FACULTY OF ENGINEERING

DEPARTMENT OF MECHANICAL ENGINEERING

A PROJECT SUBMITTED IN FULFILLMENT OF THE REQUIREMENTS FOR ME360

-PROJECT TITLE-

DESIGN AND MANUFACTURE OF AN OPTIMISED HYDROGEN BASED FUEL CELL TO POWER

A SMALL VEHICLE.

VICTOR MULOLO (R184708P)

BSc (HON) DEGREE MECHANICAL ENGINEERING PART III

SUPERVISOR ENG S. CHINGUWA

Acknowledgements

I would like to extend my gratitude to my supervisor Eng. S Chinguwa, for his guidance and

support. I cannot wish for a friendlier supervisor who understood me for being the person I

am. I would also like to thank my parents, family and friends for support and love throughout

the course of my studies.

Abstract

Fuel cells are favourable technologies in order to generate high energy efficiency and low

emissions electricity from hydrogen or other fuels. They are suitable for stationary, transport

and mobile use. A reaction occurs in a fuel cell in which hydrogen and oxygen re-combine and

thus release energy. There are two electrodes, anode and a cathode, and between them there

is a layer of electrolyte. Anode and cathode oxygen are supplied with hydrogen. The reaction

from the fuel cell produces electrons forming a current that flows through a peripheral circuit

from the anode to the cathode.

This paper not only establishes the model of a working hydrogen fuel cell stack, but also

defines the hydrogen production model, to supply the fuel cells with hydrogen. The amount

of hydrogen produced by the PV/T – PEM electrolyzers was calculated to average 4.12 kg, at

an operation temperature of 60 oC. The numerical model of the fuel cell stack that was

simulated by means of MATLAB/SIMULINK produced a power output of 50 kW with different

variations by change of parameters. The outcome has derived components that can be sized

to make a functional prototype as indicated by the drawing in SolidWorks.

Nomenclature

IT Inclined solar radiation (W/m2)

τ Transmissivity of glass
αg Absorptivity of glass
CP Water specific heat, j/kg/k
k Specific heat rate
F’ Collector efficiency factor
n Number of electron flow
δ Membrane thickness
s Membrane conductivity
r Material resistivity
i Current density
P Operating pressure
∆H Enthalpy change
∆S Entropy change
M Mass Flow Rate (kg/s)
Er Reversible voltage
Ex Exergy
∆G Gibbs free energy
P Power
R Resistance
V Voltage
η Efficiency
ρ Density (kg/m3)
A Area (m2)
Table of Figures
Figure 1: An image depicting air pollution in Zimbabwe (Dewa, 2016) .................................... 2

Figure 2 Schematic of an individual fuel cell (Bagotsky, 2012) ................................................. 5

Figure 3 Grove’s gas battery and Grove’s gas chain (Berry & Martyn, 2000) ........................... 7

Figure 4 Grove's six cell hydrogen gas battery (Liebhafsky, 1968) ............................................ 8

Figure 5 First diagram of a fuel cell with solid electrolytes (Mobius, 1997) ............................. 9

Figure 6 Basic principles of a molten carbonate fuel cell (Andujar & Segura, 2009) .............. 10

Figure 7. The first high pressure cell designed by Bacon in 1938 (Bacon, 1960) .................... 10

Figure 8 Fuel Cell Tractor Allis Chalmers (History, 2020)......................................................... 12

Figure 9 Historical Review of Fuel Cells up to their application in motor vehicles (Andujar &

Segura, 2009). .......................................................................................................................... 15

Figure 10 Principle of Alkaline Electrolysis (Sherif, et al., 2005).............................................. 15

Figure 11 Principle of solid polymer electrolysis (Barbir, 2005) .............................................. 16

Figure 12 Renewable pathways for hydrogen production (Turner, et al., 2008) .................... 17

Figure 13 the designs for the technological solutions regarding hydrogen production using

solar radiation (Badea, et al., 2017)......................................................................................... 18

Figure 14 A graphic outlook of a parabolic trough (Tributsch, 2008)...................................... 20

Figure 15 Schematic view of a parabolic saucer with a cavity-receiver at the focal point

(Tributsch, 2008). ..................................................................................................................... 20

Figure 16 Effects of insolation and temperature on the characteristics of a PV panel (Badea,

et al., 2017). ............................................................................................................................. 22

Figure 17 Central hydrogen production versus distributed versus distributed hydrogen

production through electrolysis (Turner, et al., 2008) ............................................................ 23

Figure 18 Basic Wind Machines (Mohamed & Noureddine, 2015) ......................................... 24

Figure 19 Wind Energy vs Wind Speed (Energy & Electricity, 2004) ....................................... 25

Figure 20 Principles of a monopolar design (Levene & Ramsden, 2007) ................................ 26

Figure 21 Principles of a bipolar design (Levene & Ramsden, 2007)....................................... 27

Figure 22 Illustration of carbon fibre-reinforced hydrogen reservoirs (Ang, et al., 2010) ...... 29

Figure 23 Comparison, by mass and volume, of metal hydrides, carbon nanotubes, oil and

other hydrocarbons, with the contained hydrogen. (Pukrushpan, et al., 2004) .................... 30

Figure 24 Schematic view of an acid electrolyte fuel cell (Bagotsky, 2012) ............................ 31

Figure 25 Diagram showing the basic functionality of a circulating electrolyte in an alkaline

fuel cell (Ang, et al., 2010) ....................................................................................................... 32

Figure 26 Graphic exemplification of proton exchange membrane fuel cell (Energy &

Electricity, 2004). ..................................................................................................................... 34

Figure 27 Schematic presentation of a single fuel cell (Bagotsky, 2012) ................................ 35

Figure 28 Simple series connection from the anode of one cell to the cathode of the next one

(Ernst, 2007) ............................................................................................................................. 36

Figure 29 Fuel cell stack consisting of three cells with bipolar plates (Larminie & Dicks, 2003).

.................................................................................................................................................. 37

Figure 30 Structure of a fuel cell with edge seals to prevent leakage of the gas at the edges of

the electrodes (Larminie & Dicks, 2003) .................................................................................. 38

Figure 31 Schematic diagram of the proposed system ........................................................... 42

Figure 32 Polarization curves for PEM fuel cell single cells (Song & Shin, 2001) .................... 50

Figure 33 the forces exerted by the clamped materials on the bolt and nut (Ishitani, 2001) 52

Figure 34 Disparity of incident solar radiation and ambient temperature. ............................ 63

Figure 35 I-V and P-V curves of the PV/T system: a) assessment of experimental numerical

data and simulation data; b) comparison of the efficiencies as an iterative of the ratio of (Tin -

Ta)/IT.......................................................................................................................................... 64

Figure 36 Variation of PEM over potentials. ............................................................................ 64

Figure 37 Comparison of electrolyzer system voltages with dissimilar water temperatures

contingent on current .............................................................................................................. 65

Figure 38 the disparity in the hourly electrolyzer voltage contingent on the current and water

temperature. ............................................................................................................................ 66

Figure 39 Simulated Numerical stack producing 50 kW .......................................................... 67

Figure 40. The fuel cell efficiency for the fuel stack ................................................................ 70

Figure 41 Recital of the fuel cell stack at changing operational conditions. ........................... 71
List of Tables

Table 1 Real world ventures of fuel cell vehicles and their specifications (Nan, et al., 2014) 13

Table 2 Specifications for the 2015 Hyundai and Toyota FCEV (Nan, et al., 2014) ................. 14

Table 3 Features of the photovoltaic thermal panel ............................................................... 43

Table 4 Simulation parameters for the PV/T used in the mathematical model ...................... 45

Table 5 Mock-up characteristics for the PEM electrolyzer used in the numeric model. ........ 48

Table 6 Physical Properties Used for Material Toughness and Compression Designs (Matweb,

2020) ........................................................................................................................................ 54

Table 7: Energy flow for each iteration simulated from the numerical model ....................... 66

Table 8 Simulated results to find the required parameters for the chosen case fuel cell of 50kW

.................................................................................................................................................. 67

Table 9 Measured Overall Simulation Data for Stability of the Fuel Cell Stack ....................... 69
Table of Contents

Table of Contents ........................................................................................................................ i

1 Introduction ........................................................................................................................ 2

1.1 Background.................................................................................................................. 2

1.2 Justification ................................................................................................................. 3

1.3 Problem Statement ..................................................................................................... 4

1.4 Aim .............................................................................................................................. 4

1.5 Objectives .................................................................................................................... 4

2 Literature Review ............................................................................................................... 5

2.1 What is a fuel cell ........................................................................................................ 5

2.2 History of Fuel Cells ..................................................................................................... 6

2.3 Fuel Cells in Vehicles ................................................................................................. 11

2.4 Hydrogen Production ................................................................................................ 14

2.4.1 Using solar energy to produce hydrogen .......................................................... 17

2.4.2 Using wind energy to produce hydrogen .......................................................... 22

2.4.3 Design of Electrolyzers ....................................................................................... 26

2.4.4 Hydrogen Storage methods ............................................................................... 28

2.5 Types of fuel Cells...................................................................................................... 31

2.5.1 Acid Electrolyte Fuel Cell.................................................................................... 31

2.5.2 Alkaline Electrolyte Fuel Cell .............................................................................. 32

i
 2.5.3 Proton Exchange Membrane (PEM) Fuel Cell .................................................... 33

2.6 Fuel Cell Stacks .......................................................................................................... 36

2.6.1 Simple Series Connection .................................................................................. 36

2.6.2 Bipolar Series Connection .................................................................................. 37

2.6.3 Water Management........................................................................................... 39

2.6.4 Air Management ................................................................................................ 39

2.6.5 Heat Management ............................................................................................. 40

3 Methodology .................................................................................................................... 41

4 System Configuration and Model ..................................................................................... 42

4.1 PV – Hydrogen Production Model............................................................................. 42

4.1.1 System Layout and Interconnection Design ...................................................... 42

4.1.2 Mathematical Model of the PV/T System ......................................................... 43

4.1.3 PEM Electrolyzer Mathematical Model ............................................................. 46

4.1.4 Economic Analysis .............................................................................................. 48

4.2 Fuel Cell Stack............................................................................................................ 49

4.2.1 Stack Sizing ......................................................................................................... 49

4.2.2 Number of Cells ................................................................................................. 50

4.2.3 Stack Configuration ............................................................................................ 51

4.2.4 Stack Clamping ................................................................................................... 51

5 SolidWorks CAD Design .................................................................................................... 55

ii
6 Results and Discussions .................................................................................................... 63

6.1 PV/T – PEM Hydrogen Production Scheme .............................................................. 63

6.2 Hydrogen Fuel Cell Stack ........................................................................................... 67

6.3 Bill of materials for Prototype ................................................................................... 72

7 Conclusion ........................................................................................................................ 74

8 References ........................................................................................................................ 75

9 Appendix 1 PV/T E production model MatLab load program .......................................... 85

10 Appendix 2 Hydrogen Fuel Cell Stack MatLab Load Program ...................................... 87

11 Appendix 3 List of Equations......................................................................................... 91

iii
1 Introduction

1.1 Background

According to the April 2019 Briefing report by The International Council on Clean

Transportation (Shikha & Zifei, 2019), governments with leading auto markets have declared

aggressive electrification goals – with many aiming at a 100% electric share in the 2020-2050

timeframe. This is mainly due to the advent effects of air pollution from tailpipe emissions.

Road transportation today is dominated by oil derivative fuels and internal combustion

locomotives, which makes it both unsustainable economically and environmentally. Oil price

unpredictability and stock disruptions pose a sombre threat to road transportation reliability

and affordability, the privation of which will upset most countries GDP.

Back home in Zimbabwe pollution from vehicle fumes is ranked as the second contributor to

the overall pollution in the country (Mapira, 2015), with air pollution being the dominating

contributor to respiratory diseases in the country.

Figure 1: An image depicting air pollution in Zimbabwe (Dewa, 2016)

2
The Reserve Bank of Zimbabwe governor, John Mangudya articulated to the Business Times

that the nation requires around US$100 million a month in fuel importation (Reuter, 2020) .

This is a cause for concern due to the economic situation within the country. Thus alternative,

sustainable fuel sources are of paramount importance to help sustain the country.

Addressing these issues will in the long run require the shifting away from oil-powered

energies and orthodox internal combustion engine locomotives to renewable fuels and high

efficiency powertrains. The foremost technological options being hydrogen fuel cell vehicles

and battery electric vehicles. Battery electric vehicles however pose a challenge due to the

fact that the country is facing electrical power shortages and high electricity tariffs as it is

importing most of its power (Dzirutwe, 2019). Hence fourth fuel cell vehicles would be a viable

solution, provided the fuel source is produced by renewable energy. Hydrogen can be

generated by solar photovoltaic cells using solar energy from the sun (Wilson, et al., 2013) ,

thus reducing the emission chain by up to 80%. It can also be produced from wind energy by

linking wind turbines to electrolysers which pass the wind generated electricity through water

to split the liquid into hydrogen and oxygen (Harrison, 2009).

1.2 Justification

Hydrogen fuel cells in comparison to batteries continuously produce electricity as long as the

source of fuel is being delivered. Batteries have limited ranges and require longer charging

times. In contrast the fuel cells can be refilled in minutes and have wider ranges of distance

covered (Nan, et al., 2017). Another major setback with batteries is that they lose energy

stored overtime and need periodic recharging even without use. Fuel cells do not burn fuel,

making the procedure noiseless and greenhouse gas free. Furthermore the fuel cell system

can be a zero emission system provided that the hydrogen is produced from non-polluting

3
sources. Hydrogen poses as a better carrier of energy than a battery because the energy

stored in hydrogen is much larger than the energy stored in a battery. For example 3 kg of

hydrogen is equivalent to the weight of 3 gallons of gasoline which is about 100 kWh (Burke,

2007). This is much more than the energy that can be stored in a battery for a passenger

vehicle.

1.3 Problem Statement

Solution to the energy and pollution crisis in Zimbabwe: how to achieve sustainable energy in

vehicles with zero emissions?

1.4 Aim

To design and manufacture an optimised hydrogen fuel cell that can be used to power small

vehicles.

1.5 Objectives

- To study previous models of hydrogen fuel cells and hydrogen production methods.

- Develop a mathematical model for hydrogen production to power the fuel cell using

the most suitable source of renewable energy available in Zimbabwe.

- Simulate hydrogen production model using MATLAB the various control strategies to

given specified inputs to produce 3 – 5kg of hydrogen.

- To develop a robust mathematical framework model of the hydrogen fuel cell relevant

to the small vehicle application.

- Simulate the fuel cell system in MATLAB to produce 50 kW.

- Draw design models in SolidWorks and present prototype specifications.

4
2 Literature Review

2.1 What is a fuel cell

A fuel cell is a reverse electrochemical device that continues to convert chemical energy

content of a fuel into electrical energy, water and some heat, as long as fuel and oxidant are

supplied (Revankar & Majumbar, 2014). It is different from a battery in the sense that it does

not have all the reactants initially stored, to produce electricity the fuel and oxidant are fed

continuously to the cell and the products of the reaction are removed simultaneously. Since

the intermediary stages of generating heat and mechanical work of typical most orthodox

electricity generation processes are circumvented, the thermodynamic restrictions of heat

motors like the Carnot efficiency do not limit fuel cells (EG & G Technical Services, 2004).

Furthermore, fuel cells yield power with trivial pollutants because combustion is avoided.

In the simplest case, two metallic electrodes dipping into an electrolyte solution make up a

fuel cell as displayed in figure 2 below.

Figure 2 Schematic of an individual fuel cell (Bagotsky, 2012)

5
The negative electrode, the anode, produces electrons in an operating fuel cell by burning a

fuel. In reducing an oxidizing agent, the positive electrode, the cathode, absorbs electrons.

Each is supplied to its electrode with fuel and oxidizing agent (Bagotsky, 2012). The condition

to prevent a direct mixture of the reactants or supply of the ‘wrong’ electrode is important at

this point. In these two unwanted cases, direct chemical interaction of the reactants would

begin and would produce thermal energy, stopping the production of electrical energy

completely.

2.2 History of Fuel Cells

The concept of the fuel cell has been in existence for nearly two centuries. Its foundation

being the discovery of electrolysis, a process first described by British Scientists William

Nicholson and Antony Carlisle in the 1800s (Britannica, 2020). It was later reported in 1832

by Michael Faraday that the amount of elements that can be separated by passing an

electrical current through a dissolved salt was proportional to the quantity of electric charge

passed through the circuit (Ortiz-Rivera, et al., 2007). His experiments derived the two

fundamental Laws of Electrolysis. The first law states that the mass of a substance produced

at an electrode during electrolysis is proportional to the number of moles of electrons

transferred at the electrode; the second law states that the number of Faradays of electric

charge required to discharge one mole of substance at an electrode is equal to the number

of elementary charges on that ion (Britannica, 2016).

