Excellent Thesis Design

Dynamic Simulation of
Green Ammonia
Synthesis Plant
Sampreeth Kambhampati
Delft University of Technology
Dynamic Simulation of
Green Ammonia Synthesis
Plant
by
Student Number
5466598
to obtain the degree of Master of Science

in Mechanical Engineering
at the Delft University of Technology,
to be defended publicly on Tuesday October 31, 2023 at 10:00 AM.
Thesis Committee: Prof. Dr. Ir. Earl Goetheer, TU Delft

Prof. Dr. Ir. Wiebren de Jong, TU Delft
Dr. Mahinder Ramdin, TU Delft
Dr. Gerard van Zee, Proton Ventures B.V.
Project Duration: January 2023 - October 2023
Faculty: Process & Energy Department, TU Delft
An electronic version of this thesis is available at http://repository.tudelft.nl/.

Preface
It is with immense gratitude that I present this work, a collaboration with Proton Ventures. My time
spent at this esteemed organization has been an absolute privilege which made me grow and expand
my knowledge and skills in the field. Thank you for all the chocolates, biscuits and amazing conversa-
tions at the kitchen counter everyday.
I would like to extend my sincerest appreciation to my daily supervisor Dr. Gerard van Zee, who has
been a source of guidance, encouragement and support throughout my project. Their expertise, cou-
pled with their commitment to my personal and professional development, has made this experience
unforgettable.
Moreover, I am deeply grateful to my professor Dr. Earl Goetheer at TU Delft who initially accepted
to be my supervisor from Delft, and has been a constant source of inspiration, enlightenment and al-
ways being there during my ups and downs. Their remarkable knowledge, wisdom and insights have
elevated my understanding of the subject matter to new heights. Thank you very much, Professor. I
also wholeheartedly thank Prof. Wiebren de Jong and Dr. Mahinder Ramdin for being a part of the
assessment committee.
In conclusion, I cannot overstate the importance of the support of my friends and family who made this
journey possible. Their contributions have been vital for me me to reach this point, and for that, I am
truly thankful. The two years of my time at TU Delft have been quite challenging but more fun and I am
honoured to have been a part of this experience.
Delft, October 2023
ii
Summary
The carbon emissions from human activities are causing significant harm to the planet, leading to in-
creased temperatures, melting of polar ice caps, rising sea levels, and other negative impacts on the
environment. One promising solution is the use of green hydrogen as a fuel source, which could have
a much lower carbon footprint than traditional fossil fuels. The production of hydrogen can be achieved
through various methods, including the electrolysis of water, which splits water molecules into hydro-
gen and oxygen. To mitigate these effects and ensure a sustainable future, countries are taking various
measures to reduce their carbon footprint, including increasing the use of clean energy sources and
improving energy efficiency. Hydrogen storage and transportation pose major challenges since it is the
one of the lightest gases leading to low energy densities.
Ammonia is emerging as a hydrogen carrier due to its high gravimetric storage densities of hydrogen.
It is produced through the combination of hydrogen and nitrogen using the Haber-Bosch process. Am-
monia can then be used as a clean and efficient fuel for various applications, such as transportation
and power generation. Fluctuations in the hydrogen feed flow rate, resulting from variations in renew-
able energy sources can significantly impact the pressure and operating temperature within the system.
Morocco holds significant potential for renewable energy development due to its favorable geographic
location and natural resources. The geographic location situated close to Europe makes Morocco well
positioned for exporting green hydrogen to European markets. The chosen location for the ammonia
plant is Boujdour in Morocco due to its excellent wind capacity factor of 67%.
Modern ammonia production plants employ control systems to maintain stable pressure. When there
is a reduction in hydrogen feed flow rate, these reductions result in severe pressure reductions which
would lead to metal fatigue and damage the entire production unit. Hence, these control systems re-
spond by adjusting parameters to sustain pressure within the system. Aspen Plus Dynamics has been
used in the present thesis work to model the dynamics of the ammonia synthesis plant. The varying
hydrogen feed flow rate is a consequence of renewable energy fluctuations, which is served as the
basis for modeling three distinct scenarios involving a 20%, 50%, and 70% reduction in hydrogen feed
flow rate. Three distinct control strategies were developed where each control strategy, based on con-
trolling the cooling duty of the condenser, manipulating the brake power of the recycle compressor,
and regulating the nitrogen feed flow rate, demonstrated effective stabilization of the system’s pres-
sure, even during dynamically changing input conditions. Both linear and step reduction in hydrogen
feed flow rate have been considered to gain understanding of the dynamic the behaviour of the system.
Significant outcomes were found when a reduction in hydrogen feed flow rate is imposed on all three
control strategies. For a 20% reduction in hydrogen feed flow rate, the condenser’s duty reduced from
-1.2 MW to -1.05 MW, while the brake power of recycle compressor reduced from 12.5 kW to 5.5 kW.
Furthermore, the stoichiometric ratio of H2 :N2 changed from 3 to 2.8. These changes successfully sta-
bilized the pressure in the ammonia synthesis plant under varying hydrogen input flow rate.
To determine the economic feasibility, an in-depth economic analysis was conducted, considering differ-
ent renewable energy systems—solar, wind, and hybrid (solar+wind). The analysis revealed that using
wind power is the optimal choice for ensuring a high capacity factor, thus making it the most feasible
and sustainable energy source to power the ammonia plant achieving an annual ammonia capacity
factor of 80%. It has been observed that using wind energy with oversized electrolyzer and introducing
a hydrogen buffer resulted in reduction of levelized costs of ammonia (LCOA) by 11% (USD 1.3/kg to
USD 1.15/kg). In conclusion, this thesis work underlines the viability of green ammonia production,
emphasizing effective control strategies to adapt to renewable energy fluctuations.
iii
Contents
Preface ii
Summary iii
1 Introduction 1
1.1 Impact of Carbon Emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Ammonia as a hydrogen carrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Research Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4 Objective & Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.6 Organization of this document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Literature Study 7
2.1 Fluctuations in solar and wind energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Hydrogen Production & Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.1 Alkaline Water Electrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2.2 Proton Exchange Membrane (PEM) Electrolysis . . . . . . . . . . . . . . . . . . 11
2.2.3 Solid Oxide Electrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2.4 Comparison of different electrolysis techniques . . . . . . . . . . . . . . . . . . . 14
2.2.5 Hydrogen Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3 Nitrogen Production & Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3.1 Cryogenic Distillation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3.2 Pressure Swing Adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3.3 Membrane Separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.3.4 Comparison of different air separation techniques . . . . . . . . . . . . . . . . . 20
2.3.5 Nitrogen Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4 Ammonia Production: Haber-Bosch Process . . . . . . . . . . . . . . . . . . . . . . . . 21
3 Methodology 25
3.1 Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.2 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.3 Modeling and Simulation Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4 Steady State Simulation 27
4.1 Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2 Reactor Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.3 Flash Vessel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.4 Heat Exchanger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.5 Compressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.6 Valves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.7 Condenser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.8 Other elements in MODEL-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
5 Dynamic Simulation 34
5.1 Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5.2 Dynamic Simulation Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.3 Control Philosophy-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.3.1 Step-Change in Hydrogen Feed to Control Philosophy-1 . . . . . . . . . . . . . . 36
5.3.2 Linear-Change in Hydrogen Feed to Control Philosophy-1 . . . . . . . . . . . . . 38
5.4.1 Step-Change in Hydrogen Feed Flow Rate to Control Philosophy-2 . . . . . . . 43
5.4.2 Linear Change of Hydrogen Feed Flow Rate to Control Philosophy-2 . . . . . . . 45
iv
Contents v

5.5.1 Step-Change in Hydrogen Feed Flow Rate to Control Philosophy-3 . . . . . . . 48
5.5.2 Linear-Change in Hydrogen Feed Flow Rate to Control Philosophy-3 . . . . . . . 51
6 Economic Evaluation 54
6.1 Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
6.2 Input Data to the Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
6.3 Solar PV generator with Electrolyzer and Haber-Bosch unit . . . . . . . . . . . . . . . . 56
6.4 Wind generator with Electrolyzer and Haber-Bosch unit . . . . . . . . . . . . . . . . . . 60
6.5 Hybrid Generator with Electrolyzer and Haber-Bosch unit . . . . . . . . . . . . . . . . . 63
6.6 Hydrogen Buffer Vessel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
6.7 Electrolyzer Over-sizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.8 Balancing Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
6.9 LCOH Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
7 Conclusions and Recommendations 73
7.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
7.2 Recommendations for Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
References 76
A Control Philosophy-1 82
A.1 Step Change in Hydrogen Feed Flow Rate . . . . . . . . . . . . . . . . . . . . . . . . . 82
A.2 Linear Change in Hydrogen Feed Flow Rate . . . . . . . . . . . . . . . . . . . . . . . . 86
B Control Philosophy-2 91
B.1 Step Change in Hydrogen Feed Flow Rate . . . . . . . . . . . . . . . . . . . . . . . . . 91
B.2 Linear Change in Hydrogen Feed Flow Rate . . . . . . . . . . . . . . . . . . . . . . . . 93
C Control Philosophy-3 97
C.1 Step-Change in Hydrogen Feed Flow Rate . . . . . . . . . . . . . . . . . . . . . . . . . 97
C.2 Linear-Change in Hydrogen Feed Flow Rate . . . . . . . . . . . . . . . . . . . . . . . . 101
D Economic Evaluation 106
D.1 Determination of Required Back-Up Electricity . . . . . . . . . . . . . . . . . . . . . . . 106
D.2 Hydrogen Storage Tank Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
List of Figures
1.1 Power to Hydrogen to Ammonia to usage chain [3] . . . . . . . . . . . . . . . . . . . . . 1

1.2 CO2 emissions from ammonia production from 2020 to 2050 by IEA [5] . . . . . . . . . 2
1.3 Volumetric (kg/100L) and gravimetric (wt%) hydrogen storage densities for various com-
ponents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Complete schematic of Power-to-Ammonia process . . . . . . . . . . . . . . . . . . . . 5
1.5 Ammonia Synthesis Plant Flowsheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.6 Overview of Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.7 Organization of the Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1 Illustration of wind speed in Boujdour, Morocco [12] . . . . . . . . . . . . . . . . . . . . . 8

2.2 Solar energy powered electricity generation profile in Boujdour, Morocco [19] . . . . . . 8
2.3 Wind energy powered electricity generation profile in Boujdour, Morocco [19] . . . . . . 9
2.4 Comparison of various hydrogen production sources [20] . . . . . . . . . . . . . . . . . 9
2.5 Alkaline Electrolyzer [22] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.6 PEM Electrolyzer [22] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.7 Schematic of solid oxide electrolysis [22] . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.8 Comparison of different electrolysis techniques [34] [35] [36] . . . . . . . . . . . . . . . 14
2.9 Different Technologies for hydrogen storage . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.10 Underground H2 storage facility [40] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.11 Overview of cryogenic distillation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.12 Pressure Swing Adsorption [48] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.13 Schematic of Membrane Separation nitrogen production [54] . . . . . . . . . . . . . . . 19
2.14 Various air separation techniques comparison [51] [36] . . . . . . . . . . . . . . . . . . . 20
2.15 Dewar vessels for storage of cryogens [58] . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.16 Schematic route of Haber-Bosch process . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.17 Mole fraction of ammonia at equilibrium as a function of temperature and pressure (1:3
mixture of N2 / H2 ) [64] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.18 Phase diagram of ammonia [65] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1 Conceptual Design of Ammonia Synthesis Plant . . . . . . . . . . . . . . . . . . . . . . 27

4.2 Steady-State Model from Aspen Plus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.3 Sensitivity analysis of R-PLUG reactor at 300◦ C and 150 bar . . . . . . . . . . . . . . . 28
4.4 Sensitivity analysis of R-PLUG reactor at 400◦ C and 150 bar . . . . . . . . . . . . . . . 29
4.5 Schematic of types of separator vessels . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.1 Control Philosophy-1 from Aspen Plus Dynamics . . . . . . . . . . . . . . . . . . . . . . 35

5.2 Control Philosophy-1 response to 20% step change in hydrogen feed flow rate . . . . . 37
5.3 Control Philosophy-1 response to 20% linear change in hydrogen feed flow rate . . . . 41
6.1 Three Different Options for Green Ammonia Production . . . . . . . . . . . . . . . . . . 55

6.2 Capacity Factor vs Time for Solar PV connected Electrolyzer . . . . . . . . . . . . . . . 57
6.3 Capacity Factor as a function of time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
6.4 Hourly Capacity Factor for 30 days . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
vi
List of Figures vii
6.5 Capacity Factor vs Time for Wind Generator connected Electrolyzer . . . . . . . . . . . 61

6.6 Capacity Factor vs Time for Wind Generator connected Haber-Bosch unit . . . . . . . . 62
6.7 Capacity Factor vs Time for Hybrid Generator connected Electrolyzer . . . . . . . . . . 64
6.8 Capacity Factor vs Time for Hybrid Generator connected Haber-Bosch unit . . . . . . . 66
6.9 Costs of Compressor to Compress Hydrogen Gas and Costs to Store Hydrogen . . . . 68
6.10 Capacity factor of Electrolyzer and Ammonia Plant in 30 days for 34% Oversized Elec-
trolyzer (Green: Electrolyzer, Yellow: Ammonia) . . . . . . . . . . . . . . . . . . . . . . 70
6.11 Hydrogen Buffer Vessel Storage Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
6.12 Overall breakdown of Investment Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
A.1 Control Philosophy-1 response to 50% step change in hydrogen feed flow rate . . . . . 83
A.2 Control Philosophy-1 response to 70% step change in hydrogen feed flow rate . . . . . 85
A.3 Control Philosophy-1 response to 50% linear change in hydrogen feed flow rate . . . . 88
A.4 Control Philosophy-1 response to 70% linear change in hydrogen feed flow rate . . . . 90
B.1 Control Philosophy-2 response to 50% step change in hydrogen feed flow rate . . . . . 92
B.2 Control Philosophy-2 response to 70% step change in hydrogen feed flow rate . . . . . 93
B.3 Control Philosophy-2 response to 50% linear change in hydrogen feed flow rate . . . . 94
B.4 Control Philosophy-2 response to 70% linear change in hydrogen feed flow rate . . . . 96
C.1 Control Philosophy-3 response to 50% step change in hydrogen feed flow rate . . . . . 98
C.2 Control Philosophy-3 response to 70% step change in hydrogen feed flow rate . . . . . 100
C.3 Control Philosophy-3 response to 50% linear change in hydrogen feed flow rate . . . . 102
C.4 Control Philosophy-3 response to 70% linear change in hydrogen feed flow rate . . . . 104
D.1 Hydrogen Tank Storage Level (300 kg) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

List of Tables
1.1 Characteristics of ammonia synthesis unit for present thesis work . . . . . . . . . . . . . 4
4.1 Flash Vessel Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

4.2 Heat Exchanger Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.3 Make-Up Gas Compressor Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.4 Valves Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.5 Condenser Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
5.1 Timeline of Simulation for Linear Change . . . . . . . . . . . . . . . . . . . . . . . . . . 35

5.2 Control Philosophy-1 controller parameters and details . . . . . . . . . . . . . . . . . . . 36
6.1 Solar PV connected Electrolyzer results . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

6.2 Solar PV connected Ammonia Production Plant Results . . . . . . . . . . . . . . . . . . 59
6.3 Total Capital Costs for Solar PV Connected Ammonia Plant . . . . . . . . . . . . . . . . 60
6.4 Wind Connected Electrolyzer Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
6.5 Wind connected Ammonia Production Plant Results . . . . . . . . . . . . . . . . . . . . 63
6.6 Total Capital Costs for Wind Connected Ammonia Plant . . . . . . . . . . . . . . . . . . 63
6.7 Hybrid Connected Electrolyzer Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
6.8 Hybrid connected Ammonia Production Plant Results . . . . . . . . . . . . . . . . . . . 66
6.9 Total Capital Costs for Hybrid Connected Ammonia Plant . . . . . . . . . . . . . . . . . 67
6.10 Results of Wind Connected Ammonia Plant with 34% Electrolyzer Over-Sized . . . . . . 69
6.11 LCOH comparison from different sources . . . . . . . . . . . . . . . . . . . . . . . . . . 72
viii
1
Introduction
1.1. Impact of Carbon Emissions
The impact of carbon emissions on the planet is significant and far-reaching. Climate change causes
rising temperatures, rising sea levels, and more frequent and intense weather events such as hurri-
canes, droughts, and heatwaves. These changes have a negative impact on ecosystems and wildlife,
and can lead to food and water scarcity, displacement of populations, and increased frequency of nat-
ural disasters.
To mitigate the impact of carbon emissions, various governments and organizations are taking steps
to reduce emissions and transition to clean, renewable energy sources. This includes initiatives such
as transitioning to electric vehicles, improving energy efficiency, investing in renewable energy, and
implementing carbon pricing and cap-and-trade systems. In addition, the Paris Agreement, a global
treaty signed by nearly 200 countries, pursues limitation of global warming to well below 2°C above
pre-industrial levels and pursue efforts to limit warming to 1.5°C. [1]
The idea of sustainability refers to meeting the needs of the current generation without jeopardizing
future generations’ ability to meet their own needs. The use of hydrogen as a fuel and ammonia as
an energy carrier are seen as important steps towards achieving a sustainable, zero carbon footprint
future [2]. Hydrogen is a clean, renewable fuel that can be produced through the electrolysis of water.
When burned, it produces only water and heat, making it a zero-emissions fuel. It can be used in a
variety of applications, including powering vehicles and generating electricity. Together, hydrogen and
ammonia have the potential to play a significant role in reducing greenhouse gas emissions and mov-
ing towards a more sustainable energy future. By using these clean, renewable energy sources, it is
possible to reduce dependence on fossil fuels and contribute to a zero carbon footprint future.
Figure 1.1: Power to Hydrogen to Ammonia to usage chain [3]
1
1.1. Impact of Carbon Emissions 2
Ammonia production has a significant impact on global CO2 emissions. It has been estimated that
the largest emitting product of the chemical sector is the ammonia industry. It accounts for 1.3% of
global CO2 emissions [4]. Ammonia production is carried out by Haber-Bosch process which requires
high temperatures and pressures and consumes a significant amount of power. The CO2 emissions
are projected to decrease from ammonia production from 2020 to 2050 as shown below in Figure 1.2
according to IEA [5]. These reduction in emissions could be expected due to changes in energy sector
by utilizing renewable energy generated electricity, where emissions are reduced by 95% reaching net
zero by 2040 and Net Zero Emissions by 2050 [4].
Figure 1.2: CO2 emissions from ammonia production from 2020 to 2050 by IEA [5]
Several efforts are being made to reduce the carbon footprint from ammonia production. One way to
accomplish this is to use renewable energy sources, such as solar or wind to produce hydrogen, which
is used in Haber-Bosch process. Another way to reduce the impact of carbon emissions is to capture
and store the CO2 from ammonia production. This process is known as Carbon Capture and Storage
(CCS). The use of CCS can reduce the CO2 emissions from ammonia production by 90% [4].
The hydrogen produced from water electrolysis can be used as feed stock to produce ammonia. This
method of ammonia production offers several benefits, including energy storage, carbon-free fertilizer
production for sustainable agriculture, and reduced greenhouse gas emissions. Ammonia can be easily
stored in bulk as a liquid at modest pressures or at ambient pressures if refrigerated to -33°C. Several
techniques are already in place for ammonia storage and transportation from decades.
The current research project is performed in collaboration with Proton Ventures B.V. & TU Delft. Pro-
ton Ventures is a participant of the EU research program ARENHA (Advanced materials and Reactors
for ENergy storage tHrough Ammonia). ARENHA is a European project with global impact seeking to
develop, integrate and demonstrate key material solutions enabling the use of ammonia for flexible,
safe and profitable storage and utilization of energy. One of the objectives of this program is the devel-
opment of a demonstration unit for the production of ammonia from hydrogen obtained by electrolysis
at varying capacity.
The production of hydrogen from renewable energy sources and subsequent production of ammonia is
the central theme of the current thesis project. The thesis project focuses more in detail on dynamics
in producing ammonia from renewable energy sources. The research questions and the goal of the
current thesis project have been discussed in section 1.3.
1.2. Ammonia as a hydrogen carrier 3
1.2. Ammonia as a hydrogen carrier

Handling the intermittent nature of renewable energy production is a major challenge in the 21st cen-
tury. Due to these fluctuations in space and time, energy storage techniques with high efficiencies
and capacities are required. In this regard, ammonia is considered to represent as one of the promis-
ing hydrogen carriers due to its high volumetric and gravimetric hydrogen storage densities [6]. The
comparison can be seen in the Figure 1.3 below.
Figure 1.3: Volumetric (kg/100L) and gravimetric (wt%) hydrogen storage densities for various components
The issue of hydrogen storage and delivery opens up prospects for ammonia to be viewed as an alter-
native renewable energy storage medium. Ammonia is produced from Haber-Bosch process which is
one of the most conventional processes in the world. Around 80% of produced ammonia is used as
fertilizers in the agricultural field [7]. Production of ammonia through renewable sources is not a novel
idea and isn’t fully considered over methane or coal-based ammonia. As a result, ammonia can be
produced via sustainable ways. Ammonia can then be transported as renewable energy from places
where there is cheap/excessive renewable energy availability to those places where the renewables
are limited/expensive [8]. Ammonia is a carbon free molecule and doesn’t emit any carbon when burnt
or decomposed into constituent gases. The boiling point of ammonia is around -33°C at atmosphieric
pressures which makes it relatively easier to store and transport than compared to that of hydrogen
(-253°C).
Ammonia could also be compressed to liquid at a mild pressure of 10 bar and atmospheric temperatures.
Ammonia is also the second most produced chemical in world after sulfuric acid in the world, with 180
million tonnes of ammonia being produced per year [9]. Many infrastructures and engineering practices
are already in place for the production, storage and transportation of ammonia. Ammonia has a distinct
odour that is clearly detectable by human nose at extremely low quantities, allowing preventative steps
to be implemented promptly. Furthermore, Ammonia is also non-flammable, with a comparatively low
explosive limit (16-25% in air). The likelihood of ammonia causing combustion and explosion is thought
to be lower than that of other gas and liquid fuels [9].
Some of the advantages of implementing ammonia as a hydrogen carrier are [10] [11]:
• High gravimetric hydrogen storage densities

• Mature technology of production and the second most produced chemical in the world
• Zero-carbon chemical
• Easy liquefaction by compression making it easy for transportation and storage
• Production, transportation and storage technologies already in place
1.3. Research Questions 4
1.3. Research Questions

The following research questions form the basis of the current project:
• How can a stable ammonia synthesis process be maintained under varying hydrogen feed flow
rates?
– How can the dynamic behaviour of the system be modelled?

