. These extrapolated prices will be used to show that hydrogen

buses are expected to reach a similar price to hybrid and diesel buses in the future, assuming that

these buses will still be meeting environmental regulations. The extrapolation was based on a line

of best fit generated with Python and the polyfit function. Prices are assumed to follow the same

linear evolution after 2030.

< Table 3.2: Comparing the price evolution of diesel and diesel hybrid buses >

Table 3.2 shows an increase in the forecasted prices of diesel buses with time. Diesel powertrains

are likely to become more expensive with more pollution managing technologies to be added. The

forecasted purchasing price of diesel hybrid buses appears to remain constant with time. With an

increase in production volume for hybrid buses, the selling price can be expected to decrease.

However, inflation can compensate for this decrease in selling price.

Table 3.3 below shows the price evolution of hydrogen buses.

< Table 3.3: Price evolution of hydrogen buses >

30
 Antoine Thibaut

Contrary to diesel and hybrid buses, a significant decrease in the selling price of hydrogen buses is

expected with time. The selling price of hydrogen buses is expected to drop by 22% between 2020

and 2030. This is mostly due to economies of scale, as hydrogen solutions transition to industrial

production [3.21]. It is only in recent years that bus companies have begun to benefit from

economies of scale. Hydrogen buses were initially sold in very small volumes. Only 20 2010 Winter

Olympics buses were produced for example, which led to a high £1,440,000 unit bus price [3.12,

3.2]. The relatively high purchase price of hydrogen buses can also be explained by the warranties

provided for 5 to 10 years compared to 2 years for a standard diesel bus [3.9].

Hydrogen buses are therefore forecasted to reach competitive vehicle costs. Oxford bus

companies will be able to transition to a hydrogen bus fleet with similar investment costs to those

required to renew their current fleets with diesel/hybrid technology.

3.5.2 Maintenance costs

In order to model how much bus companies are spending on bus maintenance per year, it is useful

to analyse bus maintenance costs. Section 8 will then include the maintenance costs found to

estimate the annual expenditures of bus companies.

Diesel buses have an average service cost of £0.1/km [3.22]. This relatively low maintenance cost

makes diesel buses appealing to bus operators. Diesel buses also have a relatively long lifetime of

15 years [3.23].

The service costs of hybrid buses are typically 4% lower than for diesel buses [3.24]. This

represents a service cost of £0.09/km for an average hybrid bus. A hybrid bus typically has a

lifetime of 12 years [3.25]. The maintenance and replacement of lithium ion batteries represents

most of the hybrid bus maintenance process [3.25]. The battery packs need to be replaced twice

during the bus lifetime [3.25].

Hydrogen buses are expected to have a service cost of €0.3/km [1.31] or £0.26/km [3.1]. This

service cost is three times higher than the cost found for diesel buses. However, one can estimate

the service costs of hydrogen buses to decrease over time as bus operators develop new

techniques for maintenance. The expected lifetime of a hydrogen bus is 12 years [3.9].

31
 Antoine Thibaut

3.5.3 Noise

Hydrogen buses have the additional benefit of generating significantly less noise than diesel

buses. Table 3.4 below compares the noise levels of hydrogen and diesel buses for when the

buses are stationary and moving.

< Table 3.4: Noise levels for diesel and hydrogen buses >

A reduction in noise levels represents an improvement in noise pollution in urban areas. It can be

suggested that a reduction in noise will increase certain property values in Oxford, hence providing

additional support to the financial effort of refitting Oxford fleets with hydrogen buses.

3.5.4 Conclusion on hydrogen fleet advantages

Hydrogen is a viable solution for a net zero bus fleet because it allows for zero CO2 and NOx

emissions. Due to the nature of an urban bus network, the fleet can be refuelled at specified

locations, minimising the hydrogen storage and refuelling infrastructure needs. Due to larger

production volumes, vehicle unit prices are becoming interesting for bus companies [3.9]. One

manufacturer of hydrogen buses is Wrightbus [1.31]. Wrightbus is planning to mass produce the

H2Bus, a hydrogen bus available in single and double deck configurations [1.31]. This is one of the

most popular buses considered by bus companies in the UK [3.26].

3.6 Fleet retrofitting

While it is possible to upgrade old diesel engines to lower their emission levels by installing

pollution reducing systems [3.27], it is much too complicated and expensive to completely retrofit

old diesel buses with hydrogen fuel cells and energy management systems. It is therefore

necessary to eventually replace Oxford’s entire bus fleet with hydrogen-powered vehicles in order

to transition to a net zero emission solution. This section will examine whether the Wrightbus

H2Bus will be suitable to fulfil Oxford fleet requirements.

32
 Antoine Thibaut

3.6.1 Range verification

The hydrogen buses proposed by Wrightbus have a range of 310 to 450 km [1.31] without

extended range options. As mentioned in Section 3.4.2, it is a requirement in Oxford for a bus to

have a range of at least 306 km. The 310 km range of the double deck H2Bus is therefore

insufficient to ensure that all buses can cover their routes with some safety. Extended range

options will be required for the double deck buses. This extends the range to 420 km [1.31].

3.6.2 Refuelling time verification

The refuelling times of hydrogen buses are limited due to temperatures in the on board storage

tanks rising during refuelling [3.28]. Type IV hydrogen tanks are often used for hydrogen vehicle

applications [3.29]. These tanks feature an inner plastic liner which must be kept under 85 ºC to

prevent damage [3.28, 3.29]. The extent of the temperature increase during refuelling can be studied.

Fig. 3.3 below performs a control volume analysis of refuelling a Type IV tank from an empty

vacuum state to 350 bar. 350 bar is the standard pressure used to refuel a hydrogen bus [3.30].

< Fig. 3.3: Control volume analysis for hydrogen entering a tank adapted from [3.31] >

The estimated final temperature is 𝑇2 = 𝛾𝑇1. For hydrogen, 𝛾 = 1.41 [2.1]. Assuming a filling

temperature of 25 ºC, 𝑇2 can be calculated:

𝑇2 = 1.41(25+273) = 420.2 K = 147.2 ºC

This is the temperature which would be reached in the hydrogen tank if filled from an initial

vacuum. This temperature is above 85 ºC and would therefore damage the tank. In practice the

33
 Antoine Thibaut

hydrogen tank is not filled from an empty vacuum state. However, the large pressure difference

between the tank and the fuelling station dispenser outlet justifies this approximation. In order to

refuel the buses quickly without damaging the tank, the hydrogen can be refrigerated before

transferring it to the bus [3.28]. Refuelling with refrigeration can take 5 minutes per bus [3.17].

Refuelling without refrigeration can take up to four times longer [3.32], resulting in 20 minutes per

bus. To account for a three minute time taken to move the buses to the hydrogen dispensers, the

refuelling time can be approximated as 8 or 23 minutes per bus. A model was created to determine

whether refrigeration is required before refuelling an Oxford bus fleet.

The model assumes that 148 buses are allocated to a single refuelling station. As mentioned in

Section 3.4.3, this is the maximum number of buses assigned to a single depot in Oxford. Fig. 3.4

below shows the time taken to refill all 148 buses for different amounts of hydrogen dispensers. A

red horizontal line is drawn to indicate the 8 hour refuelling period limit mentioned in Section 3.4.3.

< Fig. 3.4: Number of hydrogen dispensers required for non-refrigerated and refrigerated systems >

Fig. 3.4 shows that a single dispenser is insufficient to refuel 148 buses within an 8 hour timeframe,

regardless of whether refrigeration is used. Three hydrogen dispensers with refrigeration allows

refuelling to take place within the 8 hour period. Without refrigeration, 8 dispensers will have to be

used. This would drive the costs and space needed for the refuelling station. As a conclusion,

refrigeration will be used at all hydrogen fuelling stations to minimise the cost of additional

hydrogen dispensers and the space associated with the fuelling station.

34
 Antoine Thibaut

3.6.3 Speed verification

The Wrightbus H2Bus cannot sustain speeds higher than 40 mph on fuel cell power alone due to

limited fuel cell power output [3.26]. Additional power is provided by the electrical motor to sustain

speeds of up to 60 mph [3.8, 3.26]. Supercapacitors are used to store and deliver the electrical

energy during this intermediate power need [3.33]. These have a capacity of 20 kWh [3.33] or 72

MJ. The supercapacitors can be charged through regenerative braking or by the fuel cell when the

vehicle is not requiring the maximum fuel cell power [1.30].

To determine whether the Wrightbus H2Bus would be able to sustain 60 mph on A-type roads in

Oxford, it is necessary to first verify whether the intermediate energy storage is large enough to

provide the energy required, and secondly whether this energy can be acquired through

regenerative braking. This verification process will be applied to all Park and Ride routes in Oxford.

Park and Ride routes have been selected for this study as they span well throughout Oxford and

outside the city [1.40]. Park and Ride routes are illustrated below in Fig. 3.5. Fig. 3.5 has been

constructed by using bus timetables [3.34, 3.35, 3.36, 3.37] and Google Maps [1.40]. This figure

also features a superimposed topography map created with code borrowed from [3.38] in Python.

Points and elevations were taken from [3.39].

< Fig. 3.5: Park and Ride routes with superimposed topography map >

35
 Antoine Thibaut

Fig. 3.5 illustrates that Oxford is situated in a sink area of lower land. This is shown by the lighter

yellow colours on the outside of the figure surrounding the central blue area of the map. This

feature will play an important role in energy management.

To model energy management, a free body diagram will be considered as shown below in Fig. 3.6.

< Fig. 3.6: Free body diagram for a bus travelling at constant speed >

The energy required by a bus to sustain a speed, 𝑣, over a distance, 𝑑, will be computed. This energy

is calculated by multiplying the air resistance force, 𝐹𝑎, by the distance travelled, 𝑑. Wheel

resistance is neglected in this model. An equation for the energy required, 𝐸𝑣 , is shown below.

𝐸𝑣 = ( 1
2
𝐶𝐷𝜌𝐴𝑣 𝑑
2
) (3.1)

The additional energy required by a bus accelerating from 40 mph to 60 mph and sustaining 60 mph

over a certain distance will be computed. This amount of energy is estimated by using equation 3.1

and by adding the difference in kinetic energy, ∆𝐾𝐸, when accelerating from 40 mph to 60 mph:

∆𝐸 = ∆𝐸𝑣 + ∆𝐾𝐸

= ( 1
2
𝐶𝐷𝜌𝐴𝑣2 𝑑 −
2 1
2
2
) (
𝐶𝐷𝜌𝐴𝑣1 𝑑 +
1
2
2
𝑀𝑣2 −
1
2
𝑀𝑣1
2
)
=
1
2 (𝑣 2
2
− 𝑣1 )(𝐶 𝜌𝐴𝑑 + 𝑀)
2
𝐷
(3.2)

Where 𝑣1 corresponds to 40 mph and 𝑣2 corresponds to 60 mph. Table 3.5 below was created by

applying equation 3.2 to Park and Rides in Oxford. A single bus is considered for each route.

36
 Antoine Thibaut

< Table 3.5: Tabulating the additional energy required to obtain 60 mph >

Table 3.5 shows that all of the energies required are less than the 72 MJ capacity of the

intermediate energy storage of the Wrightbus H2Bus. This additional capacity can be used for

successive accelerations for example on high speed roads.

It will now be determined how the required energies can be generated during normal operation. Two

regenerative braking scenarios will be considered. During descent, the mechanical resistance of

the generator is used to provide braking power and maintain a constant speed while charging the

supercapacitors. When the bus decelerates to a stop, energy can also be recovered through the

generator. A free body diagram will be considered to model descent at constant speed. The free

body diagram is shown below in Fig. 3.7. A similar method can be found in the literature [3.40].

< Fig. 3.7: Free body diagram for a bus travelling at constant speed in descent >

From equation 3.3 in Fig. 3.7 the regenerative braking force, 𝐹r, can be found. To calculate the

inclination angle, θ, the elevations of different bus stops and road distances were used [3.39].

Regenerative braking systems can have an energy recovery efficiency as high as 70% [3.43]. This

efficiency will be used in the model. 𝐹r is multiplied by the distance travelled, 𝑑, and by the recovery

efficiency to calculate the energy generated from the regenerative braking system during descents.

37
 Antoine Thibaut

This method is applied to all Park and Ride routes shown in Fig. 3.5. Results are shown below in

Table 3.6, using a Python script to compile the data. Regenerative braking is considered from the

two scenarios explained above: energy recovered from descents at a constant 40 mph speed and

the energy recovered when the bus comes to a full stop, with a recovery efficiency of 70% [3.43].

Table 3.6 also shows an estimate of how much the energy recovered represents in kilograms of

hydrogen per year. This was calculated by using a hydrogen energy density of 120 MJ kg-1 [1.23]

and a conversion efficiency of 60% [2.6] for converting hydrogen into useful work. Bus frequency

data was also taken from bus timetables [3.18] to estimate how often the buses would travel per

year. Outbound routes refer to bus routes going out of the city centre. RBS stands for regenerative

braking system.

< Table 3.6: Energy recovered or hydrogen saved for all Park and Ride routes >

Table 3.6 shows that the energy recovered from regenerative braking is sufficient to provide the

intermediate energy for all routes apart from for the S5. Regenerative braking alone is therefore

insufficient to charge the intermediate energy storage in this case. Hence, the bus will have to

charge its supercapacitors by using the fuel cell power during lower speed travel. As many roads in

Oxford are limited to 20 mph [3.44], the fuel cell will be able to fulfil this role during a significant

portion of the daily route. The fuel cell power is 75 kW [3.42]. If 25% of its capacity is used to

recharge the energy storage, it will be able to provide the 8.68 MJ of additional energy required for

the S5 outbound route in around 460 seconds.

Table 3.6 also shows that more energy is recovered on inbound routes. This can be explained by

the elevation difference between the city centre and its surrounding. As shown in Fig. 3.5, the 300

and S5 routes mostly stay in regions of lower elevation. The difference between inbound and

38
 Antoine Thibaut

outbound energy recovery is therefore minimal. However, the 400 and 500 routes transition

between regions of high and low elevation. This results in a significant difference in the energy

recovered for inbound and outbound routes.

There are some limitations in the model used above. The regenerative energy recovery efficiency

may drop from 70% to 16% depending on the bus driving style [3.43]. This implies that less energy

may be recovered during descents and stopping. The model assumed 70% kinetic energy recovery

while braking from 40 mph to zero. The buses are also likely to operate at slower speeds closer to

the city centre. This would result in a lower kinetic energy recovered from stopping the bus.

In practice, when buses come to a full stop, regenerative braking systems can only slow down the

bus by 0.4g [2.6]. Beyond 0.4g, in strong braking, conventional brakes activate [2.6]. However,

passengers very rarely experience decelerations greater than 0.3g [2.6]. This implies that

regenerative braking can cover all braking situations. Fully regenerative braking is also beneficial to

a net zero solution as it drastically reduces the brake disc usage and wear. Disk pad wearing

contributes up to 20% of air pollution associated with particles smaller than 2.5 μm in diameter [3.45].

Now that the range, refuelling time and speed capabilities of the Wrightbus H2Bus have been

shown to match Oxford requirements, the financial aspects of transitioning to a hydrogen bus fleet

will be explored.

3.6.4 Financing a hydrogen bus fleet

The selling prices of the Wrightbus single deck and double deck hydrogen buses are forecasted to

be €375,000 and €410,000 respectively [1.31]. This converts to £326,250 and £356,700 [3.1]. As

shown in Table 3.1, 150 single deck and 249 double deck hydrogen buses will need to be

purchased resulting in a total investment of around £138 million. This investment would be

deployed over several years. In order to propose a replacement strategy, the age distribution of the

current diesel/hybrid fleet will be analysed.

In 2020, the average age of the 231 buses operated by the Oxford Bus Company and Thames

Travel was 7.8 years [3.46]. The average age of the 168 Stagecoach buses was 7.5 years [3.47].

39
 Antoine Thibaut

The average bus age in Oxford can then be calculated to be around 7.7 years. As mentioned in

Section 3.5.2, the diesel and hybrid buses have a lifetime of 15 and 12 years respectively.

It is expected that bus companies will start by replacing their older conventional diesel buses,

which may not be able to meet future emission regulations, and for which maintenance costs

become more significant as they reach the end of their service life. Then, in a second step, bus

companies could consider selling their mid-life buses of around 7 to 8 years of service. In order to

estimate the contribution coming from the sale, it is useful to calculate the typical residual values

for mid-life diesel and hybrid buses.

A diminishing value method can be used to calculate how much these buses have depreciated

over time, applying a high level of depreciation at the start of the bus lifetime and a progressively

lower level of depreciation towards the end of the bus lifetime [3.48]. The formula for calculating

diminishing value for a bus is provided below [3.48]:

(3.4)

Equation 3.4 can be applied to calculate the residual value of hybrid and diesel buses. Calculations

are shown below in Table 3.7. 2020 purchasing prices are used as an approximation.

< Table 3.7: Calculating the residual values of buses in Oxford >

From Table 3.7, an average diesel bus in Oxford is expected to have a residual value of around

£62,100 versus £79,500 for an average hybrid bus of a similar age. Should another bus company

be interested in making an acquisition, these amounts could definitely contribute to accelerating

the transition to a net zero fleet in Oxford.

Naturally, the most recent buses, including the more advanced hybrid versions, would remain in

service for the longest time, to finally be considered for replacement once the older, conventional

diesel vehicles are replaced. The investment timing will be further discussed in greater detail in

Section 8.

40
 Akshay Pal

4 Estimating the hydrogen demand

This section will detail the process of deriving an estimate of the annual hydrogen demand of the

system and present the results of the route modelling which was done in order to understand the

composition of the current bus network.

4.1 Route modelling

In order to begin estimating the annual hydrogen demand of the system, it was first necessary to

have a comprehensive understanding of the current bus network. The route modelling was done in

order to answer the following three questions:

1. How many different routes are provided by the three network operators identified in the

project scope?

2. What is the annual distance travelled by the bus network?

3. What is the annual diesel fuel demand and the associated carbon and financial costs?

By answering these questions, it would be possible to estimate the hydrogen demand necessary to

provide the same bus coverage and the costs of the current network, both environmental and

financial, would be quantifiable which would justify the reasons why transitioning to a green

hydrogen based transport system is required.

The first step was to identify how many bus routes fell within the scope of the project. In total, 58

routes were analysed, 25 of which are operated by OBC, 17 by Stagecoach, 8 by Thames Travel

and 8 were jointly operated by OBC and Stagecoach. The next step was to use a combination of

Google Maps [1.40] and the online timetables of the network operators [1.15, 1.16, 1.17] to

calculate the distance travelled by the bus network annually and, perhaps more importantly, to

have this data at a high resolution i.e. to have a figure for daily distance travelled on both a

weekday and a weekend. The result of this analysis for OBC is shown below in Fig. 4.1.

Generating similar results for Stagecoach and Thames Travel led to an estimate of the total

distance travelled annually by the bus network equaling 33,542,938.2 km or ~ 33.5 million km.

41
 Akshay Pal

(a)

(b)

< Fig. 4.1: Estimate of the (a) weekday and (b) weekend mileage for OBC operated routes >

Both graphs have a similar shape with the weekend mileage being lower overall as expected.

Routes 300, 400, 500 and S5 are park and ride routes and hence travel significantly more miles

during the day. The inner city routes such as 4, 4A and 4B have a much lower mileage as their

routes are much shorter. It is also important to note the peak of the LGW service to London

Gatwick which sees a sharp increase on the weekends. This is further justification for a hydrogen

based system over a battery electric one as the better range of FC buses is what will enable this

route to continue running, an electric bus would simply not have the battery capacity to complete

such a long route.