It has often been argued that Sir William Robert Grove discovered the fuel cell in 1839, but it

is described by Bossel that Professor Christian Friedrich Schoenbein presented the first

experiment that showed the fuel cell effect – in the Philosophical Magazine published in

January 1839 (Bossel, 2000). The results of Schoenbein showed the voltage and current of the

6
hydrogen and oxygen chain, which he presented to an audience in France and Switzerland

late in the summer of 1838 (Sandstede, et al., 2003).

Sir William Robert Grove discovered the principle by accident during one of his electrolysis

experiments (Cook, 2003). Grove observed a current flow in the opposite direction of his

battery after disconnecting it from the electrolyser. The gases were being simultaneously

being consumed during this process. His gaseous chain was made up of cells. The cell was

made up of a glass vessel with three necks, with the longer tubes containing a strip of

platinum each. The cell was completely filled with dilute sulphuric acid. When the process of

electrolysis was being carried out, both the tubes were filled with hydrogen and oxygen. A

potential was then measured after an interruption of the wire reaching the battery which

stems from the oxygen/hydrogen chain (Sandstede, et al., 2003). Figure 3 shows the first cell

of Grove’s gaseous chain and his first working fuel cell from his first publication in 1842.

Figure 3 Grove’s gas battery and Grove’s gas chain (Berry & Martyn, 2000)

He continued his experiments with the gaseous voltaic battery, improving his apparatus and

publishing the results (Grove, 1845), like the battery shown in Figure 4.

7
 Figure 4 Grove's six cell hydrogen gas battery (Liebhafsky, 1968)

Fifteen years after Grove had described the fuel cell system, Vergnes claimed the same system

to be his invention, going further to patent it (Sandstede, et al., 2003). More developments

were undertaken by Ludwig Mond and his assistant Charles Langer in 1889. They described

the process by increasing the surface area in contact with the electrode, gas and the

electrolyte (Chaurasia, et al., 2003). They prevented the electrodes from flooding by

introducing a matrix that soaked the electrolyte, in addition to the application of porous

electrodes. The cells operated at 0.73 V on oxygen and hydrogen, with a current density of

6.5 Ma cm-2. Most of the phosphoric acid fuel cells of today are similar in principle to Mond

and Langer cells, with the exception of modern materials.

Friedrich Wilhelm Ostwald a professor from Leipzig – Germany, an expert in energy,

experimented with the interconnection of various fuel cell components in 1893 (Boudghen &

Traversa, 2002). During this time Grove’s experiments had been considered to be speculative,

especially the specifics of where the reactions took place in his gas battery. Professor Ostwald

solved the puzzle of the gas battery, his works formed the pioneering work for future research

in fuel cells. In 1896 entrepreneur Dr. William Jacques joined the rising buzz of fuel cell

technology. He intended to use fuel cells to power ocean ships at a very large scale

(Sandstede, et al., 2003). His fuel cells had a carbon rod central anode in the electrolyte of

potassium hydroxide molten at 400-500 C – with circular iron containers being used to contain

8
the molten electrolyte (Grimes, 2001). Jacques’ fuel cells produced power to a kilowatt scale,

they were large and highly performing. It was suggested in 1898 by W. Nernst that redox fuel

cell systems could have more flexibility and more choices of reactants (Appleby, 1990). He

used zirconium as a solid electrolyte. However his experiments did not attract much attention

until the 1950’s when Professor Posner made Nernst redox systems using tin bromide

solutions. Zirconia ceramics were later used in 1937 by Baur and Preis. This mixture was not

satisfactory either because current enhanced resistance considerably by electrolytic shift

(Mobius, 1997).

In 1921 Emil Baur built a molten carbonite fuel cell (Stone & Morrison, 2002). He had taken

the suggestions of Ostwald seriously with his excitement towards fuel cells. However his fuel

cell had problems with electrical conductivity and unwanted chemical reactions. Baur worked

closely with a partner named Treadwell. Together they concluded that the most prospective

cell type was one with an aqueous alkaline electrolyte, a carbon electrode and a metal anode

(Sandstede, et al., 2003) Baur’s concepts were further developed by H. J. Broers and J. A. A.

Ketelaar, seeing the limitations of solid oxides they focused their work to using molten

carbonate salts (Andujar & Segura, 2009). Figure 5 shows the basic principles of a molten

carbonate fuel cell, and figure 6 displays the basic principles of operation of this fuel cell type.

Figure 5 First diagram of a fuel cell with solid electrolytes (Mobius, 1997)

9
 Figure 6 Basic principles of a molten carbonate fuel cell (Andujar & Segura, 2009)

It was after this that he started reflecting about such systems, in 1937 he proposed an

apparatus which could be used alternatively as a fuel cell and an electrolyser which the board

of the company he worked for at the time declined, but a year later Bacon had high pressure

cylindrical cell produced in the workshop (Bacon, 1960), as shown in Figure 7.

Figure 7. The first high pressure cell designed by Bacon in 1938 (Bacon, 1960)

10
The cell was similar to an accumulator, withstanding a pressure of up to 210 bar and a

temperature of up to 200 C. It operated as an electrolyser until 160 bars of pressure were

reached. The fuel cell was also reversible, producing a remarkable performance.

Bacon worked also worked on developing fuel cells that could be used to power the Royal

Navy submarines during the Second World War, demonstrating an alkali cell to the Britain’s

National Research Development Corporation in 1958 (Ortiz-Rivera, et al., 2007). Bacon’s fuel

cells were quite expensive but they were reliable and efficient enough to attract commercial

attention. This led to a successful demonstrations of a 5 kW fuel cell in 1959, which

orchestrated a market for fuel cells in space missions as Bacon partnered with the Pratt and

Whitney company to provide fuel cells for NASA space shuttles (Verspagen, 2007).

Teflon began to be available in 1950, it significantly changed the way in which fuel cells were

made as it helped in developing aqueous electrolyte fuel cells in their current form. Thomas

Grubb worked with the General Electric Company to modify the original design of the fuel cell

using Teflon (Andujar & Segura, 2009). They discovered a way to deposit platinum onto the

membrane so it could work as a catalyst for oxidation reactions of hydrogen and oxygen

reduction.

2.3 Fuel Cells in Vehicles

The production of fuel cell powered vehicles began in 1959 with a team led by Harry Ihrig

(History, 2020). The tractor was powered by a fuel cell that produced 15 KW of power as

shown in the image on figure 8. The fuel cell was stacked up with 1008 cells of 1 V per cell and

potassium hydroxide as the electrolyte. It used a mixture of gases as fuel, namely compressed

hydrogen, propane and oxygen acting as the oxidizing agent.

11
 Figure 8 Fuel Cell Tractor Allis Chalmers (History, 2020)

Seven years later in 1966 General Motors produced the GMC Electrovan (Nan, et al., 2014).

The vehicle was developed over a span of two year with the team being led by Dr. C. Marks.

It was made up of 32 fuel cell components with an uninterrupted power output of 32 KW and

a peak power of 160 KW. The vehicle used pure liquefied hydrogen as a fuel, together with

liquid oxygen. It managed a top speed of 70 MPH and had a range of 120 miles.

Nearly three decades later, Daimler Benz introduced NECAR 1 in 1994 (Ernst, 2007). The

vehicle utilized compressed hydrogen and oxygen. It managed a top speed of 56 MPH and a

range of 81 miles per hour. The vehicle was later modified with the second model boasting of

a fuel cell system that was one third the weight of its earlier version. With the years that

followed, major companies began to bring fuel cell projects to their manufacturing plans. In

2002 Toyota launched the first fuel cell hybrid vehicle in the Japan and the United states

(Kawai, 2004). The vehicle consisted of a nickel-metal hybrid battery and a 90 KW fuel cell.

These power sources supplied power to the propulsion system of the vehicle together when

high performance was required. This resulted in a combined fuel cell range of 155 miles.

12
Major companies began to develop their own vehicles during the beginning of the twenty first

century with real world demonstrations as illustrated in table 1, with strongest advocacy for

fuel cell vehicles coming during the Bush Administration in 2003 when it was announced $1.2

billion worth of funding was going to be issued to the United States of America Department

of Energy for the development of a controlled hydrogen fleet, infrastructure validation and

demonstration projects (Archives, 2003). Five test regions would have to test small fleets of

Fuel Cell Electric Vehicles, build fuelling stations and demonstrate the functionality of the

vehicles.

In 2014 Toyota and Hyundai pronounced that their first ever commercially obtainable fuel cell

vehicles would be available for sale in 2015 (Nan, et al., 2014). The specifications of the

vehicles have been illustrated in table 2. Both the vehicles exceeded the travelling range of

the best battery electric vehicles that were available on the market at that time. The vehicles

even had a better fuelling time in comparison to gasoline vehicles.

Table 1 Real world ventures of fuel cell vehicles and their specifications (Nan, et al., 2014)
Year Auto Model Engine FC Range Max H2 Note
Maker Power (miles) Speed Pressure
(KW) (mph) (Bar)

2002 Toyota FCHV Fuel cell/ 90 180 96 350 18 leased in California and
Battery Japan
Hybrid
2002 Daimler A-Class Fuel cell/ 85 90 87 700 60 vehicles in US, Japan,
F-Cell Battery Singapore and Europe
Hybrid started in 2003
2002 Ford Advanced Fuel cell/ 85 180 n/a 350 30 fleet vehicles in
Focus Battery Sacramento, Orlando, and
FCV Hybrid Detroit
2002 GM Advanced Fuel 94 170 100 700 6 placed in Washington DC
Hydro Cell
Gen 3
2002 Nissan X-Trail Fuel cell/ 75 n/a 78 350 3 leased to Japanese
Battery government
Hybrid
2004 Hyundai Tucson Fuel cell/ 80 185 93 350 Demonstration project in
Battery the US between 2004-
Hybrid

13
 2009 and in Korea
between 2006-2010
2004 Kia Sportage Fuel 80 185 93 n/a Demonstration project in
Cell the US between 2004-
2009 and in Korea
between 2006-2010
2006 GM Equinox Fuel cell/ 93 200 100 n/a Leasing started in 2007.
FCEV Battery 100 vehicles in California,
Hybrid New York, and
Washington DC
2007 Honda FCX Fuel 100 354 100 350 Small scale production of
Clarity Cell 200 vehicles between
2008-2010, Leasing in
Southern California and
Japan
2008 Kia Borrego/ Fuel 115 426 93 700 Leasing to Seoul, Korean
Mojave Cell/ residents starting in 2009
FCEV Super
Capacitor
2008 Toyota FCHV-adv Fuel cell/ n/a 97 n/a n/a Limited leasing in Japan
Battery started in 2008. More than
Hybrid 100 leased in Connecticut,
California and New York
2009 Daimler Mercedes Fuel 90 239 105 n/a Small series production
Benz Cell started in 2009. 70
B-Class deployed in Los Angeles
F-CELL and San Francisco by 2012
2011 Hyundai Tucson Fuel cell/ 100 403 n/a 700 Tested 50 vehicles in 2011
IX Battery
Hybrid
2012 Hyundai Ix35 Fuel cell/ 100 365 100 n/a Leasing in Sweden and
Battery Denmark started in 2012
Hybrid

Table 2 Specifications for the 2015 Hyundai and Toyota FCEV (Nan, et al., 2014)

Maker Model Fuel Cell Range Fuelling Price Note

Power (KW) (miles) Time
Hyundai Tucson Fuel 100 265 <10 min Lease at Price Includes
Cell $499/month, fuel and
$2999 due at maintenance
signing
Toyota Mirai unrevealed 435 3 min $68600 in n/a
Japan

From the discovery of fuel cells the ultimate goal has been so that they can contribute to the

energy systems of the world, especially in transportation. Figure 9 illustrates this narrative,

also showing how other fields have benefited from fuel cells.

14
Figure 9 Historical Review of Fuel Cells up to their application in motor vehicles (Andujar &
Segura, 2009).

15
2.4 Hydrogen Production

Globally renewable energies share in power generation is 13.2% of the power generated in

the world since 2010, with the majority still coming from fossil fuels such as oil (32.4%), coal

(27.3%), natural gas (21.4%) and nuclear energy (5.7%) (Mohamed & Noureddine, 2015). Car

manufacturing companies are continuing research and development on efforts on vehicles

powered with hydrogen fuel cell and sustainable renewable energy for production of the

hydrogen.

Hydrogen production can be accomplished by various practical means. For example, (1)

catalytic cracking of natural gas, however this method requires a primary source of energy,

which is not considered as renewable; (2) catalytic cracking of biogas generated domestically,

this requires a domestic biogas production plant consisting of a biogas digester, biogas

storage and a biogas catalytic cracking system; (3) electrolysis of water by using electric

energy from the grid, however electricity from the grid is usually produced by non-renewable

energy sources and quite expensive especially in developing countries like Zimbabwe; (4)

electrolysis of water by using electric energy generated by a renewable energy source, this is

the most sustainable option of the four mentioned above (Bilgen, 2004)

The electrolysis of water is an electrochemical process that requires a direct electric current.

Thus it is one of the simplest ways to yield hydrogen. At 100% efficiency, 38.7 kWh of

electricity and 8.85 litres of water are required to yield 1 kg of hydrogen at 25oC and 1

atmosphere pressure (Levene & Ramsden, 2007), with typical electrolyser system efficiencies

ranging from 57 – 74%, which corresponds to 70.2 – 53.2 kWh kg-1. They are currently two

basic types of electrolysers in the market. These are low temperature alkaline electrolysers

and low-temperature polymer electrolyte membrane electrolysers (Sandstede, et al., 2003).

14
The alkaline electrolyser typically uses an aqueous solution of water and potassium hydroxide

as an electrolyte, sodium hydroxide and sodium chloride can also be used. This liquefied

electrolyte enables the conduction of ions between electrodes and requires replenishing

periodically. The typical reactions are shown below with figure 10 illustrating the process

(Cook, 2003):

4𝑂𝐻 −1 (𝑎𝑞) → 𝑂2(𝑔) + 2𝐻2 𝑂(𝑙) (1)

2𝐻2 𝑂(𝑙) + 2𝑒 − → 𝐻2(𝑔) + 2𝑂𝐻 − (2)

Figure 10 Principle of Alkaline Electrolysis (Sherif, et al., 2005)

The polymer membrane electrolyser incorporates a solid proton-conducting membrane that

is not electrically conductive. The membrane serves as a gas separation device and as an ion

conductor. Deionized water is introduced at the anode of the cells and a potential is applied

is applied across the cells to dissociate the water (Levene & Ramsden, 2007). The protons are

pulled through the membrane under the influence of an electric field and re-join with

electrons being supplied by the power source at the cathode to form hydrogen gas. PEM

electrolysers operate at higher current densities with stack efficiency decreasing as the

15
current density increases. The proton exchange membrane's upper temperature limit is

around 125-150 °C, but most electrolysers operate at a temperature of 80 oC in the liquid

water regime. With new compound membrane materials or perhaps ceramic proton-

conducting materials, higher operative temperatures may be achieved to get higher kinetic

reactions (Sherif, et al., 2005). However additional heat is generated by higher temperatures.

Water must be provided and circulated at a comfortable rate over the anode aspect to get rid

of the waste heat. Their advantage over base-forming electrolysers embrace the flexibility to

take care of a big differential pressure across the anode and cathode; the turning away of

high pressure oxygen. The chemical reactions that transpire in this electrolyser are shown

below, figure 11 illustrating the process (Bossel, 2000).

2𝐻2 𝑂 → 4𝐻 + + 4𝑒 − + 𝑂2 (3)

4𝐻 + + 4𝑒 − → 2𝐻2 (4)

Figure 11 Principle of solid polymer electrolysis (Barbir, 2005)

16
 2.4.1 Using solar energy to produce hydrogen

Figure 12 Renewable pathways for hydrogen production (Turner, et al., 2008)

The production of hydrogen using solar energy can obtained by four technological solutions.

These are: (1) the system of producing hydrogen by water electrolysis at an ambient

temperature and electricity obtained using concentrated photovoltaic systems; (2) the

system of producing hydrogen by the electrolysis of vapour obtained using a part of the solar

radiation spectre and electricity obtained using concentrated photovoltaic systems; (3) the

system of producing hydrogen by the electrolysis of vapour obtained using concentrated

thermal solar systems and electricity obtained using hybrid systems; (4) the system of

producing hydrogen by water electrolysis at ambient temperature and electricity obtained

using crystalline photovoltaic panels (Badea, et al., 2017). Figure 13 gives a graphic description

of these processes.