– What are the suitable control philosophies to maintain pressure under dynamic input?
• How can the complete ammonia plant be optimized by introducing a hydrogen buffer for realistic
renewable power availability scenarios?
– What renewable power source can be considered to ensure ammonia production with re-
duced levelized cost of ammonia?
– What is the optimum amount of hydrogen storage capacity required for the selected ammonia
production capacity?
1.4. Objective & Scope

The major goal of the present study is to examine the dynamic response of an ammonia synthesis
process to variations in the hydrogen feed flow rate. The source of these variations is due to the
renewable power provided to an electrolyzer for hydrogen production. The goal of this project is to
demonstrate operation of an ammonia plant capable of generating 25 tonnes of ammonia per day.
This scale was selected as the plant’s capacity because it represents a scale at which the ammonia
production process can be effectively integrated with renewable energy. The feasibility and practicality
of green ammonia production is investigated for this selected scale size.
In particular, Boujdour in Morocco was chosen as the intended area of operation for the present am-
monia synthesis facility because it has excellent access to solar and wind energy resources [12], which
is consistent with a highly competitive large-scale manufacturing industry utilizing its close proximity to
the European Union.
The project merely comprises of an ammonia synthesis loop assuming that the hydrogen input is pro-
duced by an electrolyzer unit. The primary focus of the research is to create a model which can ac-
commodate dynamic fluctuations in hydrogen feed flow rate by implementation of control philosophies
to stabilize pressure in the system. The storage and transportation of produced liquid ammonia is not
modelled as it is out of scope.
Capacity 25 tonnes/day
Operating Pressure 150 bar
Feed conditions 25◦ C and 80 bar
NH3 Separation Method Cryogenic Condensation
Pressure Drop 3.5 bar
Table 1.1: Characteristics of ammonia synthesis unit for present thesis work
The predominant purpose of the research is to produce liquid ammonia, which may subsequently be
utilized to transport energy from one location to another. The classic condensation method was cho-
sen because it is consistent with the overall goal of the study and its potential influence on energy
transportation and other sectors.
An overall schematic of the process is shown in Figure 1.4. The scope of this thesis work is limited
to the dynamics of ammonia synthesis unit. Therefore, the production of hydrogen and nitrogen are
not modelled in the current thesis work and the red dashed line represents the scope boundary for
modelling of production unit.
1.5. Approach 5
Figure 1.4: Complete schematic of Power-to-Ammonia process
Figure 1.5 shows the process flow diagram of the ammonia synthesis plant that is modelled for the
current thesis project.
Figure 1.5: Ammonia Synthesis Plant Flowsheet
1.5. Approach
The following approach has been used to carry out the current thesis project as shown in Figure 1.6.
The figure below describes each step followed during the project.
Figure 1.6: Overview of Approach

1.6. Organization of this document 6
The three main elements of the performed research are:

1. Steady State Simulation
• Process design is carried out during the initial design phase in Aspen Plus
• The optimal operating conditions, equipment sizes and configurations are determined
• The cooling and heating requirements of the plant have been evaluated.
2. Dynamic Simulation
• The effect of change in hydrogen feed flow rate is examined

• Three different control strategies are developed to control the change in pressure in the
process
• Amount of excess hydrogen required and size of storage tank are determined
3. Economic Evaluation: A brief cost estimation is carried out using HySupply hydrogen and am-
monia cost tools [13] by using different sources of renewables (solar, wind and hybrid) and the
complete value chain of power-ammonia is assessed.
1.6. Organization of this document

The following section gives a brief description of each chapter that is included in this report:
Chapter 2: ”Literature Review” - This chapter provides an overview of the literature study that has been
carried out in the initial phase of the thesis project.
Chapter 3: ”Methodology” - This chapter focuses on the scenarios, assumptions made and the soft-
wares used for the current thesis project.
Chapter 4: ”Steady State Simulation” - This chapter gives an overview about the working of steady
state model and different components selected for the chemical plant.
Chapter 5: ”Dynamic simulation” - This chapter outlines the effect of variation in hydrogen feed flow
rate on the entire ammonia synthesis plant. The current chapter also covers different strategies used
to control the pressure.
Chapter 6: ”Economic Evaluation” - This chapter provides a brief overview of cost estimate of the
ammonia synthesis plant.
Chapter 7: ”Results and Conclusion” - This chapter highlights the answers to the research questions
and recommendations for the future work.
Figure 1.7: Organization of the Report

2
Literature Study
2.1. Fluctuations in solar and wind energy
The reduction of greenhouse gases is by far one of the biggest challenges ongoing. The primary rea-
sons associated behind the emission of these greenhouse gases are power generation, transportation
and industrial activities [14]. Several initiatives are being taken by various sectors and one of the most
promising and an effective action is to use the energy from renewable sources. These renewable
sources are comprised of hydropower, wind power, solar power, and geothermal power.
Renewable energy sources and their percentage in power production have steadily expanded, owing
mostly to energy regulations, markets, and environmental concerns. Wind power and photovoltaics
(Solar) are two of the most important renewable energy sources [15]. Despite the numerous advan-
tages of renewable energy sources, several shortcomings remain, such as: output discontinuity owing
to seasonal changes since most renewable energy supplies are climate-dependent, necessitating so-
phisticated design, planning, and control optimization approaches. Intermittent renewables are difficult
to implement because they disrupt traditional techniques of planning the daily functioning of the elec-
tric grid. For example, because solar energy is only accessible during the day, the grid operator must
modify the day-ahead plan to include units that can swiftly alter their power output to compensate for
the increase and decrease in solar generation [16].
On top of this, there are additional fluctuations during the day caused by the clouds. Therefore it is more
difficult for the grid operator to control the whole unit operation because of the variability by clouds.
Similarly, intra hourly fluctuations in wind energy are caused due to fluctuations in wind speed and
atmospheric instability. As a result, to maximize the utilization of the renewable energy, it requires grid
power optimizations along with energy storage technology [14]. Another promising solution is to deploy
solar/wind combined renewable energy production site in such a way that wind power compensates
the deficiency of electricity output from solar power.
Boujdour located in the Western Sahara to south-west of Morocco is chosen to be the best location
for the current thesis project because of exceptional wind energy production efficiencies. Figure 2.1
illustrates the wind speed of 10 m/s for 100m tall wind turbines. Figure 2.2 and Figure 2.3 show the
fluctuations in solar and wind energy powered electricity generation. The optimum year round tilt angle
for the solar panels is 24.4◦ from horizontal according to [17]. For the selected region, the mean capacity
factor is 22.4%. Using Siemens Gamesa SG 4.5 145 turbine at 107m hub height [18], the total mean
capacity factor is 66.7%. All the above values were obtained from renewables.ninja [19]. Mean capacity
factor is defined as the ratio of actual electricity output by maximum possible output.
7
2.1. Fluctuations in solar and wind energy 8
Figure 2.1: Illustration of wind speed in Boujdour, Morocco [12]
Figure 2.2: Solar energy powered electricity generation profile in Boujdour, Morocco [19]
2.2. Hydrogen Production & Storage 9
Figure 2.3: Wind energy powered electricity generation profile in Boujdour, Morocco [19]
2.2. Hydrogen Production & Storage

Hydrogen is produced from various sources. Hydrogen is classified into three categories, can be iden-
tified by distinct colors, based on the environmental friendliness of the manufacturing method. The
colors used for classification are green, grey and blue and depend on the primary resources used for
the production of the gas [20]. The hydrogen produced from renewable energy sources is called Green
Hydrogen. The hydrogen produced from fossil fuels is called Grey Hydrogen. The hydrogen produced
from fossil fuels with capturing and storing CO2 is called Blue Hydrogen [21].
Hydrogen can be created using a variety of renewable and non renewable energy sources. These
include fossil fuels, particularly steam methane reforming, oil reforming, coal gasification, biomass
gasification, various biological sources, and water electrolysis [22].
Nevertheless, natural gas steam reforming has been the most frequently utilized hydrogen generation
process in recent years, accounting for around 50% of hydrogen products, followed by oil reforming
and coal gasification. This classification is shown in the Figure 2.4 below:
Figure 2.4: Comparison of various hydrogen production sources [20]

Water electrolysis is a technique which is used to split water into hydrogen and oxygen gases. The
process uses electricity from renewable energy sources. The current research is mainly focused on
various techniques of electrolysis because the end goal is to produce ammonia using renewable energy
sources. The different electrolysis technologies addressed in the current thesis project are:
• Alkaline Water Electrolysis

• Proton Exchange Membrane Electrolysis
• Solid Oxide Electrolysis
2.2.1. Alkaline Water Electrolysis

The process generally consists of an anode and a cathode connected to an external power supply and
are immersed in a conducting electrolyte. The alkaline water electrolyser is made of an electrolyte in
aqueous solution which consists of either NaOH or KOH in a 20-40 wt % [23]. The operating condi-
tions typically range from temperatures of 50-90 °C and pressure upto 30 bar. The specific energy
consumption of the system is around 4.2 - 4.8 kWh / Nm3 H2 .
The electrode material chosen for the anode and cathode should be corrosion resistant, have high
conductivity, have a high catalytic effect, and be inexpensive. Although stainless steel and lead were
mentioned as inexpensive electrode materials with low overpotentials, they cannot withstand strongly
alkaline conditions. Noble metals were discovered to be prohibitively costly for usage as bulk elec-
trode materials. Therefore, Nickel was later identified as an electroactive material with high corrosion
resistance in alkaline solutions [24].
The reactions (Equation 2.1) & (Equation 2.2) occurring at anode and cathode are called Oxygen Evo-
lution Reaction (OER) and Hydrogen Evolution Reaction (HER) respectively. Hydrogen is generated
at the cathode while oxygen is generated at the cathode.
1
Anode : 2OH − −→ 2H2 O + O2 + 2e− (2.1)
2
Cathode : 2H2 O + 2e− −→ 2OH − + H2 (2.2)

The overall cell reaction of the water electrolysis process is shown in Equation 2.3
1
H2 O −→ H2 + O2 (2.3)
2
A basic schematic of alkaline water electrolyzer is shown below in Figure 2.5
Figure 2.5: Alkaline Electrolyzer [22]

A diaphragm separates two compartments in an alkaline water electrolyzer. Typically, this diaphragm
is constructed of asbestos or polyphenylene sulphide. When a direct current (DC) current is supplied
through the unit, electrons go from the DC power source to the cathode. These electrons are consumed
by the hydroxide ions to generate hydrogen gas [25]. Normally, these hydroxide ions travel from the
cathode to the anode via the diaphragm. As a result, the hydrogen and oxygen gases which are formed
at the cathode and anode respectively are collected and stored/utilized for further applications.
The minimum required cell voltage for the electrochemical reaction can be determined by thermody-
namics. The energy required to decompose one mole of water is related to enthalpy of formation of
one mole of water. The minimum amount of enthalpy that has to be applied as electrical energy is the
change in Gibbs free energy [23]. This change in Gibbs free energy is represented in terms of change
in enthalpy of reaction, entropy of reaction and the reaction temperature as shown in the Equation 2.4
∆Greac = ∆Hreac − T ∆Sreac (2.4)
At standard temperature and pressure, the reversible cell voltage (Urev ) is 1.23V. It is the minimum
voltage needed for the electrolysis when the reaction occurs ideally. This reversible voltage is usually
expressed by Nernst equation in terms of change in Gibbs free energy and the Faraday’s constant [26]
as shown in Equation 2.5
∆R G
Urev = (2.5)
n F
where, n is the equivalent number of material (n=2) and F is Faraday’s constant (F = 96500 C/mole)
There are advantages and disadvantages associated with usage of alkaline water electrolysis for the
production of hydrogen. Some of the advantages are [27]:
• It is a well established technology and has long term stability

• Relatively low manufacturing cost
• Usage of non noble catalysts
Some of the disadvantages are [28]:
• Low current densities and possibility of gas crossover

• Slow start-up and corrosion problems due to the electrolyte
• Complicated maintenance because of many components in the device
2.2.2. Proton Exchange Membrane (PEM) Electrolysis

The first PEM electrolyser was introduced by General Electric in the early 1960s through the work of
Thomas Grubb and Leonard Niedrach. The fundamental difference between the PEM electrolyzer and
the alkaline water electrolyzer is the employment of a thin-film electrode assembly (membrane elec-
trode) to generate a zero pole spacing [28]. The specific energy consumption of the system is around
4.4 - 5.0 kWh / Nm3 H2 . This technology of PEM electrolysis was invented to overcome the drawbacks
that were being faced whilst using alkaline water electrolyzer.
PEM electrolyzers consist of solid polysulfonated membranes (Nafion,fumapem) as an electrolyte which

conducts protons. The usage of membranes have various advantages such as lower gas permeability,
high proton conductivity, lower thickness and high pressure operations [22]. PEM electrolyzers can
be operated at a current density of 10000-20000 A/m2 which is around 5 times more than that of an
alkaline water electrolyzer. The operating temperatures are around 50-80°C and operating pressures
are around 80 bar. A typical schematic of a PEM electrolyzer is shown below in Figure 2.6
Figure 2.6: PEM Electrolyzer [22]
In the working of PEM electrolyzer, the deionized water is only sent to the anode side unlike the alkaline
electrolysis. The water is split into oxygen, hydrogen ions and electrons. These hydrogen ions move to
the cathode side through the proton conducting membrane. The electrons escape the anode through
the external power circuit, which provides the reaction’s driving force (cell voltage). At the cathode side,
the hydrogen ions and electrons re-combine to produce hydrogen gas. The following half-cell reactions
at the anode and cathode are shown below in Equation 2.6 & Equation 2.7 respectively [22].
1
H2 O −→ 2H + + O2 + 2e− (2.6)
2
2H + + 2e− −→ H2 (2.7)
The membrane separates hydrogen from oxygen and therefore the gases are sent to further treatment
and stored in later on stages.
The PEM electrolyzer has various advantages and disadvantages. Some of the advantages are:
• The system can be operated at high pressures
• The method has high energy efficiency due to the usage of a membrane and the absence of
solution voltage drop
• High dynamic operation i.e., certain to handling fluctuations in the electricity input
• High purity hydrogen due to high current densities
• the system has a rapid response time and corrosion-free environment
Some of the disadvantages associated are [27]:

• High costs of components due to usage of noble and scarce metal elements like iridium as catalyst
• Acidic environment since the Nafion tubing absorbs the OH − groups and leaves the excess H +
in the water causing the pH to decrease
• Low durability of the equipment
2.2.3. Solid Oxide Electrolysis

The working principle of Solid Oxide Electrolytic Cell is similar to that of Solid Oxide Fuel Cell (SOFC).
In a typical SOFC, the cell is used to generate electricity by directly transforming the chemical energy in
the fuel to electrical energy [29]. Whereas, a solid oxide electrolytic cell (SOEC) is a regenerative solid
oxide fuel cell that electrolyzes water using a solid oxide, or ceramic electrolyte to create hydrogen gas
and oxygen. These electrolyzers operate at very high temperatures. The specific energy consumption
of the system is around 2.5 - 3.5 kWh / Nm3 H2 . A typical schematic of SOEC is shown below in
Figure 2.7.
Figure 2.7: Schematic of solid oxide electrolysis [22]
SOECs consist of three components, a solid electrolyte and two porous electrodes. The most commonly
used electrolyte is Yttrium Stabilized Zirconia (YSZ). YSZ consists of zirconium dioxide doped with 8
mol% of yttrium oxide. This material is mainly used since it has properties that make it suitable for
high temperature processes. YSZ has high strength, high melting temperature and excellent corrosion
resistance. On top of that, it enables oxygen ion conduction while blocking electronic conduction [30].
The high resistance of YSZ electrolytes limits the electrolysis performance when the temperatures are
below 500°C and hence this process is carried out at high temperatures.
The operating temperatures of SOECs range from 500-850°C. Since the operating temperatures are
quite high, the water is fed in the form of steam to the porous cathode. When a voltage is applied to the
cell, the steam moves towards the cathode-electrolyte interface and is reduced to form pure hydrogen
gas and oxygen ions. These oxygen ions are sent to the anodic side of the cell through the dense
electrolyte and thereafter oxygen gas and electrons are produced. The produced electrons are again
sent back from the anode to the cathode by external power supply as a driving force [31]. The electrolyte
must be dense enough such that the steam and hydrogen gas cannot diffuse through thereby leading to
recombination of the H2 and O2− . The hydrogen evolution reaction and the oxygen evolution reaction
occurring at cathode and anode are given by (Equation 2.8) & (Equation 2.9) respectively.
H2 O + 2e− −→ H2 + O2− (2.8)
1
O2− −→ O2 + 2e− (2.9)
2
There are advantages and disadvantages as well which are associated with using SOECs for hydrogen
production. Some of the advantages are [32]:
• The need for electrolyte loss maintenance and electrode corrosion is eliminated because of all
the components being solid in the cell
• Expensive catalysts are usually avoided since the electrolyzers work at high operating tempera-
tures
• These cells usually have a better ability to tolerate the presence of impurities
• Because of high operating temperatures, faster kinetics and more favorable thermodynamics are
achieved
• High efficiency of hydrogen gas can be produced

• Very efficient when there is a source for steam available
Some of the drawbacks associated with SOECs are [33]:
• The materials used for components are thermally challenged and the material deterioration rate
is noticeable because of high operating temperatures
• Incompatibility to intermittent power resources (takes time to heat up the cell and to start the
process)
• Material fabrication is rather expensive and complicated
2.2.4. Comparison of different electrolysis techniques
Figure 2.8: Comparison of different electrolysis techniques [34] [35] [36]
2.2.5. Hydrogen Storage

To store hydrogen there are various technologies available. The three main categories associated with
the storage of hydrogen are shown in Figure 2.9: (1) hydrogen can be stored in pure form, molecular gas
or liquid with no extensive physical or chemical connection to other materials; (2) hydrogen molecules
are adsorbed onto a material which are held by weak Van der Walls bonds; (3) hydrogen molecules can
be chemically bonded [37]. Moreover, the chemical storage is divided into two subcategories based on
the chemical bonding: metal hydrides and chemical hydrides.
Since hydrogen gas has a very low density (1kg of hydrogen gas occupies 11 m3 at standard conditions),
it is therefore difficult to store with high efficiency. Hence the storage density of hydrogen must be
increased to make the storage of hydrogen remunerative. Due to the intermittency in the renewable
energy sources (wind and solar), it is therefore necessary for surplus hydrogen to make sure that the
entire ammonia plant operates continuously [38].
Figure 2.9: Different Technologies for hydrogen storage
Physical Storage
Physical hydrogen storage methods are primarily based on the compression and cooling/liquefaction
of hydrogen. The process is carried out by cooling down the hydrogen gas using heat exchangers.
The hydrogen gas is then stored in liquid hydrogen tanks, or in compressed gaseous hydrogen tanks
where the gas is compressed to high pressures [37]. Both of the technologies require infrastructure -
a compressor system for pressurization and cryogenics for liquefaction. Cryo-compressed hydrogen
storage system is a combination of liquid hydrogen and compressed gaseous hydrogen storage sys-
tems. The cryo-compressed storage system was first launched by BMW in 2012 and allows a relatively
high volumetric density (at 20K, from 70g/L at 1 bar to 87g/L at 240 bar) [39].
Compressed hydrogen gas storage
The compressed hydrogen gas storage system typically consists of two major components: the com-
pressors necessary to raise the pressure of the gas in order for the gas to be stored, and the storage
chambers. This technology of compressing the gas and storing is one of the most common technique
used to store hydrogen. The storage tanks are pressurized up to a pressure of 700 bars [38]. The
approach for high pressure compression is reasonably mature and easy in comparison to other tech-
niques. However the system’s weight and cost must be reduced more in the future.
The compressed gas can be stored either above ground or underground (below ground level). The
storage above the ground requires sophisticated infrastructure and therefore it is quite expensive in
terms of investment costs. As a result, it is possible to store the compressed gas underground by the
means of salt caverns [38]. A basic schematic of how a salt cavern storage facility looks like can be
seen in Figure 2.10.
Figure 2.10: Underground H2 storage facility [40]

2.3. Nitrogen Production & Storage 16
Metal containers are generally used to store natural gas but could also be applied for the storage of
hydrogen. Three different types of metallic containers are used for natural gas storage. They are:
• Gas holders, with storage pressures just above the atmospheric pressure
• Spherical pressure vessels, with storage pressures approximately up to 20 bar
• Pipe storages, with storage pressures up to approximately 100 bar [38]
Some of the advantages and disadvantages of storing hydrogen in compressed gaseous form are:
+ Well established technology and commercially available

- High energy input required because of use of compressors
- The storage densities are low leading to need for large storage specific volumes of storage tanks
- High investment costs and expensive to store high pressure gas
2.3. Nitrogen Production & Storage