42
 Akshay Pal

To calculate the fuel consumption from the estimates of distance, information about the mpg

ratings of the buses running on each route was required. This task proved difficult as not only were

there multiple, different, buses assigned to a single route such that there was not one single mpg

rating associated with a particular route, there was also a significant lack of mpg data available

online. This problem was partially resolved thanks to Luke Marion, who is the Finance and

Commercial Director at OBC, who was able to provide us with specific mpg ratings for the different

bus models, however, there was still the issue of multiple different mpg ratings associated with any

particular route. The solution to this problem was that, in order to avoid underestimating the annual

fuel consumption, the upper bound ‘worst case scenario’ solution would be used. This meant that

for each route, the lowest mpg rating associated with it was always used in the calculation of fuel

consumption, hence calculating the upper bound of annual fuel demand. The actual calculation

used was simply to divide the distance travelled in miles by the mpg rating as shown in equation

4.1.

fuel usage in gallons = distance travelled in miles ÷ mpg (4.1)

Again, using OBC as the example, Fig 4.2 shows the fuel consumption of each route on both a

weekday and a weekend. Similar graphs were generated for Stagecoach and Thames Travel and

the result was a diesel consumption of 13,383,242.6 L or ~ 13.4 million litres annually. This is

hugely costly, both environmentally and financially, resulting in a carbon footprint of 35,867 tonnes

and an annual fuel expenditure of £12,281,802, where the price of diesel has been taken to be

£0.9177 per litre, including fuel duty [4.1]. Transitioning to a green hydrogen based system will not

only significantly reduce carbon emissions but it will also reduce the annual fuel expenditure of the

network operators. A detailed financial analysis will be presented in Section 8.

43
 Akshay Pal

(a)

(b)

< Fig. 4.2: Estimate of the (a) weekday and (b) weekend fuel consumption for OBC operated
routes >

Appendix A provides an example of the bus route data that was collected for different routes

around Oxfordshire.

4.2 Energy equivalence method

In order to estimate the hydrogen demand, an energy equivalence method was used. The

assumption was made that the same amount of energy needed to be delivered to the wheels. This

assumption is made on the basis that the buses will be driving along the same routes and at the

same speeds (there are no changes being made to the speed limits on the roads of Oxfordshire).

Kinetic energy is proportional to velocity squared and the assumption is that this will stay constant

44
 Akshay Pal

meaning the kinetic energy will stay constant. No doubt there will be differences in the mass of a

hydrogen bus when compared to a diesel bus, however, as stated kinetic energy is dominated by

the velocity term. On top of this the overall mass depends on the passenger load and this is also

not changing meaning that the change in the mass of the bus will be negligible. To calculate the

hydrogen demand, the amount of energy provided by diesel to the wheels would be required first.

This led to the formation of equation 4.2,

𝐸 = 𝑉𝜌𝜂 (4.2)

where 𝐸 is the amount of energy delivered to the wheels, 𝑉 is the volume of fuel in litres, 𝜌 is the

volumetric energy density of the fuel and 𝜂 is the tank-to-wheel efficiency of the vehicle. The 𝑉𝜌

term represents the total stored chemical energy of the fuel and so multiplying this by the efficiency

𝜂 gives us the total useful energy i.e. the energy delivered to the wheels.

4.2.1 Accounting for efficiency

The tank-to-wheel efficiency 𝜂 incorporates all the possible sources of energy loss from the fuel

tank through to the wheels, the main losses being in the engine/fuel cell and the drivetrain. Fig. 4.3

below shows the components within the drivetrains of both a fuel cell and diesel bus respectively,

and how the tank-to-wheel efficiency is calculated.

(a)

(b)
< Fig. 4.3: Tank-to-wheel efficiency of (a) fuel cell and (b) diesel bus >

45
 Akshay Pal

The literature values for the tank-to-wheel efficiency are extremely varied for both diesel and fuel

cell vehicles giving a large range of possible efficiencies. For fuel cell vehicles, there was a range

of 30 - 53.8% [4.2, 4.3, 4.4, 4.5] and for diesel a range of 20 - 24% [4.2]. These ranges, especially

for the fuel cell vehicle, are extremely large and doing a calculation based on these numbers would

be subject to a large error and so, in order to avoid significantly underestimating the required

hydrogen demand, the decision was made to neglect the drivetrain efficiency and focus solely on

the engine/fuel cell efficiency for which there were more more universally accepted figures. This

simplification can be done as an electric drivetrain is significantly more efficient than a traditional

mechanical drivetrain [4.5, 4.6]. This means that including these figures will decrease the amount

of hydrogen required and so by excluding them a larger demand than actually necessary was

calculated, thus ensuring the system has enough hydrogen.

The efficiency of a modern PEM fuel cell sits at around 60% [2.6], 23% less than the theoretical

maximum which is due to heating effects, electrical and ionic resistance and the kinetics of the

chemical reaction [2.2]. This has been discussed in detail in Section 2.1.2. The efficiency of a

modern diesel bus operating in urban, city-driving conditions is around 41% [4.7]. The volumetric

energy density of both fuels has already been discussed. In this calculation the values are taken to

be 2.80 MJ L-1 for hydrogen (at 27°C, 350 bar) [1.23] and 38.6 MJ L-1 for diesel [1.27]. The values

used in the calculation of hydrogen demand are summarised in Table 4.1 below where 𝜂 denotes

efficiency and 𝜌 denotes volumetric energy density.

< Table 4.1: Summary of the values used in the calculation of hydrogen demand >

4.3 Results

Using the estimate for the volume of diesel fuel used annually and by applying the energy

conversion equation, 𝐸 = 𝑉𝜌𝜂, fuel consumption was converted into energy consumption and

then worked backwards to estimate the hydrogen demand. The total annual hydrogen required by

the system is 2,933,738 kg (using a mass density of 23 kg m-3 at 350 bar [2.6] to convert from

volume to mass). Thanks to the resolution of the route modelling, this figure could be broken down

46
 Akshay Pal

into the requirements on a weekday and weekend which the model calculated to be 8548.2 kg and

6838.6 kg respectively. This data was needed to design the distribution system ensuring that

hydrogen is delivered from the production site to the various bus depots in the correct quantities

each day. This will be discussed further in Section 6. The breakdown of the total annual hydrogen

demand by bus operator is shown in Fig. 4.4.

< Fig. 4.4: Annual hydrogen demand of each network operator in tonnes >

This gave a hydrogen fuel usage of 8.7 kg per 100 km, agreeing with the literature values which

are in the range 8 - 9 kg per 100 km [4.8].

4.3.1 Distribution demand

In order to keep the system net zero, it is necessary to consider the fuel required to power the

distribution of hydrogen from the production plant to the individual bus depots where refuelling will

take place. This distribution cannot be done by diesel vehicles as the emissions caused by this

would be counted in the total system carbon footprint, hence the extra demand caused by this

needs to be considered. The value for the distribution demand is 26,800 kg. This value will be

derived and discussed alongside the distribution strategy as a whole in Section 7.2.

4.3.2 Emergency demand

It is important to think about contingency plans in case of an emergency such as system failure

preventing the production or distribution of hydrogen. This could be in the form of an electrolyser

failing or trucks being unavailable to deliver the hydrogen from the production site to the bus

47
 Akshay Pal

depots. It was estimated that, in the case of an electrolyser failure, it would take a maximum of two

weeks to solve the problem and so an emergency overheard that would last that amount of time

would be required. This was simple to calculate as the weekly requirement of hydrogen (again

broken down into weekdays and weekends) was known from the route modelling. This emergency

overhead comes to 110,000 kg. However this is not included in the annual demand figure as this is

not part of the normal operation of the system, it is a contingency store. This excess will be slowly

built up during the course of operation.

4.4 System Overview

< Fig. 4.5: Green hydrogen production and storage system overview >

Fig. 4.5 above shows each component of the system which consists of a green energy production

system to supply the hydrogen production plant with green energy where possible, supplementing

with grid electricity when green production is insufficient. The hydrogen production plant itself

contains the smart electrolyser-battery system and produces the hydrogen ready for storage.

There are two forms of hydrogen storage, long term and short term. The distribution system

delivers hydrogen to the various bus depots around Oxford ready for use by the bus operators.

These systems will be expanded upon in great detail in the following sections which will explain the

48
 Akshay Pal

choices that were made and provide detail on how these systems will operate and connect with

each other to power the proposed net zero transport network.

49
 Alexander Astor

5 Social and legal study

Before the technical details of the proposed system are discussed, it is important to consider the

social and legal implications of the project. The purpose of this section is to explain regulations

around the various areas of this project, as well as to show the public response to a hydrogen

public transport network.

5.1 Social Study

FCEVs are yet to be brought into broad usage, especially in the UK. As a result of this, it is

important for this project to consider the opinions of the public on a large-scale hydrogen based

bus network.

Cummins, a global power company, recently published an article which showed the support that

the UK public have for a public transport system involving hydrogen [5.1]. Collating the opinions of

6000 respondents across the UK, Belgium and Germany, the survey found that 48% of British

citizens expressed that low-carbon technology would be vital for the UK’s economic recovery from

COVID-19 [5.1]. Furthermore, 40% of survey respondents were willing to pay up to £1 more for

their daily commute in order to reduce their carbon footprint [5.1]. This shows the extent to which

the public are willing to make sacrifices to encourage the switch to green transport, beyond simply

being in support of the idea of a low-carbon solution. Significantly, according to the Oxford Bus

Company website, passenger fares make up 80% of the financing for the bus services [5.2]. Given

that a bus commute currently costs £2.30 on their city services [5.3], the potential

passenger-supported increase in bus fare could make a huge difference to the financial prospects

of the proposed system. When discussed with Mr. Marion [3.26], the communications and financial

director of the Oxford Bus Company, he observed that these changes would certainly not be able

to be made in the immediate aftermath of COVID-19. Competition law in the UK prohibits

“anti-competitive behaviour”, such as coordination of prices [5.4]. Due to the fact that Stagecoach

Oxfordshire is an entirely independent company from the Oxford Bus Company, an increase in

ticket price by one company could lead to the public largely favouring the other with the still lower

price. Therefore, an increase in fare would be something to develop slowly, or under an exemption

50
 Alexander Astor

from the Competition Act of 1998 [5.5]. There are specific allowances for individual exemptions in

circumstances where an agreement which infringes upon the regulations set out in Chapter 1 of

the Competition Act is required to improve sustainability [5.6]. Due to the obvious environmental

benefits of this project, it is likely to meet the criteria for an individual exemption, so long as the

agreement does not permanently eliminate competition.

A more thorough study was conducted in Japan on public perception of hydrogen as an energy

source, with responses taken in 2008, 2009 and 2015 [5.7]. The following results are from the 2015

survey. Although awareness had increased from 2008 to 2015, 19.8% of the 3133 respondents

indicated that they had never heard of the use of hydrogen as an energy source [5.7]. 49.4% of

respondents believed that utilising more hydrogen energy would solve global warming and air

pollution, while 40.3% answered that they did not know [5.7]. When asked about the prospect of

fuel cell buses being introduced into their communities, 58.7% of respondents answered that they

thought it would be good, and 36.6% did not feel strongly one way or the other, leaving a very

small percentage against the idea of a hydrogen-based public transport system. Furthermore, only

8.4% of respondents indicated any form of negative reaction to the idea of a gas station near their

house also selling hydrogen [5.7]. These results show that the public are largely in favour of an

increase of hydrogen usage in energy systems and, more importantly for this project, a

hydrogen-based bus network.

5.2 Legal study

Hydrogen has a very wide range of flammability, from 4.0% to 75.0% by volume in air [5.8]. For

reference, methane is flammable from 5.3% to 15.0% by volume in air, and other fuels have yet

smaller ranges [5.8]. Hydrogen also requires a very low ignition energy. At stoichiometric ratios,

only 0.02 mJ is required [5.8]. Methane requires 0.29 mJ in these conditions, and propane 0.26 mJ

[5.8]. This means that almost any spark could cause ignition, and even mobile phones are a risk.

Hydrogen is also known to cause embrittlement of metals, and therefore storage tanks must be

regularly checked. These properties are reasons that hydrogen is classified as a hazardous

substance. It is noteworthy, however, that there are properties in which hydrogen has a safety

advantage over other substances. For instance, the low density (0.09 kg m-3 [5.8]) and very high

51
 Alexander Astor

diffusivity (0.61 cm2 s-1 [5.8]) of hydrogen mean that, so long as hydrogen is kept in a

well-ventilated area, the initially high risk caused by a leak reduces very quickly compared with

other gases.

5.2.1 Hydrogen management regulations

There is little regulation that specifically focuses on hydrogen, however there are acts and

regulations under which hydrogen must be managed. The most broad of these regulations is the

Gas Act 1986 [5.9], which applies to all gases. Any party who supplies, ships or transports a gas

must have a license to do so. Under this Act, there are several industry codes and standards which

must be complied with. Under the Gas Safety (Management) Regulations 1996 [5.10], any

transporters of gas must prepare and submit a safety case to the Health and Safety Executive

(HSE). The Pipeline Safety Regulations 1996 set out requirements for the integrity of a pipeline,

from the design through to operations, maintenance and decommissioning [5.11]. Given that the

transportation of hydrogen between production, storage and refuelling locations will be a significant

component of the proposed system, it is important that these regulations are kept in mind for

system design.

The storage of hydrogen is governed by the Planning (Hazardous Substances) Regulations 2015,

and the Control of Major Accidents Hazards Regulations 2015 (COMAH). Quantities of hydrogen in

excess of 2 tonnes must comply with the Planning (Hazardous Substances) Regulations, while

quantities in excess of 5 tonnes follow the COMAH regulations [5.12].

The Planning (Hazardous Substances) Regulations 2015 state that Hazardous Substance Consent

(HSC) is required from the Hazardous Substance Authority (HSA) for planning to be approved

[5.13]. The HSA is usually the local council, who must consult with HSE and the Environment

Agency (EA) before granting HSC. A key element of a system adhering to these regulations is

ensuring that the location of the planned site is an appropriate distance from other local

establishments, so as to minimise risk in the surrounding area.

COMAH Regulations are divided into regulations for a lower tier (5 - 50 tonnes of hydrogen) and a

top tier (in excess of 50 tonnes of hydrogen) [5.12]. All operators of establishments subject to

52
 Alexander Astor

COMAH regulations have certain duties, the main aim of which is taking all necessary measures to

prevent major accidents and their possible consequences to people and the environment [5.14].

Proper notification of site details must be provided to HSE and the EA [5.14].

Lower tier operators must prepare a Major Accident Prevention Policy (MAPP) [5.14]. This should

include a summary of the safety management system that will be put into place in order to enact

this policy [5.14]. The MAPP must also contain information regarding personnel, identification and

evaluation of major hazards, operational control, management of change and planning for

emergencies [5.14].

Top tier operators must prepare a full safety report, of which a MAPP is a component [5.14]. The

safety report must also include full site details, including a description of operating methods and all

dangerous substances involved [5.14]. It must also include relevant information about potential

hazards; this includes information on possible major accidents, potential causes (both internal and

external), and plans to mitigate and reduce consequences of these accidents [5.14]. Information on

nearby establishments which may increase or be subjected to risk must also be included, as well

as information about the emergency plan for the site which will be used by the local authority to

prepare an external emergency plan [5.14]. In addition to preparing this plan for the safety report,

the plan must be tested [5.14]. Some information must also be provided to the public, through the

HSE website [5.14].

Any facility or system containing hydrogen must also meet the standards of the Dangerous

Substances and Explosive Atmosphere Regulations 2002 (DSEAR) [5.15]. These regulations apply

to explosive substances in the workplace, including those which are not intended to be used as

such, which is the case for hydrogen in the system. These regulations implement the ATEX 137

Directive [5.16], which requires that the employer take certain risk-preventative measures. This

should include identification of risks, classification of areas with potentially explosive atmospheres,

and appropriate equipment allocation to prevent possible causes of ignition in these areas [5.16].

From all these regulations, it is clear that the amount of hydrogen stored at a given time should be

minimised, and stored in a well-ventilated area, in order to minimise risk.

53
 Alexander Astor

5.2.2 Hydrogen fleet regulations

For a bus to operate on public roads in the UK, they must follow the Road Vehicle (Construction

and Use) Regulations 1986 (RVR) [5.17]. Regulation 94(2) would seem to prevent the use of

hydrogen vehicles, indicating that no gas may be used in a gas supply system for a propelled

vehicle other than liquefied petroleum gas [5.17]. However, orders such as the Road Vehicles

(Authorisation of Special Types) (General) Order 2003 [5.18] provide specific constraints under

which vehicles not within the specification of the RVR may operate. Furthermore, Section 44 of the

Road Traffic Act 1988 [5.19] allows Special Vehicle Orders to be granted by the Vehicle

Certification Authority. In the context of hydrogen buses, this means that a fleet would be a trial,

test, or demonstration, however this would not inhibit their ability to operate. Examples of this can

be seen in Birmingham [2.31] and Aberdeen [5.20], where hydrogen bus fleets have been

operating.

5.2.3 Infrastructure and planning regulations

The proposed system for this project includes green energy generation, independent hydrogen

production, and hydrogen storage. There are therefore a number of sites which will require

extensive planning. For the required infrastructure and general site planning, smaller projects must

follow legislation in the Town and Country Planning Act 1990 [5.21], and larger projects must follow

the Planning Act 2008 [5.22].

The Town and Country Planning Act 1990 requires a two-part plan; a structure plan drawn up by

the county council, stating policies and general proposals, taking into account relevant national and

regional policies, and a local plan put together by the district council, relating the general structure

plan to the specific proposal and location, relating all planning issues to the public [5.23]. Hydrogen

pipelines are regulated by this Act.

The Planning Act 2008 (amended by the Localism Act 2011 [5.24]) specifically relates to

“Nationally Significant Developments” [5.25]. This project comes under the label of a Nationally

Significant Infrastructure Project (NSIP), as any projects related to energy and transport do. Under

the Act, a Development Consent Order (DCO) can be issued, which removes the need for several

54
 Alexander Astor

other consents, including planning permission. The process takes 12 - 15 months from the time an

application is formally submitted, alongside the appropriate pre-application consultations [5.26].

This will apply to planning for the energy source for hydrogen generation for this project, as well as

for the generation and storage facilities.

55
 Alexander Astor

6 Hydrogen production
In this section, the details of the hydrogen production components will be discussed. This will

include a summary of the decisions made, and an in-depth explanation of how the components

work.

6.1 Energy production

6.1.1 Selecting the energy source

In order to meet the system objectives of being local as well as net zero, the green energy

generation method must be located within Oxfordshire. Due to the cost and location requirements

of a hydro-electric solution, this was quickly ruled out, and the final decision came down to wind

against solar. The main criteria for this decision included factors such as running cost, reliability,

lifetime, and environmental factors.

Photovoltaic solar energy was decided upon for the energy generation component of this system

for the reasons outlined below.

Generally, wind energy does have some advantages over solar; it currently has a lower carbon

footprint [6.1], it can be produced day and night, and more consistently across the year (although

there is a trend to decrease in summer, the curve is much less harsh than that of solar). However,

there are many factors in which solar proves better.

At a large scale, solar panels require less land than wind turbines. However, of the land needed for

wind turbines, only a very small proportion is used due to the narrow nature of the tower. This

means that although the wind farm will require a larger area to be spread across, most of that area

can still be used for agriculture. The main problem with the large area required for the wind farm

comes due to the impact on the landscape. Wind farms are notorious eye sores. Taking the 500

kW EWT DW52/54 as an example, the tower stands up to 75 m from the ground, with the rotor

radial length adding a further 27 m [6.2]. Larger turbines, likely to be used in projects of this scale,

stand much taller. This makes the turbines visible from substantial distances, whereas solar panels

on a ground mount tend to have a less offensive impact on the landscape. In a place like

56
 Alexander Astor

Oxfordshire, this is likely to have an impact on obtaining planning permission, as locals are likely to

oppose this [6.3].