17
Figure 13 the designs for the technological solutions regarding hydrogen production using
solar radiation (Badea, et al., 2017)

To select a technical solution for the production and storage of hydrogen by means of solar

energy, the following steps must be taken: the assessment of the efficiency of the conversion

of solar radiation into hydrogen, the assessment of the capacity of the sub-systems, the

assessment of the necessary land area, the selection of the technical solution (Tributsch,

2008). The efficiency of photovoltaic system with concentrated radiation is calculated by

multiplying the efficiency of its components, using the following formula (Badea, et al., 2017):

𝜂𝑃𝑉 = 𝜂𝑝 × 𝜂𝑠𝑢 × 𝜂𝑠𝑒 (5)

Where 𝜂𝑃𝑉 is the efficiency of the photovoltaic system, 𝜂𝑝 is the efficiency of the

photovoltaic unit, 𝜂𝑠𝑢 is the efficiency of sun position tracking system and 𝜂𝑠𝑒 represent the

efficiency of DC electrical system for the series and parallel connection of the panels. The

18
efficiency of the photovoltaic unit is in turn calculated using the following formula (Barbir,

2005):

𝜂𝑃𝑉 = 𝛱𝜂𝑝𝑖 (6)

Where 𝜂𝑝 represents the efficiency of the photovoltaic unit component. An assessment of

the global efficiency of solar radiation conversion into chemical energy in the form of

hydrogen would show how the subsystem performs in relation to the other units of the

system. This can be calculated using the following formula (Badea, et al., 2017):

𝜂𝐺 = 𝜂𝑃𝑉+𝑇 × 𝜂𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑙𝑦𝑧𝑒𝑟 × 𝜂𝐶 (7)

Where 𝜂𝑃𝑉+𝑇 represents the efficiency of the alteration sub-system of solar radiation into

useful energy, 𝜂𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑙𝑦𝑧𝑒𝑟 is the efficiency of the electrolyser and 𝜂𝐶 is the efficiency of the

hydrogen compressor.

Solar energy is provided by solar radiation and a solar concentration system will optically

improve it. This device is either a parabolic trough or a dish that is parabolic. The geometric

layout of the hollow membrane module is extremely compatible with the tubular receivers of

parabolic troughs, rendering the integration of the two systems a natural design choice. A

parabolic dish can reach higher temperatures, but normally provides less energy (Peharz, et

al., 2007). In modules supported by simple pylons from the field, parabolic trenches are

incorporated. With reflective materials, they are made of parabolic-shaped bending boards.

When the parabola is pointed towards the light, parallel rays appear on the reflector and

reflect on a linear receiver tube positioned along the focal line (Gargallo, 2012).

The tube acts as an external surface receiver and its size is influenced by the magnitude of

the reflected sun image and manufacturing tolerance levels of the trough. The receiver

surface is usually coated with a selective coating that has a high absorption of solar radiation

19
but a low emittance of thermal radiation loss. A glass cover tube is positioned around the

receptor to reduce or remove the convective heat loss from the recipient. The glass tube is

held in a vacuum when the temperature is high (Valenzuela , et al., 2012). In general it is piped

to a Steam Generator and Super Heat Generator, where it produces steam to drive a turbine.

Heat transfer fluid that is circulating in the receiver and absorbs solar radiation. The

transmission fluid typically consists of diathermic oil (T<400°C) or a combination of the salts

of nitrates (T<600°C). A diagram view of a parabolic trough is presented in figure 14.

Figure 14 A graphic outlook of a parabolic trough (Tributsch, 2008)

The parabolic dish is usually installed in dish systems which would be constituting several

hundred of parabolic dishes. Typically, each dish has a cavity receiver with a Stirling engine

fixed at the focal point to generate electricity (Tributsch, 2008)(Figure 15).

Figure 15 Schematic view of a parabolic saucer with a cavity-receiver at the focal point
(Tributsch, 2008).
20
A major benefit of parabola trough compared to a parabola dish is that heat emitting at 800

oC is relatively low since all reflected solar irradiation is at the centre (Pitz-Paal, 2007). A 5.5

m diameter parabolic dish has a capacity output at the irradiance of 4.8 kW in design and an

output of 26.1 kW is possible on a dish with a diameter of 10 m.

Another method of producing hydrogen using solar energy is the use of photovoltaic panels.

This method is based on the use of electricity generated by PV panels to produce hydrogen

through water electrolysis. Electrolysis is conducted to generate hydrogen and oxygen by

passing direct electric current (DC) through the water (Nault, 2005). One benefit of PV-

electrolyser technology is that greenhouse gases are not emitted. The efficiencies of modern

photovoltaic systems and electrolysers are approximately 20% and 80% respectively, and the

overall effectiveness of the hydrogen transformation of solar radiant energy into energy is

nearly 16%. The power output from the PV array can be calculated using the following

equation (Nelson, et al., 2006):

𝑃𝑃𝑉 (𝑡) = 𝐼𝑛𝑠(𝑡)𝐴 𝐸∬𝑝𝑣 (8)

Where 𝐼𝑛𝑠(𝑡) is the insolation data at time t (kW/m2), A is the area of a single PV panel (m2),

and 𝐸∬𝑝𝑣 is the overall efficiency of the PV panels and the DC/DC converter. This equation

assumes the PV array has a tracking system and a maximum power point tracker.

Insolation is strongly dependent on the location and local weather. Measurements of solar

radiation are made using either a pyranometer or pyrheliometer (Meier, 2010). The features

of a 60-W, commercial PV panel with different irradiance levels (kW/m2) are shown in Figure

16. The irradiance has a very significant impact on the short-circuit current (horizontal part of

the I - V curves), while it has a rather weak effect on the open-circuit voltage (vertical arm of

21
the curves). A photovoltaic cell's maximum power (Pmax) output changes with irradiance, i.e.,

when the irradiance is higher, the cell generates more power. The temperature, as shown in

16, is another factor that affects the PV power. Both voltage and power decrease as the

temperature increases. This is an especially serious issue, because the cells are often operated

at the maximum power point.

Figure 16 Effects of insolation and temperature on the characteristics of a PV panel (Badea,

et al., 2017).

In general the system components of the PV – hydrogen concept comprise of:

- The PV panels

- The electrolyser and the compressor

- The hydrogen storage system

- The control system

2.4.2 Using wind energy to produce hydrogen

Within the last decade, wind power has become an essential source of environmental-friendly

energy. Among the benefits of using wind energy are that wind is free, clean and also an

endless source of energy. Utilizing wind energy eliminates emissions of carbon dioxide,

sulphur dioxide and other damaging wastes as in traditional carbon fuel power plants and

22
radioactive nuclear power plants. The World Wind Energy association found that in 2011 wind

power capacitated 196630 MW, with the top countries utilizing wind power being China and

the United States of America (Gong & Jing, 2012).

Wind power generation of hydrogen is considered the lowest GHG cycle of all means of

production of hydrogen, with wind energy having the lowest electricity costs under the

renewable energy option (Olateju, et al., 2014). In an energy system wind power can be stored

and transmitted via the grid to the hydrogen generation point; wind power can also be used

in the co-production of electricity from hydrogen and the grid at the wind site (Turner, et al.,

2008). Figure 17 provides an example of both of these cases.

Figure 17 Central hydrogen production versus distributed versus distributed hydrogen

production through electrolysis (Turner, et al., 2008)

A wind power conversion system is intended to extract and convert energy from the wind

into electrical energy. Typically, this conversion happens in steps. The wind turbine, made up

23
of components and controls that convert the wind's kinetic energy into useful mechanical

energy, is the first step. The turbine generally consists of an aerodynamic blade system, a

transmission of mechanical power, and various controls (Korpus & Grreniers, 2008). You can

then use the rotary motion (mechanical energy) to drive a generator to generate electrical

energy. Figure 18 shows two basic configurations of wind turbines. The horizontal type whose

blades are used to rotate around a horizontal axis has two to four blades for electrical power

generation. While the vertical axis type is less commonly used, with the blades rotating

around a vertical axis. The type of horizontal axis machine is well known for use in farms, but

its rotation speed is generally very low, making it less practical for generating electrical power

(Turner, et al., 2004). Knowing the principles behind wind energy conversion and

understanding the wind resource is critical in evaluating the potential for wind energy.

Figure 18 Basic Wind Machines (Mohamed & Noureddine, 2015)

24
Power available from the wind can be calculated based on the energy that can be extracted

as the air passes through the swept area of the wind machine blades. Data can be measured

and converted to wind energy for average wind speed every hour. The performance of power

can be determined using the following formula if the speed is between the cut in and the

rated speed of the wind turbine (Nelson, et al., 2006):

1
𝑃𝑃𝑉 (𝑡) = 2 𝜌𝐴𝑣(𝑡)3 𝐶𝑝 𝐸∬𝑎𝑑
(9)

Where 𝜌 is the density of air (kg/m3), A is the swept area of the rotor (m2), v is the wind speed

(m/s), 𝐶𝑝 is the efficiency of the wind turbine, and 𝐸∬𝑎𝑑 is the efficiency of the AC/DC

converter. It should be noted that ordinary wind machines extract a still smaller fraction of

the energy in the wind. Actual efficiencies in the range of 30 percent to 40 percent (Energy &

Electricity, 2004) can be obtained depending on size and design. Figure 3 shows how wind

energy varies with both wind velocity and machine efficiency per unit area of the turbine; this

is the basis for wind generator modelling. Note that it is obvious in the plot that the effect of

wind power is relative to the velocity of the third power.

Figure 19 Wind Energy vs Wind Speed (Energy & Electricity, 2004)

25
 2.4.3 Design of Electrolyzers

The electrolysers consist of several electrolysis cells quantities. The configuration for

monopolar and bipolar electrolyzers is two. Each electrode, as shown in Figure 20, is

connected and powered by a separate stream and has a single polarity, so it can be either a

cathode or an anode. Connected the cells in parallel, the voltage across the entire tank is the

same as the voltage across every individual cell, regardless of the number of electrodes (Zhou

& Francois, 2009). The voltage is normally 1 to 2 volts. This configuration is simple to manage

and stable. It also contains cheap components; but, because of the decrease in electricity

potential, it absorbs more energy.

Figure 20 Principles of a monopolar design (Levene & Ramsden, 2007)

Each electrode has a cathode and a side anode for the bipolar configuration as shown in figure

21. In a series, each cell is linked and its current is fed to the end plate conductors. The voltage

is equivalent to the quantity of cells multiplied by the voltage of a singular cell throughout the

whole stack (Dufo Lopez & Bernal-Agustin, 2005). This design is extra compact, additionally

efficient, operates at higher current densities and can work at higher pressures and higher

temperatures, but it is more complicated to design. Additionally, in order to use the

alternative current, each type of electrolyzers requires either rectifiers or AC/DC converters

26
(Korpus & Grreniers, 2008). In a sequence of series, the tank electrolyzers raise the overall

voltage and in parallel and in series, the filter-press electrolyzers bind the output rate and the

optimum voltage to reduce inefficiency.

Figure 21 Principles of a bipolar design (Levene & Ramsden, 2007)

The efficiency of an electrolyzer is defined as energy which can ideally be recovered by the

reoxidation of hydrogen and oxygen by the co-products in water (hydrogen heating value). In

water electrolysis, the energy consumption is electrical energy and the following formula can

determine this narrative (Revankar & Majumbar, 2014).

𝐻𝑒𝑎𝑡𝑖𝑛𝑔 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 ℎ𝑦𝑑𝑟𝑜𝑔𝑒𝑛 𝑜𝑢𝑡𝑝𝑢𝑡

𝐸𝑛𝑒𝑟𝑔𝑦 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 𝑜𝑓 𝑤𝑎𝑡𝑒𝑟 𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑙𝑦𝑠𝑖𝑠 = (10)
𝑇𝑜𝑡𝑎𝑙 𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙 𝑒𝑛𝑒𝑟𝑔𝑦 𝑖𝑛𝑝𝑢𝑡

The energy consumption of an electrolysis cell is directly proportional to the cell's voltage and

inversely proportional to its actual efficiency. The proportion of electrons theoretically

needed to generate a certain hydrogen (two electrons per formed hydrogen molecule) to the

actual electron supply from the electrical current for that hydrogen is the current efficiency

calculable using the following formula (Osaka, et al., 2005):

27
 𝑁𝑚3 𝐶 𝑚𝑜𝑙𝑒−
𝑉𝑜𝑙𝑢𝑚𝑒 𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑒𝑑 𝐻2 ( )×96487( )×2( )
ℎ 𝑚𝑜𝑙𝑒− 𝑚𝑜𝑙𝐻2
𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = 𝑠 𝑁𝑚3
(11)
𝐶𝑢𝑟𝑟𝑒𝑛𝑡(𝐴)×3600( )×0.0224( )
ℎ 𝑚𝑜𝑙𝐻2

Under real conditions, water electrolysis requires a cell voltage E that is substantially greater

than the reversible value of Erev. Electrical resistance in the electrodes, in the electrolyte

between the electrodes, and in the separator must be overcome. In addition, the excess

potentials at the electrodes, which are kinetically determined, must be covered. The voltage

of the cell then can be expressed as (Harnada, et al., 2006):

𝐸 = 𝐸𝑟𝑒𝑣 + 𝜂𝑐 + 𝜂𝑎 + 𝑗𝑅 ∗ (12)

Where 𝜂𝑐 represents the cathodic over potential, 𝜂𝑎 is the anodic over potential, j also

represents the current density and R* is the sum of the electrical resistance.

2.4.4 Hydrogen Storage methods

Hydrogen needs to be stored in order to surpass the daily and periodic differences between

energy availability and demand. It is possible to store hydrogen in gas, liquid, and also as a

hydride form. Hydrogen, with very little compound density, is very light (Kawai, 2004). This

means a large volume is occupied by a small quantity. The volumetric density of hydrogen

must therefore be increased.

Classical steel high-pressure tanks are filled up to 200.00 bar and tested up to 300.00 bar,

whereas tanks made of reinforced carbon fibre materials, shown in figure 22, are capable of

storing hydrogen up to a pressure of 600 bar (Nan, et al., 2014). Now, for common use, the

objective of the automotive industry is to store gaseous hydrogen at a pressure of 700 bar.

This necessitates innovative composites of materials that can withstand high pressure and do

28
not permit materials to become brittle (Revankar & Majumbar, 2014). The gravimetric density

increases with the higher pressure, but additional pressure control would be necessary during

a drop in pressure and the compression itself is a very complicated, hazardous and costly

process.

Figure 22 Illustration of carbon fibre-reinforced hydrogen reservoirs (Ang, et al., 2010)

With some metals and alloys, hydrogen can form metal hydrides. In the lattice of particular

metals or alloys, hydrogen in a molecular form is absorbed. In this way, the volumetric density

is comparable to the liquid form, but the gravimetric density is rather low considering the

mass of the metal. The highest achievable density is currently within the range of 0.07 kg

H2/kg of metal at high temperatures (Ang, et al., 2010). The safest way to store energy is

hydrogen in a solid form, particularly for stationary and mobile applications. The tank weight,

high desorption temperature, low desorption kinetics, long recharge time, high pressure and

heat transfer are the main problematic issues for solid storage.

There are several new hydrogen storage methods, such as on the activated carbon surface,

in micro spheres of glass or poly-hydride complexes. These technologies could improve the

hydrogen storage gravimetric and volumetric density, but they are still being developed and

29
have not been implemented in industry (Hawkes & Leach, 2005). The comparison of different

hydrogen storage materials is presented in figure 23.

Figure 23 Comparison, by mass and volume, of metal hydrides, carbon nanotubes, oil and
other hydrocarbons, with the contained hydrogen. (Pukrushpan, et al., 2004)

30
2.5 Types of fuel Cells

2.5.1 Acid Electrolyte Fuel Cell

The utmost found type of fuel cell is the acid electrolyte which is shown in figure 24. The

anodic reaction is shown below (EG & G Technical Services, 2004):

2𝐻2 → 4𝐻 + + 4𝑒 − (13)

The hydrogen gas is ionized in the ions and electrons of hydrogen as seen in equation 13. The

electrons are released by an outside load and generate a current from the reaction to the

cathode. The H+ ions pass the electrolyte, since acid is a mobile H+ ion fluid. The H+ interacts

in equation 14 with cathodic oxygen and electrons forming water (EG & G Technical Services,

2004).

𝑂2 + 4𝑒 − + 4𝐻 + → 2𝐻2 𝑂 (14)

Figure 24 Schematic view of an acid electrolyte fuel cell (Bagotsky, 2012)

The electrons are not permitted to pass through the electrolyte because no current in the

external circuit would then flow. In order to achieve greater cell performance, the operating

temperatures and acid concentrations of acid electrolyte fuel cells have increased;

temperatures of about 200 °C (392 °F) and acid concentrations of 100 percent H3PO4 are

commonly used today. Although the current practice is to operate at atmospheric pressure,

31
in the 11 MW electric utility demonstration plant, the in service pressure of the PAFCs

exceeded 8 atm, confirming an increase in power plant efficiency (Gazey, et al., 2006).

2.5.2 Alkaline Electrolyte Fuel Cell

Alkaline electrolyte fuel cell's reaction is identical to the acid fuel's electrolyte, but there are

different reactions with each electrode. Hydroxyl ions are passing ions in this case (OH-). The

hydrogen fuel reacts to the anode creating water and discharging electrical energy with OH-

as shown in equation 15 (Liebhafsky, 1968). The anode electrons react to new OH ions with

water and oxygen on the side of the cathode (Bagotsky, 2012). Even when the cathode

absorbs water, twice the amount of the anode is made. Figure 25 displays the alkaline fuel

cell's working setup.

