Simulation and Design of a Boil-Off Gas Re-liquefaction ... · Simulation and Design of a Boil-Off...

Simulation and Design of a Boil-Off Gas Re-liquefaction System

in a Small-Scale LNG Supply Chain

Case Study of Trafaria

Joana Alexandra Santos Antunes

Thesis to obtain the Master of Science Degree in

Energy Engineering and Management

Supervisor: Prof. Viriato Sérgio de Almeida Semião

Examination Committee

Chairperson: Prof. Francisco Manuel da Silva Lemos

Supervisor: Prof. Viriato Sérgio de Almeida Semião

Member of the Committee: Prof. Pedro Jorge Martins Coelho

June 2018

ii

Acknowledgments

To my parents, for their enormous and tireless help, and always being by my side.

To my Grandmother, example of resilience and strength.

To my Gabi, because you're part of me.

To my "Split", Ana, for never letting down and for fighting side by side with me.

To my dearest Emanuel, for the restless fight alongside with me, every day.

To the esteemed Professor Viriato Semião, for guiding me through this work and for his precious help and

tutoring.

To the most considered administration board of OZ Energia and all their personnel involved in this project,

in particular Dr. Micaela Silva.

To my dear cousin Amílcar, for his example of courage and support along the way, and for teaching me to

apply Aikido and shisei every day.

To my buddies at IST, Monica, and Maxime, who constantly reminded me of the importance of getting a

parachute.

iii

Abstract

Liquefied Natural Gas (LNG) supply chains have gained a relevant position in the energy market during the

last 20 years, and economic projections indicate it will continue to grow. The port of Lisbon offers

geographically worthy conditions to develop a local small-scale supply chain, and the present work

analyses the feasibility of the Trafaria fuels terminal to house a small-scale LNG (SSLNG) storage facility,

mainly dedicated to LNG ships bunkering. Thus, a preliminary study of the main characteristics of an

SSLNG terminal is made, including its BOG (Boil-off Gas) management system as a major concern within

the operation of an LNG terminal.

The location and dimensions of the chosen in-ground tank are selected, and heat input into the tank is

simulated through the definition of the dimensions and insulation of the tank, respecting the maximum

BOG rate defined, allowing also the calculation of energy use and possible operating costs.

As a method of managing the generated BOG, a re-liquefaction system based on a Nitrogen turbo-

expander cycle is studied. The system is simulated for two different configurations, allowing to determine

the most efficient one. Also, a cogeneration plant is considered as a potential solution to manage the

generated BOG; both solutions are simulated for different BOG rates, climate conditions and modes of

operation of the terminal. The cogeneration plant is found to be an interesting alternative to manage the

BOG together with a re-liquefaction system as backup, capable of processing all the BOG in the most

demanding operation scenario.

Keywords: Liquefied Natural Gas, Small-scale supply chain, Boil-off gas, BOG Re-Liquefaction, Single

Nitrogen Expansion.

iv

Resumo

As cadeias de abastecimento de Gás Natural Liquefeito (GNL) assumiram uma posição relevante no

mercado de energia nos últimos 20 anos, e as projecções económicas indicam que continuarão a seguir

esta tendência. Oferecendo o porto de Lisboa condições geográficas para desenvolver uma cadeia de

abastecimento de pequena dimensão, o presente trabalho analisa a viabilidade do terminal de

combustíveis da Trafaria albergar um terminal de GNL de pequena escala, principalmente dedicada ao

abastecimento de navios de GNL. Para este efeito, é realizado um estudo preliminar das principais

características de um terminal de GNL, incluindo o sistema de gestão de BOG (Boil-Off Gas) destacado

como questão de grande importância durante a sua operação.

A localização e dimensões do tanque enterrado são seleccionadas e a entrada de calor no tanque é

simulada após definição das dimensões e isolamento do tanque, respeitando a taxa BOG máxima definida,

permitindo o cálculo do uso de energia e possíveis custos operacionais.

Como método de gestão e recuperação de BOG é estudado um sistema de liquefacção baseado num ciclo

de turbo-expansão de Azoto. O sistema é simulado para duas configurações distintas, permitindo avaliar

qual a mais eficiente. Uma instalação de cogeração é também estudada como solução para gerir o BOG.

Ambas as soluções são simuladas para diferentes taxas de BOG, condições climáticas e modos de

operação. A instalação de cogeração constitui uma alternativa interessante e viável para gerir o BOG,

juntamente com um sistema de re-liquefação como backup, capaz de processar o BOG produzido.

Palavras-Chave: Gás Natural Liquefeito, Cadeia de abastecimento de pequena escala, Boil-off gas,

Reliquefacção de Boil-off gas, Expansão Simples de Azoto.

v

Table of Contents

1. Introduction ......................................................................................................................................... 1

1.1 Scope of the Thesis .......................................................................................................................... 1

1.2 LNG as an Alternative in the Energy Field ....................................................................................... 3

1.2.1 Historical Context ............................................................................................................... 5

1.2.2 LNG vs Other Fuels ............................................................................................................. 6

1.2.3 Economics of LNG............................................................................................................... 9

1.2.4 LNG in Portugal ................................................................................................................ 12

1.2.5 Portugal, Lisbon, as a strategic location for an LNG terminal .......................................... 12

1.3 Objectives of the Thesis ................................................................................................................ 15

1.4 Thesis Outline ................................................................................................................................ 15

2. The Case-Study .................................................................................................................................. 17

2.1 Description of the Company .......................................................................................................... 18

2.2 The Terminal – Case Study ............................................................................................................ 19

2.2.1 Characteristics of the Terminal ........................................................................................ 19

2.3 Choosing the Storage Facilities ..................................................................................................... 21

3. Literature Review .............................................................................................................................. 28

3.1 LNG Receiving Terminal................................................................................................................. 28

3.1.1 The Jetty ........................................................................................................................... 28

3.1.2 Storage ............................................................................................................................. 30

3.1.3 Materials and Insulation .................................................................................................. 32

3.1.4 Boil-Off Gas Management ................................................................................................ 34

3.1.5 LNG Regasification ........................................................................................................... 38

3.2 Liquefaction processes .................................................................................................................. 39

3.2.1 Basic Principles of Liquefaction ........................................................................................ 39

3.2.2 LNG Liquefaction Processes ............................................................................................. 42

3.2.2.1 Cascade Liquefaction Cycles ............................................................................................. 43

3.2.2.2 Mixed Refrigerant Liquefaction Cycles ............................................................................. 44

3.2.2.3 Gas Expander Cycles ......................................................................................................... 45

3.2.2.4 Comparing and Choosing the Liquefaction Technologies ................................................ 46

3.2.3 The Chosen Liquefaction Technology - N2 Expander Cycle .............................................. 47

3.3 Contributions of the Thesis ........................................................................................................... 51

4. Modelling and Simulations Methodology ........................................................................................ 52

4.1 Operation Conditions .................................................................................................................... 52

4.1.1 LNG Properties ................................................................................................................. 52

4.1.2 Climate Characterization .................................................................................................. 53

4.1.3 Tank Operation Modes ..................................................................................................... 54

4.2 Designing the Tank ........................................................................................................................ 55

vi

4.2.1 Initial Design Conditions and Intended Characteristics of the Tank ................................ 55

4.2.2 Choosing the Insulation .................................................................................................... 57

4.3 Boil-off Production ........................................................................................................................ 59

4.3.1 Defining the Maximum Boil-Off Gas Rate ........................................................................ 59

4.3.2 Heat Input to the Tank ..................................................................................................... 61

4.3.3 Heat Input through Piping and Pumping System ............................................................. 64

4.4 Simulation Scenarios ..................................................................................................................... 66

4.4.1 BOG Production ................................................................................................................ 66

4.4.2 BOG Management ............................................................................................................ 66

5. Results and Discussion ...................................................................................................................... 70

5.1 Heat Input...................................................................................................................................... 70

5.1.1 Designed Heat Ingress to the Tank ................................................................................... 70

5.1.2 Designed Heat Ingress through Piping and Pumping System .......................................... 71

5.1.3 Total Heat Ingress............................................................................................................. 72

5.1.4 Insulation Costs ................................................................................................................ 72

5.1.5 Holding Mode Operation ................................................................................................. 73

5.2 BOG Management ......................................................................................................................... 73

5.2.1 Single Expander Re-Liquefaction System ......................................................................... 73

5.2.2 Cogeneration .................................................................................................................... 76

5.2.3 Emergency System ........................................................................................................... 77

6. Concluding Remarks .......................................................................................................................... 78

6.1 Conclusions.................................................................................................................................... 78

6.2 Future Work .................................................................................................................................. 80

References .................................................................................................................................................. 81

ANNEXES .................................................................................................................................................... 88

I. Gas Infrastructure: Europe’s LNG Map of 2018 ........................................................................... 89

II. Gas Infrastructure: Portugal ......................................................................................................... 90

III. Trafaria’s Terminal – Plant and General View ......................................................................... 91

IV. Tank Projects and Examples – Determining the Dome Height and Characteristics ............... 92

V. Radiation and Temperature Data Collected ................................................................................ 96

VI. LNG Regasification ................................................................................................................... 97

VII. Other Liquefaction Cycles ...................................................................................................... 100

VIII. Tank Project – Final Drawing ................................................................................................. 102

IX. Tank Insulation Examples ...................................................................................................... 103

X. Mollier Diagram - Methane ........................................................................................................ 106

XI. Mollier Diagram - Nitrogen .................................................................................................... 107

XII. Mollier Diagram – Liquefaction Cycles A and B ..................................................................... 108

XIII. Mollier Diagram – Methane Cycle ......................................................................................... 109

vii

XIV. Simulation Results - Liquefaction Cycles A and B .................................................................. 110

XV. Re-liquefaction System – Capacity Control Scheme.............................................................. 122

Tables Table 1 - SECA obligations: Past, present and future targets [17]. ................................................................................. 4

Table 2 - Comparison of physical and chemical properties of LNG with Diesel, gasoline and LPG [25]; [26]; [27]. ....... 6

Table 3 - Comparison of Fuel Emissions (Parts per billion of energy input) ................................................................... 9

Table 4 - General vessel traffic entering the Port of Lisbon in 2015 and 2016 [47] – GT: Gross Tonnage. .................. 13

Table 5 - Travelling Distance from Lisbon to the main European ports and e Algeria ................................................. 14

Table 6 - Summary of the Advantages and Disadvantages of the Three Considered Locations for the LNG Facilities. 25

Table 7 - Main differences between SSLNG and LSLNG Terminals [63] ....................................................................... 29

Table 8 - Materials Used on the Construction of an LNG Tank and Principal Characteristics. ..................................... 34

Table 9 - Comparison of Different LNG Liquefaction Technologies [91]. ..................................................................... 47

Table 10 – LNG Composition Expected in Trafaria Terminal [113]. .............................................................................. 53

Table 11 – Methane Properties [114]. ......................................................................................................................... 53

Table 12 – LNG Tank Specifications. ............................................................................................................................. 56

Table 13 - BOG Scenarios and Results .......................................................................................................................... 70

Table 14 – Heat Ingress in the LNG System through Piping and Pumping System ....................................................... 72

Table 15 – BOG Scenarios and Results - Piping and Pumping System .......................................................................... 72

Table 16 – Heat Ingress in the LNG System during each Operation Mode .................................................................. 72

Table 17 - Tank Insulation Costs ................................................................................................................................... 72

Table 18 – Holding Mode Capacity............................................................................................................................... 73

Table 19 – Liquefaction Cycles Simulation Results ....................................................................................................... 74

Table 20 – Liquefaction Cycles Performance Assessment ............................................................................................ 75

Table 21 – Coefficients of Performance ....................................................................................................................... 76

Table 22 – Cogeneration Prospect Scenarios ............................................................................................................... 76

Table 23 – Flare Profit Loss .......................................................................................................................................... 77

Table 24 – Synthesis of the Characteristics of the Examples found in the Literature and Industrial Projects. ............ 93

Table 25 – Conclusions Regarding the Dome Height for different types of tanks, with two different approaches. ... 94

Table 26 - Dome Characteristics .................................................................................................................................. 95

Table 27 – Meteorological Data Collected – Monte de Caparica Station ..................................................................... 96

Figures Figure 1 - Portuguese Energy Dependency (2000 - 2014) [1] ......................................................................................... 1

Figure 2 - Primary Energy Consumption in 2005 ............................................................................................................ 2

Figure 3 - Primary Energy Consumption in 2014 ............................................................................................................ 2

Figure 4 - Established and under consideration Emissions Controlled Areas [16] ......................................................... 4

Figure 5 - Resulting Emissions of the Combustion of Natural Gas vs Emissions Resulting from the Combustion of

Diesel [16] ...................................................................................................................................................................... 5

Figure 6 - Levelized Cost of Energy [32] ......................................................................................................................... 8

Figure 7 - LNG Value Chain ............................................................................................................................................. 9

Figure 8 – Illustrative comparison of NG transportation: LNG carrier vessel versus Pipeline Transport of NG [21] .... 10

Figure 9 - Major import and export regions [38] ......................................................................................................... 11

Figure 10 - Main marine international traffic routes [45] ............................................................................................ 13

Figure 11 - Number of ships, per type, which entered the Port of Lisbon between 2015 and 2016 [47]. ................... 14

Figure 12 - Full Gas Field Processes .............................................................................................................................. 17

Figure 13 - Trafaria Terminal - Aerial View [50]. .......................................................................................................... 19

Figure 14 - Flow on the river estuary [52]. ................................................................................................................... 20

Figure 15 - Map of Conditioning Factors: Trafaria Terminal and Surrounding Areas [55] ........................................... 22

Figure 16 - General view of Trafaria Terminal with the three possible places to install the reservoirs (identified in

green) ........................................................................................................................................................................... 23

Figure 17 - First place considered – alongside the gasoline reservoirs (A in Figure 16) ............................................... 24

Figure 18 - Second place considered – where reservoirs T1 to T9 and T18 are placed (B in Figure 16)....................... 24