The air in the atmosphere is composed of a mixture of several gases. The most important of these are
nitrogen (N2 ) and oxygen (O2 ), which account for approximately 99.03% of the total sample volume.
There are few other gases present in trace amounts due to natural processes and human activities.
Water vapour is also present in the air. The water content of air fluctuates significantly with local ambient
conditions. Dry air contains around 78.10% nitrogen, 20.90% oxygen, and 0.933% argon by volume
along with traces of other gases like hydrogen, helium, neon, krypton, xenon, carbon dioxide etc [41].
The production of oxygen and nitrogen is accomplished by an air separation process, which involves
the separation of air into a stream containing mostly nitrogen and an oxygen enriched stream.
Air separation is typically classified as cryogenic and non-cryogenic. The non-cryogenic methods in-
clude pressure swing adsorption and membrane separation. A detailed description of each process
type is given in the following subsections.
2.3.1. Cryogenic Distillation

Gases such as oxygen, nitrogen, and argon are created from the air that surrounds us. Separation of
the naturally occurring air components allows us to develop their particular features and employ them
in a variety of applications. These gases are usually produced in Air Separation Units (ASUs) that use
cryogenic distillation. Cryogenic distillation is a process where the gases are cooled down to extremely
low temperatures therefore separating them in the form of liquid [42].
The plant consists of a collection of equipment such as
• Distillation columns
• Heat exchangers
• Adsorbers
• Gas and liquid compressors
• Turbine and supporting apparatus for control
As per the requirement, the component gases are sold and delivered to clients for a wide range of
industrial, medicinal, and other specialized uses. The core technique for large-scale air separation has
stayed stable for decades, however equipment orientation will alter depending on process needs. A
basic overview of various steps associated with the cryogenic distillation process is shown below in
Figure 2.11 based on [43].
Figure 2.11: Overview of cryogenic distillation
The various steps involved in cryogenic distillation of air are:
• Pre-treatment, Compression and cooling of air

• Removal of carbon dioxide
• Heat exchange to reduce the temperature of feed air to cryogenic temperatures
• Distillation of air
Pre-treatment, compressing and cooling the feed air is the first step in cryo-distillation air separation
unit. Air is first purified by sending it through filtration devices for removal of dirt. The feed air is then
compressed till 5-10 bar and is then sent to the next step [44]. The compression of air is a necessary
step in order to ensure the efficiency of the system and increasing the purity of air. Thereafter, the
compressed air is sent to several stages of intercoolers for cooling and much of the water vapour is
condensed and removed.
Removal of Carbon dioxide and Water vapour is a critical step to do before releasing the air down-
stream. This is done because at low temperatures, carbon dioxide and water would ultimately freeze
and block the apparatus. As a result, remaining water vapour and carbon dioxide must be eliminated
to match the product’s specifications. The most widely used purification technology is molecular sieve
adsorption, in which water and carbon dioxide molecules are adsorbed onto the surface of a molecular
sieve at near-ambient temperatures. [44].
Additional heat exchange is carried out against the cold product and waste gas streams to bring the
air feed to cryogenic temperatures. The cooling is performed by the use of Brazed Aluminum heat
exchangers, which are utilized to cool the incoming air flow to a temperature adequate for downstream
components (boiling point of nitrogen is 77 K). The outgoing gas streams are warmed to a temperature
close to that of ambient air. The heat exchange between feed and product streams reduces the plant’s
net refrigeration load and, as a result, energy consumption.
Distillation of air is done after the heat exchange. Part of the air liquefies to form a liquid that is
enriched in oxygen after the heat exchange. The remaining gas that is richer in nitrogen is distilled
to almost pure nitrogen in a distillation column. In order to separate nitrogen usually one distillation
column is needed but if the required purity of Nitrogen is very high then two distillation columns are
used.
There are some advantages and disadvantages associated with using cryogenic distillation for nitrogen
production. Some of the advantages are [45]:
• Most efficient process for large scale production of gases compared to membrane separation and
pressure swing adsorption
• Yields high purity products
• Liquid forms of cryogenic gases are easier and cheaper to transport
However, the major drawbacks associated with cryogenic distillation are [46]:
• Large scale utility and space required for economic feasibility

• High capital costs
• Longer time required for start-up or shut down
• High energy costs to cryogenically cool gases
• Indirectly causes the release of significant quantity of greenhouse gases due to huge power con-
sumption
2.3.2. Pressure Swing Adsorption

Pressure Swing Adsorption (PSA) is a non-cryogenic process used for the extraction of nitrogen from air.
These systems work on the principle of adsorption by trapping the oxygen molecules from compressed
air to produce nitrogen. The most commonly used adsorbent is Carbon Molecular sieve (CMS) for PSA
applications. Carbon molecular sieves have pores of molecular dimensions which provide a relatively
high adsorption capacity and kinetic selectivity for various gases. The PSA approach is based on
the difference in kinetics of oxygen and nitrogen adsorption, with oxygen adsorption being substantially
quicker than nitrogen adsorption [47]. Three different stages are involved in this process and are shown
below in Figure 2.12
Figure 2.12: Pressure Swing Adsorption [48]
When compressed air passes over the material, oxygen molecules are adsorbed into the pores. At
some point the carbon molecular sieve is saturated and the gas separation stops. As a result, PSA
generators are always equipped with two or more adsorption columns to ensure that the process is
continuous. These two vessels undergo switch between the separation process and regeneration
processes. The clean and dry compressed air enters the first tower (First part of Figure 2.12) and since
the oxygen molecules are smaller than nitrogen molecules, the oxygen molecules will enter the pores
of carbon sieve. Nitrogen molecules on the other hand cannot fit into the pores so they will bypass the
carbon molecular sieve. As a result, nitrogen of desired purity is obtained. This phase of separation of
nitrogen is called the adsorption or separation phase. However, the process does not stop here. Most
of the nitrogen produced in the first vessel exits the system which is ready for direct use or storage. A
small portion of the generated nitrogen is led to the second vessel from the top [49].
This flow of nitrogen into the second vessel is required to push out the oxygen that was captured in the
previous adsorption phase of second vessel. By releasing the pressure in second vessel, the carbon
molecular sieves lose their ability to hold the oxygen molecules. The process occurs in the second
part of Figure 2.12. The oxygen molecules will detach from the sieves and get carried away through
the exhaust by the small nitrogen flow coming from first tower. Hence the system makes room for
new oxygen molecules to attach to the sieves in a next adsorption phase. The process of removing
saturated oxygen particles from the tower as regeneration process [50]. After the regenerative process,
the pressure in both towers will equalize and they change stages from regenerating to adsorbing and
vice versa. The CMS in first tower gets saturated, while in second tower due to depressurization is able
to start the adsorption process again as seen in the third part of Figure 2.12.
Some of the advantages of using pressure swing adsorption for nitrogen separation are [51]:
• Low to average capital investment costs

• Easy installation and start up of the equipment
• Relatively high purity
• Low operation costs
• Easy scale-up by adding extra unit columns and the process doesn’t require any chemicals or
solvent
Some of the drawbacks associated with PSAs are [46]:
• Large maintenance equipment and production capacity is limited

• Slow cycle times can give high losses and can cause the rate change of inlet flow, leading to the
unstable pressure in the column during the plant operation [52]
• Usually requires a buffer storage vessel thus increasing the costs associated with it
2.3.3. Membrane Separation

Membrane separation is a method that separates nitrogen from air by using a selective membrane in
the form of a polymer membrane which is the heart of equipment. It is based on the notion of selective
permeation of gases through a porous medium. The membrane is made up of thousands of hollow
fibers that allow compressed air to flow through. Gas molecules can pass through the walls of each
fiber, however some gases pass through more easily than others [51]. Permeability and separation
selectivity (α) are the two main criteria used to evaluate membrane performance for gas separations.
The permeability of glassy polymeric membranes is affected by pressure and temperature when large
input flow rates and membrane surfaces are utilized. The vapor pressure and density, which also impact
permeability, are very vulnerable to temperature fluctuations [53]. The purity of the nitrogen gas that is
usually extracted from the outlet of the separator highly depends on the pore size of the membrane. A
typical schematic of membrane separation technique is shown below in Figure 2.13
Figure 2.13: Schematic of Membrane Separation nitrogen production [54]
The figure above shows the classification between fast and slow gases. Fast gases, such as oxygen,
CO2 , H2 , and water vapour, flow through the fiber walls and are vented into the environment. The
slow gas, nitrogen, moves significantly slower through the fiber wall, resulting in a high purity nitrogen
stream at the membrane exit. In the membrane separation process, there are no moving parts to the
membrane. As a result, high purity nitrogen can be obtained by controlling the pressure and flow rate
of compressed air through the membrane [55].
Some of the advantages associated with using membrane separators for nitrogen production are:
• The capital cost is inexpensive in comparison to the purity level that the generator will create, and
the generator is ready to perform the task and begin operating in a matter of seconds
• Since there are no moving parts, it is easy to control and maintain
• Can be easily expanded and no regeneration process required [56]
However, some of the drawbacks associated with the technique are:
• May cause clogging due to impurities in the gas stream thus high periodic maintenance costs
• Not suitable for very high purity outputs (99.9%)
2.3.4. Comparison of different air separation techniques
Figure 2.14: Various air separation techniques comparison [51] [36]
2.3.5. Nitrogen Storage

Nitrogen is usually stored in the liquid form since it can be transported relatively easily. Dewar vessels
and vacuum insulation were originally produced for the storage and transport of small quantities of liquid
nitrogen. A dewar is a specialized type of an insulated container which is used for storing cryogens.
All dewar vessels have multiple layers of walls and a high vacuum is maintained between the walls.
The presence of vacuum ensures that the vessel is properly insulated and no heat transfer takes place
between the interior of the vessel and surroundings. The vacuum insulation is done to reduce the boil
off rate of liquid nitrogen [57]. A general overview of how dewar vessels look like can be seen in the
Figure 2.15 below.
2.4. Ammonia Production: Haber-Bosch Process 21
Figure 2.15: Dewar vessels for storage of cryogens [58]
2.4. Ammonia Production: Haber-Bosch Process

The Haber-Bosch process which is also referred to as Haber process is the primary method applied to
in the manufacturing of ammonia. The process is named after the two German chemists, Fritz Haber
and Carl Bosch, who invented the process in early 1900s [59]. Fritz Haber first developed the process
and later this process was purchased by a German company BASF (Badische Anilin & Soda Fabrik
with headquarters in Ludwigshafen, Germany).
Later the process was scaled-up by Carl Bosch and succeeded in 1910. The main idea of the process
is that it converts nitrogen to ammonia with the presence of hydrogen under extremely high pressures
and temperatures in the presence of a metal catalyst. The reaction carried out at equilibrium is shown
below.
N2 + 3H2 −→ 2N H3 ∆H° = −91.8 KJ/mol
The reaction scheme for the catalytic ammonia synthesis comprises of number of primary steps. Ac-
cording to Ertl [60], the series of reactions are:
H2 + 2∗ ⇐⇒ 2Had
N2 + 2∗ ⇐⇒ 2Nad
Nad + Had ⇐⇒ N Had + ∗
N Had + Had ⇐⇒ N H2,ad + ∗
N H2,ad + Had ⇐⇒ N H3,ad + ∗
N H3,ad ⇐⇒ N H3 + ∗
A general overview of Haber-Bosch process is expressed as a block diagram in Figure 2.16
The below diagram represents the production of green ammonia which refers to using hydrogen pro-
duced from renewable energy sources. The process of production of ammonia is typically carried
out at pressures more than 100 bar and temperatures ranging from 400-500°C and are passed over
beds of catalyst. The recycle of the unreacted gases enhances the conversion rate to around 97% [61].
Nitrogen shares a triple covalent bond and has a very high stability. Due to the high stability, the
bond between the nitrogen molecules is strong which keeps the nitrogen molecules (N2 ) together.
Therefore, high dissociation energy of 942 KJ/mol is required to break the triple N ≡ N bond. As a
result, a catalyst is required to break the bonding between nitrogen molecules. Without a catalyst the
process would require very high temperatures and pressure thereby reducing the efficiency of the plant
[62]. Heterogeneous catalysts are used where solid catalyst is in contact with gaseous reactants. To
make the process happen in an efficient way, iron based catalysts are used in Haber-Bosch process.
Iron-based catalysts are used to increase the speed of the reaction and are cheap.
These iron based catalysts consist of promoters which usually are aluminium oxide, potassium hydrox-
ide, calcium oxide, magnesium oxide etc. These promoters are used to boost the effectiveness of the
catalyst in the process and to keep the surface area of the catalysts stable. The promoters are usually
unreactive with hydrogen and therefore do not cause any harm to the reaction [63].
Figure 2.16: Schematic route of Haber-Bosch process
The reaction is a reversible reaction, that is, it can proceed both in forward direction (ammonia syn-
thesis) and in backward direction (ammonia decomposition). The Le Chatlier’s Principle explains the
temperature and pressure dependency on the reaction and why the pressure should be high for efficient
reaction.
• Pressure: For gaseous reactions, gas pressure is related to the number of gas particles in the
system. Increasing the pressure on the equilibrium system will result in shift of equilibrium to-
wards the side with fewer molecules. Therefore, increasing the pressure gives rise to higher
mole fraction of ammonia.
• Temperature: The effect of change in temperature depends on whether the reaction is exother-
mic or endothermic. In case of ammonia synthesis reaction (exothermic), increasing the tem-
perature will result in shifting the equilibrium position to the right. Therefore, the ammonia mole
fraction decreases as the temperature is increased.
• Concentration: If the concentration of the reactant is increased, the equilibrium position shifts
to use up the added reactants by producing more products.
The pressure and temperature dependency on ammonia mole fraction is shown in Figure 2.17
Figure 2.17: Mole fraction of ammonia at equilibrium as a function of temperature and pressure (1:3 mixture of N2 / H2 ) [64]
Ammonia Recovery
Ammonia recovery is a process where ammonia is separated from the unreacted gases (i.e., hydrogen
and nitrogen) after the well known Haber-Bosch process. Two different ways to recover ammonia are
discussed in the current thesis report. One of the most commonly used method to recover ammonia is
by condensation, and the other way to recover ammonia is by absorption. The latter is still in develop-
ment phase and an extensive research is being carried out. A general overview on both the processes
is given below.
Ammonia recovery by Condensation

Ammonia recovery by condensation is one of the commonly used method to separate ammonia from the
unreacted gases in the ammonia synthesis loop. Ammonia recovery by condensation is usually carried
out by cooling the gases coming out of the reactor to temperatures ranging from -25°C to -33°C with
pressures around 140 bar [7]. The unreacted gases (H2 & N2 ) are recycled back to the reactor after
being compressed back to the reaction conditions. The mole fraction of the ammonia recovered highly
depends on the condensation temperature. The phase diagram of ammonia as shown in Figure 2.18
depicts the pressure dependency on temperature [65].
Figure 2.18: Phase diagram of ammonia [65]

The critical point of ammonia is at 132.25°C and 113.39 bar. The partial pressure is expressed in terms
of total pressure and mole fraction as:
pi = pa × X i (2.10)
where pa is the total pressure of the system and Xi is the mole fraction of the component.
The recovery rate of ammonia depends on the ammonia partial pressure and the saturation pressure.
As a result, if the total pressure is high and the saturation pressure is low, the conversion of ammonia
increases. This saturation pressure depends on temperature and the only way to reduce the pressure
is by reducing the temperature. Therefore, the plants are usually run at -25°C to -33°C [7] thereby
reducing the ammonia yield in the recycle stream.
Ammonia recovery by absorption
A recovery unit of ammonia consists of a condenser after the reactor where it is cooled down to very
low temperatures as discussed in Figure 2.4. Another alternative that can replace the condenser and
is widely being researched is the use of solid absorbents [66] [67] [68] [69]. An absorbent that collects
ammonia in a solid crystal is preferable in such a way that it can be done even at high temperatures
[66]. However, the absorption might take longer to achieve its full capacity compared to cryogenic
condensation.
The solid absorbent material used is responsible for the selectivity of the absorption where ammonia
is absorbed but the hydrogen and nitrogen gases are not. These absorbents incorporate molecules
into the crystal lattice thereby offering high capacity and being more stable at high temperatures [67].
Because of their ability to retain multiple moles of ammonia per mole of salt, metal halide salts such as
magnesium chloride or strontium chloride have been proven to be viable candidates for N H3 storage
materials.
3
Methodology
3.1. Scenarios
The ammonia production rate of a green electricity powered plant is subjected to fluctuations due to
variations in hydrogen availability. These fluctuations in hydrogen availability are caused by intermit-
tency of renewable power supply to the electrolyzer. Due to these fluctuations in hydrogen feed flow
rate, a stable and efficient production process cannot be achieved. Changes in hydrogen feed flow rate
has a direct impact on the pressure within the ammonia synthesis loop. A reduction in the hydrogen
feed flow rate can lead to a reduction in pressure and these extreme changes in pressures lead to
metal fatigue and result in damaging the components in the production unit. Therefore, the key goal
of the current thesis project is to test potentially effective control strategies that can stabilize pressure
within the ammonia synthesis loop under varying hydrogen feed flow rate.
Under various conditions, three distinct control methods and the effect of change in hydrogen feed flow
rate on the ammonia synthesis plant are examined. The effect of these variations in hydrogen feed
flow rate on ammonia synthesis plant are discussed in detail in chapter 5.
The electrolyzer and the ammonia synthesis unit are powered by renewable energy sources in order to
produce green ammonia. Three different scenarios are considered where the effect of solar, wind and
hybrid (solar+wind) on the ammonia production are studied and a cost evaluation is carried out. These
scenarios are explained in detail in chapter 6. A brief cost estimation is carried out using the HySupply
open source tool and the levelized cost of ammonia is determined for each of the selected scenarios.
The effect of introduction of hydrogen buffer on ammonia production capacity factor and levelized costs
are also examined further.
3.2. Assumptions
The following assumptions have been made for construction of the model:
• The electrolyzer is assumed to be handling the power fluctuations efficiently
• Frictional pressure drop in the synthesis loop is assumed to be constant at 3 bar
• Hydrogen coming out of the electrolyzer is assumed to be pure hydrogen
• The production and storage of feed components (H2 & N2 ) is not modelled in the simulation
• The storage of ammonia is not modelled in the present study
• An electric heater is integrated in the ammonia synthesis reactor
3.3. Modeling and Simulation Software

For the current thesis project, Aspen Plus, Aspen Plus Dynamics and HySupply Open Source Tools
[13] were used for modeling and simulation purposes. Aspen Plus and Aspen Plus Dynamics are part
of AspenTech (Aspen Technology) who is a provider of software and services for process industries.
25
3.3. Modeling and Simulation Software 26
Aspen Plus is primarily used for steady-state modeling and incorporates a comprehensive thermody-
namic property database enabling users to select appropriate models for reaction kinetics and physical
property calculations.
Aspen Plus Dynamics focuses on dynamic simulation where it allows users to model and simulate pro-
cesses over time, taking into account the transient behaviour, start-up, shut-down and control strate-
gies.
HySupply is a collaboration between Germany and Australia to investigate the feasibility of exporting
renewable energy in the form of hydrogen. As part of feasibility study, HySupply Australia developed
a series of open-source and open-access costing tools to assess the viability of the supply chain. For
the current research project, HySupply Hydrogen Analysis Tool and HySupply Ammonia Analysis Tool
were used to calculate the levelized costs of hydrogen and ammonia using renewable energy sources
as primary electricity input.
4
Steady State Simulation
The aim of the steady-state model is to find the thermodynamic limitations and sizing of the components
in the ammonia synthesis plant.
4.1. Model
The reaction kinetics and sizing of various equipment have been included in the current steady state
model. In addition to these, the heating duty, cooling duty, and heat integration possibility are investi-
gated and provided in the simulation. This specific model was built with dynamics of each component
considered along with addition of valves as necessary. Thereafter, the steady state model is sent to
the dynamic environment for simulations in Aspen Plus Dynamics. Figure 4.1 shows the conceptual
design of the ammonia synthesis unit.
Figure 4.1: Conceptual Design of Ammonia Synthesis Plant
The following Figure 4.2 shows the modelled conceptual ammonia synthesis plant that was simulated
in Aspen Plus and the configuration of each of the components shown in the figure below are described
in the following sections.
27
4.2. Reactor Configuration 28
Figure 4.2: Steady-State Model from Aspen Plus
4.2. Reactor Configuration

For the current work, R-PLUG reactor is used in Aspen Plus to model the ammonia synthesis reactor.
This specific type of reactor is used since it enables the user to model the catalyst bed. Ammonia
synthesis is a process which is carried out under the presence of an Iron based catalyst. The reaction
kinetics have been provided by Proton Ventures B.V. and the same have been implemented in the
current project work. For this purpose, Langmuir-Hinshelwood-Hougen-Watson model has been used
and the reaction kinetics are derived from [68]. For the sake of simplicity, the ratio of reactor length to
diameter is taken as 8 based on data from Proton Ventures B.V. since the reactor modelling was out of
the scope for the present work.
A sensitivity analysis has been performed on the reactor to determine the optimal length of the reactor
based on the selected flow rate as shown in Figure 4.3 and Figure 4.4.
Figure 4.3: Sensitivity analysis of R-PLUG reactor at 300◦ C and 150 bar
4.3. Flash Vessel 29
Figure 4.4: Sensitivity analysis of R-PLUG reactor at 400◦ C and 150 bar
The sensitivity analysis provides the following observations.
• Adiabatic reactor reaches the conversion much faster than isothermal reactor. This is because
there is a thermodynamic equilibrium of heat and the temperature increases in the reactor. Due
to this increase in temperature, the reaction goes faster.
• At higher temperatures, the isothermal reactor performs better due to the influence of temperature
on ammonia synthesis kinetics, whereas the adiabatic reactor performs better at lower tempera-
tures.
• The inlet temperature of the feed going into the reactor plays a major role in determining the
conversion of ammonia.
For the present work, adiabatic reactor is chosen for the reactor configuration since isothermal config-
uration is difficult to achieve and maintain. The inlet temperature of the feed is taken as 300◦ C and
the length and diameter of the reactor considered are 7.2m and 0.9m respectively from the sensitivity
analysis where a conversion of 19.7% is achieved thereby, validating the model according to [70] which
states that a single pass conversion of 18-20% is achieved.
The catalyst used for the present work is Ferrous Oxide based catalyst (ZA-5) since it has a lower
activation temperature of 300◦ C and high heat resistance (temperatures till 500◦ C) [68]. By using ZA-5
catalyst under the conditions of 8 MPa of pressure and reaction temperatures at the inlet and exit of the
reactor of 215 °C and 363 °C, respectively, the net value of ammonia is more than 10% in the industrial
process of high-purity ammonia, which has met the economic requirement for net value of ammonia
in industry [71]. Further, a bed voidage of 0.33 and particle density of 2200 kg/m3 were considered
based on literature [72].
4.3. Flash Vessel