2𝐻2 + 4𝑂𝐻 − → 4𝐻2 𝑂 + 4𝑒 − (15)

𝑂2 + 4𝑒 − + 2𝐻2 𝑂 → 4𝑂𝐻 − (16)

Figure 25 Diagram showing the basic functionality of a circulating electrolyte in an alkaline

fuel cell (Ang, et al., 2010)

32
AFCs with immobilized electrolytes are extremely sensitive to carbon dioxide (CO2). It is

possible to directly feed non-hydrocarbon hydrogen fuel or pure H2 to the anode. A carbon-

free fuel gas such as cracked ammonia (25% N2, 75% H2, and residual NH3) can be fed directly

to the cell, for instance (EG & G Technical Services, 2004). Only a very small decrease in

potential is observed due to the high diffusion rate of hydrogen compared to nitrogen, with

a hydrogen content of more than 25 percent. When H2 is produced from fuel sources

containing carbon, gas purification is necessary. There are many methods to separate CO2

from liquid or gaseous streams. The most common techniques used are physical separation

and chemical separation (Hawkes & Leach, 2005).

However, CO2 elimination by these methods involves more than one process step to reduce

CO2 to the limits needed by the fuel cell. There are two additional methods for cryogenic

separation and biological fixation. If the alkaline cell uses liquid hydrogen as a source, a heat

exchanger system is possible for the condensation of CO2 for the air stream of oxidants

(Grimes, 2001). This technique has a probable weight advantage over the soda lime scrubber.

Low-temperature distillation is widely used for the liquefaction of CO2 from high purity

sources. An investigated new, potentially successful technique uses a selective wicking to

isolate gases to create capillary condensation. Biological separation is exciting, but it has to

face the challenge of reactivation after phases have ended (Andujar & Segura, 2009).

2.5.3 Proton Exchange Membrane (PEM) Fuel Cell

At present, the Proton Exchange Membrane (PEM) fuel cell is the most useful (Bagotsky,

2012). An electrolyte is used as a solid polymer in which protons (H+) are mobile. The fact that

the electrolyte is solid and immobile makes it very simple for this cell. The operational

temperature of these cells is quite low, around 38 – 99 °C, which gives slow reaction rates a

33
problem, but the start-up is rapid at the same time. The use of more effective electrodes and

catalysts such as platinum solves the problem of slow reaction rates (Burke, 2007). The

membrane is also weak and easily breaks. The field of operation concerns primarily cars,

portable and low-power CHP applications.

The proton exchange membrane fuel cell comprises of two permeable carbon electrodes, the

anode and the cathode, disjointed by the ion-conducting proton exchange membrane of a

polymer electrolyte (PEM). A thin layer of a catalyst is integrated between each electrode and

the membrane (Andujar & Segura, 2009) as shown in figure 26. The membrane electrode

assemblage (MEA) forms the electrodes, catalyst and membrane together, and bipolar flow

field plates are placed on each side of the MEA with gas conduits exasperated into their

surface. The electrodes are connected to an external load circuit, such as the vehicle's electric

motor.

Figure 26 Graphic exemplification of proton exchange membrane fuel cell (Energy &
Electricity, 2004).

34
Another detailed graphic arrangement of a PEM fuel cell can be seen in Figure 27. In

operation, hydrogen is delivered to the anode and air to the cathode. The hydrogen molecules

dissociate into free electrons and protons in the presence of the catalyst at the anode. The

electrons are carried through the external circuit as a usable electrical current, while the

protons transfer to the cathode with water molecules through the membrane electrolyte

(Nan, et al., 2014). Oxygen from the air, electrons from the outer circuit and protons are

combined on the cathode side to form water and heat, thus concluding the total reaction.

The following equations represent the reactions occurring in the fuel cell (Andujar & Segura,

2009):

𝐻2 → 2𝐻 + + 2𝑒 − (17)
1
𝑂2 + 2𝐻 + + 2𝑒 − → 𝐻2 O (18)
2

1
𝐻2 + 𝑂2 → 𝐻2 𝑂 (19)
2

Figure 27 Schematic presentation of a single fuel cell (Bagotsky, 2012)

35
2.6 Fuel Cell Stacks

The core of a fuel-cell system is a fuel cell stack. Hundreds of fuel cells may contain a typical

fuel cell stack (Bagotsky, 2012).

2.6.1 Simple Series Connection

For purposes stated earlier, a single fuel cell is made precisely reedy. This restricts the voltage

through it, however. A characteristic value for the voltage when drawing a useful current is

about 0.7 V (Bossel, 2000), by attaching a number of cells in a succession, branded as a stack,

a higher voltage can be achieved, Figure 28. The easiest way to do this is to join the anode's

edge to the next cell's cathode. Through this arrangement, the electrons have to flow to the

edge across the electrode surface.

Figure 28 Simple series connection from the anode of one cell to the cathode of the next one
(Ernst, 2007)

36
Even if the electrodes are decent conductors, a drop in voltage will occur. In comparison to

the low cell voltage, this drop would be important in relation to low cell voltage and thus this

approach is not used until the current is very small.

2.6.2 Bipolar Series Connection

A better option for building stacks is the attachment of bipolar sheets made of materials with

a strong conductivity like graphite or stainless steel (Bagotsky, 2012). These panels bind all

over a cathode, thus bipolar, to the anode of the next cell. It needs to meet the criteria that

there should be a strong electrical connection between the electrodes and that the different

gases must be segregated.

There are usually horizontal channels used in the bipolar plates to supply oxygen to the

cathode and vertical channels to supply the anode with hydrogen, Figure 29. A solid block is

formed when the stack is assembled with bipolar plates, where the current streams more or

less straight through the cells as an alternative to across the surface, resulting in better

efficiency. As the electrodes are better supported, the block is also more robust and strong.

Figure 29 Fuel cell stack consisting of three cells with bipolar plates (Larminie & Dicks, 2003).

37
Where a balance must be considered between electrical contact and gas flux, the bipolar plate

design is very complicated. As the contact points become as wide as possible, the optimisation

of electric contact is accomplished, but the gas flow through the electrodes is decreased

(Larminie & Dicks, 2003). There are also possibilities of small contact points, but they should

be regular and make it both difficult and costly to manufacture a fragile plate. To achieve low

resistivity and a limited size of pile, the bipolar plate should be made thin. This makes it

difficult for the gas to flow across the cell, leading to narrow channels.

In reality, additional considerations need to be made when designing a stack with bipolar

plates. The challenges include the delivery and leakage of gases (Bagotsky, 2012). The

electrodes are made porous to enable the gas to pass through them, as mentioned before. A

consequence of this is that the gas can leak out of the electrode edges. Sealing the edges is

the solution to this problem. This is accomplished by making the electrodes smaller than the

electrolyte and then fitting around them a sealing gasket, Figure 30.

Figure 30 Structure of a fuel cell with edge seals to prevent leakage of the gas at the edges of
the electrodes (Larminie & Dicks, 2003)

38
 2.6.3 Water Management

A water management system is needed to humidify reactants for fuel cells that operate at

over 60°C. In order to ensure adequate conductivity and long membrane length, water must

be delivered in appropriate amounts and distributed evenly. However, the pores of the

electrodes must be equalized if flooding or water are to be prevented (Larminie & Dicks,

2003). The water balance, i.e. the amount of condensed water equal to the water needed for

humidifying, is an essential characteristic for automotive applications. The system for water

management often involves water condenses and water collection and reusability so that the

water system for the fuel cell can be water-sustaining.

The moisture mechanisms differ, from external moistures (e.g. direct water injections and

enthalpy wheels), to internal humidity such as wicks or self-moisture, to no humidification

whatsoever (Bagotsky, 2012). The elimination or reduction of external humidification will

simplify the fuel cell system in terms of room and heat supply. Internal wetting controls have

been difficult but no humidification (Harnada, et al., 2006) has been reported to increase the

fuel cell system's weight and increase market power.

2.6.4 Air Management

Depending on the necessary operating speed, a blower or a compressor supplies intake air to

the fuel cell stack. Pressurized systems make for smaller and more compact fuel cell stacks,

even at the expense of compressor power requirements (Zalba, et al., 2003). At small speeds,

the efficiency is poor and the compressor will work slowly, i.e. not respond instantly to load

changes.

Between pressure of the atmosphere and 3 bar is the operating pressure of a fuel cell stack.

The turbocharger and double-screw compressor are the most studied alternatives thanks to

39
their low weight and compact scale. (Ishitani, 2001) Found that a turbocharger is more

powerful than the twin-screw compressor, particularly at low mass flows. A neutral water

balance with a turbocharger was, however, difficult to hold than in a double-screw

configuration.

2.6.5 Heat Management

Fuel cells have an electrical efficiency of up to 60 percent, which means that the rest of the

input energy is lost as waste heat (Andujar & Segura, 2009). The waste heat developed in the

fuel cell stack is transferred to the cooling circuit in order to maintain a constant temperature

and physical stability of the fuel cell stack. Within the system, the heat rejected from the

cooling loop can be used, such as heating the passenger compartment by means of a radiator.

Proper heat exchanger sizing is important as this system appears to be voluminous with its

heat exchangers, radiators and pumps (Barbir, 2005). The integration into the vehicle chassis

is necessary here, as, for example, the heat load of the radiator is a function of the speed of

the vehicle.

40
3 Methodology

The focal portion of this work is established on computer simulations. Mathematical

simulation tools were established, mainly by using the commercially accessible software

MATLAB and Simulink, in order to deliver comprehensive models of a self-propelled fuel cell

system and hydrogen production scheme. The simulated results were used to come up with

the dimensions for the fuel cell stack model. A similar model building sequence to the one

referred to in the flow chart below was adapted for this project.

41
4 System Configuration and Model

4.1 PV – Hydrogen Production Model

4.1.1 System Layout and Interconnection Design

In this project the experimental system has been proposed to be built in Harare, Zimbabwe (-

17°47'2.40" S 31°03'10.80" E) (Latitude, 2019). The system concept consists of 2 flat-plate

PV/T panels that have a tilt angle of -22 (Jacobson & Jadhav, 2018) and a PEM electrolyzer,

figure 31. The PV/T array consists of a total of 70 cells, with the yield capacity for each PV/T

being 200W, a maximum power voltage of 36.75 V and a maximum power point current of

5.45 A (Vanilla, 2019).

Figure 31 Schematic diagram of the proposed system

To eliminate the heat from the panel and shield it against extreme temperatures, a water

pump is used to circulate water. A profitable PEM electrolyzer system is used for the

production of hydrogen for this project. For this project. The electrolyzer is able to produce

42
17.9 Nm3 of hydrogen per day without a compressor at progression pressure (Baumann , et

al., 2010). To reduce PV/voltage, T's which provides power to the electrolyzer, a step-down

DC/DC converter is required. Table 3 shows the features of the PV/T array to be used.

Table 3 Features of the photovoltaic thermal panel

Dimensions 1640 x 870 x 105

Maximum current (A) 5.45
Short Circuit current (A) 5.65
Maximum voltage(V) 36.75
Open circuit voltage (V) 45.40
Glass Superfluous Solar Cut-glass
Cell number 70
Type Monocrystalline
Thermal Power(W) 675
Active cell area (m2) 1.189

4.1.2 Mathematical Model of the PV/T System

Hybrid energy systems must be studied separately with extremely complex computational

models for each component of the system. Weather data such as hourly-based temperature

and solar radiation data are important for the measurement of solar energy systems. Analyses

such as energy and energy efficiency when processed annually are quite complicated and time

consuming. The computation and implementation of each feature have therefore been

simulated in the MatLab/Simulink setting (Index, 2020). Simulink is used especially for

evaluating and optimizing power systems and control schemes, but seldom used for

thermodynamic modelling.

The one-dimensional steady-state models of PV/T was modelled with equations from Hottel-

Whillier (Duffie & Beckman, 2013). In the thermal performance calculations of the PV/T

system, heat losses are taken into account and are dependent on variables such as solar

radiation intensity, wind speed, ambient temperature, collector structure, transparent

43
covering properties, radiation emissions and absorption value of the surface of the absorber,

coefficient of thermal conductivity, thickness and type of insulation material (Barbir, 2005).

The thermal efficiency of PV/T collectors (Duffie & Beckman, 2013):

ℚ𝑢
𝜂𝑡ℎ = (20)
𝐼𝑇

ℚ𝑢 = 𝐴𝑐 𝐹𝑅 [𝑆 − 𝑈𝐿 (𝑇𝑖𝑛 − 𝑇𝑎 )] (21)

With Formula 22 below, thermal losses from the top, bottom and border area are calculated:

𝑈𝐿 = 𝑈𝑡 + 𝑈𝑏 + 𝑈𝑒 (22)

The collector flow rate, FR:

𝐴 𝑈 𝐹𝑖
𝑚𝐶𝑝 − 𝑐 𝐿
𝑚𝐶𝑝
𝐹𝑅 = 𝐴 [1 − 𝑒 ] (23)
𝑐 𝑈𝐿

The efficiency of the collector, Fi:

1
𝑖 𝑈𝐿
𝐹 = 1 1 1
(24)
𝑊[ + + ]
𝑈𝐿 [𝐷+(𝑊−𝐷)𝐹] 𝑊∙ℎ𝑃𝑉𝐴 𝜋𝐷ℎ𝑓𝑖

F is the factor fin:

(𝑊−𝐷0 )
𝑡𝑎𝑛ℎ (𝑡𝑎𝑛ℎ(𝑀 ))
2
𝐹= (𝑊−𝐷0 ) (25)
(𝑀 )
2

The thermal efficiency is calculated using the following equation 26:

𝑇𝑎 −𝑇𝑖𝑛
𝜂𝑡ℎ = 𝐹𝑅 (𝛼𝑔 𝜏) − 𝐹𝑅 𝑈𝐿 (26)
𝐼𝑇

For the electric output and performance calculations, the mean temperature of the absorber

surface is Tpm given in the following equation 27:

𝑄𝑢
⁄𝐴
𝑇𝑝𝑚 = 𝑇𝑖𝑛 + ( 𝑈 𝐹 𝑐) (1 − 𝐹𝑅 ) (27)
𝐿 𝑅

44
The efficiency in electrical changes with the PV temperature and is found using the formula,

28 with β0 = 0.0045 (Radziemska, 2003):

𝜂𝑒 = 𝜂0 [1 − 𝛽0 (𝑇𝑝𝑚 − 𝑇𝑟𝑒𝑓 )]
(28)

The electrical output of the system is then given by the following equation, 28 (Ang, et al.,

2010):

𝑃𝑒 = 𝜂𝑒 ∙ 𝐴𝑐 ∙ 𝐼𝑇 (29)

The PV/T system's exergy analysis is based on the second thermodynamics law where the

general exergy balance is expressed in mean form, neglecting the effects of potential and

kinetic variations of energy (Fudholi, et al., 2014).

∑ 𝐸𝑥𝑖𝑛 − ∑ 𝐸𝑥0 = ∑ 𝐸𝑥𝑑 (30)

∑ 𝐸𝑥𝑖𝑛 − ∑(𝐸𝑥𝑡ℎ + 𝐸𝑥𝑒 ) = ∑ 𝐸𝑥𝑑 (31)

The exergy input 𝐸𝑥𝑖𝑛 can be calculated as follows:

4 𝑇 1 𝑇
𝐸𝑥𝑖𝑛 = 𝐴𝑐 ∙ 𝐼𝑇 ∙ [1 − 3 ( 𝑇𝑎) + 3 ( 𝑇𝑎)4 ] (32)
𝑠 𝑠

Where the temperature of the sun (Ts) is taken as 5778K. The thermal energy equation is then

expressed as follows:

𝑇 +273
𝐸𝑥𝑡ℎ = 𝑄𝑢 ∙ [1 − (𝑇𝑎+273)] (33)
0

The electrical energy produced is then calculated using the following formula:

4 𝑇 1 𝑇
𝐸𝑥𝑒 = 𝜂𝑒 ∙ 𝐴𝑐 ∙ 𝐼𝑇 ∙ [1 − 3 ( 𝑇𝑎) + 3 ( 𝑇𝑎)4 ] (34)
𝑠 𝑠

The simulation characteristics of the PV/T array are presented in Table 4

Table 4 Simulation parameters for the PV/T used in the mathematical model

Parameters Values
Diameter of the copper tube (D) 0.01m
Width of the tube spacing (W) 0.092m

45
 PV thermal conductivity (kpv) 149W/m K
PV laminate thickness (lpv) 0.0002 m
Absorber thermal conductivity (kabs) 385 W/m K
Absorber laminate thickness (labs) 0.0012 m
Insulation material thermal conductivity (ki) 0.04 W/m K
Insulation material thickness (li) 0.05 m
Transmission-absorption coefficient (ταg) 0.8 W/m K

4.1.3 PEM Electrolyzer Mathematical Model

The energy for the PEM electrolyzer consists of two components: the change in thermal

energy (T∆S) and the change in free energy from Gibbs (∆G). The Gibbs free energy value is

defined as a zero-energy point at standard pressure and temperature and is used to calculate

the change in a system's energy. It is calculated by the formula (Zhang, et al., 2010):

∆𝐻 = ∆𝐺 − 𝑇∆𝑆 (35)

In order to accurately model the complex behaviour of a PEM electrolyzer anodes, cathodes,

membrane, voltage parts and their interactions must be understood. The reversible potential

of water division can be determined using equation (36) (Siracusano, et al., 2012). Where F is

the constant of Faraday, z is the number of transmitted electrons.