Figure 19 - Third location considered - Unbuilt terrain (C in Figure 16) ....................................................................... 25

viii

Figure 20 - Calculated distance from the Jetty to Location C (273.9 metres). ............................................................. 26

Figure 21 – Final location of the LNG reservoir. ........................................................................................................... 27

Figure 22 - LNG Tanker Unloading ............................................................................................................................... 29

Figure 23 – Double Containment LNG Storage Tanks in Sines, Portugal [161]............................................................. 30

Figure 24 - Single Containment Storage Tank .............................................................................................................. 31

Figure 25 - Double Containment Storage Tank ............................................................................................................ 31

Figure 26 - Full Containment Storage Tank .................................................................................................................. 31

Figure 27 - In-ground Membrane Storage Tank ........................................................................................................... 31

Figure 28 - Process Flow of BOG Handling System [87] ............................................................................................... 37

Figure 29 - Working principle of a refrigeration / liquefaction system ........................................................................ 39

Figure 30 - Basic BOG Liquefaction .............................................................................................................................. 40

Figure 31 - Liquefaction Cycle [92] ............................................................................................................................... 40

Figure 32 - Cascade Cycle with Pure Refrigerants [95] ................................................................................................. 43

Figure 33 - Example of Coldbox [97] ............................................................................................................................ 43

Figure 34 – Pure refrigerants commonly used to produce LNG - Vapour Pressure Curves [95] .................................. 44

Figure 35 - Mixed Refrigerant Liquefaction Cycle [95] ................................................................................................. 44

Figure 36 - Qualitative Comparison of Different NG Liquefaction Technologies ......................................................... 46

Figure 37 – Reversed-Brayton Cycle............................................................................................................................. 47

Figure 38 – Single Turbo-Expander Cycle ..................................................................................................................... 48

Figure 39 - Dual Turbo-Expander Cycle ........................................................................................................................ 49

Figure 40 - Ambient Temperature during the typical days of each season- Reference Period .................................... 54

Figure 41 – Solar Radiation during the typical days of each season - Reference Period .............................................. 54

Figure 42 - Soil Temperature - Reference Period ......................................................................................................... 54

Figure 43 - Tank Project Scheme with Main Dimensions and Levels ........................................................................... 57

Figure 44 - Bottom of the tank – Composition (layers) ................................................................................................ 58

Figure 45 – Walls of the tank – Constitution (layers) ................................................................................................... 58

Figure 46 - Dome (Roof) of the tank – Constitution (layers) ........................................................................................ 59

Figure 47 - Daily percentage of BOG for different tank capacities and LNG methane content [122]. ......................... 60

Figure 48 – Linear Fitting of Ordinate Interception Values. ......................................................................................... 60

Figure 49 - Equivalent Heat Transfer Diagram – Dome ................................................................................................ 63

Figure 50 - Scheme of the Temperatures considered for the Heat Ingress Calculation in the Dome .......................... 63

Figure 51 - Single Expander Nitrogen Cycle ................................................................................................................. 67

Figure 52 - Daily Mass Flow Rate - BOG=0.039% ......................................................................................................... 70



Figure 55 - LNG Volume evaporated - BOG=0.0399% .................................................................................................. 71

Figure 56 - LNG Volume evaporated - BOG=0.050% .................................................................................................... 71

Figure 57 - LNG Volume evaporated - BOG=0.067% .................................................................................................... 71

Figure 58 - BOG Volume - BOG=0.039% ....................................................................................................................... 71



Figure 61 – Liquefaction Cycles A (in blue) and B (in orange) ...................................................................................... 74

Figure 62 - Evolution of the Flow Rate of Refrigerant Required Vs BOG inlet Temperature ....................................... 75

Figure 63 – Specific Consumption and Cost per Ton of BOG Processed ...................................................................... 75

Figure 64 - Coefficients of Performance - Cycle Comparison ....................................................................................... 76

Figure 65 – Europe’s LNG Map of 2018 [137] .............................................................................................................. 89

Figure 66 – Portuguese Gas Infrastructure [138] ......................................................................................................... 90

Figure 67 – Europe’s LNG Map of 2018 ........................................................................................................................ 91

Figure 68 – Tank Project 1 - In-ground Tank, 200 000 m3 [139] ................................................................................... 92

Figure 69 – Tank Project 2 – Above ground Tank, 15 000 m3 [140] ............................................................................. 92



Figure 72 – Tank Project 5 - In-ground Tank, 200 000 m3 [142] ................................................................................... 93

Figure 73 - Spherical Cap Illustration [143] .................................................................................................................. 95

Figure 74 - Submerged Combustion Vaporizer (SCV) ................................................................................................... 98

Figure 75 - Open Rack Vaporizer (ORV)........................................................................................................................ 98

Figure 76 - Shell and Tube Vaporiser (STV) .................................................................................................................. 98

Figure 77 - Ambient Air Vaporizer (AAV) ...................................................................................................................... 99

Figure 78 - Linde-Hampson Cycle [175]...................................................................................................................... 100

ix

Figure 79 - Claude Cycle [92] ...................................................................................................................................... 100

Figure 80 - Kapitza Cycle [156] ................................................................................................................................... 101

Figure 81 - Tank Project – Final Drawing .................................................................................................................... 102

Figure 82 - Insulation Example 1 [157] ....................................................................................................................... 103

Figure 83- Insulation Example 2 [157] ........................................................................................................................ 103





Figure 88 - Mollier Diagram – Methane [159] ............................................................................................................ 106

Figure 89 - Mollier Diagram – Nitrogen [159] ............................................................................................................ 107

Figure 90 - Liquefaction Cycles A and B [159] ............................................................................................................ 108

Figure 91 - Methane Cycles [159] .............................................................................................................................. 109

Figure 92 - Capacity Control Scheme of the Re-liquefaction System ......................................................................... 122

x

Abbreviations

AGRU – Acid Gas Removal Unit

BOG – Boil-Off Gas

CAPEX – Capital Expenditure

DGEG – Direcção Geral de Energia e Geologia

ECA – Emission Control Area

FLNG – Floating Liquefied Natural Gas

FSRU - Floating Storage and Regasification Units

GCV – Gross Calorific Value

GHG – Green House Gas

GIS – Geographical Information System

GT – Gross Tonnage

HFO - Heavy Fuel Oil

IMO – International Maritime Organization

JT – Joule-Thomson

KPI – Key Performance Indicator

LSLNG Terminal – Large-Scale Liquefied Natural Gas Terminal

LNG – Liquefied Natural Gas

LPG – Liquefied Petroleum Gas

MARPOL – Marine Pollution

MDO – Marine Diesel Oil

MRC – Mixed Refrigerant Cycles

Mtpa – Million metric tonnes per annum

NG – Natural Gas

NGL – Natural Gas Liquids

OPEX - Operational Expenditure

PM – Particulate Matter

REN – Reserva Ecológica Nacional

TEU - Twenty-foot equivalent unit

Toe – Ton of Oil Equivalent

TPES – Total Primary Energy Source

SECA – Sulphur Emission Control Area

SSLNG Terminal – Small-Scale Liquefied Natural Gas Terminal

xi

Definitions

LNG Terminal Operator: entity responsible for the maintenance and exploration of the terminal, as well

as its capacity of storage, regasification in safety conditions and quality of the service delivered.

Slot: time frame attributed to the unloading of LNG and indorsed by the LNG Terminal Operator to a

market agent or another entity on their charge, to perform the reception of the vessel and its unload

(storage and regasification of LNG).

Jetty: pier where ships berth to perform the loading or supply of LNG.

Twenty-foot equivalent unit (TEU): inaccurate unit of lad capacity of a container ship - 1 TEU is equivalent

to an ISO container of 20 feet long, 4.25 feet height, and 8 feet width. This represents roughly a container

volume of 19.26 m3.

Loaded Draft: draft (part of the ship that is immersed) that a ship undertakes while full loaded.

1

1. Introduction

1.1 Scope of the Thesis

Portugal presents a great potential on the area of renewable energies, possessing a great annual

insolation (energy per unit area), favourable wind features (onshore and offshore) and excellent orographic

characteristics for the development and existence of technologies that allow to seize the available natural

resources for power generation. Despite this fact, these renewable energy sources were only able to produce

28% of the annual energy consumption in 2014 [1] – see Figure 1. This fact results from the scarcity of

endogenous fossil energy sources, still the driving force of developed countries nowadays, resulting in a high

energy dependence on other countries, quite above EU15 average, situation which might cause severe

perturbations on the economic activity in case of political or military instability [2].

Figure 1 - Portuguese Energy Dependency (2000 - 2014) [1]

Relying on the fact that the present society and its dynamics depend on an oil-based energy economy,

there is an actual concern on diversifying the supply of the national energy mix and also on reinforcing the means

of reception and storage of the several energy sources used.

According to DGEG (Direcção Geral de Energia e Geologia), the primary energy consumption in 2014

attained a total of 20,920,916 toe, resulting in a decrease of 2.8 % when compared with 2013. It is visible that

natural gas and, more relevantly, renewable energies started contributing more to the total energy consumed,

decreasing the total oil consumed, despite the fact that this form of primary energy continues to be the main

source of primary energy. In fact, if we broaden the period of observation, say from 2005 to 2014, this becomes

even more evident as it is possible to observe in Figures 2 and 3.

2

Figure 2 - Primary Energy Consumption in 2005 Figure 3 - Primary Energy Consumption in 2014

In order to obtain all the primary energy necessary to supply the Portuguese energy demand in 2014, coal

was mainly obtained from Columbia (88%), United States of America (7%) and South Africa (4%), oil was acquired

from Angola (26%), Saudi Arabia (13%) and Algeria, Nigeria and Kazakhstan (10%), whereas natural gas was

imported mainly from Algeria (58.6%) and Qatar (13.4%).

This strong dependence on energy sources original from unstable suppliers (mostly developing countries

inserted on problematic areas or presenting high political instability), as mentioned by Chico and Mancarella

(2007), raised the issue of the risk of our current energy safety [3], leading to the development of a national

energy strategy (Estratégia Nacional para a Energia). This plan comprises a set of measures to improve the

energy panorama in Portugal by 2020 [4]:

- Reducing the overall consumption of energy by 20%;

- Improving and promoting energy utilization efficiency;

- Guaranteeing energy supply safety;

- Working on overcoming the energy dependency, in order to reduce this value to 74%;

- Support and increase the contribution of renewable energies to the overall primary energy availability.

The investment on renewable energy sources has been a core part of the Portuguese energy policy, at least

until the economic crisis changed priorities of the government [5], when the subsidy for investments in this area

was reduced. Nevertheless, knowing that Portuguese economy is still based on oil economy (74.3% TPES in 2014)

[6] and that this source of primary energy is known to account for a great responsibility on anthropogenic

environmental pollution, this is one of the several energy management and emission-control measures aiming

at sustainability that are being implemented worldwide. By analysing Figure 1 it is possible to assess that energy

dependency has now achieved values lower than 74%, a promising outcome for the national energy strategy.

This attempt to improve the Portuguese energy system also complies with more than 500 identified climate

change laws in the world’s leading economies [7], which are becoming increasingly stricter and subjecting energy

consumers to adopt less pollutant energy sources in order to decrease the associated carbon footprint. This

circumstance, alongside with a preparation for the impacts of climate change, has become a great incentive on

Oil 59%Natural Gas

14%

Renewables13%

Coal 12% Other 2%

2005

Oil Natural Gas Renewables Coal Other

Oil 43%

Natural Gas17%

Renewables26%

Coal 13%Other 1%

2014

Oil Natural Gas Renewables Coal Other

3

searching for new technologies, cleaner fuels and renewable energies, aiming at answering positively to the

imposed targets and determining this way the global evolution of technology and economy [8].

Natural gas has become one of the current and most promising alternatives to fuel oil, diesel and petrol, as

it represents a solution for the reduction of pollutants emissions, especially for NOX and PM [9], allowing to

comply with upcoming and stricter environment legislation.

It is within this scope, the urge for developing new energy strategies and mitigating climate changes, as is

the case of replacing fuel oil and/or coal by natural gas, that the present work appears.

1.2 LNG as an Alternative in the Energy Field

The world primary energy demand is anticipated to grow 1.5% annually from 2012 to 2035, while natural

gas consumption is expected to grow 1.9% per annum. With the increase of the global gas trade, projections

indicate that LNG (Liquefied Natural Gas) will account for 15% of the global gas consumption, with an annual

growth of 3.9% in LNG trade, being expected that it will play a large role on the energy economy [10].

A number of facts have combined to stimulate the proven interest in LNG: the favourable economics of

electricity generation from gas-fired combined cycles (CCGT) has made gas the fuel of choice for power

generation, and there was a substantial drop on LNG costs. This situation has made formerly uneconomic trades

appear attractive at a worldwide level. Both this factors, alongside with the interest on finding supplements for

traditional pipeline supply and regulatory obligations, have enhanced the LNG revolution [11]. Hence, as LNG

faces an increase on demand, it is clear that nowadays it will play an increasingly important role in the industry

and energy markets due to its abundance, clean-burning properties and efficiency as a fuel. Nevertheless, it is

important to safeguard that, with the overcome of, and tremendous worldwide investment in, renewable energy

sources, LNG used for energy generation on CCGT might decrease. Yet, LNG is still facing a rising on its

consumption mostly in the shipping area, as it represents a regulation-compliant and cleaner fuel option.

Whilst HFO (Heavy Fuel Oil) is the most used fuel by merchant ships nowadays, LNG has been earning a

much stronger presence worldwide on the last two decades [12], being actually widely used and with a growing

evolution in the area of shipping [13], as per the different risks it presents while being handled, as for its

regulation-compliant pollutant characteristic when compared with conventional maritime fuels such as HFO.