The flash vessel which is used as a separator to separate gas from liquid is modelled for the present
work. The dimensions of the flash vessel have been calculated as it was observed that the dimensions
play a crucial role in dynamic simulation of the model. In order to calculate the dimensions, the terminal
velocity (ut ) must be known. The terminal velocity of the stream is expressed as Equation 4.1:
√
4 × g × d (ρl − ρv )
ut = (4.1)
3 × C D × ρv
4.3. Flash Vessel 30
Bent Wiencke [73] suggested various expressions for drag coefficient (CD ) but for the present work,
the equation suggested by Brown and Lawler has been considered as shown in Equation 4.2.
24 0.407
CD = (1 + 0.150Re0.681
p ) + f or Re ≤ 2 × 105 (4.2)
Rep 1 + 8710
Rep
To perform the calculations, the density of vapor going into the separating vessel is taken from Aspen
Plus. The diameter of the droplet of ammonia is assumed to be 100µm based on the range of 50-100
µm as stated in [8]. The viscosity of the gas is calculated based on the molar composition of the gases
in the stream. Further, the volumetric flow rate of the feed stream to the vessel is also taken from Aspen
Plus.
In order to determine the Reynolds’ number, an initial guess of terminal velocity is done in order to
calculate the drag coefficient (CD ). After calculating the drag coefficient, the value is used to find the
actual terminal velocity. This process is carried out till both the predicted terminal velocity and the actual
terminal velocity match. The terminal velocity is found to be 0.083 m/s.
Once the terminal velocity is found, the diameter of the vessel is calculated by using the Equation 4.3
√
4 × Q̇
D= (4.3)
π × ut
From the above values and calculations, the diameter of the vessel is found out to be 0.54m. As a rule
of thumb, the Length:Diameter ratio for separator vessels is in the range of 3-5 [74]. For the current
project, a length-to-diameter ratio of 4 is considered. Hence, the height of the vessel is 2.16m.
(a) Horizontal Vessel (b) Vertical Vessel
Figure 4.5: Schematic of types of separator vessels
Figure 4.5 represents the basic schematic of different types of vessels. Horizontal separation vessels
are usually used when the liquid flow rate is dominant compared to the gas flow rate. In the present
work, vertical separator has been used because of liquid flow rate being less dominant and is more
efficient in terms of separation since the densities difference between the liquid ammonia and gases is
high.
The specifications of the flash vessel are described in the Table 4.1 below:
4.4. Heat Exchanger 31
Parameter Value
Duty 0 kW
Pressure Drop 0 bar
Phases Vapor-Liquid
Vessel Type Vertical & Elliptical
Length 2.1642 m
Diameter 0.54106 m
Heat Transfer Type Constant Duty
Liquid Volume Fraction 0.35
Table 4.1: Flash Vessel Specifications
4.4. Heat Exchanger

Heat exchangers are one of the key components to realistically represent the working of the production
plants. A shell and tube exchanger is commonly used in industries to transfer heat between two pres-
surized fluids while keeping them physically separate. Two fluids are involved in the process known as
the ”hot fluid” and the ”cold fluid”.
The cold fluid outlet temperature is specified as 300◦ C in design type calculation mode. In steady-
state modelling, the heat exchanger area is calculated. The calculated heat exchanger area is 11.708
m2 . This value of heat exchanger area is used as input for dynamic model. The calculation mode
is selected as simulation and the evaluated heat exchanger area is specified as unit specification for
dynamic simulation.
Based on the literature [75], the allowable pressure drop for gases is taken as 0.1 bar on both the cold
and hot side of heat exchanger. Table 4.2 shows the results obtained from steady-state simulation of
heat exchanger.
Parameter Value Unit

Heat Exchange Area 11.708 m2
Heat Transfer Coefficient 850 W /m2K
Heat Duty 1500 kW
Cold Stream Outlet Temperature 300 ◦C
Hot Stream Outlet Temperature 190 ◦C
Table 4.2: Heat Exchanger Results
4.5. Compressors
Two different compressors are used for the current ammonia synthesis plant. A reciprocating compres-
sor is used to compress the hydrogen and nitrogen feed gases. This reciprocating compressor is used
for make-up gas compression since it has better performance with high pressure ratios and intermittent
operations compared to centrifugal compressor. The make-up gas is compressed from 80 bar to 150
bar. The compressor model is selected as isentropic compressor and a pressure ratio is specified as
the input for the make-up gas compressor. The outlet temperature of the compressor is found out to be
105◦ C and therefore, only a single stage compression is used. This is because the maximum allowable
discharge temperature of the compressor is usually less than 150◦ C due to durability of the seal and
thermal stress on the compressor. The results of the make-up gas compressor are shown below in
Table 4.3.
4.6. Valves 32
Parameter Value Unit

Brake Power 87.7 kW
Isentropic Efficiency 0.72
Compressor Outlet Temperature 105 ◦C
Pressure Ratio 1.88

Outlet Pressure 150 bar
Table 4.3: Make-Up Gas Compressor Results
A centrifugal compressor is used for the recycle stream compression to bring the pressure back to the
system pressure (150 bar). A centrifugal compressor is used as it is effective for low to medium pressure
ratios and for high flow rates. The unit specification for the recycle compressor is isentropic type and
the discharge pressure has been specified as 150 bar. The brake horsepower of the compressor is
determined from the steady-state simulation as 12.5kW.
4.6. Valves
Valves are mechanical devices which are used to control the flow of fluids through a system. For the
present work, valves are used to regulate the flow rates of the fluids within the system. By adjusting
the position of the valve, the amount of fluid flowing through the pipeline can be controlled.
From Figure 4.2, H2-V and N2-V are used to control the flow rates of hydrogen and nitrogen respectively
in dynamic simulation. NH3-V is used to regulate the liquid level in the flash vessel. The PURGE-V is
used to send impurities from the system to the flare where they are burnt off. Each valve have their
own specification and are discussed in the Table 4.4 below. The calculation type used for all the valves
is adiabatic flash for specified outlet pressure (pressure changer).
Valve Specification Value (bar)

H2-V Pressure Drop 0.5
N2-V Pressure Drop 0.5
NH3-V Outlet Pressure 50
PURGE-V Pressure Drop 0.5
Table 4.4: Valves Specification
4.7. Condenser
A condenser is a device which is used in various chemical plants to convert a vapor or gas into liquid
state by releasing heat from the hot stream. In the present work, the condenser cools the ammonia
into liquid form to -20◦ C. The specifications for the condenser are temperature and pressure. In the
dynamic operation, the heater type selected is instantaneous and the heat transfer option is selected
as constant process temperature. A detailed specification of condenser can be seen in the Table 4.5
below.
4.8. Other elements in MODEL-2 33
Parameter Value
Temperature −20◦ C
Pressure Drop 0.1 bar
Cooling Duty -1209 kW
Heat Transfer Constant Process Temperature
Heater Type Instantaneous
Table 4.5: Condenser Specifications
4.8. Other elements in MODEL-2

Mixer-1 & Mixer-2: These mixers are assumed to be instantaneous in response and no other speci-
fications have been provided. The pressure drop along the mixers is assumed to be 0 bar. The valid
phases input specification for both the mixers is specified as Vapor-Only.
Pre-Heater: The pre-heater before the reactor is placed to make sure that the reactor inlet temperature
of the stream is maintained at 300◦ C. A pressure drop of 0.1 bar is specified with a vapor-only valid
phase. The heater type is assumed to be instantaneous and the heat transfer option chosen was
constant process temperature.
Splitter: A splitter is used to split the purge stream from the recycle stream. This splitter has a specified
split fraction to the purge stream as 1% of the recycle stream. The pressure drop is assumed to be 0
bar and the valid phases specification is selected as vapor-only.
5
Dynamic Simulation
The current chapter describes various control strategies implemented in Aspen Plus Dynamics and the
results obtained from each of them. The steady state model was used as the starting point and is sent
to dynamic simulation.
5.1. Scenarios
The three different scenarios that were modelled for dynamic simulation in the current thesis project
are 20% reduction, 50% reduction and 70% reduction in hydrogen feed flow rate. These scenarios are
considered to be extreme and is stopped at 70% since it is considered to be the highest ramp-down
that the compressors can handle. It is observed that there is a reduction in pressure in the system and
therefore the goal is to control the pressure back to the original value even if there is a reduction in feed
flow rates. The three different control strategies implemented are:
• Pressure control by controlling the cooling duty of the condenser

• Pressure control by controlling the brake power of the recycle compressor
• Pressure control by changing the stoichiometric ratio of N2 : H2 feed
The response of the three control strategies to 50% and 70% reduction in hydrogen feed flow rate can
be found in Appendix A, Appendix B and Appendix C.
For the present work, two different types of change in hydrogen feed flow rate are implemented in Aspen
Plus dynamics i.e., step-change and linear change (ramp). Step change is a sudden and instantaneous
change in a process variable. This abrupt change is often used to simulate disturbances that can occur
due to equipment malfunctions, sudden variations in feed conditions, or operator interventions. Step
changes can help evaluate how well the control system responds to sudden disturbances and how
quickly the system returns to a new steady state. In the current thesis work, a step-change is carried
out by changing the set point in the controller and running the simulation. The simulation is carried out
for 100 minutes when hydrogen feed flow rate is reduced as a step-change.
A ramp change, on the other hand, involves gradually changing a process variable over time. Instead
of an instantaneous jump, the variable is adjusted at a constant rate. Ramp changes can represent
scenarios where a certain parameter is slowly changing due to external factors. This type of disturbance
is useful for studying the dynamic response of the control system to more gradual disturbances. For the
current simulations, timedata in aspen plus dynamics is used to model the linear changes in hydrogen
feed flow rate. Table 5.1 shows the timeline of the dynamic linear change simulations. Four distinct
phases in time are incorporated and they are labelled as A, B, C, D in each of the plot.
34
5.2. Dynamic Simulation Preparation 35
Time (min) Action

0-30 Hydrogen feed is linearly ramped down
30-90 New steady state value is maintained for 60 min
90-120 Hydrogen feed is linearly ramped up back to original value
120-150 Simulation ends at 150 min
Table 5.1: Timeline of Simulation for Linear Change
5.2. Dynamic Simulation Preparation

The first step in the dynamic model preparation is to turn on dynamic mode in Aspen Plus and specify-
ing pressure drops at each and every component. There are two different types of dynamic simulations,
Flow-Driven Simulation & Pressure-Driven Simulation. For the current project, pressure-driven simula-
tion has been selected since manipulation of pressure is considered to be the primary concern.
5.3. Control Philosophy-1

The operational principle of control philosophy-1 is to change the condenser’s cooling duty in order
to maintain system pressure. The controller measures the downstream pressure of the reactor and
adjusts the condenser’s cooling duty to retain the pressure back to its initial value. In addition to the
pressure controller, a ratio controller is used to maintain the stoichiometric ratio of H2 :N2 at a value of
3. Figure 5.1 depicts the process flow diagram and controller addition, while Table 5.2 describes and
parameters of controllers employed in the current control philosophy.
Figure 5.1: Control Philosophy-1 from Aspen Plus Dynamics

5.3. Control Philosophy-1 36
Proportional Gain (%/%) Integral Time (min) Controller Action Controlled Variable Manipulated Variable
FLOW-C 1 1 Reverse Hydrogen Feed Flow Rate H2-V Opening
PRESSURE-C 1 1 Reverse Downstream Pressure of Reactor Condenser Duty (MW)
FLASH-LC 10 1 Direct Flash Liquid Level NH3-V Opening
RATIO-C 1 1 Reverse Stoichiometric Ratio N2-V Opening
Table 5.2: Control Philosophy-1 controller parameters and details
5.3.1. Step-Change in Hydrogen Feed to Control Philosophy-1
In the current section, an extreme scenario is assumed where there is a sudden reduction in hydrogen
production. In this scenario, it is assumed that there is no hydrogen buffer and there is an instant
reduction of hydrogen feed flow rate to the ammonia synthesis plant. Therefore, to understand the effect
of instant reduction of hydrogen, a sudden reduction in electrolyzer’s hydrogen generation capacity
to 80%, 50% and 30% were assumed. These scenarios were modelled as 20%, 50% and 70% of
hydrogen feed flow rate reduction.
Figure 5.2 illustrates different parameters plotted for 20% reduction in hydrogen feed flow rate. The
response is as follows:
(a) 20% Reduction in Hydrogen and Nitrogen Feed Flow Rates

(b) Effect of 20% Reduction of Hydrogen Feed Flow Rate on Ammonia Production Rate and Pressure at Reactor Outlet
(c) Effect of 20% Reduction in Feed Flow Rate on Cooling Duty of Condenser and Temperature of Condenser Outlet Stream
Figure 5.2: Control Philosophy-1 response to 20% step change in hydrogen feed flow rate
Results:
• Figure 5.2 (a) depicts the dynamic interaction between hydrogen and nitrogen feed flow rates
throughout a 100-minute time interval. The interaction between these two variables in response
to variations in the hydrogen feed rate is the key observation from this graph. As the hydrogen
feed rate falls, the ratio controller actively adjusts the nitrogen feed rate to maintain a constant sto-
ichiometric hydrogen-to-nitrogen ratio. This real-time control technique ensures that the system
maintains the proper chemical composition even when hydrogen input rates fluctuate.
• Figure 5.2 (b) depicts the relationship between ammonia production rate and the effect of reduc-
ing the hydrogen input flow rate on system pressure. Notably, introducing a 20% reduction in
hydrogen feed flow rate into the system results in a corresponding 20% reduction in ammonia
production rate. This relationship emphasizes the system’s sensitivity to variations in the hydro-
gen feed rate, which affects reactant availability for ammonia production. Furthermore, the graph
shows an immediate pressure reduction of 2 bar, which corresponds to a decrease in hydrogen
supply rate. However, the control system reacts quickly to this pressure drop, and the system
successfully stabilizes the pressure at the reactor exit in about 20 minutes.
• Figure 5.2 (c) depicts the relationship between the condenser’s cooling duty and the temperature
of the stream leaving the condenser. The major goal here is to control the system pressure
by adjusting the condenser’s cooling duty. The inverse relationship between cooling duty and
temperature of condenser’s outlet stream is shown in this graph. As the cooling duty is reduced
from -1.2 MW to -1.05 MW, the temperature of the outlet stream rises noticeably, from -20°C
to -1°C. Pressure is projected to decrease as the hydrogen input flow rate decreases. As a
result, the pressure control senses the deviation in pressure from the set point. Therefore, the
pressure controller reduces the cooling duty, resulting in a rise in the temperature of the outlet
condenser stream. The thermodynamic link between pressure and temperature is responsible
for this phenomena where increase in temperature resulted in increase in pressure.
5.3.2. Linear-Change in Hydrogen Feed to Control Philosophy-1
The incorporation of buffer storage between the electrolyzer and the ammonia synthesis plant repre-
sents a pivotal advancement in the approach. It serves as a crucial mechanism for gradually reducing
the hydrogen supply. In this particular section, simulations were carried out in which hydrogen feed
flow rate is lowered linearly, while observing how the system responded under the influence of control
philosophy-1.
The following Figure 5.3 shows the control philosophy-1 response to 20% reduction in hydrogen feed
flow rate linearly.
(a) Effect of 20% Linear Reduction of Hydrogen Feed Flow Rate on Nitrogen Feed Flow Rates
(b) Effect of 20% Linear Reduction of Hydrogen on Pressure at Reactor Outlet

Discussion:
• Region A: The hydrogen supply flow rate is gradually reduced during a 30-minute time period.
When the controller senses a change in the ratio, it instantly closes the nitrogen valve, causing the
nitrogen supply flow rate to drop. A minor peak in the nitrogen feed flow rate is detected, which
could be attributed to a slight delay in the controller’s response to changes in the stoichiometric
ratio. It is important to note that the greatest pressure reduction observed due to change in
hydrogen feed flow rate is essentially negligible.
• Region B: A consistent feed flow rate is maintained for 60 minutes during this simulation phase,
allowing the system to create a new steady-state equilibrium. As a result, the nitrogen and hy-
drogen feed flow rates remain constant during this time, maintaining a stoichiometric ratio of 3
(H2 : N2 ) while the system pressure remains constant. The controller responded quickly to any
disturbances or oscillations in the system, ensuring quick and efficient control.
• Region C: For a 30-minute time interval, the hydrogen feed flow rate is gradually increased back
to its initial value. To maintain the stoichiometric ratio, the nitrogen input flow rate increases
correspondingly during this process. As a result of increasing input flow rates, there is relatively
small pressure fluctuation from the set point.
• Region D: The simulation was carried out for another 30 minutes, allowing the system to gradually
revert to its original steady-state characteristics. During this time, the system’s pressure was kept
constant, ensuring that it remained stable. After 30 minutes, the simulation was terminated.
(c) Ammonia Production Rate and Temperature at Reactor Outlet

(d) Cooling Duty of Condenser
Figure 5.3: Control Philosophy-1 response to 20% linear change in hydrogen feed flow rate
Discussion:
• Region A: Because there were fewer reactants available during the steady ramp-down in hy-
drogen supply flow rate, the ammonia production rate reduced by 20%. Because the reactants
are reduced, less heat is emitted in the exothermic ammonia synthesis reaction, and thus the
temperature at the adiabatic reactor’s outlet stream lowers. The cooling duty of the condenser is
reduced to restore the pressure to its initial amount. This raises the temperature of the stream
leaving the condenser, and thus the pressure.
• Region B, C & D: The ammonia production rate and the temperature at the reactor exit remain
constant when the feed flow rates are kept constant. As the hydrogen feed flow rate is increased
back to its prior value, the heat emitted from the exothermic process increases, so does the rate of
ammonia generation. As a result of the higher feed flow rate, the pressure rises, and the cooling
duty rises to reduce the pressure. The simulation is continued for another 30 minutes before
being terminated at the 150-minute mark.
Results and Conclusion:

1. Hydrogen Flow Rate and Nitrogen flow Rate:
• The hydrogen input flow rate clearly follows a linear decline pattern across all conditions.
Simultaneously, the nitrogen input flow rate is reduced linearly to maintain the stoichiometric
ratio constant.
2. Pressure Control:
• Even when the hydrogen input fluctuates, the system successfully maintains constant pres-
sure levels. This demonstrates the usefulness of control philosophy-1, particularly in situa-
tions where hydrogen supply may be limited. As a result of this configuration, the controller
responded quickly to any disturbances or oscillations in the system, ensuring quick and effi-
cient control.
• The maximum recorded pressure reduction was 2 bar, which occurred when the hydrogen
input flow rate was reduced by 70%.
3. Reactor Outlet Temperature and Ammonia Production:
• The output temperature of the reactor’s discharge stream decreases noticeably in all cases.
This drop is due to reduced heat release from the exothermic reaction as a result of reactant
scarcity. Furthermore, the ammonia production rate falls in direct proportion to the reduction
in hydrogen feed flow rate.
4. Condenser Outlet Stream Properties:
• The temperature of the condenser’s exit stream rises, resulting in higher vapor mole fractions
and lower liquid mole fractions of ammonia. As a result, the ammonia flow rate within the
recycle stream increases.
5. Hydrogen Buffer Importance:
• It is essential to provide a hydrogen buffer between the electrolysis and ammonia production
processes. This buffer reduces abrupt pressure reductions caused by quick fluctuations in
renewable electricity output. This displays the model’s capacity to deal with modest changes
in feed flow rates.
6. Model Performance and Limitations:
• When the hydrogen feed flow rate is increased back to its initial value, the system pressure
reaches the same value by the end of 150 minutes. However, other parameters such as
nitrogen feed flow rate, condenser cooling duty, and so on do not return to their initial condi-
tions. This demonstrates that the model takes more than 30 minutes to stabilize. However,
no limitations in the model’s operation have been discovered till 70% reduction in hydrogen
feed flow rate.