∆𝐺 0
𝐸0 = − (36)
𝑧𝐹

The electrolyzer voltage (Velc) can be calculated by the following equation (Zhang, et al., 2010):

𝑉𝑒𝑙𝑐 = 𝐸𝑟 + 𝑉𝑎𝑐𝑡,𝑎 + 𝑉𝑎𝑐𝑡,𝑐 + 𝑉𝑜ℎ𝑚 (37)

Where Er is the perfect power supplies with the anode cathode, activation and loss of ohmic

value of Vact,a and Vact,c . The use of the Nernst equation is used to identify the optimal

electrical potential for water (EG & G Technical Services, 2004):

𝐸𝑟 = 1.482 − 0.000845 ∙ 𝑇 + 4.31 × 10−5 ∙ 𝑇 ∙ 𝑒 (𝑃𝐻2 ∙√𝑃𝑂2 ) (38)

46
With T being the operating temperature of the PEM electrolyzer and 𝑃𝐻2 showing the partial

pressure of hydrogen while 𝑃𝑂2 shows the partial oxygen pressure. Activation energy, which

is the result of activation overvoltage, is known as energy barriers that have to react to

reactive reactants. The over potential activation represents the electrode activity in the

electrochemical reaction and is provided by the Butler-Volmer equation (Zhang, et al., 2010):

𝑅∙𝑇 𝑖
𝜂𝐴 = 𝛼 𝑙𝑛 (𝑖 ) (39)
𝑎 ∙𝑧∙𝐹 𝑜,𝑢

𝑅∙𝑇 𝑖
𝜂𝐶 = 𝛼 𝑙𝑛 (𝑖 ) (40)
𝑐 ∙𝑧∙𝐹 𝑜,𝑐

Ohmic losses arise as a result of the ion motion in the solid electrolyte and the electric flow

through electrodes. Ohmic over potential across PEM is reverse related to ionic local

conductance and is empirically administered as (Ang, et al., 2010):

𝜙
𝑉𝑜ℎ𝑚 = ∙𝑖 (41)
𝜎

Where i represents the current density and 𝜎 is the conductivity of the membrane. The PEM

electrolyzer energy efficiency will then be given by the proportion of the exergy variance of

the reactants reacting by means of the exergy of the yields exiting the PEM electrolyzer

(Kazim, 2005).

(𝐸𝑥)𝑝 −(𝐸𝑥)𝑟 (𝐸𝐻2,𝑝 +𝐸𝑂2,𝑝 )−(𝐸𝐻2𝑂,𝑟 )

𝜂𝑒𝑥 = = (42)
𝑃𝑒,𝑡𝑛 𝑃𝑒,𝑡𝑛

Where the total energy transfer is the sum of the chemical and physical exergy:

𝐸𝑥 = 𝑒𝑥𝑝ℎ + 𝑒𝑥𝑐ℎ (43)

With the organic exergy of hydrogen being 158.125 kj/kg, that of water being 2.6 kj/kg and

oxygen being 0 (Cengel & Boles, 2008). The total exergy rate of products and reactants is then

given by the following equations:

𝐸𝐻2,𝑝 = 𝑚𝐻2,𝑝 ∙ (𝑒𝑥𝑐ℎ + 𝑒𝑥𝑝ℎ )𝐻2,𝑝 (44)

47
 𝐸𝑂2,𝑝 = 𝑚𝑂2,𝑝 ∙ (𝑒𝑥𝑐ℎ + 𝑒𝑥𝑝ℎ )𝑂2,𝑝 (45)

𝐸𝐻2𝑂,𝑝 = 𝑚𝐻2𝑂,𝑝 ∙ (𝑒𝑥𝑐ℎ + 𝑒𝑥𝑝ℎ )𝐻2𝑂,𝑝 (46)

The simulations conditions of the PEM electrolyzer used in the mathematical model are given

in Table 5.

Table 5 Mock-up characteristics for the PEM electrolyzer used in the numeric model.

Parameters Values
Transfer coefficient anode-cathode (αa, αc) 0.52
Anode-cathode electrode thickness (δa, δc) 5.1 x 10-6 m
Anode exchange current densities (i0,a) 0.0052 A/cm2
Cathode exchange current density (i0,c) 0.0009 A/cm2
Current density (i) 1.64 A/cm2
Operating pressure (P) 1.1 bar
Faraday number (F) 96,358.3 A/mol
Operating temperature (T) 302.5 K
Degree of humidification (λ) 23.8

4.1.4 Economic Analysis

The LCC method (life cycle expense), used to compare energy costs with various cost

structures, would be the preferred method of economic analysis (Ceicdata, 2020). The study

aims to quantify the cost of the service over the entire project duration, instead of just

comparing the initial capital costs or operating costs. Based on the time value of money,

consideration is made of all costs that arise during the system's lifetime (Capital Cost),

operational and maintenance costs (OMC), repair and replacement costs (RRC) (Elbaset,

2011)

𝑇𝑃𝑉 = 𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑐𝑜𝑠𝑡 + ∑ 𝑂&𝑀 + ∑ 𝑅𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡 (47)

In the following equation are calculated the real deduction rate and the retrieval factor:

𝑖 ′ −𝑓
𝑅′ = (48)
1+𝑓

48
Where i’ is the trifling discount rate and f is the inflation rate;

𝑁
(1+𝑅 ′ ) ∙𝑅′
𝐶𝑅𝐹 = (1+𝑅 ′ )𝑁 −1
(49)

The final cost of hydrogen is calculated using equation 50 (Elbaset, 2011):

$ 𝑇𝑃𝑉∙𝐶𝑅𝐹
𝐶𝑜𝑠𝑡 (𝑘𝑔𝐻 ) = 𝐴𝑛𝑛𝑢𝑎𝑙 𝐻 (50)
2 2 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛

The main cost-determining economic assumptions are as follows: (1) the economic evaluation

period (N) for PV/T panels, DC/DC converters and storage systems is 25 years and for PEM

electrolyzers is 15 years. (2) The nominal discount rate and inflation rate economic indicators

were used as 9 percent and 12 percent for 2017, respectively (Ceicdata, 2020). (3) The cost

calculation does not include installation, operation and maintenance. (4) The capital cost per

watt was 0.82, 2 and 0.9 USD/W for the PV, PV/T, and PEM electrolyzers, respectively.

The MatLab model for the PV/T arrays, PEM electrolyzers have been attached in Appendix 1

4.2 Fuel Cell Stack

Many parameters have to be taken into consideration when designing and modelling fuel

cells, as shown in the previous chapters. A fuel cell with a bipolar configuration has been

selected for the purposes of this project.

4.2.1 Stack Sizing

Two independent variables were considered, that is voltage and current. The power yield is a

direct multiplication of stack voltage and current (You & Liu, 2006):

𝑊𝐹𝐶 = 𝑉𝑆𝑇 ∙ 𝐼 (51)

The current is a multiple of the current density and the cell active area:

𝐼 = 𝑖 × 𝐴𝑐𝑒𝑙𝑙 (52)

49
While the cell voltage and current density are in relation with the polarization curve shown in

figure 32.

Figure 32 Polarization curves for PEM fuel cell single cells (Song & Shin, 2001)

4.2.2 Number of Cells

The power requirement of the fuel cell for this project is 50 kW. The total stack potential is

the result multiplying the average cell voltage and the sum of cells in the stack (Andujar &

Segura, 2009).

𝑉𝑠𝑡 = 𝑉𝑐𝑒𝑙𝑙 × 𝑁𝑐𝑒𝑙𝑙 (53)

The cell area is designed to get the required current for the stack. If the overall stack voltage

multiplies, the maximum power requirement for the stack will be reached. The average

voltage and the corresponding current density selected can greatly change the size and

efficiency of the stack. The fuel cell stack performance can be approximated by equation 54

(Zhou & Francois, 2009):

𝑉𝑐𝑒𝑙𝑙
𝜂𝑠𝑡𝑎𝑐𝑘 = 1.482 (54)

50
 4.2.3 Stack Configuration

The fuel cell stack has many cells in the traditional bipolar stack layout, series, and one cell's

cathode is linked to the next cell's anode. The layers of the fuel cell are the MEAs, gaskets,

bipolar dishes, and end plates. To clamp together the cells, the stack is detained collectively

by rods, pins, or alternative pressure device (Bagotsky, 2012). The following were considered

when the fuel cell model was being drawn: (1) the necessity to distribute fuel and oxidant

homogeneously from end to end, and across the area of its surface. (2) The temperature all

over the stack had to be uniform. (3) The membrane was not to become swamped with water.

(4) It was appropriate to keep resistive losses to a minimum. (5) To guarantee no gas seepage,

the stack was properly sealed. (6) The stack had to be robust and capable to tolerate the

required environments in which it will be used.

The drawing of the stack configuration is shown in Chapter 5.

4.2.4 Stack Clamping

The stacking design and cell assembly parameters have a major impact on the fuel cell

efficiency. Adequate contact pressure is required to keep the fuel-cell stack components

together in order to prevent leakage of the reactants and minimize the contact resistance

between layers (Jiang, et al., 2005). The needed clamping force is equal to the force necessary

to properly compress the fuel cell layers without hindering flux. The assembly pressure affects

the characteristics of the communication interfaces between components. If the insufficient

or non-uniform assembly pressure is used, stack sealing issues will occur, such as the fuel

leakage, internal burning and inappropriate contact resistance.

Too much pressure will prevent GDL flow or damage the MEA and result in broken porous

structures and blocking the gas stream (Zalba, et al., 2003). In both cases, the clamping

51
pressure will reduce the cell's efficiency. Per stack has a specific assembly pressure due to the

variations in fuel cell materials and stack design. Due to the thin dimensions and the low

mechanical resistance of the electrodes and the element layer in comparison to the gaskets,

the bipolar plates and end platforms, the most important goal of the stack design and

assembly is to achieve an accurate and even pressure distribution.

The fuel cell stack in this project was clamped by means of bounds because of the clamping

of the sides, as the edge was strengthened as a centre of reference, as the materials were

bound together to withstand the momentary loads.

The physical properties of the interface GDL and bipolar plate layers are determined by the

interaction resistance and GDL permeability (Zhou & Francois, 2009). The interaction

resistance amongst the catalyst and membrane layers is low-slung because they are merged

together, figure 33, and because the materials are typically non-porous with similar material

properties, the interaction resistance between the bipolar plates and other layers is low.

Figure 33 the forces exerted by the clamped materials on the bolt and nut (Ishitani, 2001)

52
By means of a Herzian compression equation, the quantity of firmness of the GDL and bipolar

plate material layers can be found. The calculations suppose that the contact planes are not

flawlessly smooth, that the materials' elastic boundaries are not surpassed, that the materials

are consistent and that the contact area does not have any frictional forces. For two spheres

in contact, the compression formula is (Yan, et al., 2006):

2
(3𝜋) ⁄3 2⁄ 2⁄ 1 1 1⁄
𝛼𝑐𝑜𝑚𝑝 = ∙𝐹 3 ∙ (𝑉1 + 𝑉2 ) 3 ∙ (𝐷1 + 𝐷2) 3 (55)
2

Where 𝛼𝑐𝑜𝑚𝑝 is the elastic compression at the point of interaction of two bodies, F

representing the total force applied, D being the diameter of the dynamic area of the material:

(1−𝑣 2 )
𝑉= (56)
𝜋𝐸

Where E is the Young’s modulus and v is the Poisson’s ratio.

The stiffness of the group surcharge parts in the stack is given by (You & Liu, 2006):

1 (1−𝑛)
𝑐1 = 𝑘 +𝑘 (57)
𝑏𝑜𝑙𝑡 𝑠𝑡𝑎𝑐𝑘

Where n is the coefficient of application of the operational force. The stiffness of the relieved

parts of the stack is then given by:

𝑘𝑠𝑡𝑎𝑐𝑘
𝑐2 = (58)
𝑛

The operational relieving force of the clamped parts:

𝐹×𝑐2
𝐹2 = (59)
(𝑐1 +𝑐2 )

The average contact pressure of the interface, Pavg, can be found by separating the entire

clamp force (the multiplication of the quantity of bolts and the discrete clamp force of the

bolt) with the contact area of the interface, Aint (Lee, et al., 2005):

𝑁×𝐹0
𝑃𝑎𝑣𝑔 = (60)
𝐴𝑖𝑛𝑡

53
The MATLAB models for the fuel cell stack are shown in Appendix 2.

Table 6 Physical Properties Used for Material Toughness and Compression Designs (Matweb,
2020)

Fuel Cell Layer Thickness(mm) Modulus of Young’s Poisson’s Ratio

Elasticity in Modulus
Tension (MPa) (N/mm2)
Polycarbonate 10 2896 2200 0.37
end plate
Gasket: black 1 2 100 0.48
conductive
rubber
SS flow field 0.5 206000 200000 0.31
plate
Carbon cloth 0.4 2 300 0.4
Nafion 0.05 2 236 0.487

54
5 SolidWorks CAD Design

Hydrogen Side End Plate

Electrode 1 Air Side

55
Electrode 2 None Air Side

Rubber Spacer 1 Air Side

56
Rubber Spacer 2 Non Air Side

Hydrogen Side Rubber Gasket

57
Mylar Surround

Graphite Serpentine Flow Plate Hydrogen Side

58
Graphite Combined Flow Plate

Graphite Flow Plate Air Side

59
Air Side End Plate

Fuel Cell Stack Graphical View

60
Fuel Cell Stack Assembly View

61
Fuel Cell Stack Assembly View

62
6 Results and Discussions

6.1 PV/T – PEM Hydrogen Production Scheme

In order for a valid authentication concerning the numerical model and simulation system,

the electrical, sytem arrangement efficiences were calculated using the model measurement

data given in Chapter 4 as shown in figure 34. The simulated radiation and ambient

temprature changed between 0 – 975 W/m2 and 10.2- 38.3 oC, correspondingly.

Figure 34 Disparity of incident solar radiation and ambient temperature.

Figure 35a presents the investigated PV/T arrangement current-voltage and power-voltage

characteristics for 950W/m2 solar radiation and 32 oC ambient temperature. Under these

conditions, the short circuit current of the arrangement and the open-circuit voltage were

4.86 A and 40 V, correspondingly. The electrical output of the array was therefore calculated

as 139W with 33 V. Depending on (Tin - Ta)/IT, the characteristic experimental thermal

efficiency was determined on a daily basis and compared to the simulation data under similar

climate circumstances (Figure 35b). The ophthalmic efficiency of the experimental numerical

system was obtained as 46.1% and 47.9% percent for the simulation system.

63
Figure 35 I-V and P-V curves of the PV/T system: a) assessment of experimental numerical
data and simulation data; b) comparison of the efficiencies as an iterative of the ratio of (Tin -
Ta)/IT.

In this project, the numerical model of the PEM electrolyzer was applied and confirmed with

simulation outcomes. The change in the electrolysis over potentials are shown in figure 36,

under 30 oC procedural temperature settings, with the alteration in the current. It was found

that the ohmic voltage increased when the current was increased.

Figure 36 Variation of PEM over potentials.

64
The over potential of anode stimulation was 0.37 V, while for 1600 mA/cm2 the cathode over

potential was 0.25 V. The modification in electrolysis voltage was also investigated by altering

the current and temperature in the respective ranges 200-1600 mA/cm2 and 30-65 °C. The

disparities in the voltage of the electrolyzer in relation to various current and temperature

levels interrelated soundly with the simulated outcomes as shown in figure 37.

Figure 37 Comparison of electrolyzer system voltages with dissimilar water temperatures

contingent on current

Figure 38 shows the relation between the water supply of the PV/T and the current output

between the highest iteration and lowest iteration. The change in electrolyzer voltage was

also investigated and it was found the reduction in the current and the rise in the water

temperature were instigated by the voltage of the electrolyzer decreasing.

The daily estimation of hydrogen production for the PEM electrolyzer was found between 4-

5 kg, as the total energy produced by the PV/T on an average iteration was 106W, table 7.

Showing the system can achieve the objective of attaining a hydrogen production rate of 3 –

5 kg per day of operation.