The environmental legislation imposed has become progressively stringent over the last few years in order

to overcome the climatic changes deriving from the human actions. As so, IMO (International Maritime

Organization), knowing that international shipping represents approximately 2.4% of anthropogenic GHG (Green

House Gases) emissions and that these are expected to increase in the future [14], applied its jurisdiction defining

maritime areas with limited emissions of NOX, SOX and CO2 allowed (ECA/SECA – Emission Controlled Areas /

Sulphur Emission Controlled Areas) and obliging ships navigating on those areas to comply with the defined rules

(MARPOL – Maritime Policy), meaning that these are obliged to use fuels which burning does not produce

emissions above the limits, leading towards LNG consumption instead of HFO [15] - see Figure 4.

4

Figure 4 - Established and under consideration Emissions Controlled Areas [16]

This legislation has enforced ship owners to search for new ways to comply with the obligations of maximum

emissions allowed, as clarified on Table 1, regarding the emission limits allowed inside and outside ECA/SECA

zones imposed to ships.

Table 1 - SECA obligations: Past, present and future targets [17].

Outside an ECA - established SOX limit Inside an ECA - established SOX limit

4.50% m/m prior to 1 January 2012 1.50% m/m prior to 1 July 2010

3.50% m/m on and after 1 January 2012 1.00% m/m on and after 1 July 2010

0.50% m/m on and after 1 January 2020* 0.10% m/m on and after 1 January 2015

* Depending on the outcome of a review to be concluded by 2018, as to the availability of the required fuel oil, this date could be deferred

to 1 January 2025.

All these targets have boosted, just as the equivalent for road transportations (European Emission

Standards), the urge to develop new technologies and new fuels that allow compliance with those restrictions.

Transportation firms (both shipping and bus companies) have reached a point where a choice must be made:

either an investment is made to improve the emission control systems (catalysers, filters), or new fuels and

combustion engines are developed to meet the requirements determined, making LNG to protrude amongst

other fuels as an immediate panacea for the regulatory demands [18].

5

Being so, LNG has become one of the

pronounced answers to comply with MARPOL

regulations [19]. Due to its chemical and

combustion properties, the use of LNG allows for

a significant reduction of PM, SOX, NOX and CO2,

mostly when compared with HFO: thanks to lean

combustion that occur on dual-fuel internal

combustion engines, NOX emissions are reduced

by 80-85%; as LNG does not contain precursors

of SOX, those emissions are almost inexistent;

CO2 emissions decrease 20-30%, particularly

when comparing to HFO/MDO (Marine Diesel

Oil), due to a higher hydrogen content in

molecules – see Figure 5.

However, due to a leak of unburned methane through the engine, gas engines might be affected by

methane slips [20]. According to IPCC (Intergovernmental Panel on Climate Change) Fourth Assessment Report,

methane has a 100-year global warming potential (GWP 100) that is 25 times higher than CO2, meaning that 1 kg

of methane has the same warming potential as 25 kg of CO2 sent out to the atmosphere. As so, it is necessary to

control the methane slips or the benefits associated with the replacement of HFO/MDO for LNG will be truly

reduced.

Despite the fact that there are increasingly more actions towards the mitigation of GHG emissions,

projections indicate that maritime CO2 emissions are probable to increase by 50% to 250% up to 2050, and it is

also expected that methane emissions will increase rapidly as well, since the share of LNG on the fuel mix is

increasing [14] and replacing HFO. Efficiency improvement is also expected to be one of the most important

factors contributing to mitigate emissions increase.

1.2.1 Historical Context

Natural gas has been the energy resource with most search growth on the last two decades [12], due to its

low GHG emissions as well as a great efficiency when it comes to converting in power generation, and also being

transported in safer and more efficient ways through pipelines. This mean of transportation was ideal during the

20th century, time when the gas reserves necessary to fill up the demand for natural gas were in easily accessible

places. However, the current natural gas reserves with sufficient amounts to answer for the present demand are

not always conveniently located. This means more isolated reserves are being explored, being of difficult

technical access or economically unviable, forcing the development of new solutions for LNG exploration and

commercialization.

The discovery of LNG was initiated early in the 19th century by experiments of the British scientist Michael

Figure 5 - Resulting Emissions of the Combustion of Natural Gas vs Emissions Resulting from the Combustion of Diesel [16]

6

Faraday, with the first practical compression refrigerator developed earlier by a German engineer, Karl Von Linde.

In 1912, the first LNG plant started being built in West Virginia, U.S.A., and started operating in 1917. However,

only in 1941 another LNG plant was built for commercial purposes, in Cleveland, Ohio. The world’s first LNG

tanker, “Methane Pioneer” delivered the first cargo [21].

LNG has been largely used in the past 50 years [22], especially on the U.S.A., Europe and Asia, opening this

scale economy to bigger markets and competing this way against other energy sources such as coal or oil, being

used either for energy generation or for industrial or commercial applications [23].

On the past years, only the LNG industry has been able to explore and deliver successfully natural gas

original from remote locations without the inevitability of building infrastructures as pipelines, being the

maritime transportation actually a great competitor against the market which was captured by the pipelines,

introducing more safety in the transport and commercialization of LNG and reducing the possibility of political

and geopolitical confronts concerning worldwide supplies.

1.2.2 LNG vs Other Fuels

LNG is stored at temperatures between -162⁰C and -170⁰C, slightly above atmospheric pressure and results

from the liquefaction of natural gas. This compound occupies a volume 600 times smaller than its gaseous form

and is chemically composed in average of 85 to 99% methane (CH4), and may also contain hydrocarbons as

ethane (C2H6), propane (C3H8), or butane (C4H10) and typically less than 1% nitrogen (N2) [24].

One of the main attractive features of LNG as an alternative to the conventional fuel is the fact that its

specific energy content largely overcomes the available energy contents from diesel or gasoline, which is evident

from Table 2. Also, it could be efficiently used on ships or airplanes as engine fuel owing to its high octane number

and simpler maintenance [12].

Table 2 - Comparison of physical and chemical properties of LNG with Diesel, gasoline and LPG [25]; [26]; [27].

Properties LNG Diesel Gasoline LPG

Energy Content (MJ/kg) 55 48 46.4 46.3

Auto-ignition Point (⁰C) 540 210 246 470

Stored Pressure Atmospheric Atmospheric Atmospheric Pressurized

Toxic No Yes Yes No

Carcinogenic No Yes Yes No

Health Hazards None None Eye Irritant None

Despite the fact that there are no direct health hazards associated with LNG, it is important to refer that,

once regasified and in confined spaces, it might cause asphyxiation due to the displacement of oxygen on

breathed air [25]. Also, due to its low temperatures, in case of spilling and contact with the skin, LNG causes

cryogenic freeze burns, which can be fatal.

7

As for effects on the surrounding environment, an LNG spill might cause hazards as explosion, fire and

embrittlement of metals and plastics, being a distressing issue on board vessels if such accident occurs, as it can

pierce through the hull and result in severe and dangerous fractures [28]. If spilled on the ground it will boil fast

at first and then slower as the floor cools, vaporizing completely and without leaving residue. If spilled on water,

as water is considerably warmer than LNG, it will result in natural convection currents formed by the boiling LNG,

so it will only form ice if in the presence of shallow waters. Nonetheless, while LNG is not flammable or explosive

in its liquid state, if vaporized and mixed with air it may attain the flammable range and burn if an ignition source

is present. The flammable region of an LNG vapour cloud is typically between 4% and 15% concentration of gas

in air [29], and it is usually visible as a white cloud of water vapour and ice crystals, condensed out of the air from

the cold LNG vapour, and representing an imminent danger.

These possible hazards elucidate the major importance of performing a correct handling of LNG, which has

a set of particulate procedures to be taken into account while using, and that are not necessary while using other

fuels. From the point of view of the governments, LNG storage facilities must be protected against intentional

damage (sabotage or terrorist attack) and land planning and energy infrastructure should be done minding these

possible issues [25].

It is also important to refer that while diesel, HFO, gasoline or LPG are used just as they are stored, LNG still

requires being regasified in order to be burned, meaning that there is an extra expense of energy on liquefaction

to transport and store and further regasification. Moreover, and due to the low temperatures at which LNG is

stored, one major concern is the occurrence of boil-off gas (BOG) due to heat ingress in the LNG storage facilities.

This event requires simultaneously very good insulation and efficient methods to manage BOG.

The factor that has more influence on the energy content of natural gas is the origin and therefore the

composition of the gas, which may fluctuate according to the supplier. As so, and in order to know the exact

amount of energy that is expected on a certain quantity of natural gas, the Wobbe Index, IW, is used to

characterize the gas by indicating the compatibility of the natural gas supplied and the burner [30], being one of

the key features of a gaseous fuel. Such index is expressed by equation (1), where GCV is the Gross Calorific Value

(High Heating Value) or Energy, and d is the relative density of the gas.

𝐼𝑊 =𝐺𝐶𝑉

√𝑑 (1)

In order to illustrate the importance of this index, it is known that in the U.S.A., while the usual Wobbe

index for natural gas is usually around 1310 and 1390 (Btu/scf – Btu per Standard Cubic Foot), LNG, which usually

has a higher content of ethane, can reach values high above 1400. This means that LNG IW it could be beyond the

ratings of some gas-operated equipment, which have to be adjusted, or in some cases, require mixing with other

fuel sources in order to reduce the Wobbe index [31].

In order to unveil the prices of the different energy sources, the Levelized Cost of Energy (LCOE) offers a

method to compare and standardize the prices of energy (on this specific case, in USD/MWh), as it represents

the total lifecycle costs of producing electricity (one MWh) using a specific technology [32].

8

Figure 6 - Levelized Cost of Energy [32]

The previous LCOE data presented in Figure 6 refers to US & Canada, Western Europe, China, India and

Japan, on both first and second quarters of 2013 (Q1 and Q2). It is possible to observe that natural gas presents

itself as one of the cheapest forms of energy, while the new renewable energy technologies as marine and solar

and also fuel cells are presented as the most expensive forms of energy, as they are still being improved.

While the LCOE reflects the actual costs of each technology excluding subsidies and support mechanisms,

it also does not represent the net costs faced by developers in the market. As so, and in the specific case of LNG,

it should be known that the global demand of LNG (and consequently its price) is directly influenced by oil prices,

since LNG is competing directly with petroleum in many applications. Therefore, when the oil prices increase

globally, it is expected that the LNG contracts linked to oil prices also become more expensive [33]. Regarding

the availability of the energy sources, it is difficult to predict what forms of energy will be, or not, available on a

short range, as producers tend to keep some reserves in secrecy or unveil them only when it is convenient. This

occurrence was possible to observe lately during the last two years with oil reserves, which, regardless the

scenarios, has recently been noticed as possessing increased reserves, leading to a plunge of oil prices,

consequently affecting the energy markets worldwide [34].

Regarding the associated emissions, the process of producing, transporting and exploiting any kind of

energy source has a great impact on the environment and ecosystems, especially because fossil fuels are still the

dominant energy sources. Resultant from the combustion of fossil fuels, the following chemical compounds are

9

formed: carbon dioxide (CO2), greenhouse gases; carbon monoxide (CO), resulting from incomplete combustion

of fuel, poisonous, colour and odourless; sulphur dioxide (SO2), enhancer of acid rains; and nitrogen oxides (NOX),

an indirect greenhouse gas.

Table 3 - Comparison of Fuel Emissions (Parts per billion of energy input)

Pollutant LNG Oil Coal

Carbon dioxide 117 000 164 000 208 000

Carbon monoxide 40 33 208

Nitrogen oxides 92 448 457

Sulphur dioxide 1 1112 2591

Particulate Matter 7 84 2774

Mercury 0.000 0.007 0.016

As it is possible to observe from Table 3, LNG appears as a breakthrough on the combat to particulate

emission, as it is evident a reduction of near 99%, and a great mitigator of SO2, that is reduced nearly 100%, and

NOX, that suffers a reduction of about 80% when compared with oil and coal [12].

When burned for power generation, SO2 emissions are practically extinguished and a significant reduction

of CO2 is obtained [35]. Hence, the increased use of LNG instead of other fossil fuels can potentiate significantly

the emission reduction of GHG to the atmosphere [36].

Due to the clean nature of the combustion of natural gas, heavy-duty vehicles (e.g. ships and trucks)

powered by LNG can, with the present technology, achieve low emissions rates without excessive and expensive

emission control equipment. It is estimated by Arteconi et al. (2009) that LNG can afford a 10% reduction in GHG

emissions when compared with a diesel engine. As a complement, another comparison of LNG emissions was

previously performed, which is directly connected to these heavy-duty vehicles mentioned.

1.2.3 Economics of LNG

The “Value Chain” concept was created in 1985 by Michael Porter as a competitive strategy to achieve

superior business performance [37]. This represents a combination of generic value-added activities operating

within a firm, activities that aim to cooperate in order to provide value to customers through a valuable product.

In order to better understand the LNG value chain, the process of an LNG chain should be analysed and

understood, so as to detect and evaluate what are the elements that add value to the chain – see Figure 7. To

perform this analysis, the whole process will be briefly described. A more detailed explanation is available on

Chapter 3.

Figure 7 - LNG Value Chain

10

The LNG production process starts with the exploration and production of the natural gas wells, being then

transported via pipeline to the liquefaction plant, where it is cooled to achieve -161⁰C at atmospheric pressure,

and becoming 1/600 of its original volume at gaseous state. The LNG is then loaded into tankers to be transported

by sea to the receiving countries, being then unloaded and stored as liquid or regasified to be fed into the

connecting pipelines to the natural gas grid of the accepting country.

With a brief analysis of this process, it is possible to conclude that LNG liquefaction creates the possibility

to transport a fuel with a much higher energy density per volume that it would be by pipeline in gaseous state,

which adds more value to this chain, since the same amount is transported in 1/600 of the volume, allowing

transport to and from more remote locations (where pipelines infrastructures were unfeasible), by ship or even

truck. This competence also weakens the probability of encounter geopolitical constraints, inherent to the use

of pipelines that cross countries with high instability, which may cause significant disturbance to the receiving

countries in case of war or other issues [38].

Even though pipelines deliveries continue to dominate, LNG plays an increasing role for EU natural gas

supplies and, due to the lack of capacity of international pipelines, LNG appears as the sole possibility for new

competitors to enter the market of natural gas; this situation enables traditional importers to widen their gas

suppliers assortment, considering that some producing locations of natural gas are only accessible by sea [39].