Solar panels require much lower operational expenses than wind. The predicted prices for 2025

according to a government report on the cost of energy generation [6.4] given in pounds per

megawatt per annum (£/MW/year) show how stark the difference is. The cost of operations and

maintenance (O&M) was given as £6,700/MW/year for solar, and £23,500/MW/year for wind [6.4].

Furthermore, the already low fixed O&M cost for solar is predicted to drop fairly steadily to

£5,700/MW/year by 2040, whereas the cost of wind isn’t expected to decrease. In addition to this,

the variable maintenance for solar is zero, whilst wind is expected to cost £6/MWh [6.4]. On top of

this, a solar farm has a higher expected operational lifetime than that of wind, solar being 35 years,

and wind being 25 years. Due to all these considerations, the levelized cost of electricity, or LCOE,

is much lower for solar.

However, the main problem with a wind farm

in Oxfordshire is the reliability. The wind in

Oxford can be described as “light and

sporadic” [6.3], and as shown in Fig. 5.1, this

is true of both the direction and the speed.

Note that the time scale is length of segment

as opposed to area. The annual average

wind speed in Oxford is 16.9 kmph [6.6],

translating to just under 5 m s-1. According to

the Renewables First website, it is almost

certainly not economically viable to operate a

wind farm in an area with an average wind

speed below 5 m s-1 [6.7]. Given that Oxfordshire does not meet the criteria even without taking

into account the variation in the wind direction, which also decreases efficiency, it is clear that wind

will not be an option for this project.

57
 Alexander Astor

6.1.2 Solar Energy

Once solar energy had been chosen as the power source for the proposed system, an accurate

model of the hourly energy output for the year was needed.

How a solar panel works

The following is a brief summary of how solar panels work. A solar panel is made up of a series of

photovoltaic cells. The photovoltaic cells themselves, as shown in Fig. 6.2, are made up of a

negative electrode, an anti-reflection coating, a p-n semiconductor junction, and a positive

electrode. The cell produces a current due to the photoelectric effect, by which the atoms of the

semiconductor absorb photons of light and emit electrons. With the negative and positive

electrodes on the front and back of the junction, the electrons can travel through the circuit formed

and thereby create a current. The output current of the cell is direct current. Many modules join up

58
 Alexander Astor

along a DC string to form arrays, before the current enters an inverter, which converts the direct

current to an alternating current.

Modelling the solar farm

For this project, the solar panels used in the modelling were SHARP’s NU_AC 300b, which is a

300 W panel, with 18.3% efficiency, and an area of 1.6368 m2 [6.9]. The performance ratio, which

accounts for losses in the system, including inverter losses, temperature losses, cable losses,

shading losses, and others, was taken as 0.7, being in the middle of the standard range (0.5 - 0.9)

[6.10]. The irradiance data used was from 2019, and provided an hourly average irradiance for

Oxford [6.11]. The daily irradiance across the year is plotted in Fig. 6.3.

The graphs plotted in Fig. 6.4 show the solar irradiation for June 21st and December 22nd 2019,

the longest and shortest days of that year respectively. These plots show how dramatically solar

irradiance, and therefore the power output of a solar farm, varies throughout the day, as well as

showing the stark contrast between irradiance at the height of summer and in deepest winter.

Given these seasonal, daily and hourly differences in power output, it was clear that the system

would require smart control in order to optimise the use of the locally generated power for

hydrogen production. The details of this smart system, including the optimisation process by which

it was determined, will be presented in Section 6.3.

59
 Alexander Astor

Estimating cost of installation

The cost of installation of a solar farm is made up of a variety of factors. The solar panels

themselves only take up a small percentage of the price. The rest is made up of various associated

costs including construction, land, planning, grid integration, technical analyses, as well as financial

and legal services. There is however an economy of scale when considering a solar project of a

larger size, and factors such as planning and financial/legal services are likely to have a less

significant increase than the cost of panels, which would increase approximately linearly with the

size of the project.

For a large-scale solar project, the government estimates that by 2025, the pre-construction costs

will be £50/kW, and the construction costs will be £400/kW, meaning £450,000/MW for installation

[6.4]. This is a result of a significant decrease in cost in recent years, with the cost decreasing

faster than predicted [6.12].

60
 Lingxi Tang

6.2 Hydrogen Production Components

Now that the source of power is finalised, Section 6.2 aims to examine the electrolyser and battery

to be used in the hydrogen production system. In this subsection, different electrolyser and battery

technologies will be explored before justifying the specific choices and parameters for these

components.

6.2.1 Electrolyser Type

In order to determine a suitable electrolyser model to produce green hydrogen, comparisons will be

made between the two most commercially viable electrolyser types: alkaline and PEM, as

mentioned in Section 2.1.3.

< Table 6.1: Technical Comparison between Alkaline and PEM Electrolysers [2.13, 6.13, 6.14] >

There are many literature pieces which compare the general technical specifications and costs of

alkaline and PEM electrolysers [2.13, 6.13, 6.14, 6.15, 6.16]. The relevant specifications are

compiled in Table 6.1 above. For values with different ranges in different sources, the smallest

number range is recorded in Table 6.1 to ensure that its values are included in each literature

source’s range. For example, for start-up time, alkaline electrolysers are estimated to take 20 - 60+

min [6.13], >30 min [2.13], or <60 min [6.14] depending on the literature source, thus 30 - 60 min is

the range which is included by all these sources, and will be recorded in the table above.

For lifetime and capital cost, there is no common range between the values provided in each

source. Depending on the literature source, alkaline electrolysers have a lifetime of 20 - 30 years

[2.13, 6.13], 10 years [6.16] or 60,000 - 90,000 hours [6.14] and PEM electrolysers last for 10 - 20

years [2.13], 10 - 30 years [6.13], 3 - 4 years [6.16], or 20,000 - 60,000 hours [6.14]. Despite the

widely varying ranges, it is generally agreed that alkaline electrolysers have longer operating

lifetimes than their PEM counterparts, at about 1.5 - 2 times the duration. For capital costs, alkaline

61
 Lingxi Tang

electrolysers cost £870/kW - £1044/kW [6.14] or £547 - £792 (units are not specified) [6.13] while

PEM electrolysers cost £1618/kW - £2018/kW [6.14] or £864 - £1397 [6.13]. Once again, while the

sources do not agree with the absolute values of cost, they both substantiate the fact that PEM

electrolysers cost up to 2 times the price of alkaline electrolysers.

While PEM electrolysers currently still lag behind alkaline in terms of lifetime and costs, they are

catching up quickly in both areas. According to the 2017 research paper by Schmidt [6.15], the

median PEM electrolyser will have a lifetime of 41,000 - 60,000 hours by 2020, with alkaline

electrolysers improving minimally. By 2030, PEM electrolysers will have similar lifetimes to their

alkaline counterparts.

< Fig. 6.5: Predicted Average system price and Average Order Size of ITM Power PEM
Electrolysers from 2021 to 2029 [6.15] >

In terms of pricing, ITM Power, a British manufacturer of PEM electrolysers is well on track to

producing electrolysers at a price comparable to alkaline electrolysers. This year, ITM Power

predicts to produce PEM electrolysers at just €884/kW or £769/kW . This is expected to fall over

the next decade, as the average order size of electrolysers increases from under 5 MW to nearly

45 MW by 2029.

With similar energy consumption per amount of hydrogen produced, PEM electrolysers are

superior to alkaline electrolysers in terms of load flexibility: it can start operation quickly, handle low

power production and respond to power changes quickly. These features are especially useful for

62
 Lingxi Tang

electrolysers powered by solar energy which can change quickly throughout the day and is

intermittent. Not only are PEM electrolysers technically superior, it is also fortunate that their costs

have recently been driven down to become competitive with alkaline electrolysers. This makes

PEM electrolysers the clear choice for green hydrogen production at Oxfordshire.

6.2.2 Electrolyser Model

Now that PEM electrolysers have been selected to be the desired electrolyser type, the specific

choice of electrolyser supplier and model will now be decided.

While there are many companies globally which manufacture PEM electrolysers, such as NEL

Hydrogen, Siemens and Hydrogenics, only ITM Power has provided public data on their

electrolyser capital costs (from Fig. 6.5 above), which will be essential for the financial analysis

later. It is also reasonable to assume that ITM Power’s prices are competitive, given that their

prices are even cheaper than price estimates for alkaline electrolysers mentioned in Section 6.2.1.

Additionally, with an upcoming electrolyser factory based in the UK at Bessemer Park, Sheffield

[6.15], ITM Power will be able to supply PEM electrolysers to Oxford with minimal transportation

costs and international trade taxation. As such, electrolysers from ITM Power will be used to

produce fuel to power the Green Hydrogen Mobility Project.

Specifically, ITM Power’s upcoming 5MW stack module will be used. These are currently being

developed for the Gigastack project (large scale Humber-based hydrogen generation project using

PEM electrolyser powered by wind energy) and are expected to be ready by mid-2021 [6.17]. Each

5 MW stack module can produce up to 2,100 kg of hydrogen per day [6.15] which is equivalent to a

production rate of 0.00486 kg of hydrogen produced per second per MW of electricity. ITM Power

has also advertised their electrolysers to require water of “drinking quality”, thus there will be no

additional cost of water purification. It will be assumed that the weight of water consumed is 9

times the weight of hydrogen produced [6.18]. To estimate the cost of the electrolyser, the

predicted system price in 2025, which is ~£432/kW, translates to £2,160,000 for each 5MW stack.

The electrolyser’s dynamic power range will be set at 10 - 100% of maximum power, based on a

conservative estimate of PEM electrolyser’s minimum power load (Table 5.1).

63
 Lingxi Tang

< Fig. 6.6: Evolution of ITM Power Electrolyser Stack Design [6.15] >

6.2.3 Battery Technology

Prevalence and cost of Li-ion

Accounting for nearly 90% of global large-scale battery storage additions in 2017 [6.19] and 95% of

deployed grid-scale battery systems in the US [6.20], lithium-ion (Li-ion) batteries are the dominant

technology for grid-scale energy storage. This market dominance is driven by the declining costs of

Li-ion technology which is, in turn, fuelled by the growing demand and production of Li-ion batteries

for electric vehicles [6.19] and consumer electronics [6.22].

The versatility of Li-ion batteries ensure that the cost reductions are also felt in the energy storage

market, where Li-ion technology offers the cheapest solutions for a wide range of grid-scale energy

storage solutions compared to alternatives such as vanadium-based flow batteries and zinc-based

flow batteries. Given Li-ion batteries’ low cost and applicability as energy storage for solar energy,

it is the ideal technology for the smart electrolyser-battery system.

64
 Lingxi Tang

< Fig. 6.7: Share of difference battery technologies in annual battery storage capacity additions
globally [6.19] >

< Fig. 6.8: Comparison of levelised cost of storage of different battery technologies for 3 different
applications (USD/MWh) [6.19] >

Battery Parameters

Batteries have two main design parameters: Power capacity and Energy Capacity. These

parameters will differ based on application. Power capacity refers to the maximum rate of electricity

charge or discharge of the battery and is usually expressed in kilowatts (kW) or megawatts (MW).

Energy capacity refers to the maximum amount of energy which can be stored in a battery and is

usually expressed in kilowatt-hours (kWh) or megawatt-hours (MWh).

According to the Innovation Landscape Brief on Utility Scale Batteries by IRENA, “storage systems

paired with large PV systems'' with 20 MW power capacity and 80 MWh energy capacity cost

£77.76 - £100.80 per MWh of storage [6.19]. Thus, for the purposes of this project, battery’s

power-energy capacity ratio will be fixed at 20:80 = 1:4 and battery costs will be the average of the

cost range at £89.28/MWh. Thus, the sole design parameter for the battery will be its energy

65
 Lingxi Tang

capacity in MWh. From here on, battery energy capacity will be referred to simply as battery

capacity.

6.3 Smart Electrolyser-Battery System

6.3.1 System Operations

In order to produce green hydrogen efficiently using locally generated solar power, the smart

electrolyser-battery system is proposed. The system comprises three main components, which

need to be designed: solar farm, electrolyser and battery. The electrolyser will run at its maximum

power during its operating hours since varying the electrolyser’s operating power will reduce its

lifetime. The electrolyser’s source of power will vary between combinations of solar power, battery

and the grid based on three indicators: operating hours, solar power available and battery state of

charge (SOC).

Operating hours will be fixed, for the purpose of this report, at 0700 - 1700 from 1 January to 15

April and 1 September to 31 December, and 0500 - 1900 from 16 April to 31 August; all times are

in GMT. These times are chosen for convenient operation and to coincide with daylight hours as

much as possible. They can be further tuned to changing needs in the future. The operation

flowchart of the smart system can be seen below in Fig. 6.9. The smart system monitors the three

indicators mentioned earlier in real-time using on-site sensors. which will be connected to each

other via a cloud computing service such as Amazon Web Services. The combined indicators then

determines one of the output decisions (shown in orange) based on the decision flowchart, shown

in Fig. 6.9. Ideally, timestamps are as short as possible so decisions can be made based on the

most up-to-date information.

6.3.2 System Parameter Design

In order to simulate hydrogen production from the smart electrolyser-battery system, a Python

simulation model was created to calculate annual OPEX and CAPEX based on prices shown in

Table 6.2 below. The pseudocode for the model is seen in Fig. 6.10. Note that solar power sale

revenue is negative as it reduces OPEX.

66
 Lingxi Tang

< Fig. 6.9: Operation Flowchart of Smart Electrolyser-Battery System >

< Table 6.2: Variable Unit Cost and Annual Fixed Cost of components of smart electrolyser-battery
system >

Given the operating hours and a minimum annual hydrogen demand of 2,933,738 kg (see Section

4.3), a 42 MW electrolyser (comprising eight 5 MW stacks and one 2 MW stack) will be used to

produce ~ 3,093,643 kg of hydrogen annually. This is enough to cover both the minimum and

emergency demand, which totals to 3,043,738 kg. There are two system component parameters

67
 Lingxi Tang

which are to be designed: solar farm size and battery energy capacity. The specific combination of

parameters which gives the lowest expenditures will be chosen for the system.

Initialise
Input solar irradiation data
Declare solar farm size, electrolyser power and battery energy capacity
Declare operating hours
Initialise zero arrays for hydrogen produced, battery SOC and solar power exported to grid
Main
for each timestamp
if timestamp is within operating hours
hydrogen produced based on max power
if solar power exceeds electrolyser power
battery charges at power capacity or excess solar power, whichever is smaller
if battery charges at power capacity
solar power exported to grid = solar power - battery power - electrolyser power
battery SOC increases based on power capacity, capped at energy capacity
if battery was already full in previous time stamp
solar power exported to grid = solar power - electrolyser power
battery SOC remains unchanged
else
additional power required = electrolyser power - solar power
if additional power required > battery power capacity
if battery has sufficient charge for the timestamp duration
battery SOC decreases based on discharge at power capacity
grid electricity demand = additional power required - battery power capacity
else
battery SOC is set at SOC of previous timestamp
elseif additional power required < battery power capacity
if battery has sufficient charge for the timestamp duration
battery SOC decreases based on discharge at power required
else
battery SOC is set at SOC of previous timestamp
if timestamp is outside operating hours
no hydrogen produced
if battery was already full in previous time stamp
excess solar power exported to grid = solar power
battery SOC remains unchanged
else
if solar power > battery power capacity
excess solar power exported to grid = solar power - battery power capacity
battery SOC increases based on charging at power capacity, capped at energy capacity
else
battery SOC increases based on charging at solar power, capped at energy capacity
Results
Calculate total hydrogen produced by summing hydrogen produced at each timestamp
Calculate OPEX and CAPEX
End
< Fig. 6.10: Pseudocode for Smart Electrolyser-Battery System simulation model >

To determine the desired parameters, the OPEX and CAPEX were recorded for every combination

of system component parameters where solar farm size is from 0 to 50 MW, in intervals of 1 MW

and battery capacity is from 0 to 49 MWh, in intervals of 1 MWh. Solar farm size was first tested up

to 50 MW as the upcoming South Oxfordshire Solar Farm is taken to be a feasible standard size of

solar farm in Oxfordshire. From the results of this first experiment, a 42 MW electrolyser powered

68
 Lingxi Tang

by a 50 MW solar farm and 29 MWh battery would produce the lowest OPEX of £18,499,549.

However, this is financially unattractive as this is much more expensive than the current annual

diesel spending of £12,281,802 (this value will be explained in Section 8). Thus, in order to achieve

lower OPEX, more solar farms of sizes up to 50 MW can be built to power affordable green gas

production. As such, solar farm sizes of up to 100 MW and up to 150 MW, with battery sizes up to

30 MWh, were trialled to finally obtain an annual OPEX of £9,402,416. The change of OPEX in

response to solar farm capacity and battery capacity can be seen in Fig. 6.11. It can be seen that

annual OPEX decreases sharply with increasing solar farm capacity, with minimal impact from

changing battery capacity. To see the impact from changing battery capacity, Fig. 6.12 plots OPEX

against battery capacity for solar farms of three sizes, where it can be seen that OPEX barely

decreases beyond ~20 MWh for a 50 MW solar farm and is minimised at 8 MWh and 13 MWh for

100 MW and 150 MW solar farms respectively. This also explains why battery sizes were only

trialled up to 49 MWh and 29 MWh since battery sizes above these values did not lower OPEX

further.

< Fig. 6.11: Change of Annual Operating Expenditure in response to changing solar farm size and
battery capacity >

For comparison purposes, a standalone electrolyser is also trialled. From Table 6.3, it can be seen

that the savings in OPEX can break-even the additional capital investments of solar farm,

electrolyser and battery in less than 5 years, regardless of the solar farm size. This timeframe is

well below a lifetime of 35 years for the solar farm [6.4] and 10 years for the electrolyser [2.13,

6.13]. This means that the savings from smart solar powered electrolyser-battery system will

69
 Lingxi Tang

break-even its capital costs well before any major component needs replacement due to the end of

their lifetime (cost of battery replacement will be assumed to be insignificant since even the largest

29 MWh battery tested costs only about £2,589). This not only financially justifies investing in the

smart electrolyser-battery system, it also means that OPEX minimisation is equivalent to

minimising the total cost (including CAPEX and OPEX) over the system’s lifetime.

(a)

(b)

(c)
< Fig. 6.12: OPEX vs Battery capacity for (a) 50 MW solar farm (b) 100 MW solar farm (c) 150 MW
solar farm >

< Table 6.3 : Financial information and system component parameters of lowest OPEX options >

70
 Lingxi Tang

From Table 6.3, it can be generalised that there is a tradeoff between initial CAPEX and annual

OPEX. Thus, a main factor to consider when deciding the ideal system parameters would be the

capital available, including assets and loans. Assuming unlimited capital, one might invest in a

solar project large enough for the electrolyser to be completely solar-powered, minimising

spending on grid electricity within OPEX. However, it is worth noting that there are other important

decision factors such as legal considerations of land usage in Oxfordshire for solar project

development (as mentioned in Section 5) and diminishing rate of returns from additional solar

power investments (notice the trend of increasing break-even periods for larger solar farms in

Table 6.3). Given these limitations, further testing for larger solar farms will not be considered and

a 150 MW solar farm, 42 MW electrolyser and 13 MWh battery system will be analysed for the rest

of this report.