65
Figure 38 the disparity in the hourly electrolyzer voltage contingent on the current and water
temperature.

Table 7: Energy flow for each iteration simulated from the numerical model

Iteration number Solar Energy, kWh Electrolyzer energy Hydrogen

demand, kWh Production, kg
1 65.4 17.7 4.264
2 60.5 20.1 4.846
3 86.7 20.7 4.992
4 100.3 20.7 4.264
5 135.4 20.7 4.264
6 141.1 20.7 4.264
7 135.8 20.7 4.264
8 118.5 20.7 4.264
9 106.0 20.7 4.264
10 95.0 20.7 4.264
11 98.0 20.7 4.264
12 139.0 20.7 4.264
Total Energy 1282.3 245.7 59.031

66
6.2 Hydrogen Fuel Cell Stack

The numerical model was simulated in order to find if the resultant stack would produce a 50

kW fuel cell. Figure 39 give the trade-off solutions from the simulation in Simulink MatLab.

This required resultant was found to be the chosen case, indicated in Table 8.

Figure 39 Simulated Numerical stack producing 50 kW

Table 8 Simulated results to find the required parameters for the chosen case fuel cell of 50kW

η (%) A I (A V (V) P λH2 λair Wfuel Wprs

(x104cm2) cm-2) (atm)
w=1 53.15 52.93 0.11 0.86 3.03 1.10 1.66 79.66 7.66
w= 51.33 40.56 0.14 0.87 4.91 1.10 1.53 78.44 9.47
0.85
w = 0.5 51.33 19.19 0.35 0.75 5 1.10 1.58 91.65 11.32
w= 42.20 15 0.56 0.60 5 1.10 1.46 114.96 12.71
0.15
44.18 25 0.25 0.79 2 1.25 2 50.52 6.47
Chosen
case
w=0 24.8 14.51 0.67 0.51 5 1.20 1.39 145.69 13.89

67
Figure 39 shows the highest point of the optimal solution at w = 1, which matches up to

optimizing the device performance without taking account of the size as a single goal

optimisation. This elucidation is 20% more effective, but 112% bigger than the base case. This

elucidation needs lower current (i.e. higher voltage), greater pressure and inferior

stoichiometric hydrogen and air ratios compared with the selected case. This solution's fuel

consumption is lower as a result of greater efficiency.

The scrounging loss, however, is greater owing to the upsurge in pressure. The bottommost

point in the curve, on the other hand, parallels to the optimum answer at w = 0, which is,

irrespective of efficiency, the minimization of size. This result represents a configuration that

is 42% smaller in size but 44% less proficient than the case chosen. This configuration has a

greater functioning current density (consequently, an inferior voltage) compared to the base

case, functions at greater pressure, and has minor stoichiometric ratios of hydrogen and air.

There is greater clout consumption and scrounging loss in this design. It can be concluded

from the findings that the system's efficiency and size must be optimized simultaneously. The

result is a possibly impractically large system if only the efficiency is maximized. On the other

hand, dimension optimization fallouts in a system that is almost four times smaller in size but

less than appropriate in efficiency.

The far end points of the system configuration, w = 1 and w = 0, reflect the solutions of single-

objective optimization and are therefore not projected to agree with the tendency of how

each variable performs. Overall observation, in the chosen set, poignant from a high

efficiency, bulky size elucidation to a non-dominated, low efficiency, small size solution

involves increases in the density and pressure of the operating current (thus, decreased cell

voltage).

68
To envisage the truthfulness and likelihood of miscalculations during performance, the

simulations were used to analyse each numerical solution. The number of simulation runs is

randomly listed and this is protected against certain specific constants, such as time,

temperature, humidity, etc. Directive to govern if the ideals in the mathematical design

needed alteration, several factors were considered. For example, if the maximum and

minimum data point set ratio exceeds 10, transformation is often recommended. Usually,

transformation skews the outcomes and so it is paramount to leave the outcomes in their

novel states. The measured retorts to current and voltage at different operating conditions

are shown in Table 9.

Table 9 Measured Overall Simulation Data for Stability of the Fuel Cell Stack

Factor 1 Factor 2 Factor 3 Factor 4 Respons Respons

e1 e2
St Ru Hydrogen Oxygen Hydrogen Oxygen Current( Voltage(
d n Pressure(ba Pressure(ba Flow Flow A) V)
r) r) rate(mL/mi Rate(mL/mi
n) n)
1 1 1 0.8 82.5 82.5 3.8 4.71
3 2 1 2.3 82.5 82.5 3.51 4.69
7 3 1.75 1.55 15 150 3.86 4.74
17 4 1 1.55 15 82.5 3.88 4.57
23 5 1.75 0.8 82.5 150 3.74 4.4
13 6 1.75 0.8 15 82.5 3.91 4.39
6 7 1.75 1.55 150 15 3.58 4.61
5 8 1.75 1.55 15 15 3.49 4.8
10 9 2.5 1.55 82.5 15 3.47 4.71
12 10 2.5 1.55 82.5 150 3.8 4.50
16 11 1.75 2.3 150 82.5 3.95 4.54
20 12 2.5 1.55 150 82.5 3.79 4.38
28 13 1.75 1.55 82.5 82.5 3.65 4.64
25 14 1.75 1.55 82.5 82.5 3.67 4.79
9 15 1 1.55 82.5 15 3.49 4.68
8 16 1.75 1.55 150 150 3.83 4.65
27 17 1.75 1.55 82.5 82.5 3.77 4.55
24 18 1.75 2.3 82.5 150 3.81 4.68
15 19 1.75 0.8 150 82.5 3.73 4.65
26 20 1.75 1.55 82.5 82.5 3.61 4.55

69
 19 21 1 1.55 150 82.5 3.78 4.68
18 22 2.5 1.55 15 82.5 3.76 4.67
22 23 1.75 2.3 82.5 15 3.64 4.55
14 24 1.75 2.3 15 82.5 3.95 4.92
29 25 1.75 1.55 82.5 82.5 3.66 4.66
2 26 2.5 0.8 82.5 82.5 3.69 4.69
21 27 1.75 0.8 82.5 15 3.61 4.67
4 28 2.5 2.3 82.5 82.5 3.9 4.89
11 29 1 1.55 82.5 150 3.65 4.54

The fuel cell performance was accentuated using the quadratic scheme. The proficiency of

the fuel improved with reverence to a rise in the oxygen gas velocity. Figure 40 shows it is

conceivable to attain a fuel efficiency of 29.6% at low pressures and low flow rates of

hydrogen but moderate flow rates of oxygen.

Figure 40. The fuel cell efficiency for the fuel stack

Furthermore, the best method of minimizing the costs required to operate the fuel cell stack

is to use the exact operating area, which results in the highest desired fuel cell performance.

70
The yellow regions show the zone which would produce the maximum fuel cell efficiency

under the specified operating conditions described during simulation, as shown in Figure 41.

Figure 41 Recital of the fuel cell stack at changing operational conditions.

The simulations showed that the variation of the operating constraints neighbouring the fuel

cell had an impact on the complete performance of the cell. Again, from the simulation, the

bipolar plate was observed to affect the overall performance of the cell by fluctuating the

functioning parameters neighbouring the fuel cell. Again, the design of the bipolar plate

geometry disturbs the hydrogen expended during the electrochemical reaction equally,

hence the cell's complete operating cost. Electrical conductivity must be helped by the coating

of the bipolar plate, but reducing its susceptibility to corrosion will also greatly alter the

performance of the individual cells in the stack.

71
6.3 Bill of materials for Prototype

The pricing on this bill of materials was obtained by a quotation from Electro Sales Hardware,

Avondale Branch, as 13 January 2020.

Solar Panel

Component Price ($USD)

Solar Cells 2.40
Bus Band Wire 1.25
Tab Trimming Wire 1.25
Sheet Metal 3.40
Cover (Plexiglas) 6.20
Electrically insulating underlay 0.80
Screws, bolts 4.00
Power dispersal box 7.20
Wire and terminal blocks 3.25
Schottky control diodes 9.20
Total 38.95

Electrolyzer

Components Price
11 Stainless steel nuts 4.20
1 PVC pipe, thin wall 2.10
Nickel alloy foil tabs 26.75
Nickel alloy washers
Polypropylene white. 12’’ x 72’’ 32.35
1 pipe coupler 1 15/016’ 2.20
1 pipe cap 1’ 1.25
6 stainless steel washers, ½’ 3.60
1 Rubber o-ring 1’ 1.20
Screws-pan head stainless steel, two 10/24 0.80
Monel standard grade woven wire cloth, 200 3.40
x200 mesh
2 PVC pipe caps for 3’ 3.40
Barbed hose connectors 3/8’ and 1 ¼’ 2.20
Clear silicon rubber caulking 16.20
4 Silicon rubber spacers 1/8’ x ¼’ x 6 7.20
2 silicone rubber washers’ 1’
Total 110.65

72
Gas Processing System

1 PVC nipple, threaded, 7/8x 3 2.50

Fiberglass screen, stainless mesh. 12.30
2 pieces PVC pipe 3 x12 length 1.20
4 #3 caps for 3PVC pipe 3.45
6 PVDF single-barbed tube fittings, 90 7.60
degree elbow male for 3/8 x ½
2 pieces 1x9 ½ 1.40
Clear silicon rubber caulking 12.20
1Filter with zinc body and polycarbonate 23.50
bowl, ¼
1 PVC spring-loaded ball-check valve 3/8x 13.00
3/8
2 Barb hose connectors, 3/8x ¼ 16.20
1 Threaded barb connector ½ x 3/8 8.00
2 Easy-Grip miniature ball valve 3/8x 3/8 9.45
2 Easy-Grip PVC miniature ball valve 3 way 9.45
3/8x 3/8
1 coupler, 3/8x ¼ 4.40
Plastic tubing , variety of sizes 3.00
Total 127.65

Bipolar Fuel Cell Stack

Press-n-Peel PCB transfer film 68.22

Clear silicone rubber caulking 6.25
Plating kits and supplies 22.35
Barb hose connectors, one ¼ x ½ 16.25
6 screws 8.00
6 washer 7/32
6 nuts (1/8)
3 Aluminum Channel bars, 12x3/8x3/8 12.25
Silicon rubber gasket, 0.020x12x12 24.50
12 Membrane electrode assemblies 66.00
PVC Type 1 sheet, 12 x12x1 3.40
2 Copper clad PC boards 8x10x1/16 24.50
1 Silicon rubber sheet, 3/32x12x12 40A 22.00
Ferric chloride etchant 10.00
Total 283.72

73
7 Conclusion

From this particular study, the thermodynamic scrutiny of a PV/T E system were investigated,

and the numerical model of the system was simulated to obtain measurable results. The total

PV/T energy and exergy efficiencies were calculated in the ranges of 31 - 62% and 12.9 –

16.2% respectively. Furthermore the hydrogen production performance of the system

satisfied the objectives set out in this project, averaging above 4kg per day.

The fuel stack model intended to power a small vehicle with at least 50kW. The simulation of

the model shows the fuel cell stack operated at an efficiency between 40 – 47% producing an

average of 54-63 kW for a stack containing 6 cells. The fuel efficiency was also dependant on

the way the reactants were introduced into the stack. However the design correspondingly

entails supplementary adjustment to enrich the water management of the stack itself.

74
8 References

Andujar, J. M. & Segura, F., 2009. Fuel Cells: History and updating. A walk along two centuries.

Renewable and Sustainable Energy Reviews, 13(1), pp. 2309-2322.

Ang, S. M., Brett, D. J. & Fraga, E. S., 2010. A model for the multi-objective optimisation of a

polymer electrolyte fuel cell micro-combined heat and power system.. 20th European

Symposium on Computer Aided Process Engineering, 28(1), pp. 949-954.

Appleby, J., 1990. From Sir William Grove to today: Fuel Cells and the future. Journal of Power

Sources, 29(2), pp. 3-11.

Archives, T. W. H., 2003. The White House Archives. [Online]

Available at: https://georgewbush-whitehouse.archives.gov/ceq/hydrogen_090908.html

[Accessed 16 September 2020].

Bacon, F., 1960. The High Pressure Hydrogen-Oxygen Fuel Cell. Industrial & Engineering

Chemistry, 52(1), pp. 52 - 77.

Badea, G. et al., 2017. Hydrogen production using solar energy - technical analysis. Science

Direct, 112(1), pp. 418 - 425.

Bagotsky, V. S., 2012. Fuel Cells: Problems and Solutions. 2nd ed. Moscow: John Wiley & Sons.

Barbir, F., 2005. PEM electrolysis for production of hydrogen from renewable sources. Solar

Energy, 661-669(1), p. 78.

Baumann , L., Boggasch, E., Rylatt, R. M. & Wright, A. J., 2010. Energy flow management of a

hybrid renewable energy sytem with hydrogen. Citres, IEEE Conference on Innovative

Technologies for an Efficient and Reliable Electricity Supply.

75
Berry, A. & Martyn, M., 2000. Energy through Hydrogen, Research Notes. 1st ed. Berlin:

Heliocentris Energiesysteme.

Bilgen, E., 2004. Domestic hydrogen production using renewable energy. Science Direct, 77(1),

pp. 47 - 55.

Bossel, U., 2000. The Birth of the Fuel Cell. Oberrohrdorf, The European Fuel Cell Forum.

Boudghen, S. A. & Traversa, E., 2002. Solid oxide fuel cells (SOFCs): A review of an

environmentally clean and efficient source of energy.. Renewable and Sustainable Energy

Reviews, 6(5), pp. 433 - 455.

Britannica, T. E. o. E., 2016. Enyclopaedia Britannica. [Online]

Available at: https://www.britannica.com/science/Faradays-laws-of-electrolysis

[Accessed 01 September 2020].

Britannica, T. E. o. E., 2020. Encyclopaedia Britannica, Inc. [Online]

Available at: https://www.britannica.com/technology/measurement

[Accessed 01 September 2020].

Burke, A. F., 2007. Batteries and Ultracapacitors for Electric, Hybrid, and Fuel Cell Vehicles.

Proceedings of the IEEE, 95(4), pp. 806 - 820.

Ceicdata, 2020. Ciecdata. [Online]

Available at: https://www.ceicdata.com/en/turkey/saving-discount-rate-and-interbank-

rate/central-bank-discount-rate

[Accessed 15 November 2020].

Cengel, Y. A. & Boles, M. A., 2008. Thermodynamics: An Engineering Approach. 5th ed. New

York: McGraw-Hill.

76
Chaurasia, P. B., Ando, Y. & Tanaka, T., 2003. Regenerative fuel cell with chemical reactions.

Energy Conversion and Management, 44(4), pp. 611-628.

Cook, B., 2003. Introduction to Fuel Cells and Hydrogen Technology. Engineering Science and

Education Journal, XI(6), pp. 205 - 216.

Dewa, T., 2016. Nehanda Radio. [Online]

Available at: https://nehandaradio.com/2016/09/24/big-fuel-companies-selling-

contaminated-fuel-report/

[Accessed 20 August 2020].

Duffie, J. A. & Beckman, W. A., 2013. Solar Engineering of Thermal Processes. 1st ed. Hoboken:

Wiley & Sons.

Dufo Lopez, R. & Bernal-Agustin, J. L., 2005. Design and control strategies of PV-diesel systems

using genetic algorithms.. Solar Energy, 79(1), pp. 33-46.

Dzirutwe, M., 2019. Reuters. [Online]

Available at: https://www.reuters.com/article/us-zimbabwe-power/zimbabwe-quadruples-

electricity-prices-pummelling-impoverished-consumers-idUSKBN1WO0RE

[Accessed 31 August 2020].

EG & G Technical Services, I., 2004. Fuel Cell Handbook. 7th ed. Morgantown: National Energy

Technology Laboratory.

Elbaset, A. A., 2011. Design, Modelling and Control Strategy of PV/FC Hybrid Power System.

Solar and Wind Energy, 79(1), pp. 379 - 397.

Energy , S. & Electricity, T. O., 2004. Solar and Wind Energy Sources. Wind Energy, 1(1), pp.

379-397.

77
Ernst, J., 2007. History of fuel cell development at Mercede-Benz., Stuttgart: Daimler

Communications.

Fudholi, A. et al., 2014. Performance Analysis of Photovoltaic Thermal (PVT) Water Collectors..

Energy Conversion Management, 78(1), pp. 614-651.

Gargallo, P., 2012. Archimede Power Plant - ENEL. [Online]

Available at: http://www.archimedesolarenergy.com/reference_project_1.htm

[Accessed 23 October 2019].

Gazey, R., Salman, S. K. & Aklil-D'Halluin, D. D., 2006. Afield application experience of

intergrating hydrogen technology with wind power in a remote island location. Journal of

Power Sources, 157(2), pp. 841-847.

Gong, L. & Jing, S., 2012. Applications of Bayesian methods in wind energy conversion

systems. Renewable Energy, 43(1), pp. 1 - 8.