The primary idea of control philosophy-2 is handling of the recycle compressor’s brake power as a way
of maintaining system pressure. It essentially aims to control the flow rate of the recycle stream in order
to keep system pressure constant. This is accomplished by adjusting the brake power, which in turn
exerts indirect control over the flow rate of the recycle stream, ensuring steady pressure within the loop.
In addition to the pressure controller, a ratio controller has been included to maintain the stoichiometric
ratio. The process flow diagram is visually illustrated in Figure 5.4, which has been improved by the
integration of various controllers within Aspen Plus Dynamics. Table 5.3 provides broad information
and parameters for an in-depth understanding of the controllers used in control philosophy-2.
PRESSURE-C 1 1 Direct Recycle Stream Pressure Brake Power (kW)
RATIO-C 1 1 Reverse Stoichiometric Ratio N2-V Opening
For the control philosophy-2, a step-change and linear-change of hydrogen feed flow rate is imple-
mented similar to control philosophy-1. The results are shown in the following sections.
5.4.1. Step-Change in Hydrogen Feed Flow Rate to Control Philosophy-2

A step-change in hydrogen feed flow rate is implemented similar to what was discussed in section 5.3.
The following plots show the response of control philosophy-2 to sudden reduction in hydrogen feed
flow rate.
Figure 5.5 (b) shows the response of the recycle compressor brake power and the pressure of outlet
stream of recycle compressor.
(a) Hydrogen and Nitrogen Feed Flow Rate
(b) Recycle Compressor Outlet Pressure and Brake Power
Results:
• It demonstrates that control philosophy-2 is a successful method for restoring system pressure to
its initial level when the hydrogen feed flow rate is reduced. When the hydrogen input flow rate
is reduced by 20%, an immediate pressure reeduction of 2 bar is noticed.
• The controller detects a deviation in set point, and the system is able to return to a stable and de-
sired steady-state condition after around 25 minutes. However, it should be noted that response
time is affected by control parameter adjustment.
• The recycle compressor’s brake power was lowered from 12.5kW to 5.25kW throughout the pres-
sure recovery operation. This reduction in recycle compressor brake power eventually reduces
the mass flow rate of the recycle stream.
• Because of the mass balance at the inlet of the synthesis loop and recycle, the pressure returns
to its original value. To maintain a stoichiometric inlet feed flow rate ratio of 3 (H2 : N2 ), a ratio
controller is also employed.
To get a more insightful findings about the system behaviour, a gradual change of hydrogen feed flow
rate is implemented and is discussed in the subsection below.
5.4.2. Linear Change of Hydrogen Feed Flow Rate to Control Philosophy-2
The following Figure 5.6 represents the response of control philosophy-2 to a gradual reduction in
hydrogen feed. Similar to linear reduction in hydrogen feed in Control Philosophy-1, four regions can
be observed in the plots and each region (A, B, C, D) represents similar action done as described in
Table 5.1. The results and findings are discussed after the response shown for each of the scenarios.
(a) Hydrogen and Nitrogen Feed Flow Rates

(b) Compressor Outlet Pressure and Brake Power
(c) Recycle stream Flow Rate and Ammonia Production Rate
Results:
• Region A: A similar trend has been identified when observing reduction in hydrogen and nitro-
gen flow rates, as depicted in Figure 5.3 of control philosophy-1. The compressor brake power
gradually decreases throughout the period of reduced hydrogen feed flow rate. This decrease
in compressor power corresponds to a decrease in flow rate, resulting in less gas entering the
system. As a result, the decrease in system pressure is small and exhibits only minor oscillations.
As expected, the reduction in reactants reduces ammonia production.
• Region B: The hydrogen and nitrogen feed flow rates were kept constant for 60 minutes in order
for the system to reach its new steady-state. During this simulation phase, the system pressure re-
mains constant, but the recycle flow rate does not reach a steady-state value. This demonstrates
that the system takes more than 60 minutes to achieve new steady-state conditions.
• Region C: The hydrogen supply is linearly scaled up to its initial value for 30 minutes. The
recycle flow rate increases as the flow rate increases. As a result, the compressor’s braking
power increases in order to compress more gas. The system’s pressure does not shift abruptly.
As a result of the increased reactant feed flow rate, the ammonia production rate rises.
• Region D: The system is left for 30 minutes after the linear ramp-up of the hydrogen feed flow
rate to reach its original steady-state values. Because steady-state conditions are not met, the
current model takes longer than 150 minutes.
Conclusion:
1. Compressor Brake Power:

• The model includes a recycle compressor, which is critical to the system. The compressor’s
brake power is a crucial characteristic that influences the level of gas compression. Reduced
brake power results in lower gas compression. This causes a reduction in flow rate, which
balances the recycle flow rate with the reduced feed flow rate. The result is that the pressure
is retained its initial level.
2. Ammonia Production:
• The model additionally tracks ammonia synthesis, which is an immediate consequence of the
hydrogen-nitrogen reaction process. The models provided useful insights into the temporal
variability in ammonia production rates when feed flow rates are changed. As the availability
of reactants decreased, the production rates gradually decreased over time.
3. Model Performance and Limitations:
• The model successfully runs until the hydrogen supply flow rate is reduced by 70%, and no
restrictions have been discovered. However, the brake power of the compressor is not as
flexible as it is assumed to be in Aspen Plus dynamics. When the brake power is too low,
the compressor does not work, resulting in instability in the production unit.
4. Pressure Control:
• The model includes a feedback mechanism that continuously checks the system’s pressure.
The model can respond quickly to pressure variations, ensuring that pressure fluctuations
are addressed. Based on the pressure feedback, the controller adjusts the recycle compres-
sor’s brake power to maintain a steady pressure under varying input feed flow rates.

The operating principle of control philosophy-3 is to replace the deficient hydrogen with surplus nitrogen
thereby changing the stoichiometric ratio of H2 :N2 . The pressure controller measures the downstream
pressure of the reactor and is connected to the nitrogen valve. This controller opens/closes the nitrogen
valve to reach the set-point of pressure. Figure 5.7 shows the control philosophy-3 from Aspen Plus
Dynamics and Table 5.4 illustrates the details and parameters of the controllers used for the current
philosophy 3.
PRESSURE-C 1 1 Reverse Downstream Pressure of Reactor N2-V Opening
5.5.1. Step-Change in Hydrogen Feed Flow Rate to Control Philosophy-3
Similar to Control Philosophy 1 & 2, a step-change in reduction of hydrogen feed flow rate is imple-
mented to examine the behaviour of the system. The response time of the controller and the effect of
control action are noticed and plotted below in Figure 5.8.
(b) Inlet Feed Ratio and Pressure at Reactor Outlet

Results:
• Hydrogen and Nitrogen Feed Flow Rates:

– A 20% reduction in hydrogen input flow rate causes the system to respond dynamically.
The nitrogen flow rate peaks at a given point and then gradually decreases over time. One
probable explanation is that the controller initially uses nitrogen as a substitute to replace the
deficit hydrogen feed flow rate and then gradually declines when the systems gets adapted
to reduces hydrogen feed flow rate. The pressure controller adjusts the nitrogen feed flow
rate to keep the system pressure stable.
• Inlet Ratio and Pressure at Reactor Outlet:
– The drop in hydrogen feed flow rate has an immediate effect on the stoichiometric H2 : N2
ratio, generating an initial decrease due to the addition of extra nitrogen. When the system
responds to changes in feed conditions, the ratio settles down to around 2.8.
– When the hydrogen input flow rate is reduced, the pressure at the reactor outlet rises due to
unreacted nitrogen in the reactor. An initial pressure increase of 6 bar is observed, but the
pressure gradually recovers to its initial value over time.
• Ammonia Production Rate and Recycle Flow Rate:
– The recycle flow rate has increased significantly. This is because there is more nitrogen in
the system than usual, and because the density of nitrogen is higher, the recycling flow rate
increases. This impact is caused by a decrease in the stoichiometric inlet ratio. Hydrogen
becomes the limiting reactant in ammonia production, hence a decrease in hydrogen leads
to a decrease in ammonia production rate.
5.5.2. Linear-Change in Hydrogen Feed Flow Rate to Control Philosophy-3
Instant increase in pressure was observed when there is an instant reduction in hydrogen feed flow rate.
To mitigate this, a hydrogen buffer could be added and therefore, the hydrogen ramp-down is done
linearly. The effect of the gradual reduction in hydrogen feed flow rate is shown below in Figure 5.9.
Discussion:
• Region A: During the first 30 minutes, the nitrogen feed flow rate increases while the hydrogen
feed flow rate is linearly lowered. To compensate for the system’s lack of hydrogen, the controller
pumps nitrogen. This behavior was also observed when the hydrogen input flow rate was abruptly
lowered, as illustrated in Figure 5.8 (a). As the system gets adjusted to the changing hydrogen
flow rate, the nitrogen flow rate decreases, indicating a dynamic adjustment period.
• Region B: The simulation is carried out for 60 minutes to bring the model to a new steady-state.
The hydrogen feed flow rate stabilizes but the nitrogen feed flow rate remains constant after ∼ 40
minutes. This time period represents the formation of a new equilibrium under different hydrogen
feed conditions.
• Region C: For 30 minutes, the hydrogen feed flow rate is linearly ramped up to the original value,
and a minor dip in the nitrogen feed flow rate is noticed. As a result of the increased hydrogen feed
flow rate, the controller reduces the nitrogen to compensate for any pressure increase caused by
the increased hydrogen flow rate. The nitrogen input flow rate gradually increases.
• Region D: The model is run for the final 30 minutes to recover its original steady state settings.
By the end of the 150-minute simulation, the hydrogen feed flow rate is constant, and the nitrogen
feed flow rate is nearly constant.
(b) Inlet H2 : N2 Ratio and Pressure at Reactor Outlet
(c) Recycle stream Flow Rate and Purge Flow Rate
Discussion:
• Region A: A 1 bar increase in pressure is seen at first, followed by a return to its original value.
The stoichiometric ratio of H2 : N2 fluctuates, which is controlled by nitrogen feed rate variability.
Because of the increase in nitrogen gas in the system, both recycle and purge flow rates increase
significantly.
• Region B: When the feed is kept at a steady rate, the pressure in the system remains constant,
suggesting a stable system response. The input hydrogen to nitrogen ratio stabilizes at 2.8. The
recycle and purge flow rates are also constant, resulting in a 45% increase in the recycling flow
rate.
• Region C: With only a minor decrease in pressure, the pressure practically remains constant.
The inlet ratio rises from 2.8 to 3.3, indicating an adjustment to the changing hydrogen supply
flow rate. Both the recycle and purge flow rates drop in reaction to the increasing hydrogen feed
flow rate.
• Region D: The pressure remains constant at the end of the 150-minute experiment, showing the
system’s stability. The ratio attained a value of 3. However the recycling and purge flow rates did
not reach a steady-state value, indicating that more time is needed to stabilize under the current
control parameters.
Conclusion:
1. Response to Hydrogen Reduction

• When the hydrogen feed flow rate is reduced by 20%, the nitrogen feed flow rate increases.
This spike could be the result of the pressure controller’s response, which alters the nitrogen
feed flow rate to maintain system pressure.
2. Shift in Inlet Stoichiometric Ratio
• Reduced hydrogen feed flow rate has a direct impact on the inlet stoichiometric ratio of
hydrogen to nitrogen, resulting in an initial abrupt fall. When the hydrogen supply is reduced
by 20%, the ratio settles to 2.8, stabilizing the system pressure.
3. Recycle Flow Rate Increase
• When the hydrogen feed flow rate is lowered, the recycle flow rate increases significantly.
This is because a greater amount of nitrogen is pumped into the system to compensate for
the lack of hydrogen. This rise in recycle flow rates poses issues to the manufacturing unit.
4. Model Performance and Limitations
• The model successfully runs until the hydrogen supply flow rate is reduced by 70%, and no
restrictions have been discovered. However, these reduced flow rates pose a significant
issue when managing with higher recycle flow rates.
6
Economic Evaluation
An economic evaluation has been carried out using HySupply open-source tools [13] to see the be-
haviour and the costs associated with the ammonia production plant under different scenarios. Four
different tools have been developed by the professors from the University of New South Wales. For
the current thesis project, HySupply Cost Analysis Tool and HySupply Ammonia Tool have been used
to calculate the total investment costs of the entire ammonia production unit under various conditions.
The HySupply Cost Tool is used to model the hydrogen production unit and determine the levelized
cost of hydrogen (LCOH). HySupply Ammonia Tool is used for cost analysis of ammonia plant by de-
termining the levelized cost of ammonia (LCOA). Different scenarios are considered and are shown in
section 6.1 below.
6.1. Scenarios
Five different scenarios have been considered to determine the overall costs and the feasibility of
working of the ammonia production unit. They are:
• Solar PV generator connected to electrolyzer and Haber-Bosch unit

• Wind generator connected to electrolyzer and Haber-Bosch unit
• Hybrid (Solar+Wind) generator connected to electrolyzer and Haber-Bosch unit
• Grid connected wind generator to electrolyzer and Haber-Bosch unit
• Wind generator connected to electrolyzer, Haber-Bosch unit and Battery
Due to the fluctuations in the sun and wind, generation of electricity from renewables is fluctuating.
These fluctuations depend on several factors and thus, a back-up power source is necessary in order
to maintain the ammonia synthesis plant running. Therefore, a balancing technology has to be used in
order to stabilize these fluctuations. For the sake of simplicity, the balancing technologies used for the
current thesis project are batteries and grid connected electricity. A cost analysis is carried out for the
presented scenarios. Several other scenarios could be considered depending on the desired output.
Some of the options that are generally considered are:
54
6.2. Input Data to the Tools 55
Figure 6.1: Three Different Options for Green Ammonia Production
As illustrated in the Figure 6.1, the three possibilities for manufacturing ammonia from renewable energy
sources are large scale energy storage, hydrogen storage, and flexible ammonia synthesis reactor.
Option 1 employs battery/grid electricity storage to smooth out variable electric power. This battery or
grid is used to generate uniform electric power in order to continually manufacture hydrogen, which is
then fed into the ammonia synthesis reactor, which functions on a uniform scale. Option 2 generates
hydrogen using various amounts of renewable energy and then stores it in huge, compressed gas tanks.
This stored hydrogen is utilized to balance out the fluctuating hydrogen production rate, maintaining a
steady supply of hydrogen to the ammonia synthesis unit. Option 3 uses multiple ammonia synthesis
reactors in parallel where each of the reactor is configured differently. In this way of configuration, the
ammonia production is flexible.
For the current thesis project, option 2 is considered where the effect of over sizing of electrolyzer and
introduction of hydrogen buffer storage on the levelized costs of ammonia are studied.
6.2. Input Data to the Tools

Several assumptions have been made in the cost analysis tools and are given as input to the models
in order to calculate the costs associated with each of the scenario. Some of the assumptions/input
data provided are based on the data from Open-Source tool [13]. For all the scenarios, the solar and
wind data have been extracted from renewables.ninja [19] and were exported to the custom column in
the analysis tools.
• The operational year of the plant is considered as 2030.

• The power consumption of the electrolyzer for the balance of the plant has been assumed to be
54.3 kWh/kgH2 [76].
• An 8 hours of hydrogen buffer storage at rated electrolyzer power for all configurations has been
reported [76] and therefore, the hydrogen buffer used for the current project is 8 hours but it is
highly dependant on location and availability of renewable power resources.
• The stack lifetime is assumed to be 80000 hours with a stack degradation of 1%/year.
• Electrolyzer cost is considered as USD 785/kW from the data provided by Proton Ventures B.V.
• Installation costs are 200% of CAPEX for electrolyzer, 400% for ammonia synthesis unit according
to Lange factor [77].
• Solar PV farm equipment cost is considered as USD 480/kW and Wind farm cost as USD 740/kW
(values from Proton Ventures B.V.) with a USD 15000/MW/year OPEX for solar [78] and USD
40000/MW/year OPEX for wind [79].
6.3. Solar PV generator with Electrolyzer and Haber-Bosch unit 56
• The cost of batteries is considered as USD 152/kW according to BloombergNEF [80] with a life-
time of 10 years.
• The SEC of Ammonia plant is considered as 0.41 kWh/kgN H3 and the SEC of air separation unit
is 0.23 kWh/kgN2 [13].
• The ammonia synthesis unit cost is USD 378/TN H3 and air separation unit cost as USD182/TN2
[13].
• The operating costs for electrolyzer is 2%/year of electrolyzer purchase cost. The operating costs
for ammonia synthesis unit and air separation unit are considered to be 2% of CAPEX.
• For HVAC transmission lines, a cost of USD 323000/km [81] is considered by taking 3.6% inflation
rate per year with an installation cost of 10% of CAPEX.
• According to Global Petrol Prices [82], the cost of all components of the electricity bill such as
cost of power, distribution and taxes in Morocco is taken as USD106/MWh.
• Depreciation time is taken as 20 years and a discount rate of 7% per annum for calculation of
levelized costs.
6.3. Solar PV generator with Electrolyzer and Haber-Bosch unit

In the first scenario, a solar farm is connected to the electrolyzer, air separation unit and Haber-Bosch
unit. The size of the solar farm is determined by the total amount of electricity required for each of the
processes. A calculation is made to determine the amount of solar electricity required.
Electrolyzer Capacity:
The amount of hydrogen required to produce 25 tonnes of ammonia per day is 190 kg/hr. Therefore,
the total amount of electricity required is 10.31 MW
Capacity = 190 × 54.3 = 10317 kW = 10.317 M W
Ammonia Synthesis Unit Energy Consumption:
The ammonia synthesis unit is less energy intensive process as compared to hydrogen production. As
stated in inputs that the SEC of ammonia synthesis is 0.41 kWh/kg of ammonia produced, the total
electricity requirement for the Haber-Bosch process is 0.42MW.
AnnualAmmoniaP roduction SECN H3

Required Electricity = ×
8760 1000
9000000 0.41
= ×
8760 1000
= 4212 kW = 0.42M W
In a similar way, the power demand of air separation unit is also calculated and found to be 0.19 MW.
Therefore, the total capacity of the solar farm is modelled as 11 MW.
Electrolyzer Response to Solar PV generator:
Based on the input data provided to the cost analysis tool, the following response of electrolyzer is
observed. Figure 6.2 below shows the capacity factor of electrolyzer as a function of time for the enitre
year.
(a) Capaciity Factor of Electrolyzer as a function of time
To get a clear understanding of the performance of the electrolyzer, the Figure 6.2 shows the capacity
factor of electrolyzer for 30 days.
(b) Hourly Capacity Factor of Electrolyzer for 30 days
Figure 6.2: Capacity Factor vs Time for Solar PV connected Electrolyzer
It can be observed from the graph above that the capacity of the electrolyzer drops to 0 when there is
a lack of sunlight (during the night). The capacity factor of an electrolyzer is highly influenced by the
intermittent nature of solar PV. During cloudy days, during the night, or at times in less solar irradiation,
the electrolyzer’s operation is affected leading to a lower capacity factor. This is also clear that the lack
of energy storage systems or energy supply impacted the performance of electrolyzer. The key results
obtained form the analysis tool are given in the Table 6.1 below.
Parameter Value
Total Time Electrolyzer is Operating 43%
Electrolyzer Capacity Factor Achieved 23%
Energy Consumed by Electrolyzer (MW/yr) 21000
Hydrogen Output (TPA) 325
LCOH (USD/kg) 7.2
Table 6.1: Solar PV connected Electrolyzer results
Haber-Bosch unit response to Solar PV generator:
The current section presents an analysis of ammonia production unit integrated with solar PV generator.
The hourly performance graphs indicate the dynamics between solar power generation and ammonia
production. The below Figure 6.3 show the real-time relationship between solar energy availability and
its impact on ammonia production efficiency throughout the year. The Figure 6.4 shows the perfor-
mance of ammonia production unit for 30 days.
Figure 6.3: Capacity Factor as a function of time

Figure 6.4: Hourly Capacity Factor for 30 days
The integrated solar PV generator and ammonia synthesis unit highlight significant aspects of its per-
formance. The ammonia production plant exhibits a capacity factor of 22%. Due to the limited avail-
ability of the power supply , the total ammonia production per annum reduced from 9000 tonnes to
1900 tonnes, exhibiting a reduction in 80%. Additionally, the levelized cost of ammonia stands at USD
3.3/kg. Table 6.2 below shows the results obtained from the simulation.
Parameter Value
Average Power Plant Capacity Factor 22%
Annual Capacity Factor of Ammonia Plant 20%
Average Total Time Ammonia Plant is in Operation 3220 hrs/yr
Ammonia Produced (TPA) 1900
LCOA (USD/kg) 3.3
Table 6.2: Solar PV connected Ammonia Production Plant Results
Table 6.3 shows a breakdown of capital costs associated with each of the components in the ammonia
synthesis unit. In the transmission costs, the transmission distance is assumed to be 25 km. The
indirect costs include the installation costs for the electrolyzer, ammonia synthesis unit, air separation
unit, power plant installation and transmission line installation. It is observed that the total capital costs
associated with setting up an ammonia production facility under solar PV generator is USD 62 million.
6.4. Wind generator with Electrolyzer and Haber-Bosch unit 60
Component Capital Cost (USD)

Power Plant 5.2 million
Transmission 8 million
Electrolyzer 8.1 million
Ammonia Plant 3.4 million
Air Separation Unit 1.3 million
Indirect Costs 35 million
Total Capital Costs 62 million
Table 6.3: Total Capital Costs for Solar PV Connected Ammonia Plant
6.4. Wind generator with Electrolyzer and Haber-Bosch unit

In the second scenario, a wind farm is connected to the electrolyzer, air separation unit and the ammonia
synthesis unit. The size of the wind farm is the total amount of electricity that is determined in section 6.3
is 11 MW. Both the responses of the electrolyzer and ammonia synthesis plant are recorded and the
results obtained are portrayed in each of the subsections.
Electrolyzer response to Wind generator:
In the figures below, the dynamic response of the electrolyzer when integrated with a wind generator is
observed, along with the corresponding capacity factor achieved. Figure 6.5(a) provides a detailed view
of response of the electrolyzer over a course of 30 days and Figure 6.5(b) portrays the electrolyzer’s
response and behaviour over the course of an entire year.
(a) Hourly Capacity Factor of Electrolyzer for 30 days

(b) Capacity Factor of Electrolyzer throughout the year
Figure 6.5: Capacity Factor vs Time for Wind Generator connected Electrolyzer
The figures above graphically demonstrate the significant impact of wind fluctuations on electrolyzer
performance. Variability in wind-related elements such as wind speed and turbine height have a sub-
stantial impact on daily hydrogen generation, resulting in noticeable fluctuations. Despite this unpre-
dictability, the electrolyzer’s improved performance stands out, with a capacity factor improved to 70%
when compared to a solar PV generator at 23%. Notably, this improvement in capacity factor results in
a drop in the levelized cost of hydrogen, reaching an economically advantageous USD 3.16 per kgH2
generated. Table 6.4 presents specific insights and comparative data, offering a thorough overview of
the achieved results and showing the economic viability of this integrated system.
Parameter Value
Total Time Electrolyzer is Operating 98%
Energy Consumed by Electrolyzer (MWh/yr) 63000
LCOH (USD) 3.16
Table 6.4: Wind Connected Electrolyzer Results
Haber-Bosch unit response to wind generator:

In the figures below, the dynamic response of the ammonia synthesis plant when integrated with wind
generated electricity is observed, along with the corresponding capacity factor achieved and the amount
of ammonia produced per year. Figure 6.6(a) provides a view of response of ammonia synthesis plant
over a course of 30 days and Figure 6.6(b) shows the behavior of the Haber-Bosch unit for a span of
entire one year.
(a) Hourly Capacity Factor of Ammonia Synthesis Plant for 30 days
(b) Capacity Factor of Ammonia Synthesis Plant throughout the year
Figure 6.6: Capacity Factor vs Time for Wind Generator connected Haber-Bosch unit
6.5. Hybrid Generator with Electrolyzer and Haber-Bosch unit 63
Based on the graphs above and from the simulation, it has been observed that the capacity factor of
the ammonia synthesis plant increased to 60% compared to solar PV generator. This is due to higher
availability of wind energy with an average power plant capacity factor of 66%. The ammonia plant is
observed to be running at maximum capacity for 3.5% of the entire time in an year with a production
rate of 5300 TPA. Due to the availability of wind energy, the levelized cost of ammonia reaches USD
1.3 per kg N H3 .
Parameter Value
LCOA (USD) 1.3
Table 6.5: Wind connected Ammonia Production Plant Results
Table 6.6 shows a breakdown of capital costs associated with each of the components in the ammonia
synthesis unit. In the transmission costs, the transmission distance is assumed to be 25 km. The
indirect costs include the installation costs for the electrolyzer, ammonia synthesis unit, air separation
unit, power plant installation and transmission line installation. It is observed that the total capital costs
associated with setting up an ammonia production facility under wind generator is USD 65 million.