6.3.3 System Performance Results

With system component parameters specified and using the irradiation data seen in Fig. 6.3, solar

power generation and daily hydrogen produced are plotted in Fig. 6.13. It can be observed that

solar power generation is expectedly linked closely to solar irradiation (see Fig. 6.3) and daily

hydrogen production is constant at 10.290 tonnes in summer and 7.350 tonnes in winter. The

constant production values are due to fixed operating hours.

(a) (b)
< Fig. 6.13: (a) Hourly Power Generation from a 150 MW solar farm based in Oxford (b) Daily
Hydrogen Produced >

To provide a sense of energy and power balance between the system components, the status of

solar power generation, battery SOC, grid import/export and electrolyser power consumption on

two days are plotted in Fig. 6.14. The two days are the 21-June and 22-Dec, of which the solar

71
 Lingxi Tang

irradiation plots are seen in Fig. 6.4. Thus, Fig. 6.14 shows the two extremes of solar

overgeneration and undergeneration. In summer, electricity is imported from the grid and obtained

from the battery in early morning and evening to compensate for the reduced solar power.

However, much solar power is exported to the grid throughout the day. In winter, the electrolyser is

heavily reliant on the grid, with some support from solar power during the day. The battery remains

a constant state of low charge as it does not have enough energy to discharge for an hour (see

Fig. 6.9), which is the time resolution of the irradiation data. If a dataset of higher resolution was

used, the battery would be discharged till empty. The low time resolution also explains the

staircase plot rather than a smooth curve.

(a)

(b)
< Fig. 6.14: Status of solar power generation, battery SOC, solar power exports, grid power
consumption and electrolyser power consumption based on solar irradiation on (a) 21-Jun- 2019
and (b) 22-Dec-2019 >

72
 Alexander Astor

7 Hydrogen storage and distribution

The next part of the proposed system is storing the hydrogen produced by the electrolyser, and

distributing it to refuelling sites. This section will explore the design principles of the proposed

hydrogen storage and distribution solution, as well as presenting the details of these system

components.

7.1 Hydrogen storage

In the proposed system, the seasonal nature of the production of hydrogen means that for around

half the year, more hydrogen will be produced than will be used, and for the other half more

hydrogen will be required than will be produced. This means that the system requires a significant

capacity of hydrogen storage, so that the surplus from the summer months can balance out the

deficit in the winter months.

7.1.1 Evaluation of different storage methods

Hydrogen storage, as discussed in Section 2, can be divided into three main types: gaseous

hydrogen storage, liquid hydrogen storage, and hydrogen carrier storage. Fig 7.1 shows how the

density of hydrogen varies with different methods of storage.

There are two main methods of storing hydrogen as a gas: in pressurised cylinders and in

underground reservoirs, such as salt caverns or depleted oil wells. Due to the lack of salt caverns

and oil wells in Oxfordshire, this was ruled out as an option. Hydrogen has the lowest atomic

weight of any element, and therefore has very low density. The density of hydrogen gas can be

taken as 0.090 kg m-3 [2.1]. As a result of this, to store any significant amount of hydrogen in

gaseous form in a reasonable amount of space requires pressurised storage. Compressed

hydrogen gas can be stored in cylinders. Compressed gas can be stored at various pressures,

usually 180 bar or above.

Hydrogen can also be stored in liquefied form. During liquefaction, the hydrogen must be cooled to

−253°C, as the boiling point of hydrogen at atmospheric pressure is −252.8°C [2.20]. To remain at

these temperatures, the hydrogen has to be stored in cryogenically insulated tanks. The equivalent

73
 Alexander Astor

of around 30% of the energy content of the hydrogen is required to cool the hydrogen to

temperatures as low as this [7.1]. In addition, boil-off becomes a problem when storing hydrogen

as a liquid. The temperature of the stored liquid inevitably rises, although slowed by the cryogenic

insulation. This leads to some of the hydrogen becoming a gas, which raises the pressure in the

tank. Therefore, the tank needs to release this gas, to keep the pressure low. This leads to an

inevitable loss of hydrogen. The main advantage of storing hydrogen as a liquid is that it is very

compact. The density of hydrogen in liquid state can be taken as 70.8 kg m-3 [7.2]. This is about

787 times denser than hydrogen gas, reducing the required storage space by a huge amount.

Hydrogen can also be delivered in liquid tanks. This is particularly advantageous over long

distances, at which point the benefits of the much higher density outweigh the disadvantages of the

hydrogen losses and extra energy requirements for liquefaction.

Metal hydrides are a solid form of hydrogen carrier, of which magnesium and aluminium hydrides

are examples. Metal hydrides are one of the safest ways to store hydrogen, as they take the form

74
 Alexander Astor

of stable solids. However, breaking the bonds to release the hydrogen is a slow and

energy-intensive process [2.16], which can hinder a system in which hydrogen is required

consistently in large quantities.

Liquid organic hydrogen carriers (or LOHCs) are examples of the second type of hydrogen carrier.

The most viable example is dibenzyltoluene (C21H2O) [7.3], which is currently used as a heat

transfer oil. Since the carbon involved can be recycled, this would still meet the net zero condition.

LOHCs are most useful for hydrogen which will be travelling long distances, however they are not

particularly cost effective for shorter range supply, as the system would require [7.3].

The final storage option considered was ammonia. Ammonia already plays a large part in the UK,

especially in the agricultural industry, so there is broad infrastructure to connect to. In addition to

this, ammonia has a 45% higher hydrogen density than liquid hydrogen [7.1]. Ammonia can be

stored in very mild conditions as shown in Fig. 7.1. The high hydrogen density and existing

infrastructure mean that ammonia is one of the cheapest ways to store hydrogen.

7.1.2 Storage requirements

From the route modelling in Section 4, the daily requirement of hydrogen is 8,548.2 kg on a

weekday, and 6,838.6 kg on a weekend. As discussed in Section 6, the production of hydrogen is

produced at two rates depending on the time of year. From 1st September to 15th April, when there

are fewer sunlight hours, the operating hours of the electrolyser (which track solar energy

production) are shorter, and therefore less hydrogen is produced per day. During this period of the

year, the weekly production is lower than the demand. For the remainder of the year (16th April to

31st August), the operating hours are longer, and the production exceeds the demand. Fig. 7.2

shows the difference between production and demand throughout the year.

Note that the demand during the weekend is lower than the weekdays, hence the spikes. During

the over-production period, the system must be able to store the excess, so that it can balance out

the insufficiency of production for the rest of the year.

75
 Alexander Astor

Fig. 7.3 shows the total mass of hydrogen stored on each day of the year. As large quantities of

hydrogen (up to 425,116 kg at the end of the over-production phase) will have to be stored to last

through the lower production phase of the year, a much more compact form of storage than

compressed hydrogen gas is needed. From previous evaluations, it was clear that the simplest,

and most compact, alternative method of storage is in the form of ammonia. During the

over-production phase, the excess hydrogen each day will be converted to ammonia in a synthesis

reactor. During the period of the year for which production is lacking, the required amount of

hydrogen will be provided by cracking ammonia on site.

The depots themselves will require the hydrogen in gaseous form, ready for refuelling of the

FCEVs. This means that the main method of distribution will be in gaseous form, which requires

that there be at least a day’s supply of hydrogen stored as compressed gas. In order to account for

possible complications, the hydrogen demand for two weekdays will be stored as compressed gas,

ready for distribution. This amounts to just under 17,100 kg of hydrogen, which is indicated by the

red line on Fig 7.3. This will be stored in pressurised cylinders, in an open, well-ventilated space.

76
 Alexander Astor

As discussed in Section 5, special measures must be taken for the storage of a hazardous gas

such as hydrogen in quantities of more than 5 tonnes [5.12]. Planning for an event in which the

hydrogen supply is halted, the system will contain a minimum of two weeks’ demand of hydrogen

to allow for time to fix a potential problem in the supply chain. This means that on top of the 17,100

kg as compressed gas, a further 95,380 kg of hydrogen will be stored as ammonia, so that at the

lowest point in the year, there is enough hydrogen in storage to last for two weeks of normal

operation. For the cracking of 95,380 kg hydrogen from ammonia in under 14 days, the required

cracking capacity is 6,800 kg of hydrogen per day.

During the over-production period, approximately 15,680 kg of excess hydrogen will be produced

each week. This is the amount that will be converted to ammonia weekly, which will be done at an

external ammonia synthesis plant. For this purpose, the project could connect to the plant at Ince,

which can produce just under 1,000 tonnes of ammonia per day [7.4], or the plant at Billingham,

which can synthesise up to 1,500 tonnes of ammonia per day [7.4]. Another option would be to

connect to a number of separate synthesis plants which operate more locally, in order to reduce

the distance that the hydrogen would have to travel.

77
 Alexander Astor

During the under-production phase, up to 1,400 kg hydrogen will be required in excess of what is

being produced by the electrolyser per day. This will be provided by cracking the required amount

of ammonia from the long-term storage. As the ammonia will need to be cracked to fit the daily

demand, it will be necessary to crack on site, rather than to link to an external ammonia cracker.

The capital cost of an ammonia cracking facility can be taken as $405,000 (£291,600) per tonne of

hydrogen production capacity per day [7.5]. As mentioned earlier, the cracking facility will need to

be able to crack up to 6,800 kg of hydrogen per day. This would cost approximately £1,955,340 to

set up. The energy required to crack ammonia can be taken as 96.6 kJ mol-1 of hydrogen [2.28],

which equates to 13.42 kWh kg-1 of hydrogen. This gives the yearly required running cost as

£293,519 [7.6], assuming normal running.

As discussed in Section 6, the system produces a yearly excess of 152,413.6 kg of hydrogen. Fig

7.3 shows the excess as the difference between the total hydrogen in storage at the end of the

year and total hydrogen in storage at the beginning of the year. This could be sold as ammonia or

hydrogen to boost profits from the system. The market price for hydrogen is currently £10 - £15 per

kg [7.7], meaning that the yearly excess could boost profits by upwards of £1.5 million.

78
 Antoine Thibaut

7.2 Hydrogen distribution

Now that short term and long term storage options have been explored, a hydrogen distribution

system will be designed. A general delivery strategy from the hydrogen production plant to the

refuelling stations will first be explored. Secondly, the optimal location for the hydrogen production

plant will be considered. Finally, a detailed tube trailer and truck solution will be discussed.

Hydrogen will be distributed in gaseous form at 180 bar in cylinders [7.8]. These cylinders are

carried in tube trailers [7.8] which are transported by trucks. Fig. 7.4 below illustrates the strategy

proposed for distributing hydrogen from the hydrogen production plant to the hydrogen fuelling

stations and long term ammonia storage facilities. The dotted rectangle represents the hydrogen

production plant perimeter consisting of an electrolyser, an ammonia cracking unit, a short term

hydrogen storage facility and compressors.

< Fig. 7.4: Distribution schematic adapted from [7.9, 7.10] and Fig. 2.6 >

As described in Section 7.1.2, a maximum of around 8,550 kg of hydrogen will be required by all

Oxford bus fleets in a single day. The tube trailers will have a capacity of 280 kg of hydrogen each

[7.8]. This results in around 32 tube trailer deliveries per day to fuel all 399 buses in Oxford. To

reduce costs, 16 tube trailers can be purchased and used in two batches per day. As buses

operate during the day [3.18], refuelling will take place during the evening or early morning before

the routes begin.

79
 Antoine Thibaut

While the hydrogen buses are operating passenger routes during the day, the tube trailers and

trucks will be used for distributing hydrogen from the production plant to the long term storage

facilities. As mentioned in Section 6.3.3, a maximum of 10,290 kg of hydrogen will be produced by

the electrolyser in a single day, which represents 37 deliveries, or 5 trailers in addition to the daily

fleet needs. It will then be possible to supply an average of 5 trailers per day from the hydrogen

production plant to long term storage facilities.

7.2.1 Oxford depots and general delivery strategy

There are 4 major depots owned by the Oxford Bus Company (OBC), Thames Travel and

Stagecoach in Oxford [1.40]. These depots are spread around Cowley, Didcot and Witney [1.40].

Diesel fuel is currently stored at the depot locations [3.26]. This allows buses to be refuelled at

specified locations, minimising the refuelling time.

To minimise disruptions to the refuelling schedules of the bus routes, these depots can be used as

refuelling stations for the hydrogen bus fleet as well. Seacourt Park and Ride is also planning a new

bus terminal [7.11]. It can be suggested to locate a refuelling station at Seacourt Park and Ride.

A delivery strategy from the hydrogen production plant to the depots will now be discussed.

Throughout the day, the tube trailers will visit the hydrogen production plant to be filled to maximum

capacity. In the afternoon, the tube trailers will move to the depots. In the late evening, after the

first batch of busses are refuelled, the empty trailers will come back to the hydrogen production

plant to be refilled from short term storage. When full, the tube trailers will move to deliver the

second batch to the depots. In the early morning, before the bus routes begin operating, the

remaining buses will be refuelled at the depots. In the late morning, the empty trailers will move

back to the hydrogen production plant and the cycle repeats. This cycle is illustrated in Fig. 7.5

below.

80
 Antoine Thibaut

< Fig 7.5: General delivery strategy >

7.2.2 Hydrogen production plant location

As shown in Fig. 7.4, the hydrogen production plant will consist of both the electrolyser and short

term storage facility. For determining a best possible location, the following factors will be

discussed: distance to depots, distance to residential areas and available land area for the

construction of the electrolyser and short term storage facility.

Each tube trailer will return to the hydrogen production plant typically twice a day. Using trailer

deliveries twice a day is also a strategy proposed by the Oxford Bus Company [3.26]. It is therefore

important to minimise the distances between the hydrogen production plant and the depots hosting

the fuelling stations. A model was created to optimise the hydrogen production plant location,

minimising the total distance between the plant and all depots. Coordinates were taken from [1.40].

As a first approximation, the model assumed straight line distances between the different refuelling

locations.

The results are shown in Fig. 7.6 below. The optimised hydrogen plant location coordinates are

(51.7270,-1.2325). To visualise how this solution could be implemented into Oxford’s road

structure, roads were traced between the optimised hydrogen production plant location and the

refuelling stations. This is shown in Fig. 7.7 below.

81
 Antoine Thibaut

As shown in Fig. 7.6, the optimised location is situated closely to two main bus depots for

Stagecoach and the Oxford Bus Company. The distances between the hydrogen production plant

and the depots are outlined in Appendix B.

The optimised location can be further analysed with respect to the necessary conditions required

for storing hydrogen and installing an electrolyser. Fig. 7.8 below shows a map of the proposed

area, with some key features annotated.

< Fig. 7.8: Enlarged view of the hydrogen production plant location taken from [7.12, 7.13] >

82
 Antoine Thibaut

As shown in Fig. 7.8, the optimised location is situated close to school grounds, a residential area

and a nature reserve. As discussed in Section 5.2.1, safety distances must be respected when

considering storage locations for hazardous gases. To keep some distance from these sensitive

areas, it is proposed to build the hydrogen production plant at the location marked in blue. This

location is also situated near Redbridge Park and Ride, far from residence and near main roads to

facilitate truck access.

7.2.3 Transporting hydrogen to fuelling stations

The amount of buses which can be stored inside each depot can be analysed to indicate the daily

refuelling needs of each fuelling station. The depot’s sizes will be used as an indicator of their relative

importance. Using satellite images from Google Maps [1.40], the capacity of each Oxford depot was

estimated. This was done by superimposing rectangles representing bus dimensions onto the

satellite images. The estimated size of the proposed Seacourt Park and Ride expansion terminal

was estimated by using rendered images from [7.14]. Considering that the refuelling needs are

proportional to the size of each depot, it can be estimated how many tube trailer deliveries would be

required daily to each depot. One can then suggest how many tube trailers are delivered to each

individual depot during the first and second batches of the day. This is illustrated below in Table 7.1.

< Table 7.1: Estimating the number of trailers required for each depot >

83
 Antoine Thibaut

The amount of trucks required for delivering these tube trailers will now be estimated. As shown in

Appendix B, the tube trailers mentioned in Table 7.1 will have to travel a combined distance of

around 844 km per day. This is not suitable for a single truck in an urban road network. It can be

suggested to have one truck used for each major station (Cowley OBC, Cowley Stagecoach and

Didcot Thames Travel) and one additional truck for the smaller stations (Witney Stagecoach and

Seacourt Park and Ride). This represents 4 trucks. According to the trailer delivery strategy, taking

the example of Cowley OBC, the truck would bring 4 empty trailers back to the production plant in

the afternoon, and then make 5 trips to the depot in the evening. This represents around 9 trips

during a typical truck driver’s shift. Depending on the location of long term storage, an additional

truck may also be needed for daytime deliveries of excess hydrogen production. 5 trucks will

therefore be planned for the distribution strategy.

One can also calculate how much fuel the truck fleet would consume per year. Appendix B applies

a similar model to the one used in Section 4 for determining how much diesel these trucks would

consume per year, or hydrogen if fuel cell trucks are used. In summary, around 32,000 gallons of

diesel will be required per year for a diesel truck fleet. Around 26,800 kg of hydrogen will be

required per year for a hydrogen truck fleet. Hydrogen trucks such as the Hyundai Xcient are

anticipated in the future [7.15]. These trucks will also run on hydrogen compressed to 350 bar

[7.15]. This implies that the same compressors and storage infrastructure used for buses can be

used for the trucks.

High capacity hydrogen storage is a major stepping stone for hydrogen fleets in the future.

Companies such as Everfuel develop high capacity hydrogen distribution solutions [3.17]. By

transporting larger quantities of hydrogen in one single delivery, this solution would require less

than 32 deliveries to fuelling stations per day, and therefore reduce the number of trucks needed to

operate the network.

84
 Lingxi Tang

8 Financial analysis
The aim of this section is to provide an overview of all relevant spending and costs of the Green

Hydrogen Public Mobility Project. A financial comparison will also be made between embarking on

this net zero transition and Business-As-Usual (BAU) - a scenario where the Oxford public

transport network continues to be powered by diesel.

8.1 Transition to net zero

Given the Oxford City Council’s goal of having all public buses to be zero emissions by 2035, the

transition period for this project will be from 2025 to 2035. The delay in starting time will be

reserved for planning, approval and other administrative matters. During this transition period, two

major changes will occur. Firstly, the transition of fuel usage from diesel to locally produced green

hydrogen and secondly, the transition of bus fleet technology from being powered by diesel

engines to hydrogen fuel cells. The first transition will involve investing in a solar farm, smart

electrolyser-battery system, ammonia cracker and gradually transferring expenditure on diesel to

expenditure on hydrogen production, specifically water, electricity and solar farm maintenance, and

hydrogen storage. The second transition will involve the gradual replacement of Oxford’s bus fleet

of 399 buses, by selling 40 diesel buses and purchasing 40 Wrightbus hydrogen buses every year

for 9 years, before replacing the final 39 diesel buses in the 10th year.

8.2 Expenditure Outline

In this subsection, the annual expenditures of both BAU and transitioning to the Green Hydrogen

Public Mobility Project will be examined in detail. The OPEX and CAPEX of each component and

how they contribute to cash flow will now be explained. Generally, the CAPEX of each component

is reflected by annual depreciation using the Diminishing Value Method [3.49], which is shown in

equation 3.5. For BAU, the relevant annual expenditure is assumed to be constant and includes

purchase of diesel, fleet renewal costs and vehicle maintenance. Diesel expenditure is based on

an annual demand of 13.38 million litres, as discussed in Section 4 4, and a price of £0.9177/litre

[4.1], which includes fuel duty. Annual fleet renewal expenditure is the expenditure on new buses

to keep the fleet operationally healthy. It is assumed to be constant and is based on the amount

spent by OBC (£6.9 million [1.41]) and Stagecoach (£5 million [1.42]) in 2019, adding up to a total

85
 Lingxi Tang

of £11.9 million. Maintenance cost for diesel vehicles is derived from an annual mileage of

33,542,939 km and £0.10/km (from Section 3.4.2). Putting everything together, the breakdown of

BAU’s annual expenditure is seen below in Table 8.1.