This result also enhances supply safety, since it broadens the offer and does not limit the worldwide distribution

of natural gas mostly to terrorism-susceptible pipelines.

Due to technological innovation, LNG costs have decreased significantly over time to about half the price

that was charged in 1990, and a level was reached where LNG was able to compete for pipeline supplies, as it is

shown in Figure 8, and the transportation of natural gas as LNG on carrier vessels has become preferred to

pipeline for distances larger than 2000 km [40].

Figure 8 – Illustrative comparison of NG transportation: LNG carrier vessel versus Pipeline Transport of NG [21]

Regarding the 5 main steps of LNG value chain, investment costs vary significantly amongst these:

exploration and production account for 15-20% of the total costs of the value chain; liquefaction for 30 to 45%;

shipping for 10-30%; and regasification for 15-25%. As LNG projects are very capital intensive with most projects

costing several billion dollars [41], these costs may diverge expressively from one value chain to another, as

factors such as distance, traded volumes and local conditions (construction costs, port configuration and site

11

conditions) have a great influence on the costs associated [38].

One of the problems associated with the production and use of LNG and that has been a major constraint

consists on the fact the suppliers await for a bigger number of users (ships), while ship owners do not invest on

LNG-based propulsion due to the fact that there are no sufficient suppliers or distribution that satisfy their

immediate demand [13]. This problem is being overcome by the bet from energy investors on the LNG technology

as its importance on the energy markets is actually significant and it represents a tactic to compete against the

conventional pipeline supply.

The globalization of natural gas markets has been favouring safety of supply as well as introducing

competition between previously separated regions. While the transportation of natural gas in its liquefied form

by tanker has been used for more than 40 years, only recently the industry achieved a remarkable level of global

trade [41]. During the last two decades, large investments along all the stages of the value chain have been

realized: new players (countries and companies) have entered the industry and LNG technology began to support

the globalization of formerly regional markets. Changes in the framework from monopolistic structures to

competition) have had an impact on organizational behaviour of market participants, rising strategic partnerships

and engaging both oil and natural gas producers and distributors on all stages of the LNG value chain [38].

Nowadays, LNG is responsible for supplying the US, UK, Iberian Peninsula, and Japan, among others. The

Middle East accounts for more than 40% of worldwide proven natural gas reserves and it is predictable that will

become the largest regional exporter of LNG, alongside with Australia. Also, the outlook for LNG demand in Asia-

Pacific region is very strong, as India, China, Singapore, Vietnam and Thailand are emerging as new consumers,

as Figure 9 illustrates.

Figure 9 - Major import and export regions [38]

As for Europe, the imports account for 48% of the share, and it is expected that this figure will rise to 74%

12

in 2030. Hence, the worldwide growth in the demand of natural gas has led to a major investment on LNG

projects, and projections indicate that by 2030 the demand will rise to the triple of nowadays’, and no other fossil

fuel is foreseen to grow as fast as LNG [42].

1.2.4 LNG in Portugal

Portugal relied on natural gas mainly on power plants, to fulfil the electricity demand whenever RES

(Renewable Energy Sources, wind and hydropower) were not sufficient, with a share of natural gas accounting

for about 18% of the total energy mix, and an average load factor of the gas-fired power plants of about 26% in

2012. The country is dependent on imports for about 74% of its energy needs in 2014 and 100% of natural gas

requirements, having had a total consumption of natural gas of 45.3 TWh. Considering only natural gas, 61% of

its consumption was made by the energy sector, ahead of both the industrial sector with 25% of consumption

and the Retail and Consumer sector with about 10%. As for existing natural gas storages, under Decree Law

140/2006, it is mandatory that Portugal holds reserves capable of answering, at least, for the demand for 15 days

of non-interruptible consumption on gas-fired power plants and 20 days of non-interruptible consumption of

household costumers [6]. As so, two main reserves exist on the country: LNG storage in Sines of about 240,000

m3 in combined storage of two tanks, that was recently increased by another tank of 150,000 m3, totalling

390,000 m3 (maximum storage of energy between 2874 GWh); and a storage facility on five salt caverns on

Carriço of 333 million cubic metres (total storage of energy of 3963 GWh [43]), with two additional storage

caverns foreseen to be completed soon [44]. A more detailed map of the Portuguese gas infrastructure can be

found on Annex II.

In 2016, natural gas demand in Portugal was assured 2/3 by the pipeline connection coming from Spain on

the interconnections of Campo Maior and Valença, while the remaining 1/3 was delivered on the LNG terminal

in Sines (gas original from Nigeria). The total consumption in 2016 totalized 55.8 TWh, registering a growth of

6.9% when compared with the previous year [43]. The LNG terminal received, in 2016, 26 ships (22 unloads, 3

loads and one cooling operation) and supplied 4629 tanker trucks. As for the underground storage, 45% more

natural gas was moved in, when comparing with 2015.

1.2.5 Portugal, Lisbon, as a strategic location for an LNG terminal

Taking into account the scenario for the power generation mix in Portugal up to 2030, it is expected that

the role of natural gas on power generation becomes progressively eroded by the rise of renewables and the

forthcoming low-carbon energy transition, assuming that there will be no significant ascents on power demand

[5]. Nevertheless, it is wise to observe Portugal potential to commercialize and distribute natural gas in its

liquefied form to and from the ships that travel across Portuguese waters, as about 75 000 ships pass offshore

Lisbon annually and most of them heading to or coming from ECA and SECA zones, as it is illustrated in figure 10,

where it is also possible to observe the great affluence of marine traffic existent on the Portuguese maritime

area.

13

Figure 10 - Main marine international traffic routes [45]

Knowing the crescent regulatory constraints that influence ship owners, and that these owners search for

cleaner fuels to head to northern Europe, an LNG terminal capable of supplying these vessels might be a great

contribution for the local and national economy as increasingly more ships would be expected to enter the Port

of Lisbon and use its facilities.

A terminal located in the south bank of river Tagus firth, where an important connecting link is made

between cargo coming from Spain, Northern Europe and the Americas [46], represents a promising affluence to

an LNG terminal to be built on this zone, as it is possible to see in Table 4 from the statistical data from The Port

of Lisbon, stating that about 200 ships enter the Port monthly.

Table 4 - General vessel traffic entering the Port of Lisbon in 2015 and 2016 [47] – GT: Gross Tonnage.

Traffic Monthly average January to December

2015 2016 2015 2016 Variation

Ships Entered (number)

217 192 2606 2300 11,7%

National (number)

33 37 396 441 11,4

Foreign (number)

184 155 2 210 1 859 -15,9

Total Capacity (GT)

4 153 574 3 757 375 49 842 885 45 088 498 9,5%

National (GT)

238 035 279 179 2 856 423 3 350 152 17,3

Foreign (GT)

3 915 539 3 478 196 46 986 462 41 738 346 -11,2%

Total Freight (Tons)

965 185 854 718 11 582 223 10 256 612 -11,4%

Assuming that around 205 ships pass daily offshore Lisbon (almost the same number of ships that enter the

Port monthly) [48], it is expected that the presence of an LNG terminal will make this number to increase; also

because the majority of the ships entering the Port are cargo vessels (see Figure 11), these are the potential users

of LNG as the motive fuel used, and the most probable visitors of Ports inside ECA/SECA zones.

14

Figure 11 - Number of ships, per type, which entered the Port of Lisbon between 2015 and 2016 [47].

In order to have a typical and normalized journey to consider along this document, a reference voyage for

a full tank of an LNG-powered cargo ship was considered (approximately 2000 km), and is well represented by

the trip Lisbon > Funchal > Lisbon. As so, and considering the reference travelled distance, it was possible to

estimate the distance between Lisbon and the other existing LNG storing ports where the vessels can refuel again

after the reference distance was travelled.

Table 5 - Travelling Distance from Lisbon to the main European ports and e Algeria

Departure Arrival Distance

(approximate) Country Port

LISBOA

Portugal (Madeira) Funchal 1000 km (540 Nm)

United Kingdom Southampton 1604 km (866 Nm)

Isle of Grain 1869 km (1009 Nm)

Belgium Zeebrugge 1889 km (1020 Nm)

Netherlands Rotterdam 2011 km (1086 Nm)

Algeria Arzew 1030 km (556 Nm)

Skikda 1670 km (900 Nm)

Spain

Barcelona 1508 km (814 Nm)

Cartagena 993 km (536 Nm)

Bilbao 1085 km (586 Nm)

France Montoir de Bretagne 1260 km (680 Nm)

Fos-Tonkin 1830 km (988 Nm)

It is clear from Table 5 that a ship coming to refuel LNG in Lisbon can reach the great majority of the

important LNG terminals in Europe and also Algeria, making Lisbon a strategic location for a mid-way refuelling

site between travels. For more detailed information on Europe and Northern African LNG terminals, please

consult Annex I.

0

100

200

300

400

500

Nr

of

Ship

sNational

2015

2016

0

400

800

1200

1600

Nr

of

Ship

s

Foreign

2015

2016

15

1.3 Objectives of the Thesis

As seen along this first and introductory chapter, Portugal has met the circumstances that led to enforcing

the development of a stronger primary energy supply chain, as the energy dependency on the outside should

decrease in order to ensure energy safety. Also, accompanying the market trends and the increasingly stricter

regulatory demands, Portugal gathers the conditions to develop a local LNG supply chain in the Port of Lisbon,

as it is a waypoint for ships travelling to and from ECA and SECA zones, areas where LNG is largely used as fuel.

Being so, the work presented on this thesis results from the cooperation of the author with Tecnoveritas, a

Portuguese engineering company, on the initial study and development of a small LNG supply chain for OZ

Energia, in the south bank of river Tagus, focusing on the energy usage during the energy conversion associated

to the normal operation of a small-scale LNG terminal.

The ultimate purpose of this thesis is to cooperate in the planning of a new LNG Terminal, more specifically,

to choose and develop the system of liquefaction of the boil-off gas (BOG). The boil-off gas comprises the amount

of gas that heats up back to the gaseous state either when being transported or when stored, due to imperfect

thermal insulation. This occurrence can be translated into energy losses, as the gas will necessarily have to be re-

liquefied or burned, meaning that energy will have to be spent on this process. One of the goals is to present a

solution as efficient as possible to manage the amount of BOG produced by the LNG terminal during its normal

operation.

In order to explore the most adequate solutions to perform the BOG management, it is necessary to

determine the BOG production during operation. Being a terminal yet to be built, the BOG production

determination will be based on the intended features of the terminal, the typical climate and meteorological

characteristics of the zone, the capacity of the terminal, typical insulation materials used, etc., that can influence

the BOG to manage, in order to build and study the adequate models to study the behaviour of the future

terminal.

1.4 Thesis Outline

The present thesis comprises six chapters:

Chapter 1 presents the overview of the energy scenario in Portugal and of the global LNG

development nowadays, aiming at contextualizing the following chapters.

Chapter 2 describes the case-study and the facilities where the main project is to be implemented,

characterizing the territory and its surroundings, as well as presenting the location of the LNG

storage.

In Chapter 3 it is performed a literature review on the existing LNG terminal and BOG recovery

technologies, in order to choose and dimension the more adequate and efficient system to put in

practice further on, on the terminal.

16

Chapter 4 includes a description of the methodologies and models used in the several steps of the

simulation.

Chapter 5 presents the final simulation results, for different seasons of the year, along with

alternatives to re-liquefaction that might also address the problem of BOG management.

Chapter 6 presents the general conclusions, as well as the perspectives to continue the present

work.

17

2. The Case-Study

The process of obtaining and commercializing natural gas in the liquid state comprises a value chain that is

of major importance to understand and deepen the knowledge of it up to the stage where the regasification

terminal is used, one of the key steps on the LNG technical and economic cycles. A simplified scheme of the full

process is schematized in Figure 12, which contemplates the 5 stages of the LNG value chain: exploration,

production, transportation, storage and commercialization of liquefied natural gas.

Figure 12 - Full Gas Field Processes

The process begins with the capture of gas and liquid (crude) from the natural reservoirs (wells), through

production separation, which will be treated differently. The liquid phase of the collected assets is subjected to

few treatments in loco, being transported to refineries to continue to further crude processing. As for the gaseous

part, until it is ready to be shipped and sold, there are important stages through which the gas passes in order to

collect other valuable components from the original gaseous mixture, and finally attain the purity levels required

on LNG. On Gas Conditioning (see Figure 12), the gaseous mixture is then prepared to be liquefied and later sold

this way. On this step, at the inlet of the liquefaction process, the impurities that are removed comprise non-

hydrocarbon contaminants as water, mercury, solids and acid gases. For acid gases, there is a dedicated removal

unit (AGRU) to collect CO2 that freezes and blocks the liquefaction section, and H2S that has toxic and corrosive

characteristics. On the refrigeration, heavy hydrocarbons and LPG are recovered from fractionation and stored

material, providing then the required characteristics of less than 0.1% of water content, less than 50 mol. ppm

of CO2 and less than 4mol. ppm of H2S, required to proceed to liquefaction [49].

Whilst liquefaction occurs, the NGL (Natural Gas Liquids) components are also separated as these are

valuable as separate products, meaning that it is profitable to remove them from the mixture. In NGL it is possible

to find propane, butane and ethane, which are also removed to obtain a rich and pure mixture of natural gas to

be liquefied, reaching the best possible liquid to gas volume ratio when liquefied. This process is of major

18

importance, since it allows the volume of natural gas to be 600 times less than the same mass of natural gas at

room temperature [12], proceeding then to storage of natural gas liquefied at -161⁰C. According to the laws of

thermodynamics, even with the latest insulation technologies, it is impossible to have a perfect storage with no

losses by regasification. This regasified gas constitutes the so-called the boil-off gas.

With the desired amounts of gas stored, it is then possible to ship LNG to distant places (see Figure 12), the

main method of transportation being the LNG marine carriers, which come in different classes with distinct

volume capacities (small, conventional, Q-Flex and Q-Max carriers). Later on, the LNG is unloaded on

regasification facilities and then distributed to the consumers.