Electrolyzer 8.1 million
Air Separation Unit 1.3 million
Indirect Costs 35.3 million
Table 6.6: Total Capital Costs for Wind Connected Ammonia Plant
6.5. Hybrid Generator with Electrolyzer and Haber-Bosch unit

In the current section, a hybrid generator (combination of wind and solar) is connected to the electrolyzer
and ammonia synthesis plant and the performance of the plants are observed. A hybrid system is usu-
ally considered since it provides a more reliable and consistent energy output. Since the optimization
of renewable energy plants is out of the scope, 25% of the farm is considered solar and the rest 75%
of the farm is considered wind. Based on the depicted hybrid generator split, the performance of elec-
trolyzer and ammonia synthesis plant are shown below along with a cost estimation at the end of the
current section.
Electrolyzer Response to Hybrid Generator:

In the figuers below, the dynamic response of the electrolyzer when connected to a hybrid system is
modelled and observed, along with the costs associated with it and capacity factor achieved. Similar
to previous sections, Figure 6.7(a) shows the performance for 30 days and Figure 6.7(b) shows the
response of the electrolyzer for the entire year.
(a) Hourly Capacity Factor of Electrolyzer for 30 days
(b) Capacity Factor of Electrolyzer throughout the year
Figure 6.7: Capacity Factor vs Time for Hybrid Generator connected Electrolyzer
It has been observed from the results that the hybrid generator does not have a major impact on increase
of electrolyzer performance. The electrolyzer capacity factor achieved was 57% with a total hydrogen
production of 830 tonnes per year. The levelized cost of hydrogen also increased to USD 3.5 per kgH2
compared to wind connected electrolyzer (USD 3.16/kg). The results obtained are demonstrated below
in Table 6.7.
Parameter Value
Total Time Electrolyzer is Operating 99.2%
Energy Consumed by Electrolyzer (MW/yr) 51500
LCOH (USD) 3.5
Table 6.7: Hybrid Connected Electrolyzer Results
Haber-Bosch unit Response to Hybrid Generator:
Based on the simulation results from the model analysis tool for ammonia, there has been a reduction
in the annual capacity factor of ammonia production unit. The intermittent response of the ammonia
synthesis plant to hybrid electricity generator is plotted in Figure 6.8.
(a) Hourly Capacity Factor of Ammonia Synthesis Plant for 30 days

(b) Capacity Factor of Ammonia Synthesis Plant throughout the year
Figure 6.8: Capacity Factor vs Time for Hybrid Generator connected Haber-Bosch unit
Based on the results obtained, the average power plant capacity factor is 55% and the ammonia plant
capacity factor achieved is 52%. These values are comparatively lower than those obtained from wind
only connected ammonia plant. It has been observed that the annual ammonia production reduced by
48% compared to the nominal production capacity (9000 tonnes per year).
Parameter Value
LCOA (USD) 1.55
Table 6.8: Hybrid connected Ammonia Production Plant Results
The capital costs of the hybrid connected ammonia production plant were found to be USD 63 million.
This value is slightly lower than the total capital costs associated with wind operated ammonia plant.
This is due to the partial usage of solar PV which result in reduction in reduction of capital costs due to
low CAPEX/kW compared to wind. A breakdown of components is given in the Table 6.9 below.
6.6. Hydrogen Buffer Vessel 67

Indirect Costs 35 million
Table 6.9: Total Capital Costs for Hybrid Connected Ammonia Plant
Conclusion:
Boujdour in Morocco presents as exceptional wind profile, making it the ideal location for integrating
wind power with the ammonia production unit. The wind-connected plant demonstrates the highest
capacity factor the ammonia plant achieved at 60%. This shows the reliability of wind power in driving
the ammonia production process. Moreover, the wind-connected plant showed the lowest levelized cost
of ammonia (LCOA) signifying the cost effectiveness and competitiveness (USD 1.3/kg). However, the
plant demonstrates that it cannot reach full capacity (100%) to produce 9000 tonnes of ammonia per
year. Therefore, extra measures need to be considered in order to achieve maximum desired capacity.
The measures stated below are applied to wind operated ammonia synthesis plant.
• Over-sizing of electrolyzer
• Introducing a hydrogen buffer between ammonia synthesis unit and the electrolyzer
• Usage of balancing technology (Batteries or Grid Electricity) to supply back-up electricity
6.6. Hydrogen Buffer Vessel

The use of a hydrogen buffer between the electrolyzer and the ammonia synthesis plant plays a crucial
role in ensuring a smooth and consistent operation of the entire ammonia production process. Some
of the benefits of implementing hydrogen buffer storage are:
1. Stabilizing Hydrogen Supply:

• The hydrogen buffer acts as an intermediate storage facility, ensuring a continuous and sta-
ble supply of hydrogen to the ammonia synthesis plant. This is important as the electrolyzer
produces hydrogen intermittently due to variation in renewable energy sources.
2. Optimizing Electrolyzer Operations:
• The buffer storage allows the electrolyzer to operate at its optimal efficiency, irrespective of
the variations in renewable energy availability. The excess hydrogen can be stored during
periods of high production and utilized during low production, thereby enhancing the overall
efficiency of the ammonia synthesis plant.
The hydrogen buffer vessel storage size is modelled as 8 hours of storage based on [76]. Eichman
et al. [76] from National Renewable Energy Laboratory (USA) mentioned that the hydrogen storage
size is 8 hours for all rated power of electrolyzer for all configurations except islanded. If the green
ammonia plant is built in an island, then the hydrogen storage size would be 168 hours. But these
values are subjected to change i.e., the size of the hydrogen buffer storage can be more or less than 8
hours based on the location and availability of renewable energy sources.
6.6. Hydrogen Buffer Vessel 68
Cost of Hydrogen Buffer Storage:
In order to determine the size and cost of the hydrogen buffer vessel, the pressure of hydrogen gas
plays a vital role. The higher the pressure of hydrogen, the lower the size of the vessel could be
since the density increases when the pressure is increased. Based on literature data [83], the costs of
compressor required to compressed hydrogen gas and the cost associated to store hydrogen per kg
are shown in figure below.
(a) Costs of Compressor based on [83]
(b) Costs to Store Hydrogen in Pressurized Vessels [83]
Figure 6.9: Costs of Compressor to Compress Hydrogen Gas and Costs to Store Hydrogen
It is assumed that hydrogen is stored at room temperatures and the storage pressure is considered to
be 200 bar. Therefore, the density of hydrogen at 25◦ C and 200 bar is 14.48 kg/m3 . The amount of
hydrogen produced for the chosen electrolyzer capacity and the ammonia production unit requirement
is 190 kg/hr. Therefore, for 8 hours of storage the total hydrogen storage is 1520 kg. The size of the
vessel is determined by:
6.7. Electrolyzer Over-sizing 69
T otal Hydrogen Storage in kg

Size =
Density of hydrogen at 200 bar
The total size of the hydrogen buffer storage tank to store 1520 kg of hydrogen at 200 bar is 104.9 m3 .
An aspen simulation has been carried out to determine the compressor work in order to pressurize
the hydrogen gas to 200 bar. Giving an inlet pressure of 80 bar (assuming the outlet pressure of
hydrogen from PEM electrolyzer is 80 bar) and a discharge pressure of 200 bar, the work required for
compression is 100 kW. Therefore, based on the above graphs, the specific costs of compressor are
∼ 2100 USD/kW. For a 100 kW compressor power, the cost of compressor is 210,000 USD. Since
the paper was published in the year 2003 and the values were reference points from the year 2003,
an inflation rate of 3.6% per year is applied in order to calculate the cost of compressor in 2023. It is
determined that in 2023 the cost of compressor is USD 361,000.
From the above graphs, the cost of compressed hydrogen storage container at a pressure of 200 bar is
∼ USD 500/kg H2 . Adding the inflation costs, the cost to store compressed hydrogen in 2030 is USD
986 / kg H2 . Therefore, to store 1520 kg hydrogen in a container, the cost of the buffer vessel is USD
1,498,720.
6.7. Electrolyzer Over-sizing

The current section illustrates the effect of over-sizing of the electrolyzer on the ammonia production.
As discussed in section 6.6, the excess hydrogen required is 1520 kg when 8 hours of hydrogen buffer
is considered. Thus, producing this excess hydrogen needs an electrolyzer capacity of 13.8 MW. For
the current thesis project, the goal is to get the annual ammonia plant capacity factor to 80% since start-
up and shut-down times were not modelled in the present study. However, the option of hot stand-by
could be used and is discussed in section 6.8.
Based on the new electrolyzer capacity required to store 8 hours of hydrogen buffer, the electrolyzer
needs to be over sized by 34%. This value is given as input parameter to the ammonia cost tool. It has
been observed that the ammonia plant capacity factor achieved increases to 82% with an additional
increase in capital costs by 14.7 million USD. A breakdown of results and costs is given below in
Table 6.10
Component Value
Ammonia plant annual capacity factor achieved 82%
Time ammonia plant is at maximum capacity 63%
LCOH (USD/kg) 3.15
LCOA (USD/kg) 1.19
Electrolyzer Cost (USD) 10.8 million
Hydrogen Storage Costs (USD) 1.5 million
Total Capital Costs 79.7 million
Table 6.10: Results of Wind Connected Ammonia Plant with 34% Electrolyzer Over-Sized
6.7. Electrolyzer Over-sizing 70
Figure 6.10: Capacity factor of Electrolyzer and Ammonia Plant in 30 days for 34% Oversized Electrolyzer (Green:
Electrolyzer, Yellow: Ammonia)
Figure 6.11: Hydrogen Buffer Vessel Storage Level
The capacity factor profile of both electrolyzer and ammonia production plant is shown in Figure 6.10
above and Figure 6.11 shows the level of hydrogen storage tank. It can be observed that there is not
much of excess hydrogen produced and therefore, there are instances in the year where the ammonia
plant goes to a capacity factor of 0. This could be solved either by over-sizing the electrolyzer more or
by providing more electricity.
It can also be seen from the hydrogen storage level that the tank is at its 100% level only for a very
few days in average. This shows that the hydrogen produced from electrolyzer is directly used by the
ammonia plant and huge amounts of excess hydrogen is not produced. Therefore, the size of the
hydrogen storage tank can be reduced. For the current thesis project, an overall ammonia production
capacity factor of 80% is targeted considering the cost optimization.
6.8. Balancing Technology 71
In order to achieve a capacity factor of 80%, the size of the hydrogen storage tank can be reduced to
300 kg. This value is obtained by doing trial and error input in the open-source ammonia tool. The
profile of the hydrogen storage tank level of 300 kg size as a function of time is shown in Figure D.1.
The overall capital costs of 76 million is observed considering 34% oversized electrolyzer. The LCOA
calculated is USD 1.15/kg.
Based on the above results, the following conclusion have been drawn:
1. Effect of Electrolyzer Over-Sizing on Ammonia Production

• Achieving an annual ammonia production capacity factor of 80%, the electrolyzer capacity is
increased by 34% than nominal capacity. In order to store the excess the excess hydrogen
produced by oversized electrolyzer, a hydrogen storage tank of 1520 kg was considered
leading to an increase in capital costs by USD 14.7 million compared to wind connected
ammonia synthesis plant.
2. Efficient Hydrogen Storage and Cost Savings
• It has been observed that the hydrogen produced by the electrolyzer is utilized by the ammo-
nia synthesis unit without any excess hydrogen produced. This showed that the hydrogen
storage tank size can be reduced to 300 kg to optimize the cost savings. This led to a re-
duction in capital costs by USD 3.7 million compared to using 1520 kg of hydrogen storage
tank. The total capital costs calculated are USD 76 million and the reduced size of hydrogen
storage tank is 20.7 m3 .
3. LCOA and LCOH
• The LCOA where 300 kg of hydrogen storage used is found to be minimum at USD 1.15/kg
compared to USD 1.3/kg for wind connected Haber-Bosch unit where an 11% reduction
is levelized cost of ammonia is observed. The LCOH reduced from 3.16 USD/kg to 3.02
USD/kg.
6.8. Balancing Technology

Balancing technology is crucial in managing and mitigating fluctuations in renewable intermittent elec-
tricity, ensuring a stability and reliability in the electrical power grid. Renewable energy sources like
wind and solar are intermittent and variable in nature which are influenced by weather conditions. This
intermittency can cause imbalances between electricity supply and demand, leading to instability in the
grid. Therefore, balancing technology is used as back-up electricity in order to supply electricity under
renewable electricity unavailability.
Some of the balancing technologies usually used are batteries, gas turbines, grid electricity etc [13].
For the current thesis project, batteries and grid electricity have been considered and the cost analysis
has been carried out. The capacity of the batteries and grid electricity is determined by finding the
energy required to heat the catalyst bed in the ammonia synthesis reactor to maintain it in hot stand-by
and the heat loss from the reactor to the surroundings. Based on the calculation, it was observed that
the electricity input to the heater in the reactor is 233 kWhr. The detailed calculation of the derivation
of the exact number is shown in section D.1.
Batteries
Lithium-ion batteries are considered for the current application of balancing technology. According to
BloombergNEF [80], the cost of batteries is considered to be $151/kWh as an input to the cost analysis
model. In addition, a battery rated power of 0.116 MW is considered with 2 hours of charge/discharge
time. Therefore, the rated energy capacity of the battery is 0.233 MWh. Using the model with 34%
oversized electrolyzer with 300 kg hydrogen buffer storage, it has been observed that the total capital
costs associated with usage of batteries increase by USD 90,000
6.9. LCOH Comparison 72
Grid Electricity
Grid electricity could also be used to mitigate the fluctuations in intermittent renewable electricity but
at a cost. This grid electricity is produced results in emission of carbon with a carbon footprint of 296
kg CO2 per MWh [84]. The cost of grid connection is USD 44.5/kW [85] and the cost of electricity in
Morocco is USD 106/MWh for businesses [82]. This leads to an overall increase in capital costs to USD
10,600. This is significantly cheaper than using batteries as a balancing technology. But the carbon
emissions related to using 0.23 MW of grid electricity is estimated to be 68 kg CO2 .
6.9. LCOH Comparison

The following table below shows the LCOH values reported by various sources and are compared to
the current work.
Source Value (USD/kg H2)

PV Magazine [86] 2.75-4.08
[87] 3.23-13.70
NREL [88] 4.50
IEA [89] 3.20-7.70
Bloomberg NEF [90] 4.30
GEP [91] 3-6
Current Work 3.02-7.20
Table 6.11: LCOH comparison from different sources
An overall breakdown of total investment costs is shown in the pie chart (Figure 6.12) below:
Figure 6.12: Overall breakdown of Investment Costs

7
Conclusions and Recommendations
This chapter summarizes the main conclusion of the study and provides recommendations for future
work.
7.1. Conclusions
This section answers the research questions that were presented in chapter 1.
1. How can a stable ammonia synthesis process be maintained under varying hydrogen feed
flow rates?
(a) How can the dynamic behavior of the system be modelled?
The dynamic behaviour of the system is captured through Aspen Plus Dynamics, a simulation
tool for dynamic interactions. Initially, a steady-state model is created in Aspen Plus by determin-
ing the boundaries and operating conditions. This steady-state model is transferred to dynamic
mode and the model is sent to Aspen Plus Dynamics. For this purpose, the dynamics of electricity
input are translated into fluctuations in hydrogen feed flow rate, as the electrolyzer is connected
to renewable energy source.
Three distinct scenarios were simulated, each representing a linear and step reduction of 20%,
50% and 70% in the hydrogen feed flow rate. These scenarios provided an understanding about
the response of the plant to the changes in hydrogen feed flow rate.
(b) What are the suitable control philosophies to maintain pressure under dynamic input?
As the hydrogen feed flow rate reduces due to intermittent electricity, there is a direct impact
on the pressure within the system. Effective control strategies were implemented to main stable
pressure levels during these variations in order to prevent the damaging of production unit and
ensure consistent operation of ammonia synthesis process.
Three different control strategies were implemented and each strategy is designed to efficiently
regulate system pressure under dynamic input conditions. The components used to control the
system pressure in three different control strategies were
• cooling duty of condenser

• brake power of recycle compressor
• nitrogen feed flow rate
73
7.1. Conclusions 74
It has been observed that all three control strategies perform till 70% reduction of hydrogen feed
flow rate. The challenge associated when controlling the system pressure using brake power of
compressor is that the compressor is not as flexible as it is assumed to be for simulation pur-
poses. The compressor does not work where the brake power is too low, resulting in instabilities.
The challenge associated with using nitrogen to substitute deficit hydrogen is that the recycle
flow rate increases drastically. Therefore, the control strategy involving usage of cooling duty of
compressor in terms of stability and flexibility in controlling pressure of the system.
2. How can the complete ammonia plant be optimized by introducing a hydrogen buffer for
realistic renewable power availability scenarios?
(a) What renewable power source can be considered to ensure ammonia production with re-
duced levelized cost of ammonia?
Various possibilities were explored to achieve the desired operation and maximize ammonia pro-
duction. The analysis revealed that connecting the plant to wind power exhibited exceptional
results due to Boujdour in Morocco having high mean capacity factor for wind. The overall am-
monia production capacity factor achieved is 60% with a LCOA of USD 1.3/kg ammonia produced.
A 0.23 MW of back-up electricity is considered to maintain the reactor in hot stand-by during times
when the renewable power production is insufficient. This step could ensure several advantages
such as reduction in start-up times, process flexibility, reduction in thermal stress and improved
energy efficiency. This 0.23 MW electricity is provided either by batteries or grid electricity. It
has been observed that batteries cause an increase in capital costs by USD 90,000 whereas grid
electricity results in increase of capital costs by USD 10,600. But this comes with carbon emis-
sions of 68 kg.
It was also observed from dynamic simulations that by using control strategy 1 for pressure sta-
bilization, the system responds to stabilizing the initial conditions within 100 minutes. Therefore,
the system can be shut down during times when there is no wind and can be made run again
in 100 minutes. It can also be said that hydrogen buffer storage is not necessary if the system
responds instantly to changes in feed flow rates.
(b) What is the optimum amount of hydrogen storage capacity required for the selected am-
monia production capacity?
The hydrogen buffer acts as a storage unit that can stock excess hydrogen during periods of
abundant electricity, which can later be utilized when there is a deficit in power supply. This en-
sures a consistent hydrogen availability for the ammonia synthesis plant. An 8 hours of hydrogen
buffer is considered resulting in the requirement of oversizing of electrolyzer by 34%. However, it
is observed that 8 hours of excess hydrogen storage cannot be incorporated when the electrolyzer
is oversized by 34% due to all hydrogen being utilized by ammonia synthesis plant. This required
hydrogen buffer is specific to location and production capacity. Therefore, a 300 kg of hydrogen
buffer is found to be beneficial for the selected ammonia capacity plant to achieve a capacity fac-
tor of 80% with a reduction of levelized cost of ammonia by 11% from USD 1.3/kg to USD 1.15/kg.
However, designing an efficient and cost-effective storage system is a key challenge. Due to
the lower densities of hydrogen gas at standard temperatures and pressures, there is a need for
compression. This balance of costs between the required compressor work and size of hydrogen
buffer vessel is crucial. Other than costs associated with compression and storage vessel, safety
is another predominant factor that should be considered when storing hydrogen at high pressures.
7.2. Recommendations for Future Work 75
7.2. Recommendations for Future Work

The following section discusses about the recommendations for future study based on the current
research.
• Control Philosophies:
– Examining and fine tuning of the control parameters within the controllers would be a rec-
ommendation since the tuning was out of the scope. This provides a deeper insights in
the system behaviour and performance by better understanding and reproducing realistic
dynamics in the ammonia plant under varied inputs.
– Investigate and execute a control method that focuses on managing the feed-to-recycle ratio.
Understanding and adjusting this ratio has the potential to improve ammonia production
efficiency.
• Considering compressor curves to get an in detail understanding on the dynamics associated

with change in brake power of compressor in the control strategy
• Extend the techno-economic analysis to study the scalability of the production unit and analyze
the economic viability, costs, and benefits associated with scaling up the ammonia production
plant, considering different sizes and capacities
• Pressure Drop and Dynamic Effects:

– Implement a pressure drop correlation study to fully understand how different pressure drops
effect the ammonia production plant’s dynamics. This is essential for improving and stabiliz-
ing the system’s performance.
• Start-Up and Shut-Down Dynamics:
– Conducting a detailed analysis focusing on the start up and shut down times would be bene-
ficial. This could be possible by modelling parallel Haber-Bosch reactors with different con-
figurations. Using parallel reactors provides a high turn-down ratio of ammonia production
and therefore, a further cost optimization can be carried out.
• Optimization in Hydrogen Storage Compression:

– Optimize the hydrogen storage compression process by finding an appropriate balance be-
tween compression work and hydrogen storage costs. Energy reductions and cost-effectiveness
in hydrogen storage can be achieved through efficient compression.
References
[1] Jos Delbeke et al. “The paris agreement”. In: Towards a Climate-Neutral Europe: Curbing the
Trend (2019), pp. 24–45. DOI: 10.4324/9789276082569-2.
[2] J S Wallace and C A Ward. “H Y D R O G E N As a Fuel”. In: Int. J. Hydrogen Energy, 8.4 (1983),
pp. 255–268.
[3] Muhammad Aziz, Agung TriWijayanta, and Asep Bayu Dani Nandiyanto. “Ammonia as effective
hydrogen storage: A review on production, storage and utilization”. In: Energies 13.12 (2020),
pp. 1–25. ISSN: 19961073. DOI: 10.3390/en13123062.
[4] International Energy Agency. “Ammonia Technology Roadmap Towards more sustainable nitro-
gen fertiliser production”. In: (). URL: www.iea.org/t&c/.
[5] Direct CO� emissions from ammonia production, 2020-2050 – Charts – Data & Statistics - IEA.
URL: https://prod.iea.org/data-and-statistics/charts/direct-co-emissions-from-
ammonia-production-2020-2050.
[6] Yoshitsugu Kojima and Masakuni Yamaguchi. “Ammonia as a hydrogen energy carrier”. In: Inter-
national Journal of Hydrogen Energy 47.54 (2022), pp. 22832–22839. ISSN: 03603199. DOI: 10.
1016/j.ijhydene.2022.05.096. URL: https://doi.org/10.1016/j.ijhydene.2022.05.096.
[7] Collin Smith, Alfred K. Hill, and Laura Torrente-Murciano. “Current and future role of Haber-Bosch
ammonia in a carbon-free energy landscape”. In: Energy and Environmental Science 13.2 (2020),
pp. 331–344. ISSN: 17545706. DOI: 10.1039/c9ee02873k.
[8] Muhammad Heikal Hasan et al. A comprehensive review on the recent development of ammonia
as a renewable energy carrier. 2021. DOI: 10.3390/en14133732.
[9] Zhijian Wan et al. “Ammonia as an effective hydrogen carrier and a clean fuel for solid oxide fuel
cells”. In: Energy Conversion and Management 228.September 2020 (2021), p. 113729. ISSN:
01968904. DOI: 10 . 1016 / j . enconman . 2020 . 113729. URL: https : / / doi . org / 10 . 1016 / j .
enconman.2020.113729.
[10] Grigorii Soloveichik. “Ammonia as a virtual Hydrogen Carrier”. In: US Department of Energy 95.5
(2016), pp. 364–370. ISSN: 0916-8753. URL: https://www.energy.gov/sites/prod/files/
2016/12/f34/fcto_h2atscale_workshop_soloveichik.pdf.
[11] H. Ishaq and I. Dincer. “Design and simulation of a new cascaded ammonia synthesis system
driven by renewables”. In: Sustainable Energy Technologies and Assessments 40.April (2020),
p. 100725. ISSN: 22131388. DOI: 10.1016/j.seta.2020.100725. URL: https://doi.org/10.
1016/j.seta.2020.100725.
[12] Global Wind Atlas. URL: https://globalwindatlas.info/en.
[13] Jack Shepherd et al. “Open-source project feasibility tools for supporting development of the
green ammonia value chain”. In: Energy Conversion and Management 274.June (2022), p. 116413.
ISSN: 01968904. DOI: 10.1016/j.enconman.2022.116413. URL: https://doi.org/10.1016/
j.enconman.2022.116413.
[14] Hirokazu Kojima et al. “Influence of renewable energy power fluctuations on water electroly-
sis for green hydrogen production”. In: International Journal of Hydrogen Energy 48.12 (2022),
pp. 4572–4593. ISSN: 03603199. DOI: 10 . 1016 / j . ijhydene . 2022 . 11 . 018. URL: https :
//doi.org/10.1016/j.ijhydene.2022.11.018.
[15] M. Anvari et al. “Short term fluctuations of wind and solar power systems”. In: New Journal of
Physics 18.6 (2016). ISSN: 13672630. DOI: 10.1088/1367-2630/18/6/063027.
[16] Phebe Asantewaa Owusu and Samuel Asumadu-Sarkodie. A review of renewable energy sources,
sustainability issues and climate change mitigation. 2016. DOI: 10.1080/23311916.2016.11679
90.
76
References 77
[17] Solar Panel Tilt Angle Calculator - Footprint Hero. URL: https://footprinthero.com/solar-
panel-tilt-angle-calculator.
[18] Siemens Gamesa renewable energy. “SG 4.5-145 New SGRE turbine with the best-in-class LCoE
>4 MW”. In: (). URL: https://www.siemensgamesa.com/en- int/- /media/siemensgamesa/
downloads/en/products-and-services/archive/siemens-gamesa-onshore-wind-turbine-
sg-4-5-145-en.pdf.
[19] Renewables.ninja. URL: https://www.renewables.ninja/.
[20] Mengdi Ji and Jianlong Wang. Review and comparison of various hydrogen production methods
based on costs and life cycle impact assessment indicators. 2021. DOI: 10.1016/j.ijhydene.
2021.09.142.
[21] Marcus Newborough and Graham Cooley. “Developments in the global hydrogen market: The
spectrum of hydrogen colours”. In: Fuel Cells Bulletin 2020.11 (2020). ISSN: 14642859. DOI:
10.1016/S1464-2859(20)30546-0.
[22] S. Shiva Kumar and V. Himabindu. Hydrogen production by PEM water electrolysis – A review.
2019. DOI: 10.1016/j.mset.2019.03.002.
[23] Md Mamoon Rashid et al. “Hydrogen Production by Water Electrolysis: A Review of Alkaline
Water Electrolysis, PEM Water Electrolysis and High Temperature Water Electrolysis”. In: Inter-
national Journal of Engineering and Advanced Technology 3 (2015), pp. 2249–8958.
[24] Diogo M.F. Santos, César A.C. Sequeira, and José L. Figueiredo. “Hydrogen production by alka-
line water electrolysis”. In: Quimica Nova 36.8 (2013). ISSN: 01004042. DOI: 10.1590/S0100-
40422013000800017.
[25] Jörn Brauns and Thomas Turek. “Alkaline water electrolysis powered by renewable energy: A
review”. In: Processes 8.2 (2020). ISSN: 22279717. DOI: 10.3390/pr8020248.
[26] L. W. Hourng, T. T. Tsai, and M. Y. Lin. “The analysis of energy efficiency in water electrolysis
under high temperature and high pressure”. In: IOP Conference Series: Earth and Environmental
Science. Vol. 93. 1. 2017. DOI: 10.1088/1755-1315/93/1/012035.
[27] Marko Vuksic, Siniša Zorica, and Ivan Zulim. “Evaluation of DC-DC Resonant Converters for
Solar Hydrogen Production Evaluation of DC-DC resonant converters for solar hydrogen produc-
tion based on load current characteristics”. In: Contemporary Issues in Economy and Technol-
ogy June 2014 (2014), pp. 121–133. URL: https : / / www . researchgate . net / publication /
263470190_Evaluation_of_DC-DC_Resonant_Converters_for_Solar_Hydrogen_Production_
Based_on_Load_Current_Characteristics.
[28] Yujing Guo et al. “Comparison between hydrogen production by alkaline water electrolysis and
hydrogen production by PEM electrolysis”. In: IOP Conference Series: Earth and Environmental
Science 371.4 (2019), pp. 0–5. ISSN: 17551315. DOI: 10.1088/1755-1315/371/4/042022.
[29] Xiufu Sun et al. “Durability of Solid Oxide Electrolysis Cells for Syngas Production”. In: Journal
of The Electrochemical Society 160.9 (2013), F1074–F1080. ISSN: 0013-4651. DOI: 10.1149/
2.106309jes.
[30] Meilin Liu et al. “Understanding the phase formation and compositions of barium carbonate mod-
ified NiO-yttria stabilized zirconia for fuel cell applications”. In: International Journal of Hydrogen
Energy 40.45 (2015). ISSN: 03603199. DOI: 10.1016/j.ijhydene.2015.09.092.
[31] Greg Gege Tao, Anil Virkar, and Roxanne Garland. “A Reversible Planar Solid Oxide Fuel-Assisted
Electrolysis Cell and Solid Oxide Fuel Cell for Hydrogen and Electricity Production Operating on
Natural Gas/Biogas”. In: FY Annual Progress Report (2006).
[32] Pierre Millet and Sergey Grigoriev. “Water Electrolysis Technologies”. In: Renewable Hydrogen
Technologies: Production, Purification, Storage, Applications and Safety. 2013. DOI: 10.1016/
B978-0-444-56352-1.00002-7.
[33] Aldo Vieira da Rosa and Aldo Vieira da Rosa. “Chapter 10 – Hydrogen Production”. In: Funda-
mentals of Renewable Energy Processes. 2009.
References 78
[34] Greig Chisholm, Tingting Zhao, and Leroy Cronin. Hydrogen from water electrolysis. Elsevier
Inc., 2022, pp. 559–591. ISBN: 9780128245101. DOI: 10.1016/B978-0-12-824510-1.00015-5.
URL: http://dx.doi.org/10.1016/B978-0-12-803440-8/00016-6.
[35] S. Shiva Kumar and Hankwon Lim. “An overview of water electrolysis technologies for green
hydrogen production”. In: Energy Reports 8 (2022), pp. 13793–13813. ISSN: 23524847. DOI:
10.1016/j.egyr.2022.10.127. URL: https://doi.org/10.1016/j.egyr.2022.10.127.
[36] Agustin Valera-Medina and Rene Banares-Alcantara. Techno-Economic Challenges of Green
Ammonia as an Energy Vector. 2020, pp. 41–52. DOI: 10.1016/B978-0-12-820560-0.01001-8.
URL: https://books.google.nl/books?id=W2XRDwAAQBAJ&lpg=PA27&dq=DOI%3A10.1016%
2FB978-0-12-820560-0.00003-5&pg=PP1#v=onepage&q&f=false.
[37] Yu Wang et al. “Storage system of renewable energy generated hydrogen for chemical industry”.
In: Energy Procedia 29.February 2014 (2012), pp. 657–667. ISSN: 18766102. DOI: 10.1016/j.
egypro.2012.09.076.
[38] Joakim Andersson and Stefan Grönkvist. “Large-scale storage of hydrogen”. In: International
Journal of Hydrogen Energy 44.23 (2019), pp. 11901–11919. ISSN: 03603199. DOI: 10.1016/j.
ijhydene.2019.03.063.
[39] Patrick Preuster, Alexander Alekseev, and Peter Wasserscheid. “Hydrogen storage technolo-
gies for future energy systems”. In: Annual Review of Chemical and Biomolecular Engineering 8
(2017), pp. 445–471. ISSN: 19475438. DOI: 10.1146/annurev-chembioeng-060816-101334.
[40] Hydrogen Storage Underground? URL: https://fuelcellsworks.com/subscribers/hydroge
n-storage-underground/.
[41] M L Williams. “CRC Handbook of Chemistry and Physics, 76th edition”. In: Occupational and
Environmental Medicine 53.7 (1996). ISSN: 1351-0711. DOI: 10.1136/oem.53.7.504.
[42] W. F. Castle. “Air sepration and liquefaction: Recent developments and prospects for the begin-
ning of the new millennium”. In: International Journal of Refrigeration 25.1 (2002), pp. 158–172.
ISSN: 01407007. DOI: 10.1016/S0140-7007(01)00003-2.
[43] Cryogenic Process of Air Separation – IspatGuru. URL: https://www.ispatguru.com/cryogen
ic-process-of-air-separation/.
[44] Ahmed Khalil, Ahmed M Khalil, and Ahmed Mohame Khalil. “Production Of Nitrogen from the Air
During Cryogenic Process”. In: January (2018).
[45] N. K. Mohalik et al. “Application of nitrogen as preventive and controlling subsurface fire - Indian
context”. In: Journal of Scientific and Industrial Research 64.4 (2005). ISSN: 00224456.
[46] Mehran Moazeni Targhi, Mohammadreza Malek, and Pooyan Shams. “Simulation, Sensitivity
Analysis and Introducing New Valid Process Cases in Air Separation Units”. In: Journal of Applied
Chemistry 10.September (2017), pp. 45–60. DOI: 10.9790/5736-1009024560.
[47] C. R. Reid and K. M. Thomas. “Adsorption of gases on a carbon molecular sieve used for air
separation: Linear adsorptives as probes for kinetic selectivity”. In: Langmuir 15.9 (1999). ISSN:
07437463. DOI: 10.1021/la981289p.
[48] Generating Nitrogen with Pressure Swing Adsorption (PSA) Technology - Atlas Copco UK. URL:
https : / / www . atlascopco . com / en - uk / compressors / wiki / compressed - air - articles /
pressure-swing-adsorption-generator.
[49] Svetlana Ivanova and Robert Lewis. “Producing nitrogen via pressure swing adsorption”. In:
Chemical Engineering Progress 108.6 (2012), pp. 38–42. ISSN: 03607275.
[50] K. Seshan. “Pressure swing adsorption”. In: Applied Catalysis 46.1 (1989), p. 180. ISSN: 01669834.
DOI: 10.1016/S0166-9834(00)81410-4.
[51] Zykamilia Kamin, Mohd H V Bahrun, and Awang Bono. “A short review on pressure swing ad-
sorption ( PSA ) technology for nitrogen generation from air A Short Review on Pressure Swing
Adsorption ( PSA ) Technology for Nitrogen Generation from Air”. In: 050006.August (2022).
[52] Soonchul Kwon et al. CO2 Sorption. Elsevier Inc., 2011, pp. 293–339. ISBN: 9780815520498.
DOI: 10.1016/b978-0-8155-2049-8.10010-5. URL: http://dx.doi.org/10.1016/B978-0-
8155-2049-8.10010-5.
References 79
[53] Lei Wang, Cheng Shao, and Hai Wang. “Operation optimization of a membrane separation pro-
cess through auto-controlling the permeate gas flux”. In: Separation and Purification Technology
55.1 (2007). ISSN: 13835866. DOI: 10.1016/j.seppur.2006.10.015.
[54] Nitrogen & Oxygen Gas Generators - Pneumatic Engineering. URL: https://www.pneumatic.
com.au/gases/nitrogen-and-oxygen-gas-generators/.
[55] Moises A. Carreon. Molecular sieve membranes for N2/CH4 separation. 2018. DOI: 10.1557/
jmr.2017.297.
[56] Lakshminarayana Kudinalli Gopalakrishna Bhatta et al. “Progress in hydrotalcite like compounds
and metal-based oxides for CO2 capture: A review”. In: Journal of Cleaner Production 103 (2015).
ISSN: 09596526. DOI: 10.1016/j.jclepro.2014.12.059.
[57] M. G. Kaganer et al. “Vessels for the storage and transport of liquid oxygen and nitrogen”. In:
Chemical and Petroleum Engineering 3.12 (1967). ISSN: 15738329. DOI: 10.1007/BF01136404.
[58] Cryogenic Dewar Flasks. URL: https : / / www . tedpella . com / cryo - supplies _ html / cryo -
dewar.aspx.
[59] Max Appl. “The Haber-Bosch Process and the Development of Chemical Engineering”. In: A
Century of Chemical Engineering. 1982. DOI: 10.1007/978-1-4899-5289-9{\_}3.
[60] G. Ertl. “Primary steps in catalytic synthesis of ammonia”. In: Journal of Vacuum Science & Tech-
nology A: Vacuum, Surfaces, and Films 1.2 (1983). ISSN: 0734-2101. DOI: 10.1116/1.572299.
[61] Max Appl. “Chapter 4 Production”. In: Ullmann’s Encyclopedia of Industrial Chemistry: Ammonia
[M]. 2006.
[62] Home | ARENHA. URL: https://arenha.eu/.
[63] Role of Promoters in Haber’s Process. URL: https://unacademy.com/content/jee/study-
material/chemistry/role-of-promoters-in-habers-process/.
[64] Jayant M Modak. “Haber Process for Ammonia Synthesis”. In: August (2002).
[65] Ammonia - Properties at Gas-Liquid Equilibrium Conditions. URL: https : / / www . engineerin
gtoolbox.com/ammonia- gas- liquid- equilibrium- condition- properties- temperature-
pressure-boiling-curve-d_2013.html.
[66] Mahdi Malmali et al. “Better Absorbents for Ammonia Separation”. In: ACS Sustainable Chemistry
and Engineering 6.5 (2018), pp. 6536–6546. ISSN: 21680485. DOI: 10.1021/acssuschemeng.
7b04684.
[67] Matthew J. Kale et al. “Optimizing Ammonia Separation via Reactive Absorption for Sustain-
able Ammonia Synthesis”. In: ACS Applied Energy Materials 3.3 (2020), pp. 2576–2584. ISSN:
25740962. DOI: 10.1021/acsaem.9b02278.
[68] Marlinda Bauer. “Dynamic behavior of a flexible smallscale ammonia synthesis process”. In:
(2022). URL: http://repository.tudelft.nl/..
[69] K.H.R. Rouwenhorst et al. Ammonia Production Technologies. Elsevier Inc., 2021, pp. 41–83.
ISBN: 9780128205600. DOI: 10.1016/b978-0-12-820560-0.00004-7. URL: http://dx.doi.
org/10.1016/B978-0-12-820560-0.00004-7.
[70] Cheng Liang. “First Workshop ARENHA project : “ Introduction to novel technologies related to
ammonia- based energy storage ” Advanced Ammonia Synthesis Technologies Overview Index
1 . About Proton Ventures BV”. In: (2022), pp. 1–18. URL: https : / / arenha . eu / content /
consortium-workshops/.
[71] Huazhang Liu. “Ammonia synthesis catalyst 100 years: Practice, enlightenment and challenge”.
In: Cuihua Xuebao/Chinese Journal of Catalysis 35.10 (2014), pp. 1619–1640. ISSN: 02539837.
DOI: 10.1016/S1872-2067(14)60118-2. URL: http://dx.doi.org/10.1016/S1872-2067(14)
60118-2.
[72] Antonio Araújo and Sigurd Skogestad. “Control structure design for the ammonia synthesis pro-
cess”. In: Computers and Chemical Engineering 32.12 (2008), pp. 2920–2932. ISSN: 00981354.
DOI: 10.1016/j.compchemeng.2008.03.001.
References 80
[73] Bent Wiencke. “Fundamental principles for sizing and design of gravity separators for indus-
trial refrigeration”. In: International Journal of Refrigeration 34.8 (2011), pp. 2092–2108. ISSN:
01407007. DOI: 10.1016/j.ijrefrig.2011.06.011. URL: http://dx.doi.org/10.1016/j.
ijrefrig.2011.06.011.
[74] Simon Learman. “Horizontal Gas-Liquid Separator Calculator”. In: (2011). URL: www.blackmonk.
co.uk.
[75] Rajiv Mukherjee. “TEMA designations for shell-and-tube heat exchangers”. In: February (1998).
URL: http://www.torr-engenharia.com.br/wp-content/uploads/2011/05/exchanger.pdf.
[76] Joshua. D. Eichman et al. “Optimizing an Integrated Renewable-Electrolysis System”. In: United
States National Renewable Energy Laboratory March (2020). URL: https://www.nrel.gov/
docs/fy20osti/75635.pdf%0Ahttp://www.osti.gov/servlets/purl/1606147/.
[77] Lang Factor and Return on Investment – Foundations of Chemical and Biological Engineering I.
URL: https://pressbooks.bccampus.ca/chbe220/chapter/lang- factor- and- return- on-
investment/.
[78] Ryan Wiser, Mark Bolinger, and Joachim Seel. “Benchmarking Utility-Scale PV Operational Ex-
penses and Project Lifetimes: Results from a Survey of U.S. Solar Industry Professionals”. In:
Eletricity Markets & Policy 1.34078 (2020), pp. 1–8. URL: https://eta- publications.lbl.
gov/sites/default/files/solar_life_and_opex_report.pdf.
[79] Tyler Stehly and Patrick Duffy. “2020 Cost of Wind Energy Review”. In: Nrel December (2021),
p. 68.
[80] Lithium-ion Battery Pack Prices Rise for First Time to an Average of $151/kWh | BloombergNEF.
URL: https://about.bnef.com/blog/lithium-ion-battery-pack-prices-rise-for-first-
time-to-an-average-of-151-kwh/.
[81] Brian Cory. “Pricing in electricity transmission and distribution”. In: Proceedings of the Mediter-
ranean Electrotechnical Conference - MELECON 1.April (1996), pp. 19–25. DOI: 10 . 1109 /
melcon.1996.550955.
[82] Morocco electricity prices, December 2022 | GlobalPetrolPrices.com. URL: https://www.globa
lpetrolprices.com/Morocco/electricity_prices/.
[83] Peteves Stathis. HYDROGEN STORAGE : STATE-OF-THE-ART AND FUTURE PERSPECTIVE
HYDROGEN STORAGE : STATE-OF-THE-ART AND FUTURE PERSPECTIVE. April. 2016. ISBN:
9289469501. URL: https : / / www . researchgate . net / publication / 255636014 _ HYDROGEN _
STORAGE_STATE-OF-THE-ART_AND_FUTURE_PERSPECTIVE.
[84] Nicolae Scarlat, Matteo Prussi, and Monica Padella. “Quantification of the carbon intensity of
electricity produced and used in Europe”. In: Applied Energy 305 (2022). ISSN: 03062619. DOI:
10.1016/j.apenergy.2021.117901.
[85] How much does it cost to sign up for an electricity or gas supply?| Endesa. URL: https://www.
endesa.com/en/advice/processes- contracts/doubts- with- your- contract/how- much-
does-it-cost-to-register-for-a-new-contract.
[86] The Hydrogen Stream: LCOH of subsidized green hydrogen under $2/kg – pv magazine Interna-
tional. URL: https://www.pv- magazine.com/2023/04/14/the- hydrogen- stream- lcoh- of-
subsidized-green-hydrogen-under-2-kg/.
[87] Leonhard Povacz and Ramchandra Bhandari. “Analysis of the Levelized Cost of Renewable
Hydrogen in Austria”. In: Sustainability 2023, Vol. 15, Page 4575 15.5 (Mar. 2023), p. 4575. ISSN:
2071-1050. DOI: 10.3390/SU15054575. URL: https://www.mdpi.com/2071-1050/15/5/4575/
htm%20https://www.mdpi.com/2071-1050/15/5/4575.
[88] T. Ramsden, D. Steward, and J. Zuboy. “”Análisis del costo nivelado de producción de hidrógeno
centralizado y distribuido””. In: September (2009). URL: www.nrel.gov/docs/fy09osti/46267.
pdf.
[89] Global average levelised cost of hydrogen production by energy source and technology, 2019 and
2050 – Charts – Data & Statistics - IEA. URL: https://www.iea.org/data-and-statistics/
charts/global- average- levelised- cost- of- hydrogen- production- by- energy- source-
and-technology-2019-and-2050.
References 81
[90] 2023 Hydrogen Levelized Cost Update: Green Beats Gray | BloombergNEF. URL: https : / /
about.bnef.com/blog/2023-hydrogen-levelized-cost-update-green-beats-gray/.
[91] Green & Blue Hydrogen: Current Levelized Cost of Production & Outlook | GEP Blog. URL: https:
//www.gep.com/blog/strategy/Green-and-blue-hydrogen-current-levelized-cost-of-
production-and-outlook.
[92] Heat Capacities for Some Select Substances. URL: https://gchem.cm.utexas.edu/data/
section2.php?target=heat-capacities.php.
A
Control Philosophy-1
A.1. Step Change in Hydrogen Feed Flow Rate
Without the addition of controllers to the control philosophy-1, when the hydrogen feed flow rate was
reduced by 50%, a pressure reduction of 46 bar was observed. Figure A.1 shows the behaviour of
control philosophy-1 under 50% reduction in hydrogen feed flow rate and the response is plotted below:
82
A.1. Step Change in Hydrogen Feed Flow Rate 83
(b) Ammonia Production Rate and Pressure at Reactor Outlet
(c) Cooling Duty of Condenser and Temperature of Condenser Outlet Stream
Figure A.1: Control Philosophy-1 response to 50% step change in hydrogen feed flow rate
Results:
• As observed in Figure A.1 (a), the nitrogen flow rate reduces when the hydrogen feed flow rate is
reduced and reaches steady state value where the stoichiometric ratio of H2 : N2 is maintained to
a value of 3. The small peak in the nitrogen feed in the beginning could be due to slower response
of ratio controller.
• A pressure reduction of ∼ 9 bar is observed and then increases back to the desired set point
in around 15-20 minutes. A small overshoot is also observed. As expected, 50% reduction in
hydrogen feed flow rate resulted in 50% of ammonia production rate.
• The condenser cooling duty reduces from 1.2MW to 0.85MW i.e., a reduction of 29% is observed.
Hence, the outlet stream of condenser increases from -20◦ C to 17◦ C.
For 70% reduction in hydrogen feed flow rate, a pressure reduction of 68 bar was observed before the
addition of controllers in control philosophy-1. Figure A.2 shows the behaviour of the model under 70%
reduction in hydrogen feed flow rate.