< Table 8.1: Expenditure related to Business-As-Usual >

For the net zero system, the CAPEX of each component under Hydrogen Production (see Table

8.2) and the solar farm’s annual OPEX is based on a 150 MW solar farm, a 42 MW electrolyser

and a 13 MWh battery and derived from the values in Table 5.2. The electrolyser’s annual OPEX is

output from the Python simulation model seen in Fig. 5.11. The lifetime of the electrolyser in years

is calculated based on the fixed operating hours. Based on a predicted average operating lifetime

of 50,500 hours [6.14] and an annual operation time of 4,205 hours, the electrolyser is expected to

last for 12 years. The expenditures for hydrogen storage were discussed in Section 7.1.2. The total

CAPEX of a fleet of 399 hydrogen buses has been discussed in Section 3.5.4 and the

maintenance cost of hydrogen buses was discussed in Section 3.4.2. The revenue from the sale of

the diesel bus fleet is obtained by multiplying the value of (diesel or hybrid) buses in Table 3.8 and

the number of (diesel or hybrid) buses in Table 3.5. Hydrogen bus fleet renewal expenditure is

estimated using the cost of purchasing 40 new hydrogen buses and subtracting it with the

depreciated value of 40 hydrogen buses after 12 years of service. Using the Diminishing Value

method again, with an initial value of £304,500 (see Table 3.2), a hydrogen bus depreciates to a

value of ~£34,152 after 12 full years of service. Thus, annual hydrogen fleet renewal will cost

£10,813,920. This expenditure drops to £10,543,572 every 10 years when only 39 buses need to

be renewed. It is important to note that the hydrogen fleet renewable expenditure will only begin in

the 13th year, or year 2038, after the first batch of hydrogen bus reaches its lifetime of 12 years.

This is why the 2030 predicted price, and not the current price, of hydrogen bus is used. A

summary of expenditures related to the Green Hydrogen Public Mobility Project is seen below in

Table 8.2.

86
 Lingxi Tang

< Table 8.2: Expenditure related to Green Hydrogen Public Mobility Project >

While the depreciations of the solar farm, electrolyser, battery and cracker are based on the

diminishing value method, the purchase and sale of buses is constant for the first ten years. As

mentioned earlier, 40 diesel or hybrid buses will be replaced by 40 hydrogen buses every year (39

in the final year) during the 10-year transition period from BAU to net zero. Thus, one-tenth of

hydrogen fleet purchase costs will be accounted for annually in the first ten years to reflect annual

purchases of hydrogen buses. Diesel buses will be replaced first before hybrid buses begin to be

40
replaced. During this period, as the number of diesel bus decreases by 399
or ~ 10%, so does the

annual OPEX of BAU. Simultaneously, hydrogen bus maintenance costs will increase by 10%

40
every year, starting from £8, 754, 707 × 399
= £877, 664. 86 in the first year, up to a constant

annual maintenance expenditure of £8,754,707 from the 10th year onwards once the full hydrogen

fleet is purchased.

8.3 Comparison with Business As Usual

Combining the expenditures discussed and applying a real discount rate of 3.5% [8.2], the net

present values (NPV) of the annual expenditure for both Green Hydrogen and BAU up to 35 years

after the transition begins are tabulated in Fig. 8.1 below. The difference between the two is also

plotted to show how much more spending is expected due to the transition. The compiled

numerical data for the first ten years can be found in Appendix C.

87
 Lingxi Tang

< Fig. 8.1: Net Present Value of annual expenditure over the first 35 years of transition >

Generally, the value of expenditure NPV decreases over the years due to discounting. Note that

the cost of maintaining BAU is constant every year but decreases in NPV due to discounting. Also,

there will always be the cost of asset depreciation, namely due to the solar farm, electrolyser,

battery and ammonia cracker, for as long as the project stays online. This cost is based on the

assumption that each component is replaced immediately after their lifetime. From Fig. 8.1, it can

be observed that the purchase of the hydrogen fleet will incur a large additional annual expenditure

for the first ten years, resulting in the sudden decrease in expenditure in year 11. From year 13

onwards, the net zero transition then begins to cost more as hydrogen fleet renewal cost sets in.

Once the transition process is fully complete by year 23, maintaining the green hydrogen public

mobility system will cost ~ £6.39 million more than BAU annually, which translates to ~ £2.82

million with discounting. Given that producing hydrogen is already cheaper than buying diesel, this

additional cost will come from purchasing the solar farm, smart electrolyser-battery system,

hydrogen storage and maintenance of the hydrogen bus fleet (which is more expensive than that of

diesel fleet).

While this long-term additional cost may seem financially unattractive, there are two important

financial arguments which further justifies the Green Hydrogen Public Mobility Project. First of all, a

net zero transition based on green hydrogen is likely to be cheaper than that based on electric

vehicles due to the higher costs of electric buses and charging infrastructure, making this the better

alternative for an essential transition to net zero. Secondly, the cost analysis of BAU assumes that

diesel costs and diesel bus maintenance costs remain constant up till 2060, which is unlikely to be

88
 Lingxi Tang

true. According to the future price assumptions by the Department of Business, Energy and

Industrial Strategy (BEIS), the “Central Price” of oil is expected to increase from $63/bbl (barrel) in

2019 to $90/bbl in 2035 [8.3]. Diesel bus maintenance costs are also likely to increase due to more

intensive fleet retrofitting to meet low emissions standards imposed by the ZEZ.

8.4 Financing the transition

Now that the additional expenditure required for this green transition is ascertained, this subsection

aims to take revenue into account to complete a cash flow analysis.

Due to the sensitive nature of financial information, exact revenue and profits are difficult to obtain.

Nonetheless, profits of the main bus companies will be estimated based on the approximate

percentage of reserves within OBC’s revenue (see Fig. 8.2).

< Fig. 8.2: Breakdown of revenue spending by Oxford Bus Company [5.2] >

As such, it will be assumed that 5% of revenue is profit. From Fig 8.2, 4% of revenue is spent on

fuel duty, which costs £0.5795/litre of diesel [8.4]. The annual diesel consumption of 13,383,242.6

litres (mentioned in Section 4.3) translates to an annual spending of £7,755,589.09 on fuel duty.

Thus, this amounts to an annual revenue of ~ £193,842,750 and annual profits of £7,753,710

across all three major bus companies in Oxford.

Since the UK Government plans to have 5000 MW of “low carbon” hydrogen production capacity

by 2030 [8.5], it will be assumed that the solar farm, electrolyser and battery would be publicly

financed. Thus, the following cash flow analysis will not take into account the costs of those

components. The annual cash flow is calculated by subtracting profits by the difference between

BAU and Green Hydrogen transition, the latter shown in Fig. 8.1. The yearly cumulative cash flow

89
 Lingxi Tang

is tabulated in Fig. 8.3 below and specific values for up to year 10 can be found in Appendix D.

Without any additional sources of funding, the bus companies will face a maximum negative

balance of ~ − £6,976,441 in Year 4 and break-even occurs in Year 9.

8.4.1 Loan

To relieve this financial debt, a £5 million business loan is proposed. Assuming an annual interest

rate of 7.5%, a repayment period of 10 years and using the amortisation formula:

𝑟=
𝑃𝑖 (8.1)
(1 − (1 + 𝑖)−𝑁)
where 𝑟 is the yearly repayment, 𝑃 is the initial loan amount, 𝑖 is the interest rate and 𝑁 is the

repayment duration in years, the yearly repayment amount is ~ £728,430. Taking the initial loan

and yearly repayments into account, the maximum negative cash balance is delayed to year 5 and

reduced to ~ − £5,104,011, as can be seen in Fig. 8.3. This debt amount is reduced even further if

the future is discounted. Taking the loan into account, the updated yearly cumulative cash flow up

to Year 10 is in Appendix D.

< Fig. 8.3: Cash Flow of Green Hydrogen transition with and without an initial loan >

8.4.2 Increase of Passenger Fare

Despite the substantial loan amount, the negative cumulative cash balance is still financially

unattractive. Increasing passenger ticket prices would be a viable method to increase revenue;

after all, 40% of respondents are willing to pay up to £1 (see Section 5), which is ~ 43.5% of the

current average Oxford bus ticket price of £2.30. Currently, revenue from passenger fares is at

£193,842,750 × 80% = £155,074,200. If passenger fares increase by just 5% (11.5p for average

90
 Lingxi Tang

Oxford bus ticket), this would amount to an additional yearly revenue of ~ £7,753,710 and thus

annual profits of £15,507,420. This increase in revenue will negate the need for a loan since the

profits are enough to cover the additional cost of transition to green hydrogen from the start of the

transition period. Yearly cumulative cash flow, with just the expenditure, increased revenue and

without the loan can be found in Appendix D. Having said this, it is important to note that increasing

fares might face social and political barriers due to its unpopular nature, and also legal barriers, as

explained in Section 5. While possessing its challenges, this solution is incredibly lucrative, making

it a necessary financial solution for this sustainable transition.

8.4.3 Grants

Given the UK government’s general support for sustainability and more specifically, its plan to

potentially ban sales of new diesel buses [8.6], additional financial support can be expected from

the Government to assist transition towards a hydrogen bus fleet in Oxford. This is on top of the

Climate Emergency Budget mentioned in Section 1.2.

The Department for Transport (DfT) has recently announced the Zero Emission Buses Regional

Area (ZEBRA) scheme, which makes up to £120 million available for local transport authorities to

bid for the purchase of zero-emission buses [8.7]. Applications are now open for this funding and

end on 21 May 2021. This initiative is part of the Bus Back Better campaign, backed by £3 billion,

to accelerate the transition to greener and more sustainable transport in the UK. This campaign

aims to deliver 4,000 electric or hydrogen buses and to ban sales of new diesel buses [8.8]. As

such, the goals of this campaign ties in closely with this project.

The new ZEBRA scheme is just one of many funding campaigns to support clean and green local

public transport systems from the DfT, with past campaigns such as the Ultra Low Emissions Bus

scheme which awarded ~ £48 million to local authorities in 2018 [8.9] and the Low Emissions Bus

scheme which awarded more than £30 million in 2015 [8.10]. Oxford City Council had also been

awarded £2.3 million [8.11] from the Clean Bus Technology Fund in 2018 to 2019 to reduce air

pollution from public buses. Thus, extrapolating from history, financial support from the

Government to transform public transport to become powered by green gas is likely.

91
 Lingxi Tang

9 Conclusion
To summarise, this report proposes an Oxford public transport system comprising a bus fleet of

399 hydrogen buses, with fuel production powered by a local 150 MW solar farm, 42 MW PEM

electrolyser and a 13 MWh Li-Ion battery. This will cost an additional ~ £126.8 million over the

ten-year transition period of 2025 to 2035, which can be financed through loans, grants or increase

in passenger fares. Net zero is achieved by using zero-emissions hydrogen buses and clean

renewable energy to power the hydrogen production process, which is made more efficient using

the proposed smart electrolyser-battery system. This system benefits the national energy system

by exporting excess solar power to the National Grid. Production, storage and distribution needs

were also examined to ascertain technical feasibility, while social, legal and financial analysis of the

proposal have also been conducted.

9.1 Project limitations

This subsection aims to point out possible limitations which may affect the results of this project

proposal. First of all, the COVID-19 pandemic has brought unprecedented change to the world and

affected public transport operations deeply. Thus, this report has taken this to be an anomaly and

based calculations on pre-COVID levels. However, it is uncertain when or if public transport will

return to pre-COVID operation frequencies, especially with the current trend of working from home.

This may result in a long-term decrease in public transport demand, thus reducing hydrogen

demand.

Secondly, there is room for improvement in terms of availability and suitability of data collected for

calculations. Certain specific data is difficult to locate, such as compressor capital cost or ammonia

cracker lifetime while mpg data can differ greatly based on a variety of factors such as bus model,

driving habits, local topography, etc. As such, these may contribute to slight inaccuracies in the

results.

Finally, since hydrogen technology is still an emerging technology in the transportation industry,

there is a lack of laws and regulation on hydrogen in the UK. While there are global standards and

an ongoing national project to develop hydrogen laws, there will still be slight uncertainty on the

outcome.

92
 Antoine Thibaut

9.2 Next steps

The framework of the proposed net zero solution can be taken further to all public transport

systems including taxi services in Oxford. ​There are 3 main taxi services in Oxford [9.1, 9.2, 9.3].

These taxi services could become net zero emitters by replacing their current vehicles with

hydrogen vehicles. Hydrogen cars typically use a higher 700 bar pressure for storing hydrogen

[3.30]. This would require upgraded compressors, refrigeration systems [3.30] and storage vessels

for the system.

Additionally, due to the versatility of hydrogen, it is applicable to many sectors. These sectors

include food processing, fertilisers and metal treatment industries [9.4]. Future hydrogen pipelines

could also be installed to simplify hydrogen distribution. This can provide potential solutions for

house heating for example.

Alternative electrolyser power sources may be investigated. There is an ongoing hydroelectric

project in Kennington close to Oxford which may provide power for the electrolyser [9.5]. The

Gigastack project also aims for cheaper and larger electrolysers [9.6]. Solar farms could also be

further integrated into grid electricity networks.

In conclusion, a smart local energy transport system can be extended to many wider applications.

The storage, distribution and production infrastructure required for a hydrogen net zero transport

system can be extended for other uses. Hydrogen is truly a versatile gas which can be produced

from different renewable sources. Developing a zero emission bus network is only the first step of a

wide transformation project that will benefit many Oxford industries, households and local

businesses.

93
 Alexander Astor

References
The link to the code repository for this report is
https://github.com/akshaypal123/Green-Hydrogen-Public-Mobility-Repository

Section 1 [1.8] Oxford City Council (2020). Oxford City

Council announces city has achieved 2020
[1.1] Oxford City Council (2020). Oxford’s first emissions reduction target and sets date for
electric open-top double decker bus has Zero Carbon Oxford summit. [online]. Available
arrived. [online]. Available at: at:
https://www.oxford.gov.uk/news/article/1342/oxf https://www.oxford.gov.uk/news/article/1654/oxf
ord_s_first_electric_open-top_double_decker_b ord_city_council_announces_city_has_achieve
us_has_arrived [Accessed 7th of April 2021]. d_2020_emissions_reduction_target_and_sets
_date_for_zero_carbon_oxford_summit
[1.2] United Nations (2015). The Paris [Accessed 7th of April 2021].
Agreement. [pdf]. Available at:
https://unfccc.int/sites/default/files/english_paris [1.9] Oxford City Council (2021). Zero Carbon
_agreement.pdf [Accessed 7th of April 2021]. Oxford Summit. [online]. Available at:
https://www.oxford.gov.uk/info/20291/climate_e
[1.3] United Nations Climate Change (UNFCC) mergency/1431/zero_carbon_oxford_summit
(2016). The Paris Agreement. [online]. Available [Accessed 7th of April 2021].
at:
https://unfccc.int/process-and-meetings/the-pari [1.10] Oxford City Council (2019). New data on
s-agreement/the-paris-agreement [Accessed carbon emissions in Oxford. [online]. Available
7th of April 2021]. at:
https://www.oxford.gov.uk/news/article/1184/ne
[1.4] UK Government (2019). The Climate w_data_on_carbon_emissions_in_oxford
Change Act 2008 (2050 Target Amendment) [Accessed 7th of April 2021].
Order 2019. [online]. Available at:
https://www.legislation.gov.uk/ukdsi/2019/97801 [1.11] Oxford City Council (2019). New Low
11187654 [Accessed 7th of April 2021]. Emission Zone for buses in Oxford. [online].
Available at:
[1.5] Energy & Climate Intelligence Unit (n.d). https://www.oxford.gov.uk/news/article/1088/ne
How is the UK tackling climate change? w_low_emission_zone_for_buses_in_oxford
[online]. Available at: [Accessed 7th of April 2021].
https://eciu.net/analysis/briefings/uk-energy-poli
cies-and-prices/how-is-the-uk-tackling-climate-c [1.12] Oxford Bus Company (2021). Oxford Bus
hange [Accessed 7th of April 2021]. Company Fleet List. [pdf]. Available at:
https://assets.goaheadbus.com/media/cms_pag
[1.6] Ipsos MORI (2019). Oxford Citizens’ e_media/72/OBC_Fleet_210204.pdf [Accessed
Assembly on Climate Change. A summary 7th of April 2021].
report prepared for Oxford City Council. [online].
Available at: [1.13] Thames Travel (2021). Thames Travel
https://www.oxford.gov.uk/downloads/download/ Fleet List. [pdf]. Available at:
1132/oxford_citizens_assembly_on_climate_ch https://assets.goaheadbus.com/media/cms_pag
ange_report [Accessed 7th of April 2021]. e_media/423/TT_Fleet_210204.pdf [Accessed
7th of April 2021].
[1.7] Oxford City Council (2019). City Council
responds to Oxford Citizens’ Assembly on [1.14] Stagecoach Oxfordshire (2020). About
Climate Change and outlines £19m climate Stagecoach in Oxfordshire. [online]. Available
emergency budget. [online]. Available at: at:
https://www.oxford.gov.uk/news/article/1275/city https://www.stagecoachbus.com/about/oxfordsh
_council_responds_to_oxford_citizens_assembl ire#tab5 [Accessed 7th of April 2021].
y_on_climate_change_and_outlines_19m_clim
ate_emergency_budget [Accessed 7th of April
2021].

94
 Akshay Pal

[1.15] www.oxfordbus.co.uk. (n.d.). Times&Tcks [1.25] Ammonia Energy Association. (n.d.).