In this chapter, it is intended to describe the basic characteristics of an existing terminal and elucidate the

relevant planning factors, as they will play an important part on the subsequent development of a local LNG

small-scale supply chain. The characteristics of the terminal, its surroundings and also the operations that are

already performed in it will be described, so that the possibility of retrofitting the existing terminal also to an

LNG receiving terminal is assessed.

2.1 Description of the Company

The present thesis was developed under a project in progress by TecnoVeritas1, a 20-years-old Portuguese

engineering company, for OZ Energia, the owner of the existing terminal and of the project in development.

OZ Energia is a Portuguese company (former ESSO Portuguesa) that has been operating in the market for

more than 40 years. The company has as main driver the target of being a reference brand in the market of

energy and services, looking always for innovative opportunities and the generation of value in the energy chain.

OZ Energia is positioned on the top four largest gas bottle distributors in Portugal, also competing on the

distribution of propane gas for domestic and industrial use, aviation fuels, and also production of pellets as

substitute of firewood.

Following the strong core values of continuously improving and adding value to their business, OZ Energia

decided to explore the possibility of benefiting from their Terminal of Trafaria and assess, alongside with

Tecnoveritas, if it is suitable to develop a Small-scale LNG Chain on the existing terminal.

1 TecnoVeritas has a large technological scope of expertise on the areas of marine industry, mechanics, chemistry and

electronics, and a high focus on R&D, being a technology provider for national and international companies. Composed by a skilled and experienced team, TecnoVeritas has earned several recognition prizes throughout the world, such as the Green Project Awards in 2013 and the Seatrade Awards, in 2012, for Clean Shipping. This visibility and experience on the field of energy engineering and maritime industry has led TecnoVeritas to participate on this project to provide consultancy and technical support on the planning and further assembly of an LNG terminal located on the south bank of river Tagus, in Lisbon, which will be held during the next years, alongside with a major stakeholder in the area of Oil & Gas.

19

2.2 The Terminal – Case Study

2.2.1 Characteristics of the Terminal

Trafaria Terminal, located on the south bank of River Tejo, Murcafém, comprises a total area of 79,825 m2,

having the possibility of an expansion zone southeast – see Figure 13. The terminal possesses on its premises 38

metallic reservoirs, some of them deactivated, where oil derivate products – such as gasoil, biodiesel, lubricants

or LPG – are stored, gasoil being the product with the biggest nominal storage capacity on the terminal.

Figure 13 - Trafaria Terminal - Aerial View [50].

As relevant information regarding the adjacent areas, river Tagus comprises the northern frontier whereas

the remaining frontier is a natural reserve, REN – Ecological National Reserve, which imposes some restrictions

regarding, respectively, the conservation of river species on the estuary and the natural habitats existing within

the Ecological Reserve.

Also important to refer is that the terminal is located near sensitive areas, namely the passenger pier of

Tagus’ Ferry, Basic School and Kindergarten of Trafaria, and also a Health Centre; these facilities represent public

organizations that are subject to be severely affected in case of a malfunction of the LNG terminal, and should

be taken seriously into account while conceiving the safety measures of the terminal during its planning.

Being on the bank of the river Tagus, the terminal possesses a jetty with 170 metres long and the

bathymetric on the riverbed allows it to support ships that have up to 9 metres of sea-gauge, which means that

20

the maximum loaded draft of the incoming ships cannot exceed 8.5 metres as, otherwise, safety of both the

vessels, the jetty and the personnel might be at risk [51].

Figure 14 - Flow on the river estuary [52].

Consulting the Portuguese Hydrographic Centre, it was possible to apprehend that, while at the entrance

of the estuary the medium amplitude of the tide reaches 2 metres, upstream the river (entering the estuary) the

tide is amplified to 3.5 metres, providing good sea-gauge conditions for the ships unloading on Trafaria terminal.

It is also possible to verify in Figure 14 that the terminal is sheltered from the major flow, providing good stability

to the jetty and to the careful operations to be performed [52].

In order to determine the consumption of a ship that could be representative for this particular study, a

search was made for the type of ships that enter the most in the Port of Lisbon (see Figure 11), and, within these,

which ones would represent the most probable client for the infrastructures we intend to study and build, and

also what could be the size of the main supplier ships, according to the characteristics of the jetty and the actual

bathymetric.

According to the data that was possible to gather on this matter (see Chapter 3, section 3.1.1), and with the

characteristics of Trafaria terminal jetty, the most adequate type of ship considered for the initial assumptions is

a small LNG carrier ship (supplier), capable of docking on the terminal transporting the maximum amount of fuel

of approximately 30,000 m3, or a container ship (client) of up to 1500 TEU, assuring, this way, that the vessels

are compatible with the bathymetric of the river along the jetty [53].

Due to legal restrictions, only REN (National Energy Network) can commercialize natural gas in gaseous

state. This implies that the facilities to be built cannot include a regasification unit. So said, natural gas shall only

be delivered from the terminal in liquid state, either to fuel ships or to supply LNG carrier trucks, providing,

therefore, LNG for both land and sea/river users. This circumstance demonstrates the major importance and role

21

that the boil-off recovery system will have to play on the facilities, as all the BOG (Boil-off Gas) will have to be

either used to unload LNG from incoming cargo (situation that will probably not occur as desirable in using the

BOG due to the need of sharing the jetty slots with other users) or reliquefied in order to be stored once again

in its liquid form.

LNG is often used to supply industries (more often, cogeneration), mainly where there are no natural gas

pipelines, finding, this way, another interesting market niche for exploiting the facilities while the marine LNG

market for the terminal is being formed (see Chapter 3, section 3.1.4). In fact, small industries exist near to the

terminal, that might represent potential clients and should be options to explore by the LNG terminal operator:

a food dry bulk storage and receiving terminal – around 650 metres away (might benefit from refrigeration

coming from the LNG terminal), NATO Ammunitions Dock (whose pipes pass beneath the terminal about 350

metres away) and a petrol station about 2 kilometres away.

One of biggest present difficulties on creating new LNG terminals to supply ships resides on the fact that

the market for LNG fuelled ships is still under development, as ship owners do not invest on these vessels because

LNG terminals are not abundant. As for investors, an LNG terminal destined to fuel ships is still a risky investment,

as the market is still not mature enough to guarantee the profitability of the terminals. This vicious cycle still

constitutes a great barrier to the development of LNG chains and to the use of LNG as a marine fuel, and can

only be solved by new investments on this area that can promote this business. As so, the present project

represents an opportunity to contribute to the development of a stronger LNG market in Portugal, also creating

a chance to improve and advance maritime economy, one of the present targets on the national strategic

development plans [54], attaining the economical, geostrategic and geopolitical potential that Portugal

possesses, and creating conditions to attract national and international investments.

2.3 Choosing the Storage Facilities

In order to determine later on the most suitable LNG liquefaction system to be installed on the terminal,

and previously to any other further studies, it was necessary to assess which will be the better place on the

existing terminal to place/build the storage facilities, taking into account all the restrictions regarding other

products that are handled in the terminal, and also the legislation regarding the identified sensible surrounding

areas mentioned above.

As a first step to visualize and learn the surrounding premises, the municipal plans of Almada were

consulted, accessing the GIS system of the city council, in order to assess what type of terrains will have to be

taken into account – see Figure 15.

22

Figure 15 - Map of Conditioning Factors: Trafaria Terminal and Surrounding Areas [55]

It is possible to recognise that there are three sensitive areas around the terminal: the river Tagus, a

National Ecological Reserve and a Historical Centre on the vicinity. As so, in order to plan the location of the

storage tanks on the terminal, some thoughtfulness is required on this matter while the studies are developed,

implying that the inherent restrictions of this special areas must be acquainted and applied.

To complement the study of the terrain, Trafaria facilities were visited in loco, and the following available

places for LNG storage were identified on building plans, through the use of AutoCAD software to identify and

measure the available areas, as depicted in Figure 16.

The identified regions in Figure 16 as A, B and C, are the available and possible locations for the LNG tank

and systems, and are the options under consideration:

A: Located near the Gasoline Reservoirs;

B: Located where reservoirs T1 to T9 and T18 are placed;

C: Located on unbuilt land.

A more detailed general view of Trafaria Terminal can be found on Annex III.

23

Figure 16 - General view of Trafaria Terminal with the three possible places to install the reservoirs (identified in green)

The three identified locations for the storage facilities are presented and discussed next.

24

A - Storage facilities located near the Gasoline Reservoirs

Figure 17 - First place considered – alongside the gasoline reservoirs (A in Figure 16)

B - Storage facilities located where reservoirs T1 to T9 and T18 are placed

Figure 18 - Second place considered – where reservoirs T1 to T9 and T18 are placed (B in Figure 16)

25

C - Storage facilities located on unbuilt land

Figure 19 - Third location considered - Unbuilt terrain (C in Figure 16)

Table 6 summarizes the advantages and disadvantages found for each possible location, in order to choose

the best location for the LNG storage facilities.

Table 6 - Summary of the Advantages and Disadvantages of the Three Considered Locations for the LNG Facilities.

Option Advantages Disadvantages

A

999 m2

Proximity to the reception jetty.

Reduced boil-off losses while being stored.

Basic Infrastructures to receive the storage tanks are already built.

Reduced area.

Possibility of building constraints regarding safety distances to be kept from the other existing reservoirs.

B

1966 m2

Biggest area among the considered options.

Existing tanks and structures in this area are currently out of service.

Urge to demolish/remove the existing tanks and structures already existent in this area.

Necessity to create infrastructures capable of transporting LNG to a higher elevation.

C

1125 m2

Unnecessary to remove equipment or infrastructures already existent.

No safety restrictions regarding the proximity to reservoirs of other fuels.

Proximity to a National Ecological Reserve (REN), implying a restriction buffer to any construction to be made on its vicinities.

Unavoidability to create infrastructures to transport LNG to a high elevation (∆h ≈ 50 metres).

Farthest location from the jetty.

Urge to perform excavations, earthworks and creation of basic infrastructures (roads, etc.) that already exist on the other options.

26

Available area is definitely a major constraint to consider: whether is the place chosen, as the BOG recovery

system will be an installation in the surroundings, and the available space should be as large as possible. Also,

the distance to the jetty is an important factor as there is heat ingress through the piping system, and the farthest

the storage tank is, the higher is the heat ingress, and thus larger is the BOG production while loading or

unloading LNG. The proximity to other storage tanks in the limited space of the terminal, where trucks often pass

carrying other fuels, highlight the option of building an in-ground tank; this technology offers two significant

advantages – effective land use and structural safety [56] – which meets the requirements of the terminal.

Building an in-ground tank is only possible on the second or third locations considered - locations B and C.

This alternative is considered as it might be sufficient to supplant the issues related to the proximity to the special

areas surrounding and the other products storage tanks, since it has almost no landscape impacts or leakage

problems if any situation occurs. It is important to notice that, in case of choosing this location for the

implementation of an in-ground storage tank (see Chapter 3, section 3.1.2), the buffer from the National

Ecological Reserve might not be necessary, enlarging the available area of the third location to the triple of its

actually defined value.

Figure 20 - Calculated distance from the Jetty to Location C (273.9 metres).

It is also known that, when the LNG is unloaded from ships to the storage facilities, after approximately 250

metres of transport through pipelines the regasification losses (BOG production) are significant [57].

Acknowledging that the maximum distance covered by pipeline inside the terminal might be greater than this

value - maximum height (jetty to third location) of 50 metres and maximum distance approximately 270 metres

(see Figure 20) – and disregarding the possibility of installing an intermediate boil-off recovery system, if the

27

chosen storage location is the farthest from the jetty, the production of extra boil-off gas while the operations

occur must be considered when dimensioning the boil-off recovery system.

The different hypothesises were subjected to the appreciation of the responsible people for the project and

after consideration of all the details, option B – Storage facilities located where reservoirs T1 to T9 and T18 are

placed - was chosen as the most suitable site to place the LNG facilities. This option overcomes the eventual

licensing problems regarding construction near to the National Ecological Reserve (REN) and it is closer to the

jetty than Option C would be (approximately 220 meters). Simultaneously, it does not present the space issues

option A has, and allows the construction of an in-ground tank. Also, it was decided that the inactive tanks should

be removed in a very near future, therefore leaving the place available to implement the LNG storage facility and

also the required BOG recovering unit.

Being so, Figure 21 demonstrates the preview of the final location of the in-ground LNG storage tank to be

built on Trafaria Terminal.

Figure 21 – Final location of the LNG reservoir.

As the characteristics of the terminal and the site options are presented, and in order to better understand

the several processes that shall be expected on Trafaria Terminal, on the next chapter a bibliographic review

regarding LNG terminals will be performed, having as objective the clarification the main operations on the

terminal and the assessment of the options and constraints probable to encounter during the stages of planning

and development, deepening later on the need for liquefaction units or other technologies able to recover the

boil-off gas produced on such facilities.

28

3. Literature Review

As the project under development comprises solely technologies within LNG storage and re-liquefaction,

the current chapter will focus essentially on this type of facilities and on the associated boil-off gas recovery

technologies, as well as practices to apply later on the liquefaction system.

A more detailed explanation of the most significant infrastructures related to the regasification and/or re-

liquefaction plant can be found next, which comprises the scenery of the second chapter of the present thesis

and explains in detail each section that will be part of the operations on Trafaria Terminal.

3.1 LNG Receiving Terminal

The receiving terminal (for regasification, storage and/or re-liquefaction) comprises the final step of the

LNG supply chain before the commercialization, as that is where LNG is delivered to the end users, and includes

the LNG unloading jetty (berth), the storage and boil-off recovery, and finally the send out facility, where LNG is

either heated back to its gaseous state to be sent through pipelines, or delivered to tankers on liquid form. The

following sections approach each of these steps with more detail.

3.1.1 The Jetty

The positioning of a jetty at an LNG marine terminal is a crucial factor in determining the overall risk on the

transfer operation of ship/shore and its position should be studied on the conceptual stage of the project [58].