(b) Ammonia Production Rate and Pressure at Reactor Outlet
(c) Cooling Duty of Condenser and Temperature of Condenser Outlet Stream
Figure A.2: Control Philosophy-1 response to 70% step change in hydrogen feed flow rate
A.2. Linear Change in Hydrogen Feed Flow Rate 86
Results:
• Based on the data shown in Figure A.2 (a) it can be observed that when the hydrogen feed flow
rate is reduced there is a decrease, in the nitrogen flow rate. Eventually it reaches a point where
the stoichiometric ratio of hydrogen to nitrogen (H2 : N2 ) remains at 3.
• A pressure drop of ∼ 13 bar is observed, then rises again to the desired set point within 10-15
minutes. As expected, reducing the hydrogen supply by 70% resulted in a reduction ammonia
production rate of 70%.
• In a similar way as the other two scenarios, the temperature of condenser outlet stream is in-
creased to increase the pressure in the loop. The cooling duty reduces from 1.2MW to 0.72MW
and the temperature increases from -20◦ C to 33◦ C.
A.2. Linear Change in Hydrogen Feed Flow Rate

Figure A.3 shows the response of control philosophy-1 to 50% gradual change in hydrogen feed flow
rate.

(b) Nitrogen Feed Flowrate and Pressure at Reactor Outlet

Figure A.3: Control Philosophy-1 response to 50% linear change in hydrogen feed flow rate
Discussion:
• Region A:When the controller detects a change in the ratio, it immediately closes the nitrogen
valve, resulting in a decrease in the nitrogen supply flow rate. A small peak is observed in the
nitrogen feed flow rate due to delay in controller response and a pressure reduction of 1 bar is
observed. The cooling duty of condenser is reduced to increase the pressure.
• Region B: Both the nitrogen and hydrogen feed flow rates are unchanging during this period,
maintaining a stoichiometric ratio of 3 (H2 : N2 ) while the system’s pressure remained constant
as well. The temperature of outlet of reactor reduces from 455◦ C to 375◦ C along with ammonia
production reduced by 50%.
• Region C: The hydrogen feed flow rate is gradually ramped up again to its original value for a 30
minute time interval. In this process, the nitrogen feed flow rate also increases proportionally to
uphold the stoichiometric ratio. As a result in increased feed flow rates, there is a very negligible
amount of deviation in pressure from the set point. During this phase, the system responds to
increase in pressure and therefore, the temperature of the outlet condenser stream is reduced by
increasing the cooling duty again.
• Region D: The simulation was run for 30 minutes more, allowing the system to gradually return
to its original steady-state conditions. During this period, the system’s pressure was maintained
at constant level, ensuring that it remains stable. The simulation was stopped after 30 mins. It
was observed that the system requires more than 30 minutes to reach the initial steady state
conditions.
Figure A.4 shows the response of control philosophy-1 to 70% linear reduction in hydrogen feed flow
rate.
(b) Nitrogen Feed Flowrate and Pressure at Reactor Outlet

Figure A.4: Control Philosophy-1 response to 70% linear change in hydrogen feed flow rate
Similar to 20% and 50% reduction in hydrogen feed flow rates, the system responds to 70% linear
reduction in hydrogen feed flow rate. The temperature of the outlet reactor stream drops from 455◦ C
to 345◦ C. The reduction in hydrogen feed flow rate also led to 70% reduction in ammonia production
rate. The cooling duty of condenser goes to -0.7MW thereby increasing the condenser outlet stream
temperature by 50◦ C.
B
B.1. Step Change in Hydrogen Feed Flow Rate
The following Figure B.1 shows the response of the control philosophy-2 to a 50% reduction in hydrogen
feed flow rate.
91
B.1. Step Change in Hydrogen Feed Flow Rate 92
Figure B.1: Control Philosophy-2 response to 50% step change in hydrogen feed flow rate
It was observed that there is an initial pressure reduction of ∼ 8 bar. The controller responds to change
in pressure and therefore rectifies the pressure back to the set point in ∼ 15 min. To increase the
pressure back to the initial value, the recycle compressor brake power reduces from 12.5kW to 1kW.
The following Figure B.2 shows the response of control philosophy-2 to 70% reduction in hydrogen feed
flow rate.

B.2. Linear Change in Hydrogen Feed Flow Rate 93
Figure B.2: Control Philosophy-2 response to 70% step change in hydrogen feed flow rate
It was noticed that for 70% reduction in hydrogen feed flow rate, the brake power of the compressor
almost reaches to a negligible value. For the current scenario, a pressure reduction of ∼ 12 bar was
observed. To retain the system pressure, the brake power of the recycle compressor reduces from
12.5kW to 0.15kW.
B.2. Linear Change in Hydrogen Feed Flow Rate

Figure B.3: Control Philosophy-2 response to 50% linear change in hydrogen feed flow rate
Results:
• Region A: A similar trend has been identified when observing reduction in hydrogen and nitrogen
flow rates, as depicted in Figure A.3 of control philosophy-1. When there is a linear reduction in
hydrogen feed flow rate by 50%, there is a gradual decrease in compressor brake power. This
reduction in compressor power correlates with a decrease in flow rate, resulting in a less inflow of
gas into the system. Therefore, the reduction in system pressure remains minimal and exhibits
only minor fluctuations. The ammonia production also reduces by 50% due to reduction in feed
flow rates. It is also observed that the brake power of the compressor also reduces gradually to
1 kW thereby, reducing the recycle flow rate too.
• Region B: For a span of 60 minutes, the hydrogen and nitrogen feed flow rates have been main-
tained constant in order to make the system reach its new steady-state as described in Table 5.1.
The pressure of the system remains constant during this phase of simulation, but the recycle flow
rate did not emerge to a steady-state value. This shows that the system requires more than 60
minutes to reach new steady-state conditions. The ammonia production rate also reduced by
50%.
• Region C: Hydrogen feed flow rate is ramped up back to original value for 30 mins and as a
result, the recycle flow rate also increases. In order to compress this increased recycle flow rate,
the brake power of the compressor also increases. The pressure in the system does not fluctuate
a lot.
• Region D: The system is left for 30 minutes to reach its original steady-state conditions. The cur-
rent model requires more time than simulated since the steady-state conditions are not reached
within 150 minutes.

Figure B.4: Control Philosophy-2 response to 70% linear change in hydrogen feed flow rate
A similar set of results are obtained as observed in 50% gradual reduction in hydrogen feed flow rate.
It has been observed that the brake power of the compressor almost reaches to 0 kW indicating the
compressor is shut down. The pressure in the system is regulated by this reduction in brake power and
the recycle flow rate also reduces. The flow rate of the recycle stream also reduces from 5000 kg/hr to
1000 kg/hr.
C
C.1. Step-Change in Hydrogen Feed Flow Rate
A step-change in reduction of hydrogen feed flow rate is implemented to examine the behaviour of
the system. This step-change is carried out by reducing hydrogen feed flow rate by 50% and 70%.
The response time of the controller and the effect of control action are noticed and plotted below in
Figure C.1.
97
C.1. Step-Change in Hydrogen Feed Flow Rate 98
Figure C.1: Control Philosophy-3 response to 50% step change in hydrogen feed flow rate
Results:
• When a 50% reduction in hydrogen feed flow rate is imposed, the nitrogen feed flow rate increases
by 30 kmol/hr initially and again reduces. This response is observed as nitrogen is pumped to
cover up the deficiency in hydrogen in the system.
• Due to this initial increase in the nitrogen feed flow rate, a peak in the pressure of the system is
also observed. A pressure increase of 19 bar is observed but later the pressure stabilizes to its
initial value.
• After the pressure is stabilized, the stoichiometric inlet ratio of hydrogen to nitrogen reduces from
3 to 2.56. This causes an increase in the nitrogen gas in the ammonia synthesis loop. Additionally,
the recycle flow rate increases by 5000 kg/hr due to excess nitrogen in the system.
In a similar way, a step change of reduction of reduction of 70% of hydrogen feed flow rate is carried
out and the results are shown below in Figure C.2.

Figure C.2: Control Philosophy-3 response to 70% step change in hydrogen feed flow rate
C.2. Linear-Change in Hydrogen Feed Flow Rate 101
A similar pattern is observed compared to 20% and 50% step reduction in hydrogen feed flow rate.
The nitrogen feed flow rate peaks its value to 75 kmol/hr from 31.5 kmol/hr in order to replace the
deficit hydrogen in the beginning. As the system gets adapted to reduced hydrogen feed flow rate,
the nitrogen feed flow rate again reduces to ∼ 12.5 kmol/hr. An initial pressure increase of 26.5 bar is
observed and thereafter, the pressure is maintained constant after the reduction in nitrogen feed flow
rate. Ammonia production rate also reduces by 70% due to less amount of reactants present. The
stoichiometric ratio of H2 : N2 in the inlet changes from 3 to 2.27 in order to maintain the pressure in
the ammonia synthesis loop. Due to the excess nitrogen, the recycle flow rate increases significantly
by 6200 kg/hr which possesses a challenge to performance of the system.
C.2. Linear-Change in Hydrogen Feed Flow Rate

The effect of 50% and 70% gradual reduction in hydrogen feed flow rates are shown in the following
section. Figure C.3 shows the effect of 50% gradual reduction in hydrogen feed flow rate.

Figure C.3: Control Philosophy-3 response to 50% linear change in hydrogen feed flow rate
Discussion:
• Region A: A small increase in nitrogen feed flow rate is observed similar to the system’s response
to step change in hydrogen feed flow rate. After 10 minutes, the system again reduces the ni-
trogen feed flow rate. A 1.5 bar pressure increase in observed and in 20 minutes the controller
stabilizes the pressure to its original value.
suggesting a stable system response. The input hydrogen to nitrogen ratio stabilizes at 2.55 re-
sulting in excess nitrogen in the system than at normal conditions. The recycle flow rate increases
by 5000 kg/hr.
• Region C: With only a minor decrease in pressure, the pressure practically remains constant.
The inlet ratio rises from 2.8 to 4, indicating an adjustment to the changing hydrogen supply flow
rate. Both the recycle and purge flow rates drop in reaction to the increasing hydrogen feed flow
rate.
• Region D: All the parameters do not reach to their initial values and therefore, a 150 minute
simulation timeline is not sufficient for the system to reach the original steady state conditions.
The model’s response to 70% gradual reduction in hydrogen feed flow rate is shown the Figure C.4
below.

Figure C.4: Control Philosophy-3 response to 70% linear change in hydrogen feed flow rate
Discussion:
• Region A: A small increase in nitrogen feed flow rate is observed similar to all other responses.
After 10 minutes, the system again reduces the nitrogen feed flow rate. A 2 bar pressure in-
crease in observed and in about 15 minutes the pressure reduces and the controller stabilizes
the pressure to its original value
suggesting a stable system response. The input hydrogen to nitrogen ratio stabilizes at 2.27 re-
sulting in excess nitrogen in the system than at normal conditions. The recycle flow rate increases
by 6500 kg/hr.
• Region C: The pressure practically remains constant with minor fluctuations. The inlet ratio rises
from 2.8 to 4.3, indicating an adjustment to the changing hydrogen supply flow rate. The recycle
flow rate drops due to hydrogen being pumped into the system and almost reaches to its initial
value.
• Region D: The simulation needs to be carried out for more than 150 minutes since the parameters
do not reach their initial steayd state conditions. In this region the inlet ratio reduces from 4.5 to
3 due to the response of the ratio controller.
D
Economic Evaluation
D.1. Determination of Required Back-Up Electricity
The amount of heat gained/lost by a substance or sample is represented as:
Q = m × Cp × (Toperating − Tambient ) (D.1)
The above Equation D.1 is used to calculate the amount of heat required to heat the catalyst bed in the
reactor where m is the mass of the catalyst in kg and Cp is the specific heat capacity of the catalyst. In
order to calculate the amount of heat required to heat the catalyst, the mass of catalyst present in the
reactor should be determined. This is calculated by:
Wc = ρB × VR (D.2)
where,
ρB = (1 − ϵ)ρp
ρB is the bed density of catalyst

ρp is the particle density of catalyst (2200 kg/m3 )
ϵ is the bed voidage which is the percentage of bed volume uncovered by the particles (0.33)
VR is the volume of the reactor (= 4.58 m3 with 0.9 m diameter and 7.2 m length of reactor)
Substituting the above values in the above Equation D.2, the mass of catalyst is 6751 kg.
The thermal conductivity of iron is 0.451 kJ/kg◦ C [92]. The operating temperature of the reactor is
300◦ C and ambient temperature is considered as 25◦ C. Substituting all the values in the Equation D.1,
the heat required to heat up the iron catalyst bed in the reactor from 25◦ C to 300◦ C is 835436.25 KJ
which translates into 232 kWhr.
The heat transfer between the reactor and surroundings i.e., the heat loss is calculated by:
Q = U × A × (Toperational − Tambient ) (D.3)
where,
”U” is the heat transfer coefficient of insulation material. Glass wool is considered as insulation
material with a thickness of 150 mm (0.2 W/m2 K) (thermal conductivity of glass wool is 0.023-
0.040 W/mK)
”A” is the surface area of the reactor (21.62 m2 )
106
D.2. Hydrogen Storage Tank Level 107
Substituting the above values in Equation D.3, the total heat lost is 1189.6 W, i.e., 1.18 kW. It is assumed
that the dispersion of heat inside the reactor is much faster than loss of heat to the surroundings.
Therefore, the total amount of back-up electricity required to keep the reactor warm is 233.18 kW
D.2. Hydrogen Storage Tank Level

The following image below shows the storage level of the 300 kg hydrogen buffer considered as a
function of time for 30 days.
Figure D.1: Hydrogen Tank Storage Level (300 kg)

Excellent Thesis Design

Uploaded by

Copyright:

Available Formats

Excellent Thesis Design

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Excellent Thesis Design

Uploaded by

Copyright:

Available Formats

Dynamic Simulation of

to obtain the degree of Master of Science

Thesis Committee: Prof. Dr. Ir. Earl Goetheer, TU Delft

An electronic version of this thesis is available at http://repository.tudelft.nl/.

5.5 Control Philosophy-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

1.1 Power to Hydrogen to Ammonia to usage chain [3] . . . . . . . . . . . . . . . . . . . . . 1

2.1 Illustration of wind speed in Boujdour, Morocco [12] . . . . . . . . . . . . . . . . . . . . . 8

4.1 Conceptual Design of Ammonia Synthesis Plant . . . . . . . . . . . . . . . . . . . . . . 27

5.1 Control Philosophy-1 from Aspen Plus Dynamics . . . . . . . . . . . . . . . . . . . . . . 35

6.1 Three Different Options for Green Ammonia Production . . . . . . . . . . . . . . . . . . 55

6.5 Capacity Factor vs Time for Wind Generator connected Electrolyzer . . . . . . . . . . . 61

D.1 Hydrogen Tank Storage Level (300 kg) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

1.1 Characteristics of ammonia synthesis unit for present thesis work . . . . . . . . . . . . . 4

4.1 Flash Vessel Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

5.1 Timeline of Simulation for Linear Change . . . . . . . . . . . . . . . . . . . . . . . . . . 35

6.1 Solar PV connected Electrolyzer results . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

Figure 1.1: Power to Hydrogen to Ammonia to usage chain [3]

1.2. Ammonia as a hydrogen carrier

• High gravimetric hydrogen storage densities

1.3. Research Questions

– How can the dynamic behaviour of the system be modelled?

1.4. Objective & Scope

Figure 1.4: Complete schematic of Power-to-Ammonia process

Figure 1.5: Ammonia Synthesis Plant Flowsheet

Figure 1.6: Overview of Approach

The three main elements of the performed research are:

• The effect of change in hydrogen feed flow rate is examined

1.6. Organization of this document

Figure 1.7: Organization of the Report

Figure 2.1: Illustration of wind speed in Boujdour, Morocco [12]

2.2. Hydrogen Production & Storage

Figure 2.4: Comparison of various hydrogen production sources [20]

• Alkaline Water Electrolysis

2.2.1. Alkaline Water Electrolysis

Cathode : 2H2 O + 2e− −→ 2OH − + H2 (2.2)

A basic schematic of alkaline water electrolyzer is shown below in Figure 2.5

Figure 2.5: Alkaline Electrolyzer [22]

∆Greac = ∆Hreac − T ∆Sreac (2.4)

• It is a well established technology and has long term stability

Some of the disadvantages are [28]:

• Low current densities and possibility of gas crossover

2.2.2. Proton Exchange Membrane (PEM) Electrolysis

PEM electrolyzers consist of solid polysulfonated membranes (Nafion,fumapem) as an electrolyte which

Figure 2.6: PEM Electrolyzer [22]

Some of the disadvantages associated are [27]:

2.2.3. Solid Oxide Electrolysis

Figure 2.7: Schematic of solid oxide electrolysis [22]

H2 O + 2e− −→ H2 + O2− (2.8)

• High efficiency of hydrogen gas can be produced

Some of the drawbacks associated with SOECs are [33]:

2.2.4. Comparison of different electrolysis techniques

Figure 2.8: Comparison of different electrolysis techniques [34] [35] [36]

2.2.5. Hydrogen Storage

Figure 2.9: Different Technologies for hydrogen storage

Compressed hydrogen gas storage

Figure 2.10: Underground H2 storage facility [40]

+ Well established technology and commercially available

2.3. Nitrogen Production & Storage

2.3.1. Cryogenic Distillation