: Oxford Bus Company. [online] Available at: Ammonia for Power: a literature review. [online]
https://www.oxfordbus.co.uk/timetables-tickets/. Available at:
[Accessed 2nd of April 2021] https://www.ammoniaenergy.org/articles/ammon
ia-for-power-a-literature-review/ [Accessed 25th
[1.16] www.stagecoachbus.com. (n.d.). Bus of March 2021].
Times & Timetables | Stagecoach. [online]
Available at: [1.26] Clean Energy Institute, University of
https://www.stagecoachbus.com/timetables. Washington, (n.d.). Lithium-Ion Battery | Clean
[Accessed 2nd of April 2021]. Energy Institute. [online] Available at
https://www.cei.washington.edu/education/scien
[1.17] www.thames-travel.co.uk. (n.d.). ce-of-solar/battery-technology/#:~:text=They%2
Timetables & Fares : Thames Travel. [online] 0have%20one%20of%20the. [Accessed 25th of
Available at: March 2021]
https://www.thames-travel.co.uk/timetables-fare
s/. [Accessed 2nd of April 2021] [1.27] Berkeley.edu. (2019). List of common
conversion factors (Engineering conversion
[1.18] SMMT. (2016). December 2015 – EV factors) - IOR Energy Pty Ltd. [online] Available
registrations. [online] Available at: at:
https://www.smmt.co.uk/2016/01/december-201 http://w.astro.berkeley.edu/~wright/fuel_energy.
5-ev-registrations/. [Accessed 25th of March html. [Accessed 25th of March 2021]
2021].
[1.28] Nunno, R. (2018). Fact Sheet: Battery
[1.19] SMMT. (2017). December 2016 – EV Electric Buses: Benefits Outweigh Costs | White
registrations. [online] Available at: Papers | EESI. [online] Eesi.org. Available at:
https://www.smmt.co.uk/2017/01/december-201 https://www.eesi.org/papers/view/fact-sheet-ele
6-ev-registrations/. [Accessed 25th of March ctric-buses-benefits-outweigh-costs. [Accessed
2021] 25th of March 2021]​
[1.20] SMMT. (2018). December - EV [1.29] Wrightbus showcases first hydrogen fuel
registrations. [online] Available at: cell double-decker bus. (2018). Fuel Cells
https://www.smmt.co.uk/2018/01/december-ev-r Bulletin, 2018(12), p.3. Available at
egistrations/. [Accessed 25th of March 2021]‌ https://doi.org/10.1016/S1464-2859(18)30441-3
[1.21] SMMT. (2019). December – EV [1.30] Fuel Cell Electric Buses. (2015). About
registrations - SMMT. [online] Available at: Fuel Cell Electric Buses. [online] Available at:
https://www.smmt.co.uk/2019/01/december-ev-r https://www.fuelcellbuses.eu/wiki/fuel-cell-electri
egistrations-2/. [Accessed 25th of March 2021] c-buses-fuel-cell-electric-buses/about-fuel-cell-e
lectric-buses. [Accessed 25th of March 2021]
[1.22] SMMT. (2019). December – EV
registrations - SMMT. [online] Available at: [1.31] h2bus.eu. (n.d.). Offering. [online]
https://www.smmt.co.uk/2019/01/december-ev-r Available at: https://h2bus.eu/offering
egistrations-2/. [Accessed 25th of March 2021] [Accessed 25th of March 2021]​
[1.23] Hwang, H.T. and Varma, A. (2014). [1.32] Fuel Cell Electric Buses. (2015). Different
Hydrogen storage for fuel cell vehicles. Current types of buses. [online] Available at:
Opinion in Chemical Engineering, 5, pp.42–48. https://www.fuelcellbuses.eu/wiki/concepts-fuel-
Available at cell-electric-buses/different-types-buses
https://doi.org/10.1016/j.coche.2014.04.004 [Accessed 25th of March 2021].
[1.24] Valera-Medina, A., Xiao, H., [1.33] Mario (2021). Fuel cell bus projects in the
Owen-Jones, M., David, W.I.F. and Bowen, P.J. spotlight: fleets, manufacturers, trends. [online]
(2018). Ammonia for power. Progress in Energy Sustainable Bus. Available at
and Combustion Science, [online] 69, https://www.sustainable-bus.com/fuel-cell/fuel-c
pp.63–102. Available at ell-bus-hydrogen/ [Accessed 25th of March
https://doi.org/10.1016/j.pecs.2018.07.001 2021].

95
 Akshay Pal

[1.34] FuelCellsWorks (n.d.). Skeleton [1.42] Annual Report (2019). (2019). [online]
Technologies’ Next Generation High Energy Stagecoach Oxfordshire. Available at:
Ultracapacitors To Power Wrightbus’ Hydrogen https://www.stagecoachbus.com/media-library/fi
Buses - FuelCellsWorks. [online] Available at: les/oxfordshire/oxf%20annual%20report%2020
https://fuelcellsworks.com/news/skeleton-techn 19%202020.pdf?fbclid=IwAR2b4qKgPVdBfni1ki
ologies-next-generation-high-energy-ultracapaci SfBKl6PrW93RryZoCc0vrDlt0VRvNlYmEfNCm
tors-to-power-wrightbus-hydrogen-buses/ wRPM [Accessed 2nd of April 2021].
[Accessed 25th of March 2021].

[1.35] IER. (2020). The Environmental Impact

of Lithium Batteries. [online] Available at
https://www.instituteforenergyresearch.org/rene
wable/the-environmental-impact-of-lithium-batte
ries/. [Accessed 25th of March 2021]

[1.36] Katwala, A. (2018). The spiralling

environmental cost of our lithium battery
addiction. [online] Wired.co.uk. Available at
https://www.wired.co.uk/article/lithium-batteries-
environment-impact. [Accessed 25th of March
2021]

[1.37] Maisel, D. (2018). Desolation Desert.

[Photography] Available at
https://davidmaisel.com/works/desolation-desert
/ [Accessed 25th of March 2021]

[1.38] United States Geological Survey (2021).

Cobalt Statistics and Information. [online]
Available at
https://pubs.usgs.gov/periodicals/mcs2021/mcs
2021-cobalt.pdf [Accessed 25th of March 2021].

[1.39] Amnesty.org. (2016). Amnesty

International. [online] Available at:
https://www.amnesty.org/en/latest/news/2016/0
1/child-labour-behind-smart-phone-and-electric-
car-batteries/. [Accessed 25th of March 2021]

[1.40] Google Maps. (n.d.). Google Maps.

[online] Available at:
https://www.google.co.uk/maps/. [Accessed 2nd
of April 2021].

[1.41] Sustainability Report 2019. (2019).

[online] The Oxford Bus Group. Available at:
https://assets.goaheadbus.com/media/cms_pag
e_media/70/The_Oxford_Bus_Group_Sustaina
bility_Report_2019.pdf. [Accessed 2nd of April
2021].

96
 Akshay Pal

Section 2 www.greentechmedia.com. Available at

https://www.greentechmedia.com/articles/read/g
[2.1] Howatson, A.M., Lund, P.G. and Todd, reen-hydrogen-explained. [Accessed 26th of
J.D. (2009). Engineering tables and data. March 2021].
Department of Engineering Science, University
of Oxford. [2.12] İbrahim Dinçer and Calin Zamfirescu
(2016). Sustainable hydrogen production.
[2.2] Frano Barbir (2013). PEM fuel cells : Boston: Elsevier.
theory and practice. Waltham, Ma: Elsevier
Academic Press. [2.13] Hall, W., Spencer, T., Renjith, G., and
Dayal, S. (2020). The Potential Role of
[2.3] Larminie, J. and Dicks, A. (2003). Fuel cell Hydrogen in India: A pathway for scaling-up low
systems explained. Chichester, West Sussex: J. carbon hydrogen across the economy. New
Wiley. Delhi: The Energy and Resources Institute
(TERI). Available at
[2.4] Barbir, F. (1997). Efficiency and economics https://www.teriin.org/sites/default/files/2020-12/
of proton exchange membrane (PEM) fuel cells. Report%20on%20The%20Potential%20Role%2
International Journal of Hydrogen Energy, 0of%20Hydrogen%20in%20India%20%E2%80
[online] 22(10-11), pp.1027–1037. Available at %93%20%27Harnessing%20the%20Hype%27.
https://doi.org/10.1016/S0360-3199(96)00175-9‌ pdf
[2.5] Litster, S. and McLean, G. (2004). PEM [2.14] Norsk Hydrogenforum. (n.d.). EvoBus
fuel cell electrodes. Journal of Power Sources, (Mercedes-Benz). [online] Norsk
[online] 130(1-2), pp.61–76. Available at: Hydrogenforum. Available at:
https://doi.org/10.1016/j.jpowsour.2003.12.055. https://www.hydrogen.no/kjoretoy/hydrogenbus
[2.6] Folkson, R. and Woodhead Publishing ser/evobus-daimler [Accessed 30th of March
(2014). Alternative fuels and advanced vehicle 2021].‌
technologies for improved environmental [2.15] H2.LIVE. (n.d.). Toyota MIRAI. [online]
performance : towards zero carbon Available at:
transportation. Amsterdam: Woodhead https://h2.live/en/wasserstoffautos/toyota-mirai
Publishing/Elsevier, Cop. [Accessed 30th of March 2021].
[2.7] Fuel Cell & Hydrogen Energy Association. [2.16] Andersson, J. and Grönkvist, S. (2019).
(n.d.). Fuel Cell Basics. [online] Available at: Large-scale storage of hydrogen. International
https://www.fchea.org/fuelcells#:~:text=In%20a Journal of Hydrogen Energy, 44(23),
%20fuel%20cell%2C%20hydrogen. [Accessed pp.11901–11919. Available at
8th of April 2021] https://doi.org/10.1016/j.ijhydene.2019.03.063.
[2.8] Liu, K., Song, C. and Subramani, V. eds., [2.17] Klerke, A., Christensen, C.H., Nørskov,
(2009). Hydrogen and Syngas Production and J.K. and Vegge, T. (2008). Ammonia for
Purification Technologies. Hoboken, NJ, USA: hydrogen storage: challenges and
John Wiley & Sons, Inc. opportunities. Journal of Materials Chemistry,
[2.9] Newsome, D.S. (1980). The Water-Gas 18(20), p.2304. Available at
Shift Reaction. Catalysis Reviews, 21(2), https://doi.org/10.1039/B720020J
pp.275–318. Available at [2.18] Kang, D.W. and Holbrook, J.H. (2015).
https://doi.org/10.1080/03602458008067535. Use of NH3 fuel to achieve deep greenhouse
[2.10] Rapier, R. (n.d.). Estimating The Carbon gas reductions from US transportation. Energy
Footprint Of Hydrogen Production. [online] Reports, 1, pp.164–168. Available at
Forbes. Available at: https://doi.org/10.1016/j.egyr.2015.08.001
https://www.forbes.com/sites/rrapier/2020/06/06
/estimating-the-carbon-footprint-of-hydrogen-pr
oduction/?sh=719f2dc724bd [Accessed 15th of
May 2021].

[2.11] Deign, J. (2020). So, What Exactly Is

Green Hydrogen? [online]

97
 Lingxi Tang

[2.19] Kobayashi, H., Hayakawa, A., /09724-industrialimpact.pdf [Accessed 7th of

Somarathne, K.D.Kunkuma A. and Okafor, April 2021].
Ekenechukwu C. (2019). Science and
technology of ammonia combustion. [2.27] Pattabathula, V. and Richardson, J.
Proceedings of the Combustion Institute, (2016). Introduction to Ammonia Production.
[online] 37(1), pp.109–133. Available at: [online] Aiche.org. Available at:
https://doi.org/10.1016/j.proci.2018.09.029. https://www.aiche.org/resources/publications/ce
[Accessed 7th of April 2021] p/2016/september/introduction-ammonia-produ
ction. [Accessed 7th of April 2021].
[2.20] Office of Energy Efficiency and
Renewable Energy. (n.d.). Hydrogen Storage. [2.28] Jackson, C., Fothergill, K., Gray, P.,
[online] Available at: Haroon, F., Makhloufi, C., Kezibri, N., Davey, A.,
https://www.energy.gov/eere/fuelcells/hydrogen- Lhote, O., Zarea, M., Davenne, T., Greenwood,
storage#:~:text=Hydrogen%20can%20be%20st S., Huddart, A., Makepeace, J., Wood, T.,
ored%20physically. [Accessed 7th of April David, B. and Wilkinson, I. (n.d.). Ammonia to
2021]. Green Hydrogen Project: Feasibility Study.
[online] Available at:
[2.21] Erdemir, D. and Dincer, I. (2020). A https://assets.publishing.service.gov.uk/govern
perspective on the use of ammonia as a clean ment/uploads/system/uploads/attachment_data/
fuel: Challenges and solutions. International file/880826/HS420_-_Ecuity_-_Ammonia_to_Gr
Journal of Energy Research, 45(4), een_Hydrogen.pdf [Accessed 7th of April 2021]
pp.4827–4834. Available at
https://doi.org/10.1002/er.6232 [Accessed 7th of [2.29] Wrightbus. (n.d.). Wrightbus - driving a
April 2021]. Greener Future. [online] Available at:
https://www.wrightbus.com/ [Accessed 7th of
[2.22] Jeerh, G., Zhang, M. and Tao, S. (2021). April 2021].
Recent progress in ammonia fuel cells and their
potential applications. Journal of Materials [2.30] Intelligent Transport. (n.d.). Aberdeen
Chemistry A, [online] 9(2), pp.727–752. becomes home to world’s first
Available at hydrogen-powered double decked bus. [online]
https://doi.org/10.1039/D0TA08810B [Accessed Available at:
7th of April 2021]. https://www.intelligenttransport.com/transport-n
ews/109315/aberdeen-becomes-home-to-world
[2.23] Lan, R. and Tao, S. (2014). Ammonia as s-first-hydrogen-powered-double-decked-bus/.
a Suitable Fuel for Fuel Cells. Frontiers in
Energy Research, 2. Available at [2.31] Birmingham City Council. (n.d.).
https://doi.org/10.3389/fenrg.2014.00035 Hydrogen buses coming to Birmingham next
[Accessed 7th of April 2021]. Spring [online] Available at:
https://www.birmingham.gov.uk/news/article/727
[2.24] US Department of Energy (n.d.). /hydrogen_buses_coming_to_birmingham_next
Potential Roles of Ammonia in a Hydrogen _spring [Accessed 7th of April 2021].
Economy: A Study of Issues Related to the Use
Ammonia for On-Board Vehicular Hydrogen [2.32] H2-View (2021). Liverpool set for
Storage. [online] . Available at: state-of-the-art hydrogen buses. [online]
https://www.energy.gov/sites/prod/files/2015/01/ Hydrogen View. Available at:
f19/fcto_nh3_h2_storage_white_paper_2006.pd https://www.h2-view.com/story/liverpool-set-for-
f. [Accessed 7th of April 2021] state-of-the-art-hydrogen-buses/ [Accessed 7th
of April 2021].
[2.25] Krietsch, L. (2019). Taking the CO2 out
of NH3. [online] Chemical & Engineering News.
Available at:
https://cen.acs.org/environment/green-chemistr
y/Industrial-ammonia-production-emits-CO2/97/i
24. [Accessed 7th of April 2021].

[2.26] Brunning, A. (2019). Environmental

Impacts of Industrial Reactions. [online]
Available at:
https://cen.acs.org/content/dam/cen/97/24/WEB

98
 Antoine Thibaut

Section 3 909_FINAL_Bus_Study_Report_OUT_0.PDF.
[Accessed October 31st 2020]
[3.1] TransferWise Ltd. (2021). Euro to British
Pound Sterling. [Online]. Available at: [3.10] FuelCellStore. (2018). Fuel Cell Stack
https://transferwise.com/gb/currency-converter/ Temperature in Mid-to-High Temperature Fuel
eur-to-gbp-rate#rate-alerts. [Accessed 28th of Cells. [Online]. Available at:
February 2021] https://www.fuelcellstore.com/blog-section/fuel-c
ell-stack-temperature-in-mid-to-high-temperatur
[3.2] TransferWise Ltd. (2021). Us Dollar to e-fuel-cells. [Accessed 1st of April 2021]
British Pound Sterling. [Online]. Available at:
https://transferwise.com/gb/currency-converter/ [3.11] Ballard. (2018). Fuel Cell Electric Buses,
usd-to-gbp-rate. [Accessed 28th of February Cold Weather Operation. [Online]. Available at:
2021] https://www.ballard.com/docs/default-source/mo
tive-modules-documents/technical-note---bus-c
[3.3] ACEA, European Automobile old-weather-operation---final.pdf?sfvrsn=4&sfvr
Manufacturers Association. (2021). Diesel sn=4. [Accessed 1st of April 2021]
Exhaust Fluid – AdBlue. [Online]. Available at:
https://www.acea.be/industry-topics/tag/categor [3.12] California Hydrogen Business Council.
y/diesel-exhaust-fluid-adblue. [Accessed 16th of Fuel Cell Electric Buses for Zero-Emission
March 2021] Public Transit. [Pdf]. Available at:
https://www.californiahydrogen.org/wp-content/
[3.4] Growmark, Inc. (2011). Soot in Diesel uploads/2017/10/CHBC-CaFCP-Fuel-Cell-Elect
Engines. [Pdf]. Available at: ric-Bus-Fact-Sheet.pdf. [Accessed October 31st
https://www.gofurthergofs.com/Portals/0/Assets/ 2020]
Knowledge/Whitepapers/Soot-in-Diesel-Engine
s.pdf. [Accessed 6th of April 2021] [3.13] Sustainable Bus. (2019). Pau (France)
welcomes the launch of the world’s first
[3.5] Fast Fit. (2021). AdBlue. [Online]. hydrogen BRT. [Online]. Available at:
Available at: https://www.sustainable-bus.com/news/pau-wel
https://www.fastfit.ie/services/adblue.php. comes-the-launch-of-the-worlds-first-hydrogen-
[Accessed 16th of March 2021] brt/. [Accessed 4th of April 2021]
[3.6] Oxford Bus Company. About us. [Online]. [3.14] Stagecoach Oxfordshire. (2020). Annual
Available at: report. [Pdf]. [Accessed 26th of February 2021]
https://www.oxfordbus.co.uk/about-us/oxford-bu
s-company/. [Accessed 3rd of March 2021] [3.15] UK Buses. (2021). Stagecoach
Oxfordshire. [Pdf]. Available at:
[3.7] J. Partridge, W. Wu, R. Bucknall. (2016). https://www.ukbuses.co.uk/fleet/stagecoachoxfo
Development of bus Drive Technology towards rdshire.pdf. [Accessed 14th of March 2021]
Zero Emissions: A Review. [Online]. Available
at: [3.16] Google. (2021). Google Images. [Online].
https://www.intechopen.com/books/hybrid-electr Available at:
ic-vehicles/development-of-bus-drive-technolog https://www.google.com/imghp?hl=EN.
y-towards-zero-emissions-a-review. [Accessed [Accessed 14th of March 2021]
16th of March 2021]
[3.17] H2Bus. (2019). Affordable Truly
[3.8] U.S. Department of Energy. How Do Fuel Zero-Emission Buses [PowerPoint
Cell Electric Vehicles Work Using Hydrogen?. presentation]. Available at:
Alternative Fuels Data Center. [Online]. https://zero.no/wp-content/uploads/2019/09/29-
Available at: Lars-Jakobsen-Nel-H2BusEurope.pdf.
https://afdc.energy.gov/vehicles/how-do-fuel-cell [Accessed 4th of March 2021]
-electric-cars-work. [Accessed 3rd of February
2021] [3.18] Oxford bus company. When2Travel.
[Online]. Available at:
[3.9] Roland Berger. (2015). Fuel Cell Electric https://www.oxfordbus.co.uk/covid-19/when-2-tr
Buses – Potential for Sustainable Public avel/. [Accessed 20th November 2020]
Transport in Europe. [Pdf]. Available at:
https://www.fch.europa.eu/sites/default/files/150 [3.19] Holts. (2021). How to Tell What Type of
Road You’re Driving on in the UK. [Online].