The jetty should also be planned according to the bathymetric, tide, wind and currents [59] that characterize the

docking area as well as being provided with the necessary stability to perform the loading and unloading

operations [60]. The jetty design should be projected according to the planned volume capacity for the storage

facilities at the terminal, as its characteristics will determine the ships that are allowed to moor and load/unload

on the terminal, and its dimensions will determine the volume the moored tanker can carry. As an advised

minimum water depth for a small-scale LNG terminal, 10 metres are usually sufficient for the operations to be

carried out on the terminal [57].

With the purpose of clarifying the meticulous operations that take place at the jetty, the unloading and

loading operations will be briefly explained.

The LNG is usually unloaded from the tanker vessel through articulated unloading arms (cryogenic hoses)

to an unloading pipeline that drives the LNG to the tank – see Figure 22. It is common that two different types of

arms are connected to the vessel tank: one for unloading LNG and the other to return natural gas in its gaseous

form to the tank, in order to avoid vacuum and aid on the removal of LNG. Another hypothesis, not always viable,

prescind from the return blower and perform the return the gas to the tanker through a pressure difference

29

between the storage tank and the vessel [61]. This

procedure is usually temporary as it does not always

ensure that the maximum volume of LNG is removed

from the tanker.

As for the loading procedures, LNG is loaded into

each of the tankers tanks through the arms to liquid

header pipes and finally to the tank bottom. The gas

displaced by the LNG, as well as the BOG generated

during the loading process, must be returned to shore

installations and are controlled by the safety devices that

control cargo vapour pressures and liquid levels. Overfilling is also controlled by self-closing valves, automatically

activated whenever the predefined levels are reached. During both processes of loading and unloading, it is

compulsory that the ballasting (while emptying) and deballasting (while loading) occur, while the vessel draught,

stability and longitudinal bending are carefully monitored onboard [62].

According to SIGTTO’s safety procedures [60], it is of major importance that, during all the loading and

unloading procedures, the arms and all the process are handled in a manner that eliminates any risk of liquid

release and reduces the cargo vapour leaks to the atmosphere to an absolute minimum, in order to ensure the

operations safety, as well as guarantee the safety of all the personnel executing the activities on the jetty.

When it comes to planning the receiving terminal, care should be taken to make sure that there will not be

oversizing or undersizing of the different facilities inside the terminal, as these will be working directly with each

other and depend on each other. Being so, there is an important connection between the cargo ships, the jetty

and the storage. If there are no storage restrictions, these should be planned according to the possible

dimensions of the jetty and vice-versa. Also, the size of the LNG carriers should not exceed the maximum value

allowed on the jetty. Andrieu [63] refers in his research the target dimensions of LNG carriers, the storage

facilities, and also the working pressures and flow rates for a small-scale receiving terminal (SSLNG) and a large-

scale receiving terminal (LSLNG), which are reproduced in Table 7:

Table 7 - Main differences between SSLNG and LSLNG Terminals [63]

Terminal Functions Characteristic SSLNG

Terminal

LSLNG

Terminal

Unloading LNG Carriers Size From 7,500 to 35,000 m3 From 70,000 to 265,000 m3

Storage LNG Storage Tank

Capacity From 20,000 to 50,000 m3

> than 160,000 m3, up to

millions of m3

Send-out

Flowrate From 0.2 mtpa* to 1 mtpa > 2 mtpa

Pressure From few barg to 25 - 40

barg

Gas Network Pipeline

pressure (typically between

55 and 90 barg)

*1 mtpa ≈ 170 Million m3/h

Figure 22 - LNG Tanker Unloading

30

As it is possible to observe in Table 7, the main operational resides on a reduced send-out flow rate on the

SSLNG, which can be 5 to 20 times lower than on a LSLNG terminal. Also, due to the storage capacity and the

send-out flowrate, SSLNG terminals, contrary to LSLNG that can supply several costumers, are usually dedicated

to few or even just one consumer, which, due to the reduced number of users, increases the availability and

flexibility of the SSLNG terminal. Similarly, the send-out pressures on the SSLNG terminal are customizable for

the specific consumer needs, whilst on LSLNG terminals the pressure is usually determined by the natural gas

network connected to it [63].

3.1.2 Storage

The types of storage facilities depend mainly on whether they are supposed to be used to meet winter

shortages of gas (or seasonal fluctuating gas demand) or to supply baseload gas by long-distance shipment. An

LNG import terminal developer faces two important decisions related to storage: how much to build and the

type of storage tank more adequate to the planned dynamics of the terminal. The selection of a tank design as

well as its complementary foundations are influenced and should also account for the different characteristics

of the terrain, such as the topography, geology (soil conditions), seismic concerns, regional safety regulations (as

a vapour or liquid leakage should be concerned and prevented), and, finally, exclusion and protected zone

requirements [64].

On board ships, apart from the indispensable insulation to minimize evaporation losses, it is necessary that

the LNG cargo is kept away from the ship structure as it might lead to a disastrous situation in case of contact

with the ship structure, as mild steel becomes brittle below -50⁰C. Provided that insulation is adequate,

evaporation losses (boil-off gas) might be as low as 0.1% per day for the tank contents [12].

According to the European Standards, an LNG

containment should be designed to safely contain

the cryogenic temperature, allow safe filling and

removal of LNG as well as the boil-off gas to be

safely removed, minimize the rate of heat in leak,

and also prevent the ingress of moisture and air,

except as a last resort to prevent vacuum conditions

in the NG vapour space, as the tank should be

prevented from going into negative relative

pressure beyond the permissible limit.

LNG onshore can be contained in double-walled metal tanks (similar to those used on board ships) - see

Figure 23 - as aluminium or nickel steel inner vessels or membranes, surrounded by insulation and weather-

proofing materials. Another option is to erect pre-stressed concrete tanks with extra resistance above the ground

or cast them below the ground surface. Occasionally, it is possible to adapt already existing underground spaces

for LNG storage, e.g. on salt caverns or depleted oil reservoirs, as Wang and Economides [65] refer on their study.

Figure 23 – Double Containment LNG Storage Tanks in Sines, Portugal [161]

31

In order to better organize the different tank categories, these are presented in four main categories: single,

double and full containment, for above-ground storage tanks, and membrane, for in-ground tanks.

Single Containment

This type of tanks (see Figure 24) is the fastest and easiest to build as

the structure to assemble is not very intricate. These tanks require a

large area to build in, as they require leak prevention dykes to secure the

LNG on this pre-designated area [66] as there is only one outer wall made

of carbon steel [67]. Although it is cheaper and has a simple construction,

it is the most difficult type of containment to get approved by the

regulatory entities, as it requires a larger safety distance and it comprises

the less safe choice among the several tank options.

Double Containment

Comparing with single containment tanks, these tanks (see Figure 25)

allow a closer spacing between tanks, as they do not require a

prevention dyke. In case of leakage, the LNG is contained on a second

bund wall of pre-stressed concrete.

Full Containment

For this type of tanks (see Figure 26), no bund wall is required, as, in case

of failure, both natural gas and LNG are contained by the concrete walls,

making this the safer kind of tank above ground and also the most used

nowadays, alongside with the in-ground membrane tanks [68].

In-ground Membrane Tanks

In-ground tanks (see Figure 27) are usually membrane tanks and

consist of a pre-stressed concrete outer wall and an inner layer of

insulating load-bearing foam, over which is laid a thin cryogenic steel

membrane with 9% nickel [67] that will be in direct contact with the LNG,

assuring, this way, that no intrusions of water enter the tank [64]. It is

also possible to build in-ground spherical membrane tanks, which have

similar characteristics to the cylindrical tanks and are suited to areas that

have a high seismic potential, as described on the European Standards for LNG tanks design.

Figure 27 - In-ground Membrane Storage Tank

Figure 25 - Double Containment Storage Tank

Figure 26 - Full Containment Storage Tank

Figure 24 - Single Containment Storage Tank

32

The main advantage of in-ground tanks, both concrete or natural, lies in the fact that the building of

containment dykes is not mandatory, as the risk of leaking is low and there is no danger associated to the burst

containers, meaning that there will not be any leakage products to collect. On the other hand, above ground

containers are attractive for the fact that they allow repairing and there is a much stricter control of the

temperature and behaviour of the tanks insulation.

The safety characteristics of in-ground tanks, as well as the protection from solar radiation to the walls that

this technology provides, made this type of tank the chosen to store LNG in Trafaria Terminal. Also, the reduced

aesthetic damages of the landscape are a plus on this type of tanks, as only the dome and some auxiliary

equipment is visible.

Another way of storing LNG is on floating storage regasification units (FSRU), special moored vessels capable

of storing, regasifying and, if necessary, transport LNG onboard [69]. These facilities are either converted from a

traditional LNG vessel or it can be purpose-built, always requiring an offshore terminal, which is connected by

undersea pipelines to transport regasified LNG to shore or to a receiving terminal. The FSRU is a solution for

smaller or seasonal markets, and can be developed cost-effectively and considerably faster than onshore

facilities. For that reason, this is often a temporary solution while onshore facilities are constructed, as it can be

easily redeployed elsewhere and used just as effectively as before and in short time [70].

As these floating facilities do not provide a direct analogy with the industry case-study at issue, they will

not be discussed further.

3.1.3 Materials and Insulation

Insulation has become a theme where a great effort of research is put on, as many uprising fields of

engineering that use cryogens are growing, such as energy storage, superconductivity and even space

technology.

When it comes to maintaining such temperatures as the ones used to operate and store LNG or other

cryogenic fluids it is of utmost importance to have a correctly dimensioned insulation in order to reduce the heat

that enters the tank. Consequently, insulation is one of the major determining factors that influence the

production of BOG in an LNG system, and which will be analysed and applied later on.

Nevertheless, the performance of the insulation must always justify the cost [71], which means that, in the

present case study, pondering must be done regarding the achieved daily BOG rate and its management and the

cost of erecting the tank with an improved insulation. Therefore, insulation of the tank will be dimensioned

according to the thermal performance requirements, limited by the calculated BOG rate.

The three elementary factors that determine the overall suitability of the insulation materials are the

thermal conductivity, density/weight, and the cost of labour and materials. These factors will be presented for

every insulation material studied, especially for those usually utilized on LNG in-ground tanks.

33

LNG tanks usually present different materials for the bottom, walls and roof, being a studied assortment of

materials that provide structural strength to the tank and thermal insulation to the cryogenic fluid.

Below follows a short description of materials typically used in LNG tanks to insulate and for structure

purposes:

Concrete: it is meant to provide structural strength to the tank and resistance to weather conditions.

9% Nickel Steel: standard ferric structural steels are not suitable for extreme cryogenic temperatures and

LNG’s, as there is an increased risk of brittle fractures and not enough toughness. Alloy 9% Nickel steel is

the material that must be in direct contact with LNG [67], achieving the mechanical and physical properties

required for building storage tanks and cryogenic pipes, withstanding temperatures down to -196°C while

still offering the structural integrity required [72].

Sand: a sand layer is usually placed between the nickel steel layer and the foam glass insulation.

Asphalt: a thin layer of asphalt is commonly laid between the layers of foam glass insulation (occasionally

also with a felt layer) [73] in order to seal and drawn tight the layers so that the system is vapour sealed

and no additional vapour barrier is required [74].

Foam Glass: this impermeable material is designed for industrial applications on with a high load-bearing

requirement, having a combination of high compressive strength and low thermal conductivity. Its

attributes make this material suitable for cryogenic tanks bases [75]. A possible substitute for foam glass as

insulation is polyurethane but it usually turns out to be more expensive as this material is highly flammable

and thus needs fire proof coating in order to be used in a hazardous environment (such as an LNG terminal)

and comply with all the safety regulations applied.

Expanded Perlite: being a naturally occurring siliceous volcanic rock, perlite can be expanded from four to

twenty time its original volume by heating it to above 900°C with two to six percent combined water,

causing entrapped water molecules to turn to steam and creating countless tiny glass bubbles that allow

the crude rock to pop and expand [76]. Expanded perlite, in addition to its thermal properties, is easy to

install, relatively low costly, non-combustible, and does not shrink, swell, warp or slump [77], being often

the main insulation material chosen for LNG tanks.

Aluminium: this material is usually an alloy that receives a special treatment to develop a temper that

guarantees the required intergranular and exfoliation corrosion, as well as temperature resistance without

risking brittle fractures. It is often used on LNG tanks as the supporting material of the insulation platform

(deck) hanging from the roof, often used as an alternative to welded nickel steel, for it represents a lighter,

safer and with better strength-to-weight ratio than former technologies [78].

According to the particularities of these materials, the tank characteristics will be chosen and planned later

on in order to project the BOG that shall be produced on the tank. All the described materials will be used in the

34

Trafaria LNG tank project, according to examples of actual tanks built and information available from industrial

projects. According to these examples, the typical order of placement and thickness of each material will be

taken into consideration, as there is no rule or standard for the choice of the materials of an LNG tank, and each

contractor has their own developed technology and method. As so, the basic characteristics of these materials

are presented in Table 8, for further reference and guidance, as well as reference prices for each material.

Table 8 - Materials Used on the Construction of an LNG Tank and Principal Characteristics.

Material Conductivity Emissivity Density Price per m3

W/m.K kg/m3 €

Concrete 1,80 0,97 - 79,93 2

9% Nickel Steel 17,18 - - 125,18 3

Sand 0,25 - 1 682 9,96 4

Asphalt 0,75 - - 1,88 5

(price per m2)

Foam Glass 0,02 - - 888,46 6

Expanded Perlite

0,03 - 32 1 544,37 7

Aluminium 71,21 - 2 700 43,21 8

The prices displayed in Table 8 were obtained from manufacturers of the presented materials, which divulge

the prices publicly, and are used only as a reference. To note that for the actual project planning and budget

phase it is necessary to resort to official quotation requests to adjust these values to the actual market

conditions, eventual bulk quantities special prices, location influence and final tank project.