Grimes, P., 2001. Historical Pathways for Fuel Cells. IEEE Aerospace and Electronic Systems

Magazine, 15(12), pp. 7 - 10.

Grove, W. R., 1845. Voltaic Action of Phosphorus, Sulphur and Hydrocarbons. Philosophical

Transactions of the Royal Society of London, 135(1), pp. 351-361.

Harnada, Y. et al., 2006. Operating results and simulations on a fuel cell for residential energy

systems.. Energy Conversion and Management, 47(20), pp. 3562-3571.

Harrison, K., 2009. The Natio Renewable Energy Laboratory Website. [Online]

Available at:

https://webcache.googleusercontent.com/search?q=cache:DthXgJHHHlwJ:https://www.nre

78
l.gov/hydrogen/wind-to-hydrogen.html+&cd=10&hl=en&ct=clnk&gl=zw

[Accessed 31 August 2020].

Hawkes, A. & Leach, M., 2005. Solid oxide fuel cell systems for residential micro-combined

heat and power in the UK: Key economic drivers.. Journal of Power Sources, 149(1), pp. 72-

83.

History, N. M. o. A., 2020. Smithsonian. [Online]

Available at: https://www.si.edu/object/allis-chalmers-fuel-cell-tractor:nmah_687671

[Accessed 16 September 2020].

Index, S., 2020. Simulink. [Online]

Available at: https://www.mathworks.com/help/simulink/index.html

[Accessed 21 November 2020].

Ishitani, H., 2001. Fuel Cells in Japan. Stockholm, Royal Swedish Academy of Engineering

Science.

Jacobson, M. Z. & Jadhav, V., 2018. World Estimates of PV optimal tilt angle and ratios of

sunligh incident upon tilted and tracked PV panels relative to horizontal panels. Solar Energy,

169(1), pp. 55-66.

Jiang, Z., Gao, L. & Dougal, R. A., 2005. Flexible multi-objective control of power converter in

active hybrid fuel cell/battery power sources.. IEEE Transactions on Power Electronics, 20(1),

pp. 244-253.

Kawai, T., 2004. Chapter 4 - Fuel Cell Hybrid Vehicles: The Challenge for the Future. The

Hydrogen Energy Transtion: Moving Toward the Post Petroleum Age in Transportation, 1(1),

pp. 59 - 71.

79
Kazim, A., 2005. Exergoeconomic analysis of a PEM electrolyser at various operating

tempratures and pressures.. International Journal of Energy Resources, 29(1), pp. 539-548.

Korpus, M. & Grreniers, C. J., 2008. Opportunities for hydrogen production in connection with

wind power in weak grids. Renewable Energy, 33(1), pp. 1199-1208.

Larminie, J. & Dicks, A., 2003. Fuel cell systems explained. 1st ed. New York: John Wiley & Sons

Limited.

Latitude, 2019. Latitude. [Online]

Available at: https://latitude.to/articles-by-country/zw/zimbabwe/7584/university-of-

zimbabwe

[Accessed 20 11 2020].

Lee, S. J., Hsu, C. D. & Haung, C. H., 2005. Analysis of the fuel cell stack assembly pressure..

Journal of Power Sources, 145(1), pp. 353-361.

Levene, J. & Ramsden, T., 2007. Summary of Electrolytic Hydrogen Production, Golden,

Colorado: National Renewable Energy Laboratory.

Liebhafsky, H. A. &. C. E. J. (. J., 1968. Fuel Cells and Fuel Batteries. 1st ed. New York: John

Wiley & Sons.

Mapira, J., 2015. Air Pollution In Zimbabwe: An Environmental Health Challenge.. Journal Of

Social Science Research, IX(1), pp. 1750-1758.

Matweb, 2020. Matweb. [Online]

Available at: http://www.matweb.com/

[Accessed 15 December 2020].

80
Meier, A., 2010. Task II: Solar Chemistry Research. 1 ed. Villigen: Paul Scherrer Institute.

Mobius, H.-H., 1997. On the history of solid electrolye fuel cells. In: Rudolf-Breitschied-

Strasse, ed. Soli State Electrochem. Greifswald: Springer, pp. 2-16.

Mohamed, D. & Noureddine, S., 2015. Estimation of hydrogen production using wind energy

in Algeria. Scienc Direct, 74(1), pp. 981 - 990.

Nan, Q., Brooker, P. R. & Raissi, A., 2017. Fuel Cell Vehicle Technologies, Infrastracture and

Requirements, Washington DC: Electric Vehicle Transportation Centre.

Nan, Q., Raissi, A. & Brooker, P., 2014. Analysis of Fuel Cell Vehicle Developments, Florida:

Electric Vehicle Transportation Center.

Nault, R. M., 2005. The Basic Energy Sciences Workshop on Solar Energy Utilization, Berkley:

Science Workshop on Solar Energy Utilization.

Nelson, D. B., Nehrir, M. H. & Wang, C., 2006. Unit sizing and cost analysis of stand alone

hybrid/wind/PV/fuel cell power generation systems. Renewable Energy, 31(1), pp. 1641-

1656.

Olateju, B., Monds, J. & Kumar, A., 2014. Large scale hydrogen production from wind energy

for the upgrading of bitumen from oil sands. Applied Energy, 118(1), pp. 48 - 56.

Ortiz-Rivera, E. I., Reyes-Hernandez, A. L. & Febo, R. A., 2007. Understanding the history of

fuel cells. New Jersey, IEEE Power Engineering Society.

Osaka, N. et al., 2005. Development of residential PEFC co-generation systems.. Palm Springs,

Fuel Cell Seminar and Exposition.

81
Peharz, G., Dimorth, F. & Wihstadt, U., 2007. Solar hydrogen production by water splitting

with conversion efficiency of 18%. International Journal of Hydrogen Energy, 32(1), pp. 3248-

3252.

Pitz-Paal, R., 2007. High Temprature Solar Concentration, Oxford: Institute of Technical

Thermodynamics.

Pukrushpan, J. T., Stefanopoulou, A. G. & Peng, H., 2004. Control of fuel cell power systems:

principles, modelling, analysis and feedback design.. 1 ed. London: Springer-Verlag.

Radziemska, E., 2003. Thermal performance of Si and GaEs based solar cells and modules: A

review. Prog. Energy Combust. Sci, 29(1), pp. 407 - 424.

Reuter, S., 2020. The Zimbabwe Mail. [Online]

Available at: https://www.thezimbabwemail.com/economic-analysis/zimbabwe-fuel-

imports-gobble-us1-2-billion/

[Accessed 23 August 2020].

Revankar, S. & Majumbar, P., 2014. Fuel Cells: Principles Design and Analysis. 1st ed. Florida:

Taylor & Francis Group.

Sandstede, G., Cairns, E. J., Bagotsky, V. S. & Weisener, K., 2003. History of low temprature

fuel cells. In: A. L. H. A. G. Wolf Vielstich, ed. Handbook of Fuel Cells - Fundamentals,

Technology and Applications. Chichester: John Wiley & Sons, pp. 145 - 218.

Sherif, S. A., Barbir, F. & Vezigrolu, T. N., 2005. Wind energy and the hydrogen economy-

review of the technology. Solar Energy, 78(1), pp. 647-660.

Shikha, R. & Zifei, Y., 2019. Overview of global zero-emission vehicle mandate programs. The

International Council On Clean Transportation, 24 April, pp. 1-15.

82
Siracusano, S., Baglio, V. & Briguglio, N., 2012. An electrochemical study of a PEM stack for

water electrolysis. International Journal on Hydrogen Energy, 37(1), pp. 1939-1946.

Song, R. H. & Shin, D., 2001. Influence of CO concentration and reactant gas pressure on cell

performance in a PAFC. International Journal of Hyrdogen Energy, 26(1), pp. 1259-1262.

Stone, C. & Morrison, A. E., 2002. From curiosity to power change the world. Solid State Ionics,

152(1), pp. 1-13.

Tributsch, H., 2008. Photovoltiac hydrogen production. International Journal of Hyrdogen

Energy, 33(1), pp. 5911-5930.

Turner, J. A., Williams, M. C. & Rajeshwar, K., 2004. Hydrogen based Economy based on

Renewable Energy Sources, Fall: The Electrochemical Society Interface.

Turner, J. et al., 2008. Renewable Hydrogen Production. International Journal Of Energy

Research, 32(1), pp. 379 - 407.

Valenzuela , L., Zarza, E., Berenguel, M. & Camacho, E., 2012. Control schem for direct steam

generation in parabolic troughs under recirculating operation mode.. Solar Energy, 80(1), pp.

1-70.

Vanilla, 2019. Wind & Sun. [Online]

Available at: https://forum.solar-electric.com/discussion/354752/how-much-current-from-

a-200-watt-panel-is-mine-fake

[Accessed 23 November 2020].

Verspagen, B., 2007. Mapping Technological Trajectories as Patent citation Networks: a study

on the History of Fuel Cell Research. Advances in Complex Systems (ACS), 10(1), pp. 93 - 115.

83
Wilson, J. F. et al., 2013. Hydrogen Production Using Solar Energy. Journal of Undergraduate

Research in Physics, 27(1), pp. 1-2.

Yan, Q., Toghiani, H. & Wu, J., 2006. Investigation of water transport through membrane in a

PEM fuel cell water balance experiments. Journal of Power Sources, 158(1), pp. 316-325.

You, L. & Liu, H., 2006. A two-phase flow and transport model for PEM fuel cells.. Journal of

Power Sources, 155(1), pp. 219-230.

Zalba, B., Marin, J., Cabeza, L. & Mehling, H., 2003. Review on thermal energy storage with

phase change: Materials, heat transfer analysis and applications.. Applied Thermal

Engineering, 23(1), pp. 251-283.

Zhang, H., Lin, G. & Chen, J., 2010. Evaluation and calculation on the efficiency of a water

electrolysis system for hydrogen production. International Journal of Hydrogen Energy, 35(1),

pp. 10851-10858.

Zhou, T. & Francois, B., 2009. Modelling and control design of hydrogen production process

of an active hydrogen/wind hybrid power system.. International Journal of Hydrogen Energy,

34(1), pp. 21-30.

84
9 Appendix 1 PV/T E production model MatLab load program

% program PV/T E hydrogen production scheme

% author: Victor Mulolo
% Tel: +2630774037984
%-----------------PV/T E hydrogen production scheme---------

function
[MH2_Production_1,MH2_total,Pelzrequired,PtoElectolyzer,surp,d
eficit,MH2_def,Sum_Positive_Power,Sum_Negative_Power]=Pw_Load_
Electrolyser_grid_gss2(PLoad,Pwg)
% the system solar/H2
SPCHe=39.4;% Kwh/kg Means HHV of H2=142 MJ/kj which is Equal
to 39.4kWh/kg for the electrolyser
SPCHec=(SPCHe+SPCHc); % Kwh/kg for the electrolyser.
%for an hourly hydrogen requirement of 0.208kg, the SPCH are
multiplied by
%0.208 to get the combined max power Pelzrequired.
Pelzrequired=8.66;% kw maximum electrolyser power
minReq_Elec_Power=0.30*Pelzrequired;% kw minimum electrolyser
power since

%......input.......load demand for the

site.........................

clear all
PLoad=[
21.6,21.6,17.5,18.5,18.9,19.2,19.9,21.6,20.9,20.6,18.5,15.7,19
.2,17.8,18.2,18.5,18.9,19.2,17.5,19.6,19.9,20.9,20.9,18.2;
18.5,18.9,19.2,19.6,19.6,19.9,24.4,24.4,24.7,21.3,18.9,17.5,15
.7,18.2,19.2,18.5,18.2,18.9,18.2,20.3,18.2,19.2,19.2,18.2;
21.7,20.3,20.9,22.4,23.1,25.5,25.8,26.0,17.8,17.5,17.5,18.2,18
.5,20.3,17.5,18.9,19.2,19.6,17.5,19.2,19.6,20.3,20.6,20.9;
21.3,22.0,22.0,22.1,22.0,22.7,24.1,24.1,26.2,26.9,26.9,18.9,18
.5,18.5,19.2,19.2,17.8,19.2,20.3,20.3,20.6,18.9,20.3,20.6;
23.4,23.8,24.0,24.0,24.4,27.9,30.1,35.0,32.9,27.9,20.3,19.9,19
.6,17.5,20.3,19.6,19.9,18.5,18.9,18.9,19.9,22.0,22.4,23.1;
23.7,24.0,24.4,24.4,27.9,29.7,32.5,33.6,27.9,30.1,20.9,20.9,17
.5,17.5,20.6,19.9,20.3,18.9,19.2,19.6,20.3,22.7,23.1,23.4;
23.9,24.1,24.4,24.1,25.5,25.8,26.1,31.5,33.2,30.1,30.1,18.5,18
.2,17.5,19.2,17.5,17.5,19.2,19.6,19.9,20.6,22.7,23.0,23.4;
22.0,26.6,26.9,27.2,25.5,24.8,25.2,29.0,25.8,27.9,24.5,17.5,19
.2,18.2,18.9,19.2,18.2,18.5,18.2,17.8,18.2,22.0,22.6,22.4;
20.6,18.9,18.9,18.2,19.6,22.7,20.9,32.2,33.2,29.8,26.6,17.5,17
.5,18.9,17.5,20.6,20.6,17.5,18.5,18.9,19.2,19.6,19.9,19.9;

85
22.1,22.4,22.7,22.4,23.0,23.0,22.4,25.9,26.2,26.5,17.8,18.5,17
.5,19.2,18.9,18.5,19.6,19.6,19.6,19.6,19.6,19.6,19.2,22.4;
26.2,25.2,25.2,24.5,24.2,23.1,22.7,25.9,25.2,23.1,18.2,17.8,17
.8,18.2,18.2,18.2,18.5,18.9,19.2,18.5,19.9,17.8,17.8,22.7;
26.2,26.5,25.7,26.2,27.3,27.6,29.7,31.1,31.1,22.7,17.8,19.2,18

.2,18.5,17.5,18.2,19.9,20.3,20.6,22.0,24.1,24.1,24.4,26.2]*1e-

1;

for n=1:12;
for fg=1:24
% Eyload(fg)=trapz(n,PLoad(fg,:)) % Eyload total Kwh generated
for each iteration for load
p_load= PLoad(n,fg)
end
end
% total genrated power Etotal_Load=sum(Eyload) per cyle
% Eytotal_load=sum(Eyload)*30.41666/1000
% P_Load=[n,PLoad(fg,:)]
n=1:24;
subplot(2,2,1);
plot(n,PLoad(3,:))
title('Load curve for Iteration');
ylabel('Power,kW');
xlabel('TIME, iteration number');

86
10 Appendix 2 Hydrogen Fuel Cell Stack MatLab Load Program

% program hydrogen fuel cell stack

% author: Victor Mulolo
% Tel: +2630774037984
%-----------------hydrogen fuel cell stack model---------

% UnitSystem SI %
% Inputs %
Power = 50000; % required stack power (W) %
Voltage = 48; % Stack Voltage (V) %
Vcell = 0.7; % Cell voltage %
i = 0.5; % Current Density (A/cm^2) %
% Calculate the required stack current
I = P/V;
% Assume that the fuel cell voltage is 0.7 V %
N_cells = Voltage/Vcell;
% Assume the current density is 0.5 A/cm^2, thus %
I_cell = I/N_cells;
% The area required per cell %
A_cell = i_cell/i; % cm^2

end

% UnitSystem SI %
% Inputs
%Design of optimal bolt connection through stack %
%The bolt is connected with a through bolt, and the loading is
in the bolt axis. The course of loading is static. clear; %
N = 4; % number of bolts
% Material properties of the bolt %
% Material of bolt
E_bolt = 200 000; % Young’s Modulus of the bolt (N/mm) %
v_bolt = 0.31; % Poisson’s ratio of the bolt %
Em_bolt= 210 000; % Modulus of elasticity in tension (MPa) or
30 000 ksi %
% Bolt and Thread Parameters
dia_bolt = 4.826; % Thickness %
dia_rt = 3.451; % bolt thread root diameter (mm) %
thread_pitch = 1.058; % Thread Pitch %
pitch_dia = 4.139; % Pitch diameter %
dbolthead = 8; % Bolt head diameter (mm) %
hnut = 3; % Nut thickness (0 if threaded into flange) %
hbolthead = 5; % Thickness of bolt head (mm) %
Lbolt = 35; % Bolt length %
L_bolt = 25.4; % Bolt length between the bolt head and nut %
% Geometry of the bolt connection – Calculated for annulus
seating face
seatDe = 7.925; % Outer diameter of the seating face (mm) %
87
seatDi = 5.232; % Inner diameter of the seating face (mm) %

% bolt hole
dbore = 5.232; % Bolt clearance hole (mm) %
alpha = 45; % Cone angle %
qa = 0.5; % Desired coefficient of tightness %
n = 0.5; % Coefficient of implementation of the operational
force %
mi = 0.150; % Friction coefficient in thread %
mc = 0.150; % Friction coefficient in seating face of head
(nut) of the bolt %