99
 Antoine Thibaut

Available at: electric vehicles. [Pdf]. Available at:

https://www.holtsauto.com/holts/news/how-to-tel https://www.osti.gov/pages/servlets/purl/138963
l-what-type-of-road-youre-driving-on-in-the-uk/. 5. [Accessed May 6th 2021]
[Accessed 3rd of May 2021]
[3.29] US. Department of Energy, Energy
[3.20] Hydrogen Europe. (2017). Hydrogen Efficiency and Renewable Energy, Fuel Cell
buses. [Online]. Available at: Technologies Office. (2017). Hydrogen storage.
https://www.hydrogeneurope.eu/hydrogen-buse [Pdf]. Available at:
s. [Accessed 3rd of March 2021] https://www.energy.gov/sites/prod/files/2017/03/
f34/fcto-h2-storage-fact-sheet.pdf. [Accessed
[3.21] Ballard. (2020). Fuel Cell Price to Drop 19th of March 2021]
70-80% as Production Volume Scales. [Online].
Available at: [3.30] D. Reuter, M. Faltenbacher, O. Schuller,
https://blog.ballard.com/fuel-cell-price-drop. N. Whitehouse, S. Whitehouse. (2017). New
[Accessed 4th of March 2021] Bus ReFuelling for European Hydrogen Bus
Depots. [Pdf]. Available at:
[3.22] Innovative UK, Technology Strategy http://newbusfuel.eu/wp-content/uploads/2017/0
Board, Equipmake. Cost Effective Electric Bus. 3/NewBusFuel_D4.3_Guidance-document-for-l
[PowerPoint presentation]. Available at: arge-scale-hydrogen-refuelling_final.pdf.
https://www.cenex-lcv.co.uk/storage/seminar-pr [Accessed 5th of April 2021]
ogramme/sessions/presentations/ian_foley_the
_cost_effective_electric_bus_1537363682.pdf?f [3.31] B. Williams, M. Davy. (2019).
bclid=IwAR2z6fAAUBmbH1q7gTFV8MkkscfF0 Engineering Science Prelims, P4
TjzSaDVNMouJ-OrP0Rkjw8NWuWjPdo. Thermodynamics. Department of Engineering
[Accessed 14th of March 2021] Science, University of Oxford. [Accessed 19th
of March 2021]
[3.23] Depreciationrates.net. (2020). ATO
Depreciation Rates 2020. [Online]. Available at: [3.32] A. Elgowainy, K. Reddi, M. Aly, Argonne
https://www.depreciationrates.net.au/bus. National Laboratory. (2018). Hydrogen
[Accessed 3rd of March 2021] Refueling Analysis of Fuel Cell Heavy-Duty
Vehicles Fleet. [Pdf]. Available at:
[3.24] Environmental and Energy Study https://www.energy.gov/sites/prod/files/2018/08/
Institute. Hybrid Buses Costs and Benefits. f54/fcto-truck-workshop-2018-9-elgowainy.pdf.
[Pdf]. Available at: [Accessed 7th of May 2021]
https://www.eesi.org/files/eesi_hybrid_bus_032
007.pdf. [Accessed 1st of April 2021] [3.33] Fuel Cells and Hydrogen Joint
Undertaking, CHIC. (2016). Clean Hydrogen in
[3.25] Anders Nordelöf, Mia Romare, Johan European Cities. [Pdf]. Available at:
Tivander. (2019). Life cycle assessment of city https://fuelcellbuses.eu/sites/default/files/docum
buses powered by electricity, hydrogenated ents/CHIC_publication_final_0.pdf. [Accessed
vegetable oil or diesel. [Pdf]. Available at: 20th of March 2021]
https://doi.org/10.1016/j.trd.2019.08.019.
[Accessed November 13th 2020] [3.34] Oxford Bus Company. (2020).
park&ride500 Covid-19 Timetables from 1st
[3.26] L. Marion. (2021). Interview with Green June 2020. [Pdf]. Available at:
Hydrogen Public Mobility Project. Interviewed https://assets.goaheadbus.com/media/cms_pag
by A. Astor, A. Pal, L. Tang, A. Thibaut. 9th of e_media/8103/200601_parkride500_COVID-19
February 2021 _Coronavirus_Timetable.pdf. [Accessed 26th of
March 2021]
[3.27] Low Carbon Vehicle Partnership. (2019).
The Clean Vehicle Retrofit Technology Guide. [3.35] Oxford Bus Company. (2020).
[Pdf]. Available at: park&ride400 Covid-19 Timetables from 24th
https://www.zemo.org.uk/assets/reports/LowCV October 2020. [Pdf]. Available at:
P_Clean_Vehicle_Retrofit_Technology_Guide_ https://assets.goaheadbus.com/media/cms_pag
Final.pdf. [Accessed 1st of April 2021] e_media/8103/201024_parkride400_COVID-19
_Coronavirus_TimetableV2.pdf. [Accessed 26th
[3.28] K Reddi, A. Elgowainy, N. Rustagi, E. of March 2021]
Gupta. Impact of hydrogen SAE J2601 fueling
methods on fueling time of light-duty fuel cell

100
 Antoine Thibaut

[3.36] LiveBus.org. (2021). City Centre and East%20Area%20Committee/200903171800/A

Pear Tree Park and Ride. [Online]. Available at: genda/81837Item5.pdf. [Accessed 2nd of April
http://www.livebus.org/oxfordshire/routes/obc/3 2021]
00/city-centre-and-pear-tree-park-and-ride/.
[Accessed 26th of March 2021] [3.45] FleetNews. (2020). Brake pad pollution
as bad as that from diesel exhaust. [Online].
[3.37] Stagecoach. (2021). S5 Bus Route & Available at:
Timetable: Oxford-Bicester. [Online]. Available https://www.fleetnews.co.uk/news/environment/
at: 2020/01/10/brake-pad-pollution-as-bad-as-that-f
https://www.stagecoachbus.com/routes/oxfords rom-diesel-exhaust. [Accessed 1st of April
hire/s5/oxford-bicester/xoas005.o. [Accessed 2021]
26th of March 2021]
[3.46] The Oxford Bus Group. (2020).
[3.38] F. Faraj, Towards Data Science. (2020). Sustainability report 2020. [Pdf]. Available at:
Plotting Regional Topographic Maps from https://assets.goaheadbus.com/media/cms_pag
Scratch in Python. [Online]. Available at: e_media/423/2020_Corporate_and_Social_Res
https://towardsdatascience.com/plotting-regiona ponsibility_Report_-_OBC.pdf. [Accessed 15th
l-topographic-maps-from-scratch-in-python-845 of March 2021]
2fd770d9d. [Accessed 26th of March 2021]
[3.47] Stagecoach Yorkshire. (2020). Annual
[3.39] FreeMapTools. (2019). Elevation Finder. report. [Pdf]. [Accessed 15th of March 2021]
[Online]. Available at:
https://www.freemaptools.com/elevation-finder.h [3.48] Australian Government, Australian
tm. [Accessed 26h of March 2021] Taxation Office. (2020). Guide to depreciating
assets 2020. [Pdf]. Available at:
[3.40] M. Islameka, E. Leksono, B. Yuliarto, https://www.ato.gov.au/uploadedFiles/Content/I
Journal of Physics: Conference Series. (2019). ND/Downloads/guide-to-depreciating-assets-20
Modelling of regenerative braking system for 20.pdf. [Accessed 4th of March 2021]
electric bus. [Pdf]. Available at:
https://iopscience.iop.org/article/10.1088/1742-
6596/1402/4/044054. [Accessed 20th of March
2021]

[3.41] The Engineering ToolBox. (2004). Drag

Coefficient. [Online]. Available at:
https://www.engineeringtoolbox.com/drag-coeffi
cient-d_627.html. [Accessed 22nd of March
2021]

[3.42] Hydrogen Mobility Australia. Capability

Statement, Hydrogen Fuel Cell Buses. [Pdf].
Available at:
https://static1.squarespace.com/static/5a668f10
80bd5e34d18a7e76/t/5c9401cd71c10b2160f88
c2f/1553203697824/Hydrogen+Fuel+Cell+Buse
s.pdf. [Accessed 22nd of March 2021]

[3.43] A. Doyle, T. Muneer. (2017). Traction

energy and battery performance modelling.
[Online]. Available at:
https://www.sciencedirect.com/topics/engineerin
g/regenerative-braking. [Accessed 24th of
March 2021]

[3.44] Oxfordshire Country Council. (2009). The

Oxfordshire County Council (City of Oxford) (20
MPH Speed Limit) Order 200. [Pdf]. Available
at:
https://mycouncil.oxford.gov.uk/Data/North%20

101
 Akshay Pal

Section 4

[4.1] RAC Foundation. (n.d.). Wholesale v retail

fuel prices. [online] Available at:
https://www.racfoundation.org/data/wholesale-f
uel-prices-v-pump-prices-data [Accessed 9 Mar
2021]

[4.2] Kobayashi, S., Plotkin, S. and Ribeiro,

S.K. (2009). Energy efficiency technologies for
road vehicles. Energy Efficiency, [online] 2(2),
pp.125–137. Available at:
https://link.springer.com/article/10.1007/s12053-
008-9037-3.

[4.3] www.volkswagenag.com. (n.d.). What’s

more efficient? Hydrogen or battery powered?
[online] Available at:
https://www.volkswagenag.com/en/news/stories
/2019/08/hydrogen-or-battery--that-is-the-questi
on.html#. [Accessed 6th of April 2021].

[4.4] Thomas, C.E. (2009). Fuel cell and battery

electric vehicles compared. International
Journal of Hydrogen Energy, 34(15),
pp.6005–6020. Available at
https://doi.org/10.1016/j.ijhydene.2009.06.003

[4.5] Li, M., Zhang, X. and Li, G. (2016). A

comparative assessment of battery and fuel cell
electric vehicles using a well-to-wheel analysis.
Energy, 94, pp.693–704. Available at
https://doi.org/10.1016/j.energy.2015.11.023

[4.6] Eaves, S. and Eaves, J. (2004). A cost

comparison of fuel-cell and battery electric
vehicles. Journal of Power Sources, [online]
130(1-2), pp.208–212. Available at
https://doi.org/10.1016/j.jpowsour.2003.12.016

[4.7] Rosero, F., Fonseca, N., López, J.-M. and

Casanova, J. (2020). Real-world fuel efficiency
and emissions from an urban diesel bus engine
under transient operating conditions. Applied
Energy, 261, p.114442. Available at
https://doi.org/10.1016/j.apenergy.2019.114442

[4.8] Fuel Cell Electric Buses - Potential for

Sustainable Public Transport in Europe A Study
for the Fuel Cells and Hydrogen Joint
Undertaking. (n.d.). [online] . Available at:
https://www.fch.europa.eu/sites/default/files/150
909_FINAL_Bus_Study_Report_OUT_0.PDF.
[Accessed 14th of April 2021]

102
 Alexander Astor

Section 5 [5.10] UK Government (2011). Gas Safety

(Management) Regulations 1996. [online].
[5.1] Cummins Inc. (2020). Momentum for Available at:
hydrogen transit grows as European consumers https://www.legislation.gov.uk/uksi/1996/551/co
prioritize climate before cost when it comes to ntents/made [Accessed 7th of April 2021].
travel. [online]. Available at:
https://www.cummins.com/news/releases/2020/ [5.11] UK Government (2011). The Pipelines
11/24/momentum-hydrogen-transit-grows-europ Safety Regulations 1996. [online]. Available at:
ean-consumers-prioritize-climate-cost https://www.legislation.gov.uk/uksi/1996/825/co
[Accessed 7th of April 2021]. ntents/made [Accessed 7th of April 2021].

[5.2] Oxford Bus Company (n.d.). How are your [5.12] Newsholme, G. (Health and Safety
services funded. [online]. Available at: Executive) (n.d.). Risk management strategy for
https://www.oxfordbus.co.uk/about-us/how-are- hydrogen economy projects. [pdf]. Available at:
your-services-funded/.[Accessed 7th of April http://www.hysafe.org/science/eAcademy/docs/
2021]. 1stesshs/presentations/Riskmanagementstrate
gyforhydrogeneconomyprojects.pdf [Accessed
[5.3] Oxford Bus Company (n.d.). Timetables & 7th of April 2021].
fares : city. [online]. Available at:
https://city.oxfordbus.co.uk/timetables-fares/ [5.13] UK Government (2015). The Planning
[Accessed 7th of April 2021]. (Hazardous Substances) Regulations 2015.
[online]. Available at:
[5.4] Office of Fair Trading (n.d.). A quick guide https://www.legislation.gov.uk/uksi/2015/627/co
to competition and consumer protection laws ntents/made [Accessed 7th of April 2021].
that affect your business. [pdf]. Available at:
https://assets.publishing.service.gov.uk/govern [5.14] Health and Safety Executive (2015).
ment/uploads/system/uploads/attachment_data/ Control of Major Accident Hazards Regulations
file/284428/oft911.pdf. [Accessed 7th of April 2015 (COMAH). [online]. Available at:
2021]. https://www.hse.gov.uk/comah/background/com
ah15.htm [Accessed 7th of April 2021].
[5.5] UK Government (2020). Competition Act
1998. [online]. Available at: [5.15] UK Government (2011). The Dangerous
https://www.legislation.gov.uk/ukpga/1998/41/co Substances and Explosive Atmospheres
ntents. [Accessed 7th of April 2021]. Regulations 2002. Available at:
https://www.legislation.gov.uk/uksi/2002/2776/c
[5.6] UK Government (n.d.). Environmental ontents/made [Accessed 7th of April 2021].
sustainability agreements and competition law.
[online]. Available at: [5.16] Health and Safety Executive (2019).
https://www.gov.uk/government/publications/env ATEX and explosive atmospheres – Fire and
ironmental-sustainability-agreements-and-comp explosion. [online]. Available at:
etition-law/sustainability-agreements-and-comp https://www.hse.gov.uk/fireandexplosion/atex.ht
etition-law [Accessed 7th of April 2021]. m [Accessed 7th of April 2021].

[5.7] Itaoka, K., Saito, A., Sasaki, K. (n.d.). [5.17] UK Government (2011). The Road
Public Perception of Hydrogen Infrastructure in Vehicles (Construction and Use) Regulations
Japan. [pdf]. 1986. [online]. Available at:
https://h2tools.org/sites/default/files/ICHS_impo https://www.legislation.gov.uk/uksi/1986/1078/c
rt/paper_284.pdf. [Accessed 7th of April 2021]. ontents/made [Accessed 7th of April 2021].

[5.8] Health and Safety Laboratory (2010). [5.18] UK Government (2011). The Road
Hazards of liquid hydrogen: Position paper. Vehicles (Authorisation of Special Types)
[pdf]. Available at: (General) Order 2003. Available at:
https://www.hse.gov.uk/research/rrpdf/rr769.pdf https://www.legislation.gov.uk/uksi/2003/1998/c
[Accessed 7th of April 2021]. ontents/made [Accessed 7th of April 2021].

[5.9] UK Government (2020). Gas Act 1986. [5.19] UK Government (2015). The Road Traffic
[online]. Available at: Act 1988.
https://www.legislation.gov.uk/ukpga/1986/44/co https://www.legislation.gov.uk/ukpga/1988/52/S
ntents. [Accessed 7th of April 2021]. ection/44 [Accessed 7th of April 2021].

103
 Alexander Astor

[5.20] Aberdeen City Council (2020). The

world’s first hydrogen-powered double decker
bus arrives in Aberdeen. [online]. Available at:
https://news.aberdeencity.gov.uk/the-worlds-first
-hydrogen-powered-double-decker-bus-arrives-i
n-aberdeen/ [Accessed 7th of April 2021].

[5.21] UK Government (2019). Town and

Country Planning Act 1990. [online]. Available
at:
https://www.legislation.gov.uk/ukpga/1990/8/con
tents [Accessed 7th of April 2021].

[5.22] UK Government (2019). Planning Act

2008. [online]. Available at:
https://www.legislation.gov.uk/ukpga/2008/29/co
ntents [Accessed 7th of April 2021].

[5.23] Department for Environment, Food and

Rural Affairs (2011). Town and Country
Planning Act 1990 – Summary. [online].
Available at:
http://adlib.everysite.co.uk/adlib/defra/content.a
spx?doc=19277&id=19279 [Accessed 7th of
April 2021].

[5.24] UK Government (n.d.). Localism Act

2011. [online]. Available at:
https://www.legislation.gov.uk/ukpga/2011/20/co
ntents/enacted [Accessed 7th of April 2021].

[5.25] Shepherd and Wedderburn (2009). The

Planning Act 2008: an overview. [online].
Available at:
https://shepwedd.com/knowledge/planning-act-
2008-overview [Accessed 7th of April 2021].

[5.26] Smith, L. (House of Commons Library)

(2017). Planning for Nationally Significant
Infrastructure Projects. [online]. Available at:
https://commonslibrary.parliament.uk/research-
briefings/sn06881/ [Accessed 7th of April 2021].

104
 Alexander Astor

Section 6 /04/1909_NUAC300B_Mono_Datasheet_EN.pd
f [Accessed 7 Apr. 2021].
[6.1] Evans, S. (2017). Solar, wind and nuclear
have “amazingly low” carbon footprints, study [6.10] Photovoltaic-software.com (2019). How
finds. [online]. Available at: to calculate output energy of PV solar systems?
https://www.carbonbrief.org/solar-wind-nuclear- [online]. Available at:
amazingly-low-carbon-footprints [Accessed 7 https://photovoltaic-software.com/principle-ress
Apr. 2021]. ources/how-calculate-solar-energy-power-pv-sy
stems [Accessed 7 Apr. 2021].
[6.2] Renewables First - The Hydro and Wind
Company (n.d.). How large are wind turbines for [6.11] Renewables.ninja (2019).
farms. [online]. Available at: Renewables.ninja. [online]. Available at:
https://www.renewablesfirst.co.uk/windpower/wi https://www.renewables.ninja/ [Accessed 7 Apr.
ndpower-learning-centre/how-large-are-farm-wi 2021].
nd-turbines-medium-wind-turbines/ [Accessed 7
Apr. 2021]. [6.12] Stoker, L. (2018). UK solar costs
plummeting beyond forecasts, as cheap as
[6.3] CPRE Oxfordshire (n.d.). What we think £40/MWh by 2030. [online]. Available at:
about wind turbines. [online]. Available at: https://www.solarpowerportal.co.uk/news/uk_sol
http://www.cpreoxon.org.uk/campaigns/energy- ar_costs_plummeting_beyond_forecasts_as_ch
and-water/energy/wind-turbines/item/2139-what eap_as_40_mwh_by_2030 [Accessed 7 Apr.
-we-think-wind-turbines [Accessed 7 Apr. 2021]. 2021].

[6.4] Department for Business, Energy & [6.13] Ruth, M., Mayyas, A. and Mann, M.
Industrial Strategy (2020). Electricity Generation (2017). CEMAC -Clean Energy Manufacturing
Costs 2020. [pdf]. Available at: Analysis Center Manufacturing Competitiveness
https://assets.publishing.service.gov.uk/govern Analysis for PEM and Alkaline Water
ment/uploads/system/uploads/attachment_data/ Electrolysis Systems Fuel Cell Seminar and
file/911817/electricity-generation-cost-report-20 Energy Expo CEMAC -Clean Energy
20.pdf [Accessed 7 Apr. 2021]. Manufacturing Analysis Center 2. [online] .
Available at:
[6.5] meteoblue (n.d.). Climate Oxford. [online]. https://www.nrel.gov/docs/fy19osti/70380.pdf
Available at: [Accessed 7 Apr. 2021].
https://www.meteoblue.com/en/weather/historyc
limate/climatemodelled/oxford_united-kingdom_ [6.14] Schmidt, O., Gambhir, A., Staffell, I.,
2640729 [Accessed 7 Apr. 2021]. Hawkes, A., Nelson, J. and Few, S. (2017).
Future cost and performance of water
[6.6] World Weather Online (n.d.). Oxford electrolysis: An expert elicitation study.
Monthly Climate Averages. [online]. Available International Journal of Hydrogen Energy,
at: [online] 42(52), pp.30470–30492. Available at:
https://www.worldweatheronline.com/oxford-we https://www.sciencedirect.com/science/article/pii
ather-averages/oxfordshire/gb.aspx [Accessed /S0360319917339435 [Accessed 1 Apr. 2020].
7 Apr. 2021].
[6.15] ITM Power. Interim Results Presentation
[6.7] Renewables First - The Hydro and Wind Jan 2021. [online]. Available at:
Company (n.d.). How much wind for a wind https://www.itm-power.com/images/Interim_Res
turbine. [online]. Available at: ults_Presentation_Jan_2021.pdf [Accessed 7
https://www.renewablesfirst.co.uk/windpower/wi Apr. 2021].
ndpower-learning-centre/how-windy-does-it-hav
e-to-be [Accessed 7 Apr. 2021]. [6.16] Guo, Y., Li, G., Zhou, J. and Liu, Y.
(2019). Comparison between hydrogen
[6.8] Eco2Solar (n.d.). How does Solar PV production by alkaline water electrolysis and
work. [online]. Available at: hydrogen production by PEM electrolysis. IOP
https://www.eco2solar.co.uk/solar-electricity/ho Conference Series: Earth and Environmental
w-does-solar-pv-work/ [Accessed 7 Apr. 2021]. Science, 371, p.042022. [Accessed 7 Apr.
2021].
[6.9] UPS Solar (2020). NU-AC300B. [online].
Available at:
https://ups-solar.co.uk/wp-content/uploads/2020

105
 Lingxi Tang

[6.17] Gigastack. (n.d.). Gigastack -

Demonstrating renewable hydrogen for a net
zero future. [online] Available at:
https://gigastack.co.uk/. [Accessed 7 Apr.
2021].