3.1.4 Boil-Off Gas Management

Due to temperature differences between tanks, pipes and the surrounding atmosphere, whenever the

tanks are filled, LNG tankers are loaded or unloaded, or due to heat exchange [58], boil-off gas is generated,

usually with an assumed volume 600 times greater than that in liquefied form. This makes essential the existence

of an adequate treatment of the boil-off gas.

As seen before, LNG is stored and transported in cryogenic tanks in liquid form, i.e. liquid at a temperature

below its boiling point. Nevertheless, due to heat transfer, LNG continuously evaporates. Inside the tanks, LNG

exists in thermodynamic equilibrium, where liquid and vapour coexist, their masses depending on the operating

pressure and temperature. Having a low pressure on the tanks, it is possible to apply Raoult’s law to the multi-

2 [167] Orçamentos 2009-2017, “Orçamentos e Orçamentação na Construção Civil,” 2014. [Online]. Available: http://orcamentos.eu/precos-de-betao-pronto/.

[Accessed 21 September 2017]. 3 [169] Alibaba Group, 2017. [Online]. Available: https://www.alibaba.com/product-detail/Nickel-Steel-Plates-9-Inox-

Stainless_60545095394.html?spm=a2700.7724857.main07.21.4e9e903cCp6LP&s=p. [Accessed 21 09 2017]. 4 [168] Alibaba Group, “Alibaba,” 2017. [Online]. Available: https://www.alibaba.com/product-detail/Constructions-

Sand_117792245.html?spm=a2700.7724838.2017115.104.7d299801xNSsdC. [Accessed 21 September 2017]. 5 [170] Alibaba Group, “Alibaba,” 2017. [Online]. Available: https://www.alibaba.com/product-detail/Bitumen-Asphalt-Roofing-Felt-For-

Waterproof_60617179669.html?spm=a2700.7724838.2017115.1.23861b7cKyvS5A. [Accessed 21 September 2017]. 6 [171] Alibaba Group, “Alibaba,” 2017. [Online]. Available: https://www.alibaba.com/product-detail/Foam-Glass-Board-for-Oil-

Platform_60389703830.html?spm=a2700.7724838.2017115.1.156e29e9WyyJzI. [Accessed 21 September 2017]. 7 [172] U.S. Geological Survey, “Mineral Commodity Summaries,” U.S.A., 2016. 8 [173] Alibaba Group, “Alibaba,” 2017. [Online]. Available: https://www.alibaba.com/product-detail/5383-5454-5456-aluminum-alloy-

plain_1657854430.html?spm=a2700.7724838.2017115.65.23861b7cKyvS5A. [Accessed 21 September 2017].

35

component mixture system [79], as follows:

NG Vapour at T, p, yi.

LNG at T, p sat, xi.

𝑦𝑖 . 𝑝 = 𝑝𝑖𝑠𝑎𝑡(𝑇). 𝑥𝑖

(2)

𝐾𝑖 =

𝑝𝑖𝑠𝑎𝑡(𝑇)

𝑝=

𝑦𝑖

𝑥𝑖 (3)

In equations (2) and (3), p is the total vapour pressure, and pisat represents the saturation pressure of a pure

component i in the liquid phase at temperature T, yi and xi are respectively the volume fractions of component i

in vapour phase and liquid phase, and Ki is the dimensionless equilibrium ratio, the vapour-liquid equilibrium

between an ideal gas and a liquid.

Any heat entering the tank, will cause the evaporation of the liquid without visible bubbles, due to the low

pressure registered inside. Also, while the liquid is stored on cryogenic insulated tanks at constant pressure, it

will remain at nearly constant temperature; this process is called auto-refrigeration. As long as the vapour boil-

off is allowed to leave the tank in a controlled and safe manner, the temperature will be kept constant by the

auto-refrigeration phenomenon [80].

It is necessary to continuously remove the boil-off gas both to keep the auto-refrigeration phenomenon

running and to ensure the tanks safety. The increase of BOG in storage also increases the LNG operating tank

pressure, which should be maintained inside its nominal operating range. As so, on a regasification terminal, the

BOG leaving the tank can either be reliquefied, be injected on the natural gas network, or even be used as a fuel

to the LNG carrier vessels in port or adjacent industrial installations.

Another consequence of the occurrence of BOG is the change of the energy value of the fuel during this

process. In fact, the liquid is composed by several components with different phase change temperatures (at the

operating pressure) that can vary widely from – 196 ⁰C to + 36 ⁰C [40]; while methane and nitrogen have higher

evaporation rates, C2+ components do not share the same characteristics [81]. This slow process is called ageing

or weathering of LNG, as its properties and quality change as time passes [82].

At the same time, BOG results in a reduction the quantities of the cargo carried on LNG carriers, making this

event a crucial factor on the technical and economic valuation of the LNG supply chain [40].

While the process of BOG reduces the quantity of cargo, it also increases its density since the lighter

elements are those that are more volatile. Consequently, as LNG enters the storage tanks in the terminal, if the

densities of the incoming and already stored LNG are different, stratification may occur, creating a steady

interface layer. When the densities of the layers approach, this layer might become unstable and mix suddenly.

This process is known as the rollover phenomenon. While mixing, a large amount of BOG is abruptly released,

quickly increasing the pressure in the tank, which might origin irreversible damage [83]. The BOG production rate

during the rollover can be 10 to 30 times greater than the tank nominal values [40]. The rollover event might be

the source of enormous economic losses for the LNG terminal operator and presents a considerable risk to the

36

terminal [84].

i) Boil-off on Storage Tanks: Holding Mode

The holding mode refers to the interval between loading and unloading operations, where LNG is stored

[81]. Despite the fact that LNG tanks are designed and prepared to sustain the cryogenic temperatures of LNG

through the purposely-made multi-layered insulation, heat transfer still occurs through the walls, roof and floor

of the tanks, by convection, radiation and conduction. Thus, although these are meant to keep the temperature

of the liquid so that vaporisation is less than 0.05% of the total tank content per day, the variation of BOG daily

rate can be between 0.02 and 0.1% [85].

During the holding mode, in order to maintain the cryogenic temperature on the loading/unloading system,

a small portion of the stored LNG is circulated through the pipelines, pumps and all the LNG circuit, absorbing

the heat from the surroundings and also the heat generated by pumping, turbulent flow and line friction [40].

While LNG circulates, more BOG is formed in the tanks (also known as tankage BOG [82]), which may cause a

drop in pressure, lowering the boiling point of the liquid. To compensate this pressure drop, the temperature in

the tank should decrease by approximately 0.1 ⁰C per each 0.01 bar the pressure drops [81]; nevertheless, this

option potentiates the BOG rate, as the only way to decrease the temperature is to regasify some of the liquid.

ii) Boil-off on Loading/Unloading Mode

This period corresponds to the time frame on which the LNG carrier is moored to the jetty unloading LNG

to the storage tanks or, on the contrary, the terminal is supplying LNG to a vessel or truck, being the connection

made through insulated pipelines and loading arms [82]. As Benito (2007) states in his study, the BOG generated

during the process of loading and unloading can reach values 8 to 10 times above that observed for the terminal

operating in holding mode. This increase is due to the need of a return line of vapour from the reaccepting facility

that while emptying either the storage tanks or the LNG carrier tanks.

An LNG carrier is unloaded by the ship pumps and loaded by the terminal pumps [40]. While such operations

are performed, as large quantities of LNG are pumped, a rapid change in pressure might take place. To prevent

this effect, the gaseous volume is displaced by the equivalent volume of liquefied gas, in order to maintain the

nominal operating pressures on both the tanks (the ship and terminal tanks). This way, and under normal

circumstances, no BOG is released to the atmosphere, neither liquefied gas is leaked [86].

iii) Using the Boil-off Gas at Receiving Terminals

The generated boil-off gas is collected through the utilization of a compressor and is then directed to one

of many possible utilizations. The utilization of the BOG is usually made by the following order of prioritization

[81]:

Returned to the ships while these are unloaded;

Used as a fuel on-site: it can be used as a heat source for the vaporization, as for example in the SVC

37

technology (Submerged Combustion Vaporizers – see Annex VI), and might require pre-heating before

entering the fuel gas system;

Recondensed into LNG: vapour is collected in the boil-off header that connects to the boil-off compressor

suction drum; it is then liquefied on a “recondenser”. While compressed, it is mixed with subcooled LNG

capable of absorbing the natural gas and holding it in liquefied form. As a reference, for each kg of LNG

absorption of 0.1 kg of boil-off gas from the discharge of the boil-off compressor is possible [87] – see

Figure 27.

Figure 28 - Process Flow of BOG Handling System [87]

Compressed and fed into pipeline of the natural gas network: as compressing large volumes of gas to high

pressure is usually an expensive process, this option is typically undertaken when there is a very low

internal demand for the boil-off gas;

Reliquefied and stored: this option is usually carried out when large volumes of BOG are produced in the

terminal or when there is no connection to a natural gas network or internal needs for natural gas. This

option will be target of further analysis on section 2.2.

Flare: a flare is mandatory in all terminals for safety reasons [88]. It is used for emergencies only, such as

the sudden liberation of BOG on the storage tanks on a rollover event, where the natural gas is burned in

a controlled way.

Regarding the flaring option, this safety measure might, at first sight, be difficult to understand, as it would

be simpler to just release natural gas into the atmosphere. This can be straightforwardly explained as an

environmental protection policy: since methane (CH4) has a global warming potential of 25 times the global

warming potential of CO2 [89] (1 kg of released methane is equivalent to releasing 25 kg of CO2), and the burning

of methane originates CO2 and 2H2O, it can simply be concluded that the combustion of the natural gas produces

less ozone-depleting substances than releasing it directly to the atmosphere.

Despite the disadvantages and safety issues that the BOG might introduce on the processes in the terminal,

38

it is clear that there are several ways to recover it and make it useful for the LNG chain or to nearby installations,

representing potential energy savings and increased efficiency to the holistic process.

Regarding the potential and characteristics of Trafaria terminal, and acknowledging the incapability of the

terminal operator to commercialize natural gas in its gaseous form, compressing the boil-off gas and sending it

out the to the natural gas network is not a possibility in this case. Also, taking into account that, especially at the

beginning of the operation, the interested customers in acquiring NG from the terminal might be scarce, it is

important that the boil-off recovery system is powerful enough to endure long working periods on holding mode.

For these reasons reliquefaction and storage will be the chosen method to manage BOG.

Also, by choosing the location identified as number 2 for the storage tanks (section 2.3), as the distance

LNG is supposed to travel is larger than 250 metres, the BOG recovery system should also be able to withstand

the larger amounts of gas that should be formed during loading and unloading processes.

The boil-off issue has been object of several attempts of reducing it or even to prevent this phenomenon

from happening at all, not only on the LNG field but in cryogenics in general. Plachta and Guzik (2013) have

developed a zero-boil-off cryogenic propellant storage for NASA (National Aeronautics and Space

Administration), using multi-layer insulation and reverse-Brayton cycle turbo-cryocoolers. Li et al. (2012) have

improved the recondensation system on a Chinese LNG terminal by maintaining the BOG recondenser pressure

via the condensating LNG, and also found that this operation can save energy by installing a pre-cooling system

on the beginning of the process. Also on jetties during loading and unloading operations, the resulting BOG rate

is as big as the length of the jetty and of piping from the ship to the tank, resulting in high-pressure pumping and

consequent heat leakage, which is transferred to the LNG. However, lowering the pressure results on monetary

costs, as the bunkering operations should be as prompt as possible [90].

Taking into account all the BOG production means the literature review, it is possible to perceive that

several improvements can still be achieved, namely on the efficiency of the re-liquefaction cycles, insulation

materials development and optimization of the cryogenic processes, both on the LNG carriers and on the

terminal facilities. These facts are precursors of the search for improvements on liquefaction cycles, and are the

main reason to perform an adequate choice on the technology to be implemented on the terminal, as it will

determine its overall efficiency.

3.1.5 LNG Regasification

Liquefied for a more effective shipment, LNG has to be later regasified at a facility close to the natural gas

consumers. The regasification procedure is a heating process that typically uses ambient temperature heat

sources in order to pressurize LNG back to its gaseous state and feed it for distribution into the send-out pipelines

of the NG transmission network. As this segment of the LNG chain will not be necessary on the project in

development, more information regarding this topic can be found in Annex VI, for reference.

39

3.2 Liquefaction processes

One of the major problems found in the LNG industry is due to the heat exchanged by the various

components of the supply chain with the surroundings. While being handled or stored, and since insulation is

not perfect, the heat transferred from the surroundings to the systems yields a small but yet significant part of

the liquefied gas that phase change, assuming a

volume of about 600 times the corresponding

value at PTN conditions. LNG supply chains are

equipped with re-liquefaction plants at different

points, as seen in section 3.1.3, namely, at

storage, on board the LNG vessels and at the

terminals. At all those points, heat transfer from

the surroundings and non-ideal heat insulation

will make liquefied gas to evaporate. The

vaporised gas can either be used as fuel (for the

facilities or ship propulsion) or it can be re-

liquefied. This process represents significant

energy losses since it embodies a loss of work

done to liquefy the gas on the first time, which

has to be performed once again to restore the

boil-off gas on liquid form. A number of re-liquefaction plant cycles exist and a short introduction to those plants

will be addressed on the foregoing paragraphs.

The system of re-liquefaction is required due to the phenomenon of evaporation of LNG, occurring at

different places in the LNG supply chain. In order to better elucidate the liquefaction process verified on the

terminal, the basic operation principles of liquefaction will be introduced, leading to the identification of the

several liquefaction cycles applicable to BOG recovery. Figure 29 illustrates the basic working principle of a

liquefaction system – a refrigeration cycle.

3.2.1 Basic Principles of Liquefaction

In order to liquefy a gas, it is required that the BOG, at a low pressure, is compressed and then cooled at a

higher pressure (compressor discharge pressure). This process may be repeated several times until the fluid, in

its liquid form at desired pressure and temperature, is obtained. For storage tanks, typically at approximate

atmospheric pressure, the temperature should be – 164 ⁰C.