% Material properties of end plate (polycarbonate) %

Em_end = 2896; % Modulus of elasticity in tension (MPa) %
thick_end = 10; % Thickness (mm) %

%Material properties of rubber gasket %

Em_gask = 2; % Modulus of elasticity in tension (MPa) %
thick_gask = 1; % Thickness (mm) %

% Material properties of stainless steel %

Em_nia = 206 000; % Modulus of elasticity in tension (MPa) or
30 000 ksi %
thick_nia = 0.5; % Thickness (mm)%

% Material properties of carbon cloth %

Em_cc = 2; % Modulus of elasticity in tension (MPa) or 0.3 ksi
%
thick_cc = 0.4; % Thickness (mm)%

% Material properties of Nafion %

Em_naf = 2; % Modulus of elasticity in tension (MPa) or 0.3
ksi %
thick_naf = 0.05; % Thickness (mm) %

% Stiffness calculations %
% Total thickness of parts
tot_thick = (thick_end + thick_gask + thick_nia + thick_cc +
thick_naf + thick_cc +
thick_nia + thick_gask + thick_end);
De = seatDe + (tot_thick ∗ 0.2)/2;

% Bolt stiffness
boltdia = (pitch_dia + dia_rt)/2;
nut =1.8/(pi ∗ pitch_dia);
head = 1.5/(pi ∗ boltdia);
kboltshaft = (4 ∗ L_bolt)/(pi ∗ boltdia^2);
cb = Em_bolt/(kboltshaft + head + nut);

88
% Calculate stiffness of each part
kcomp_end=(4 ∗ thick_end)/(pi ∗ (De^2-dbore^2) ∗ Em_end);
kcomp_gask =(4 ∗ thick_gask)/(pi ∗ (De^2-dbore^2) ∗ Em_gask);
kcomp_nia = (4 ∗ thick_nia)/(pi ∗ (De^2-dbore^2) ∗ Em_nia);
kcomp_cc = (4 ∗ thick_cc)/(pi ∗ (De^2-dbore^2) ∗ Em_cc);
%multiply by the area%
kcomp_naf = (4 ∗ thick_naf)/(pi ∗ (De^2-dbore^2) ∗ Em_naf);

% Total stiffness of the clamped parts

cm1 = 1/(kcomp_end + kcomp_gask + kcomp_nia + kcomp_cc +
kcomp_naf +
kcomp_cc + kcomp_nia + kcomp_gask + kcomp_end);
cm = N/(kcomp_end + kcomp_gask + kcomp_nia + kcomp_cc +
kcomp_naf +
kcomp_cc + kcomp_nia + kcomp_gask + kcomp_end);
kjoint = cb + cm1; % total layers stiffness

% Force calculations %
d_gdl = 40; % diameter (mm)
d_bpp = 40; % diameter (mm)
rho_GDL = 1.1e4; % Electrical resistivity through plane (800
u-ohm-m) uohm mm %
rho2 = 300; % Electrical resistivity through plane of bipolar
plate (190 uohm-m) uohm mm %
void = 0.8; % GDL porosity %
E_gdl = 3; % Young’s modulus of carbon cloth (3 GPa) %
E_bpp = 200; % Young’s modulus of stainless steel %
v_gdl = 0.4; % Poisson’s ratio of Carbon Cloth %
v_bpp = 0.31; % Poisson’s ratio of stainless steel %
V_gdl =(1-(v_gdl^2))/(pi ∗ E_gdl);
V_bpp =(1-(v_bpp^2))/(pi ∗ E_bpp);

% Amount of compression in microns

alpha = (((3 ∗ pi)^(2/3))/2 ∗ (F_bolt_max^(2/3)) ∗ ((V_gdl +
V_bpp)^(2/3)) ∗ ((1/d_gdl)+(1/d_bpp))^(1/3)) ∗ 10

% Electrical constriction resistance of the single contact

rho1 = rho_GDL ∗ (1 − void)^1.5; % transverse electrical
resistivity of GDL
r = d_gdl/2;
Resist = (rho1 + rho2)./(4 . ∗ r) % Resistivity

% Maximum force
F_bolt_max = 310; % compression is 75.92 um GDL, Resistivity =
16.05

89
% Tightening torque calculation %
c1 = 1/((1/cb) + ((1-n)/cm)); % Resulting stiffness of the
group of surcharged parts of the joint %
c2 = cm/n; % Resulting stiffness of the group of relieved
parts of the joint %
F2 = F_bolt_max ∗ c2/(c1 + c2); % Part of operational force
relieving clamped parts %
mseat1 = mc ∗ (seatDe + seatDi)/2;
F0T = 0; % Change of prestressing due to the heating of
the connection %
F0L = 0; % Loss of prestressing due to the deformation
of the connection %
F0 = qa ∗ F_bolt_max + F2 + F0T − F0L + 0.5 % Assembly
prestressing of the joint
(N)–based upon axial load only
Mseat = (mseat1 ∗ F0)/(2 ∗ 1000);

% Tightening torque
M = F0 ∗ pitch_dia ∗ (thread_pitch ∗ pi ∗ pitch_dia ∗ mi)/(pi ∗
pitch_dia − thread_pitch ∗ mi)/
(2 ∗ 1000)+ Mseat

% N-m
Mm = 141.61 ∗ M % oz-in %

% Average interface contact pressure using bolt clamp force

F = F0/1000; % Force in Kn %
w = 8/100; % width in m %
l = 8/100; %length in m %
A_int = w ∗ l; %m^2
P_avg_MPa = (N ∗ F /A_int)/1000 % MPa
P_avg_bar = P_avg_MPa ∗ 10

90
11 Appendix 3 List of Equations

4𝑂𝐻 − 1𝑎𝑞 → 𝑂2(𝑔) + 2𝐻2𝑂(𝑙) (1)............................................................................. 15

2𝐻2𝑂(𝑙) + 2𝑒− → 𝐻2(𝑔) + 2𝑂𝐻 − (2) ....................................................................... 15

2𝐻2𝑂 → 4𝐻 + + 4𝑒 − + 𝑂2 (3)..................................................................................... 16

4𝐻 + + 4𝑒−→ 2𝐻2 (4) .................................................................................................. 16

𝜂𝑃𝑉 = 𝜂𝑝 × 𝜂𝑠𝑢 × 𝜂𝑠𝑒 (5) ............................................................................................. 18

𝜂𝑃𝑉 = 𝛱𝜂𝑝𝑖 (6) .......................................................................................................... 19

𝜂𝐺 = 𝜂𝑃𝑉 + 𝑇 × 𝜂𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑙𝑦𝑧𝑒𝑟 × 𝜂𝐶 (7) ....................................................................... 19

𝑃𝑃𝑉𝑡 = 𝐼𝑛𝑠𝑡𝐴 𝐸∬ 𝑝𝑣 (8) .................................................................................................. 21

𝑃𝑃𝑉𝑡 = 12𝜌𝐴𝑣(𝑡)3𝐶𝑝𝐸∬ 𝑎𝑑 (9) .................................................................................. 25

𝐸𝑛𝑒𝑟𝑔𝑦 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 𝑜𝑓 𝑤𝑎𝑡𝑒𝑟 𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑙𝑦𝑠𝑖𝑠 =

𝐻𝑒𝑎𝑡𝑖𝑛𝑔 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 ℎ𝑦𝑑𝑟𝑜𝑔𝑒𝑛 𝑜𝑢𝑡𝑝𝑢𝑡𝑇𝑜𝑡𝑎𝑙 𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙 𝑒𝑛𝑒𝑟𝑔𝑦 𝑖𝑛𝑝𝑢𝑡 ............................. (10)

27

𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = 𝑉𝑜𝑙𝑢𝑚𝑒 𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑒𝑑 𝐻2(𝑁𝑚3ℎ) × 96487(𝐶𝑚𝑜𝑙𝑒−) × 2(𝑚𝑜𝑙𝑒 −

𝑚𝑜𝑙𝐻2)𝐶𝑢𝑟𝑟𝑒𝑛𝑡(𝐴) × 3600(𝑠ℎ) × 0.0224(𝑁𝑚3𝑚𝑜𝑙𝐻2) .........................................................

(11) ........................................................................................................................................... 28

𝐸 = 𝐸𝑟𝑒𝑣 + 𝜂𝑐 + 𝜂𝑎 + 𝑗𝑅 ∗ (12)...................................................................................... 28

2𝐻2 → 4𝐻 + +4𝑒 − (13) ................................................................................................. 31

𝑂2 + 4𝑒 − +4𝐻+→ 2𝐻2𝑂 (14) ....................................................................................... 31

2𝐻2 + 4𝑂𝐻−→ 4𝐻2𝑂 + 4𝑒 − (15) ................................................................................. 32

𝑂2 + 4𝑒 − +2𝐻2𝑂 → 4𝑂𝐻 − (16) .................................................................................. 32

𝐻2 → 2𝐻 + +2𝑒 − (17).................................................................................................... 35

12𝑂2 + 2𝐻 + +2𝑒−→ 𝐻2O (18) ..................................................................................... 35

𝐻2 + 12𝑂2 → 𝐻2𝑂 (19) ................................................................................................. 35

91
𝜂𝑡ℎ = ℚ𝑢𝐼𝑇 (20) ............................................................................................................. 44

ℚ𝑢 = 𝐴𝑐𝐹𝑅𝑆 − 𝑈𝐿(𝑇𝑖𝑛 − 𝑇𝑎) (21) .................................................................................. 44

𝑈𝐿 = 𝑈𝑡 + 𝑈𝑏 + 𝑈𝑒 (22) ................................................................................................. 44

𝐹𝑅 = 𝑚𝐶𝑝𝐴𝑐𝑈𝐿1 − 𝑒 − 𝐴𝑐𝑈𝐿𝐹𝑖𝑚𝐶𝑝 (23) ..................................................................... 44

𝐹𝑖 = 1𝑈𝐿𝑊1𝑈𝐿𝐷 + 𝑊 − 𝐷𝐹 + 1𝑊 ∙ ℎ𝑃𝑉𝐴 + 1𝜋𝐷ℎ𝑓𝑖 (24) .......................................... 44

𝐹 = 𝑡𝑎𝑛ℎ(𝑡𝑎𝑛ℎ𝑀𝑊 − 𝐷02)(𝑀𝑊 − 𝐷02) (25) ............................................................. 44

𝜂𝑡ℎ = 𝐹𝑅𝛼𝑔𝜏 − 𝐹𝑅𝑈𝐿𝑇𝑎 − 𝑇𝑖𝑛𝐼𝑇 (26) ........................................................................... 44

𝑇𝑝𝑚 = 𝑇𝑖𝑛 + 𝑄𝑢𝐴𝑐𝑈𝐿𝐹𝑅1 − 𝐹𝑅 (27) ............................................................................ 44

𝜂𝑒 = 𝜂01 − 𝛽0(𝑇𝑝𝑚 − 𝑇𝑟𝑒𝑓) (28) .................................................................................. 45

𝑃𝑒 = 𝜂𝑒 ∙ 𝐴𝑐 ∙ 𝐼𝑇 (29) ........................................................................................................ 45

𝐸𝑥𝑖𝑛 − 𝐸𝑥0 = 𝐸𝑥𝑑 (30) ................................................................................................... 45

𝐸𝑥𝑖𝑛 − 𝐸𝑥𝑡ℎ + 𝐸𝑥𝑒 = 𝐸𝑥𝑑 (31) ..................................................................................... 45

𝐸𝑥𝑖𝑛 = 𝐴𝑐 ∙ 𝐼𝑇 ∙ 1 − 43𝑇𝑎𝑇𝑠 + 13(𝑇𝑎𝑇𝑠)4 (32)........................................................... 45

𝐸𝑥𝑡ℎ = 𝑄𝑢 ∙ 1 − 𝑇𝑎 + 273𝑇0 + 273 (33) ..................................................................... 45

𝐸𝑥𝑒 = 𝜂𝑒 ∙ 𝐴𝑐 ∙ 𝐼𝑇 ∙ 1 − 43𝑇𝑎𝑇𝑠 + 13(𝑇𝑎𝑇𝑠)4 (34) ................................................... 45

∆𝐻 = ∆𝐺 − 𝑇∆𝑆 (35) ........................................................................................................ 46

𝐸0 = −∆𝐺0𝑧𝐹 (36) .......................................................................................................... 46

𝑉𝑒𝑙𝑐 = 𝐸𝑟 + 𝑉𝑎𝑐𝑡, 𝑎 + 𝑉𝑎𝑐𝑡, 𝑐 + 𝑉𝑜ℎ𝑚 (37) ................................................................. 46

𝐸𝑟 = 1.482 − 0.000845 ∙ 𝑇 + 4.31 × 10 − 5 ∙ 𝑇 ∙ 𝑒(𝑃𝐻2 ∙ 𝑃𝑂2) (38)............................ 46

𝜂𝐴 = 𝑅 ∙ 𝑇𝛼𝑎 ∙ 𝑧 ∙ 𝐹𝑙𝑛𝑖𝑖𝑜, 𝑢 (39) ..................................................................................... 47

𝜂𝐶 = 𝑅 ∙ 𝑇𝛼𝑐 ∙ 𝑧 ∙ 𝐹𝑙𝑛𝑖𝑖𝑜, 𝑐 (40) .................................................................................... 47

𝑉𝑜ℎ𝑚 = 𝜙𝜎 ∙ 𝑖 (41) ......................................................................................................... 47

𝜂𝑒𝑥 = (𝐸𝑥)𝑝 − (𝐸𝑥)𝑟𝑃𝑒, 𝑡𝑛 = 𝐸𝐻2, 𝑝 + 𝐸𝑂2, 𝑝 − (𝐸𝐻2𝑂, 𝑟)𝑃𝑒, 𝑡𝑛 (42) .................. 47

𝐸𝑥 = 𝑒𝑥𝑝ℎ + 𝑒𝑥𝑐ℎ (43) .............................................................................................. 47

92
𝐸𝐻2, 𝑝 = 𝑚𝐻2, 𝑝 ∙ 𝑒𝑥𝑐ℎ + 𝑒𝑥𝑝ℎ𝐻2, 𝑝 (44) .................................................................... 47

𝐸𝑂2, 𝑝 = 𝑚𝑂2, 𝑝 ∙ 𝑒𝑥𝑐ℎ + 𝑒𝑥𝑝ℎ𝑂2, 𝑝 (45)..................................................................... 48

𝐸𝐻2𝑂, 𝑝 = 𝑚𝐻2𝑂, 𝑝 ∙ 𝑒𝑥𝑐ℎ + 𝑒𝑥𝑝ℎ𝐻2𝑂, 𝑝 (46) ............................................................. 48

𝑇𝑃𝑉 = 𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑐𝑜𝑠𝑡 + 𝑂&𝑀 + 𝑅𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡 (47) ........................................................ 48

𝑅′ = 𝑖′ − 𝑓1 + 𝑓 (48)........................................................................................................ 48

𝐶𝑅𝐹 = 1 + 𝑅′𝑁 ∙ 𝑅′1 + 𝑅′𝑁 − 1 (49) ............................................................................ 49

𝐶𝑜𝑠𝑡 $𝑘𝑔𝐻2 = 𝑇𝑃𝑉 ∙ 𝐶𝑅𝐹𝐴𝑛𝑛𝑢𝑎𝑙 𝐻2 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛 (50) ............................................ 49

𝑊𝐹𝐶 = 𝑉𝑆𝑇 ∙ 𝐼 (51) ........................................................................................................ 49

𝐼 = 𝑖 × 𝐴𝑐𝑒𝑙𝑙 (52) ......................................................................................................... 49

𝑉𝑠𝑡 = 𝑉𝑐𝑒𝑙𝑙 × 𝑁𝑐𝑒𝑙𝑙 (53).................................................................................................. 50

𝜂𝑠𝑡𝑎𝑐𝑘 = 𝑉𝑐𝑒𝑙𝑙1.482 (54) .............................................................................................. 50

𝛼𝑐𝑜𝑚𝑝 = (3𝜋)232 ∙ 𝐹23 ∙ (𝑉1 + 𝑉2)23 ∙ (1𝐷1 + 1𝐷2)13 (55) .................................... 53

𝑉 = (1 − 𝑣2)𝜋𝐸 (56) ....................................................................................................... 53

𝑐1 = 1𝑘𝑏𝑜𝑙𝑡 + 1 − 𝑛𝑘𝑠𝑡𝑎𝑐𝑘 (57) .................................................................................... 53

𝑐2 = 𝑘𝑠𝑡𝑎𝑐𝑘𝑛 (58) .......................................................................................................... 53

𝐹2 = 𝐹 × 𝑐2(𝑐1 + 𝑐2) (59)............................................................................................. 53

𝑃𝑎𝑣𝑔 = 𝑁 × 𝐹0𝐴𝑖𝑛𝑡 (60) ................................................................................................ 53

93