[6.18] Clean Energy Partnership. (n.d.).

Hydrogen production and storage. [online]
Available at:
https://cleanenergypartnership.de/en/faq/hydrog
en-production-and-storage/. [Accessed 7 Apr.
2021].

[6.19] International Renewable Energy Agency

(n.d.). UTILITY-SCALE BATTERIES
INNOVATION LANDSCAPE BRIEF. [online] .
Available at:
https://www.irena.org/-/media/Files/IRENA/Agen
cy/Publication/2019/Sep/IRENA_Utility-scale-ba
tteries_2019.pdf. [Accessed 7 Apr. 2021].

[6.20] Windpower Engineering & Development.

(n.d.). How three battery types work in
grid-scale energy storage systems. [online]
Available at:
https://www.windpowerengineering.com/how-thr
ee-battery-types-work-in-grid-scale-energy-stor
age-systems/. [Accessed 7 Apr. 2021].

[6.21] Grand View Research (n.d.). Lithium-ion

Battery Market Size & Share Report,
2020-2027. [online] Available at:
https://www.grandviewresearch.com/industry-an
alysis/lithium-ion-battery-market. [Accessed 7
Apr. 2021].

[6.22] Troiano, J. (2013). How do Lithium Ion

Batteries Work? A Nanotechnology Explainer.
[online] Sustainable Nano. Available at:
https://sustainable-nano.com/2013/10/15/how-d
o-lithium-ion-batteries-work/ [Accessed 7 Apr.
2021].

[6.23] Business Electricity Prices. (2017). Water

Deregulation Pricing Guide for Businesses.
[online] Available at:
https://www.businesselectricityprices.org.uk/wat
er-prices/. [Accessed 7 Apr. 2021].

[6.24] Bulb. Business. [online] Available at:

https://bulb.co.uk/business/ [Accessed 7 Apr.
2021].

[6.25] Bulb. Export Payments. [online] Available

at: https://bulb.co.uk/export/ [Accessed 7 Apr.
2021].

106
 Alexander Astor, Antoine Thibaut

Section 7 https://www.pinterest.com/pin/80959247050030
4870/. [Accessed 2nd of April 2021]
[7.1] Hydrogen Europe (n.d.). Hydrogen
Storage and Distribution. [online]. [Accessed [7.11] Oxford City Council. (2020). Seacourt
2nd of November 2021]. Park and Ride extension update – July 2020.
[Online]. Available at:
[7.2] College of the Desert (2001). Module 1: https://www.oxford.gov.uk/news/article/1250/sea
Hydrogen Properties. [pdf]. Available at: court_park_and_ride_extension_update_-_july_
https://www1.eere.energy.gov/hydrogenandfuel 2020. [Accessed 18th of March 2021]
cells/tech_validation/pdfs/fcm01r0.pdf
[Accessed 11th of April 2021] [7.12] Google. Google Earth. [Online]. Available
[7.3] Hurskainen, M., Ihonen, J. (2020). at: https://www.google.com/earth/. [Accessed
Techno-economic feasibility of road transport of 18th of March 2021]
hydrogen using liquid organic hydrogen
carriers. [online]. Available at: [7.13] Berks, Bucks & Oxon Wildlife Trust
https://doi.org/10.1016/j.ijhydene.2020.08.186 (BBOWT). Iffley Nature Reserve, Where
Thames Smooth Waters Glide. Available at:
[7.4] CF Industries (n.d.). 2018 Annual Report. https://thames.me.uk/s01627.htm. [Accessed
[pdf]. Available at: 12th of April 2021]
https://www.cfindustries.com/globalassets/cf-ind
ustries/media/documents/reports/annual-reports [7.14] Oxford Mail. (2017). ‘Cunning plan’ to
/cf-industries-2018-annual-report.pdf [Accessed repaint park and ride bays ‘could save £4m
11th of April 2021] Seacourt expansion’. [Online]. Available at:
https://www.oxfordmail.co.uk/news/15708913.c
[7.5] Papadias, D.D., Peng, J-K., Ahluwalia, unning-plan-repaint-park-ride-bays-could-save-
R.K. (2019). H2@Scale: Outlook of Hydrogen 4m-seacourt-expansion/. [Accessed 18th of
Carriers at Different Scales. [pdf]. Available at: March 2021]
https://www.energy.gov/sites/prod/files/2020/03/
f72/fcto-hydrogen-carriers-workshop-2019-anl.p [7.15] Interreg, North-West Europe, H2-Share,
df [Accessed 7th of April 2021]. European Regional Development Fund.
Hyundai: 1.600 H2 Xcient trucks in Switzerland.
[7.6] Bulb (n.d.). Our business tariff. [online]. Available at:
Available at: https://bulb.co.uk/business/tariff/ https://fuelcelltrucks.eu/project/hyundai-1-600-h
[Accessed 7th of April 2021]. 2-xcient-trucks-in-switzerland/. [Accessed 11th
of April 2021]
[7.7] RAC Drive (2020). Hydrogen Cars: Are
they the future? [online]. Available at:
[7.16] NextExit Logistics. (2021). Semi-Truck
https://www.rac.co.uk/drive/advice/buying-and-s
Aerodynamics: How it Works. [Online].
elling-guides/hydrogen-cars [Accessed 14th
Available at:
May 2021].
https://nextexitlogistics.com/semi-truck-aerodyn
[7.8] Office of Energy Efficiency & Renewable amics-works/. [Accessed 15th of May 2021]
Energy, Hydrogen and Fuel Cell Technologies
Office. Hydrogen Tube Trailers. [Online].
Available at:
https://www.energy.gov/eere/fuelcells/hydrogen-
tube-trailers?fbclid=IwAR28Tpu4RT5HZW_0pS
yxrz511PLEE-9NT3WOJAsln_Plv2vackvytWKN
gdE. [Accessed 11th of April 2021]

[7.9] Flaticon. (2021). Electrolyser Premium

icon. [Online]. Available at:
https://www.flaticon.com/premium-icon/electroly
ser_2017946. [Accessed 2nd of April 2021]

[7.10] Pinterest. Compressor Station. [Online].

Available at:

107
 Lingxi Tang

Section 8 [8.9] GOV.UK (2019). Ultra-low emission bus

scheme: successful bidders. [online] Available
[8.1] Brandon, N P, J S Edge, M Aunedi, E R at:
Barbour, P G Bruce, B K Chakrabarti, T Esterle, https://assets.publishing.service.gov.uk/govern
et al. ‘UK RESEARCH NEEDS IN GRID SCALE ment/uploads/system/uploads/attachment_data/
ENERGY STORAGE TECHNOLOGIES’, n.d., file/774207/ultra-low-emission-bus-scheme-win
92. ning-bidders.csv/preview [Accessed 12 Apr.
2021].
[8.2] HM Treasury (n.d.). THE GREEN BOOK
CENTRAL GOVERNMENT GUIDANCE ON [‌ 8.10] GOV.UK (2017).Low emission bus
APPRAISAL AND EVALUATION 2020. [online] . scheme 2015: successful bidders - GOV.UK.
Available at: [online] Available at:
https://assets.publishing.service.gov.uk/govern https://assets.publishing.service.gov.uk/govern
ment/uploads/system/uploads/attachment_data/ ment/uploads/system/uploads/attachment_data/
file/938046/The_Green_Book_2020.pdf. file/541695/low-emission-bus-scheme-winning-
[Accessed 11th of April 2021] bidders.csv/preview [Accessed 12 Apr. 2021].
[8.3] Department of Business, Energy and [8.11] Oxford City Council (2019). City Council
Industrial Strategy. (2020). BEIS 2019 FOSSIL secures further £700,000 to upgrade buses and
FUEL PRICE ASSUMPTIONS. [online] . cut air pollution. [online] www.oxford.gov.uk.
Available at: Available at:
https://assets.publishing.service.gov.uk/govern https://www.oxford.gov.uk/news/article/1039/city
ment/uploads/system/uploads/attachment_data/ _council_secures_further_700000_to_upgrade_
file/863717/beis-2019-fossil-fuel-price-assumpti buses_and_cut_air_pollution [Accessed 7 Apr.
ons.pdf. [Accessed 11th of April 2021] 2021].
[8.4] Government Digital Service (2012). Tax on
shopping and services. [online] GOV.UK.
Available at:
https://www.gov.uk/tax-on-shopping/fuel-duty.
[Accessed 14 Apr. 2021].

[8.5] Argus Media. (2020). UK targets 5GW

hydrogen production by 2030. [online] Available
at:
https://www.argusmedia.com/en/news/2161130-
uk-targets-5gw-hydrogen-production-by-2030
[Accessed 14 Apr. 2021].

[8.6] GOV.UK. (2021). Ending the sale of new

diesel buses. [online] Available at:
https://www.gov.uk/government/consultations/e
nding-the-sale-of-new-diesel-buses [Accessed
7 Apr. 2021].

[8.7] GOV.UK. (2021). Multi-million pound

scheme for zero-emission buses across
England launched. [online] Available at:
https://www.gov.uk/government/news/multi-milli
on-pound-scheme-for-zero-emission-buses-acr
oss-england-launched [Accessed 7 Apr. 2021].

[8.8] GOV.UK (2021). Prime Minister launches

3 billion bus revolution. [online] Available at:
https://www.gov.uk/government/news/prime-min
ister-launches-3-billion-bus-revolution
[Accessed 11th of April 2021]

108
 Antoine Thibaut

Section 9

[9.1] Go Green Taxis Ltd. Fleet. [Online].

Available at:
https://www.gogreentaxisltd.co.uk/our-fleet/.
[Accessed 6th of April 2021]

[9.2] 001 Taxis. (2021). Environmentally friendly

taxis – why we’re going greener. [Online].
Available at:
https://www.001taxis.com/about/environment.
[Accessed 6th of April 2021]

[9.3] Royal Cars. (2020). About Royal Cars

Oxford. [Online]. Available at:
https://royal-cars.com/royal-cars-oxford-taxi-co
mpany/. [Accessed 6th of April 2021]

[9.4] U.S. Energy Information Administration.

(2021). Hydrogen explained, Use of hydrogen.
[Online]. Available at:
https://www.eia.gov/energyexplained/hydrogen/
use-of-hydrogen.php. [Accessed 6th of April
2021]

[9.5] Kennington Village. (2021). Sandford

Hydro. [Online]. Available at:
https://kennington-pc.gov.uk/sandford-hydro/.
[Accessed 6th of April 2021]

[9.6] Gigastack. Renewable Hydrogen. [Online].

Available at:
https://gigastack.co.uk/about/renewable-hydrog
en/. [Accessed 6th of April 2021]

109
 Akshay Pal

Appendix A: Bus route data

Bus Route Name [1.15, Operator [1.15, 1.16, 1.17] Weekday Distance Weekend Distance Lowest mpg Rating [3.25]
1.16, 1.17] (miles) [1.40] (miles) [1.40]

6 - Magdalen Street to OBC 347.8 310.8 5.3

Clifford Place

10 - JR Hospital to Stagecoach 1150.2 1107.6 7

Speedwell Street

X2 - Oxford Railway Thames Travel 1540 1540 9.1

Station to Orchard Centre,
Didcot

3A - Westgate to Kassam OBC/Stagecoach 130 120 9

Stadium

This is an example of the data that was collected for bus routes in Oxfordshire. Overall 58 bus routes were analysed, 25 of which were operated by

OBC, 17 by Stagecoach, 8 by Thames Travel and 8 were jointly operated by OBC and Stagecoach. Both the weekday and weekend distance was

taken into account and the output of this analysis was a hydrogen requirement of 8548.2 kg on a weekday and 6838.6 kg on a weekend. As expected

the weekend demand is lower due to the reduced services on the weekend. The Python source code developed for the route modelling can be found

in the GitHub repository.

110
 Antoine Thibaut

Appendix B: Truck trailer distribution data

Data used:

Average truck mpg rating = 6 [7.16]

Fuel economy of a hydrogen vehicle = 8.7 kg per 100km (from Section 4.3 findings)

111
 Alexander Astor

Appendix C: 10 Year Transition Period Expenditure

Year 1 2 3 4 5 6 7 8 9 10
Business-As-Usual
Diesel fuel cost -12281801. -12281801 -1228180 -1228180 -12281801.7 -12281801. -12281801. -12281801.7 -12281801. -1228180
73 .73 1.73 1.73 3 73 73 3 73 1.73
Diesel fleet -3354293.8 -3354293. -3354293 -3354293. -3354293.88 -3354293.8 -3354293.8 -3354293.88 -3354293.8 -3354293
maintenance 8 88 .88 88 8 8 8 .88
Diesel fleet -11900000 -11900000 -1190000 -1190000 -11900000 -11900000 -11900000 -11900000 -11900000 -1190000
renewal 0 0 0
Total Cash Flow -27536095. -27536095 -2753609 -2753609 -27536095.6 -27536095. -27536095. -27536095.6 -27536095. -2753609
for BAU 61 .61 5.61 5.61 1 61 61 1 61 5.61

Green Hydrogen Public Mobility Project

Solar Farm -3857142.8 -3636734. -3428921. -3232982 -3048241.042 -2874055.8 -2709824.0 -2554976.98 -2408978.3 -2271322
Depreciation 57 694 283 .924 4 78 8 03 .4
Electrolyser -3024000 -2520000 -2100000 -1750000 -1458333.333 -1215277.7 -1012731.4 -843942.901 -703285.75 -586071.
Depreciation 78 81 2 1 4592
Battery -232.128 -185.7024 -148.5619 -118.8495 -95.0796288 -76.063703 -60.850962 -48.6807699 -38.944615 -116.064
Depreciation 2 36 04 43 5 96
Solar Farm O&M, -942598.13 -1885196. -2827794. -3770392 -4712990.697 -5655588.8 -6598186.9 -7540785.11 -8483383.2 -9402416
Water Costs, 93 279 418 .557 36 75 5 54 .44
Electricity Costs,
Solar Energy Sold
Diesel fuel cost -11050543. -9819285. -8588026. -7356768 -6125510.138 -4894251.8 -3662993.5 -2431735.18 -1200476.8 0
41 096 776 .457 19 61
Cracker -391068 -312854.4 -250283.5 -200226. -160181.4528 -128145.16 -102516.12 -82012.9038 -65610.323 -52488.2
Depreciation 2 816 22 98 3 07 5845

112
 Alexander Astor

Compressor -314420.90 -628841.8 -943262.7 -1257683 -1572104.503 -1886525.4 -2200946.3 -2515367.20 -2829788.1 -3136348
running cost 06 012 018 .602 04 04 5 06 .484
Cracker OPEX -293,518.7 -293,518. -293,518. -293,518. -293,518.75 -293,518.75 -293,518.75 -293,518.75 -293,518.75 -293,518.
5 75 75 75 75
Purchase of -13810105. -1381010 -1381010 -1381010 -13810105.26 -13810105. -13810105. -13810105.2 -13810105. -1346485
Hydrogen Fleet 26 5.26 5.26 5.26 26 26 6 26 2.63

Sale of Diesel 2484000 2484000 2484000 2484000 2484000 2484000 2484000 2779800 3180000 3100500
Fleet
Hydrogen fleet -846946.59 -1634606. -2366093. -3044373 -3672275.743 -4252495.3 -4787600.9 -5280039.92 -5732143.3 -6130766
Maintenance 49 928 528 .673 11 71 8 47 .149
Diesel fleet -2912392.9 -2497316. -2107726. -1742349 -1399967.189 -1079416.9 -779590.19 -499429.063 -237924.21 0
Maintenance 83 139 993 .875 13 61 9 29
Diesel Fleet -10332271. -8859707. -7477565. -6181319 -4966651.744 -3829438.2 -2765745.5 -1771820.26 -844081.71 0
Renewal 93 325 209 .899 43 39 2 58
Hydrogen Fleet 0 0 0 0 0 0 0 0 0 0
Renewal
Total Cash Flow -45802335. -4437384 -4305759 -4183574 -40693435.87 -39618312. -38600063. -37334285.6 -36005190. -3486134
for Green 68 0.56 2.19 0.48 87 63 3 55 1.52
Hydrogen
Difference -18266240. -1683774 -1552149 -1429964 -13157340.25 -12082217. -11063968. -9798190.02 -8469094.9 -7325245
between BAU 06 4.94 6.58 4.87 26 02 34 .902
and Green
Hydrogen

Discount Factor 0.965 0.931225 0.898632 0.867180 0.836828700 0.80753969 0.77927580 0.752001153 0.72568111 0.700282
125 0006 6 61 67 5 31 2742
NPV of -17626921. -1567972 -1394811 -1240036 -11010439.95 -9756870.0 -8621882.6 -7368250.19 -6145862.2 -5129739
dIfference 66 9.03 5.45 6.05 51 04 7 39 .859

113
 Lingxi Tang

Appendix D: 10 Year Transition Period Cash Flow Analysis

Year 1 2 3 4 5 6 7 8 9 10
Cash Flow
Difference in -1098639 -9946250. -8979515. -8079119 -7239129.6 -6454509 -5720937. -4812221.9 -3887322.7 -3128676.3
expenditure between 4.8 838 669 .848 04 .4 698 12 35 12
BAU and Green
Hydrogen (without solar
farm)
Profits 7753710 7753710 7753710 7753710 7753710 7753710 7753710 7753710 7753710 7753710
Annual Cash Flow -3232684. -2192540. -1225805. -325409. 514580.395 1299200. 2032772.3 2941488.08 3866387.26 4625033.68
(without loan) 802 838 669 8475 9 6 02 8 5 8
Cumulative Cash Flow -3232684. -5425225. -6651031. -697644 -6461860.7 -5162660 -3129887. -188399.77 3677987.49 8303021.18
(without loan) 802 64 309 1.157 61 .161 859 05 5 3
Cash Flow With Loan
Loan 5000000 0 0 0 0 0 0 0 0 0
Loan Repayment -728430 -728430 -728430 -728430 -728430 -728430 -728430 -728430 -728430 -728430
Annual Cash Flow (with 1038885. -2920970. -1954235. -105383 -213849.60 570770.5 1304342.3 2213058.08 3137957.26 3896603.68
Loan) 198 838 669 9.848 41 999 02 8 5 8
Cumulative Cash Flow 1038885. -1882085. -3836321. -489016 -5104010.7 -4533240 -3228897. -1015839.7 2122117.49 6018721.18
(with Loan) 198 64 309 1.157 61 .161 859 71 5 3
Cash Flow with Fare Increase
Profits after fare 15507420 15507420 15507420 1550742 15507420 1550742 15507420 15507420 15507420 15507420
increase 0 0
Annual Cash Flow (with 4521025. 5561169. 6527904. 7428300 8268290.39 9052910. 9786482.3 10695198.0 11620097.2 12378743.6
fare increase) 198 162 331 .152 6 6 02 9 7 9
Cumulative Cash Flow 4521025. 10082194 16610098 2403839 32306689.2 4135959 51146082. 61841280.2 73461377.4 85840121.1
(with fare increase) 198 .36 .69 8.84 4 9.84 14 3 9 8

114