Figure 29 - Working principle of a refrigeration / liquefaction system

40

One option to liquefy BOG would be to use

the BOG as working fluid, and improving the cycle

by adding a heat exchanger such as exemplified in

Figure 30. However, liquefaction of the BOG is

commonly made resourcing to external

refrigerant fluids with complementary cycles, as

this technology attains better results and in a

much more efficient way [91].

Analysing a liquefaction system, whichever

working fluid is utilized, it is known that isentropic

compression of a gas has as result its temperature

increase. Analogously, an isentropic expansion

results in the cooling of the gas, according to

equation (4), where T and p are the temperature

and pressure of the gas, subscripts 1 and 2 the

initial and final state, and k is the isentropic

coefficient calculated as the ratio between CP (specific heat at constant pressure) and CV (specific heat at constant

volume):

𝑇2

𝑇1

= (𝑝2

𝑝1

)

𝑘−1𝑘

(4)

To improve the efficiency of the cycles, it is required that the work

performed on the expansion is applied and used externally, in order to

minimize the energy used for liquefaction.

The condenser is typically a heat exchanger, which the nominal

capacity must be sufficient to dissipate the heat of the various stages of

compression as well as to dissipate the heat removed from the BOG, as

sketched in Figure 31.

For the natural gas liquefaction system shown in that figure, the first law (energy balance) equation is

written as:

�̇�𝐹(ℎ𝐺 − ℎ𝐿) = �̇�0 − (�̇�𝐶 − �̇�𝐸) , (5)

where �̇�𝐹 and h are respectively the mass flow rate and specific enthalpy of natural gas feed (gaseous and

liquid), and �̇�0 and (�̇�𝐶 − �̇�𝐸) are the heat rejected and the net power input (compression) [92]. It is also

assumed that no mass losses occur during the process. Note that, according to equation (5), �̇�𝑒 is negative.

A liquefaction system is generally evaluated in terms of work input by unit mass of liquefied gas,

as (�̇�𝐶 − �̇�𝐸)

�̇�𝐹⁄ .

Figure 31 - Liquefaction Cycle [92]

Figure 30 - Basic BOG Liquefaction

41

From the second law of Thermodynamics, the entropy balance equation for the liquefaction system can be

written as:

�̇�𝐹(𝑠𝐺 − 𝑠𝐿) + �̇�𝑔𝑒𝑛 =�̇�0

𝑇0⁄ , (6)

where s is the specific entropy of natural gas (gaseous and liquid), 𝑇0 is the ambient temperature, and �̇�0

is the heat rejected. The entropy generation rate, �̇�𝑔𝑒𝑛, is zero in a reversible system, but assumes positive values

on practical systems. As so, substituting �̇�0 from equations (5) and (6), the work input is given by:

�̇�𝐹[(ℎ𝐿 − ℎ𝐺) − (𝑠𝐿 − 𝑠𝐺) 𝑇0] + �̇�𝑔𝑒𝑛𝑇0 = �̇�𝐶 − �̇�𝐸 (7)

Since �̇�𝑔𝑒𝑛 is non-negative, the minimum work required to perform the liquefaction is:

�̇�𝑚𝑖𝑛 = (�̇�𝐶 − �̇�𝐸)𝑚𝑖𝑛

= �̇�𝐹[(ℎ𝐿 − ℎ𝐺) − (𝑠𝐿 − 𝑠𝐺) 𝑇0] , (8)

which represents the absolute minimum given by the thermodynamic limit. In brackets, in equation (8), is

the flow availability of exergy of LNG.

The figure of merit (FOM) is expressed through an index and it is used to quantify and characterize the

performance of systems or devices in a practical and efficient way, allowing to compare different technologies

used for the same purpose [93]. The FOM of a liquefaction system is defined as the ratio of the minimum work

to the actual work:

𝐹𝑂𝑀 =

�̇�𝑚𝑖𝑛

(�̇�𝐶 − �̇�𝐸)=

(ℎ𝐿 − ℎ𝐺) − (𝑠𝐿 − 𝑠𝐺) 𝑇0

(�̇�𝐶 − �̇�𝐸)�̇�𝐹

⁄ (9)

The entropy generation can be given by the difference between the actual and minimum works multiplied

by the ambient temperature, and it is called the irreversibility of the loss of availability, or exergy:

(�̇�𝐶 − �̇�𝐸) − �̇�𝑚𝑖𝑛 = 𝑇0�̇�𝑔𝑒𝑛 (10)

Another possible index, in this case, a key performance indicator (𝐾𝑃𝐼), which can complement the FOM

index expressed by equation (10) is the volume of liquid produced per unit time, and spent energy, as expressed

by equation (11), where VLNG is the volume of LNG, t is time, and E the energy spent:

𝐾𝑃𝐼 =

𝑉𝐿𝑁𝐺𝑡⁄

𝐸 (11)

The importance of this KPI resides on the fact that it can evaluate the energy efficiency of the performed

cycle per unit of volume of LNG produced.

Being also an important index, Çengel & Boles [94] present the coefficient of performance (COP) for a

refrigeration (or liquefaction) cycle, relating the amount of heat transferred by the hot source (𝑄𝑒𝑣𝑎𝑝) with the

work performed (𝑊𝑐𝑜𝑚𝑝), as shown in equation (12):

𝐶𝑂𝑃 =𝑄𝑒𝑣𝑎𝑝

𝑊𝑐𝑜𝑚𝑝

(12)

By applying the First Thermodynamics Law, it is possible to achieve the relation between heat (including

42

also Qcond, condenser heat) and work performed on the cycle, resulting in equation (13):

𝑄𝑒𝑣𝑎𝑝 = 𝑄𝑐𝑜𝑛𝑑 − 𝑊𝑐𝑜𝑚𝑝 (13)

By substituting equation (14) in equation (13), it is possible to obtain equation (14), below:

𝐶𝑂𝑃 =

𝑄𝑐𝑜𝑛𝑑

𝑊𝑐𝑜𝑚𝑝

− 1 (14)

According to equation (14) it is possible to infer that in order to maximize the attained COP, it is necessary

to minimize the work performed on the compression stage, or, as an alternative, assure that the amount of heat

taken in the condenser to cool the refrigerant is increased.

The previous equations constitute the basic and general thermodynamic model for further analysis to apply

to the chosen natural gas liquefaction cycle to use on the present project.

Determining the most suitable refrigeration cycle for a system of natural gas is a difficult, yet important,

task during the terminal planning, since a number of practical factors have to be taken into consideration, such

as the amount, composition and pressure of the NG, constraints on size and weight, safety, simplicity of the

operations, statutory requirements, and even the local climate. Knowing there are several options when

choosing liquefaction cycles to obtain LNG, the following section intends to explore these methods, presenting

the cycle that will be used to simulate the recovery of BOG by re-liquefying it.

3.2.2 LNG Liquefaction Processes

In order to better evaluate and quantify the performance of the liquefaction cycle chosen to recover BOG,

one of the assessment factors will rely on the performance achieved by each technology. Later on, to perform

the evaluation of the chosen cycle, equation (9) provides an efficient and practical form of assessing the FOM of

the considered liquefaction systems, constituting one of the important factors to take into account, as well as

the KPI considered on equation (11), that can directly determine the amount of energy spent in order to produce

a determined volume of LNG. Also, the COP, equation (14), will be used to assess the performance of the

liquefaction system chosen. Nevertheless, particular attention should be paid to the expected investment on

equipment, especially due to the small-scale dimension of the terminal, which implies that not all systems are

feasible nor economically viable.

Trigilio et al. [95] suggest that refrigeration cycles can be categorized in two main groups: those that use

pure refrigerant fluids and those that use mixed refrigerant fluids (mixtures of pure components). Nevertheless,

and in order to enlighten and analyse this issue in the scope of the current LNG technologies, the liquefaction

processes used on LNG plants will be classified in three main groups of processes in the next section: cascade

liquefaction, mixed refrigerant and expansion-based processes.

Some other liquefaction cycles and options used to achieve liquefied natural gas are explained and explored

in Annex VII, for reference.

43

3.2.2.1 Cascade Liquefaction Cycles

Demanding very low temperatures, LNG liquefaction will difficultly succeed with a simple compression

cycle, as the temperature difference between the surrounding medium and the expected LNG temperature is

too high; this would require, subsequently, a large pressure difference during the cycle. Alternatively,

liquefaction can be achieved by using several cycles, i.e., by operating two or more liquefaction cycles in series,

named cascade liquefaction cycles [94], and attaining progressively lower temperatures with moderate pressure

ratios.

Cascade cycles are based on the use of a number of refrigerants that present different, yet constant, boiling

temperatures, thus reducing irreversible heat exchange losses [91] by guaranteeing a minimum area in the heat

exchangers. It is common to utilize pure refrigerants such as methane, ethylene and propane (see Figure 32) –

first, natural gas is cooled to -35°C in the propane cycle, then to -90°C in the ethylene cycle, and finally liquefied

to -155°C in the methane cycle [96]. Each refrigerant can be controlled separately, ensuing flexibility of operation

to the liquefaction system as a function of the natural gas flow rate. However, these systems are very sensitive

to the natural gas composition, exhibiting problems in adapting to composition fluctuation.

Figure 32 - Cascade Cycle with Pure Refrigerants [95]

The heat exchangers in these complex liquefaction systems, called “coldboxes”, are very complex equipment

designed to enhance the heat transfer between two or more fluids, and are considered the heart of such

installations. Coldboxes usually have multi-stream capability (see Figure

33), which means that the process streams inside the coldbox can have

different configurations, allowing the operator to optimize the cooling

curves of the refrigerants [97], in order to achieve the maximum process

efficiency possible.

The high efficiency and low required power of these systems are

achieved with a large number of equipment installed to do so, implying a

considerable investment and automatically discarding non-large-scale

facilities as users of such a system, as it does not represent a viable

solution for them [91]. Figure 33 - Example of Coldbox [97]

44

Being a mixture mainly composed of methane but with traces of several other components, natural gas

does not present a phase change at a constant temperature, being important that the refrigerant cooling curve

presents a similar cooling curve to natural gas, thus reducing the energy consumption due to the exergy of the

process [95].

Figure 34 – Pure refrigerants commonly used to produce LNG - Vapour Pressure Curves [95]

It is then important to choose the adequate refrigerants and their vapour pressure curves, as exemplified

in Figure 34, in order to make the mentioned curve approximation. In alternative to the pure components, the

refrigeration fluids used can be a mixture of components with cooling curve approaching the natural gas curve.

3.2.2.2 Mixed Refrigerant Liquefaction Cycles

Similarly to the fundamentals behind the use of refrigerants in

cascade liquefaction systems, the mixed refrigerant liquefaction cycles

intend to use a blend of fluids that mimic the cooling behaviour of

natural gas, thus optimizing the energy usage and the heat exchangers

size [91]. One of the advantages of these cycles is the fact that less

equipment is required when compared to cascade cycles and,

depending on the power required, there is a possibility to use a sole

compressor for the refrigeration process [95] – see Figure 35.

Mixed refrigerant cycles also allow a more precise control of the

process, as the refrigerant blend can be adjusted to the gas

composition. On the other hand, it also results in a slower start of the

operation due to the fine adjustments required. In Figure 35 a double passage of the refrigerant on the

evaporator is illustrated, as a mean of solving the restarting time hindrance at each adjustment action, by

controlling the temperature profile using intermediate separators for the resulting vapour, and assuring

simultaneously the adjustment and the uninterrupted operation of the plant [95].

Figure 35 - Mixed Refrigerant

Liquefaction Cycle [95]

45

According to IGU 2017 World LNG Report [98], in 2016 the biggest share of installed liquefaction units was

held by the C3-MR technology (propane precooled mixed refrigerant), accounting for 61% of existing LNG trains,

which represents 69.9 MTPA of the total 114.6 MTPA.

i) Single Mixed Refrigerant Cycle (SMR)

Based on the reverse Rankine cycle, these systems are based on cooling and liquefaction in a single heat

exchanger [91]. SMR cycles often present low efficiency when compared with other options, therefore being

more common to find these systems in midsize to small-scale LNG plants.

ii) Dual mixed Refrigerant Cycle (DMR)

These processes achieve the intended liquefaction by using two different mixed refrigerant cycles: the first

is usually a heavier mixed refrigerant and ensures the precooling of natural gas, while the second stream of

refrigerant is a lighter mixed refrigerant and intends to condense the cooled natural gas in a second heat

exchanger [91]. A broad utilization of this cycle is the C3-MR technology which employs as first refrigerant a

single propane refrigerant, and a mixed refrigerant on the second refrigerant stream [91]. The disadvantage of

this system resides in a higher complexity of the process as well as higher inherent cost in equipment.

3.2.2.3 Gas Expander Cycles

Liquefaction using a gas expander cycle is based on the circulation of a fluid, produced by means of a turbo-

expander, compressing and expanding it in order to generate the required refrigeration (see Section 3.2.3). The

fluid that is commonly used as cooling fluid is nitrogen (or methane, in alternative), due to the characteristic low

temperature it can achieve. This process is usually suitable for small-scale facilities as it is of simple operation,

low-maintenance and presents a fast start up response [99]. The simplicity is related to the fact that, by using

pure components as refrigerant, it is not necessary to adjust its composition according to the NG composition.

Plus, the temperature control is facilitated, resulting in a more stable cycle over a range of conditions of

liquefaction [100].

Having a considerably lower investment associated when compared with other technologies available, the

gas expander cycles have the disadvantage of presenting a low-efficiency operation when compared with

cascade of mixed refrigerant liquefaction systems. The fact that these systems require lower investments makes

gas expander cycles more suitable for small-scale LNG plants, peak shaving plants or BOG recovery and re-

liquefaction systems, where lower natural gas flow rates need to be processed, not being appropriate for load-

base plants [91].

Contrary to cascade and mixed refrigerant cycles, expansion-based liquefaction systems using nitrogen

have the advantage of being inherently safe, as there is no hydrocarbon liquid inventory on the installation [91].

Also, emergency venting of refrigerant in gas expander systems does not represent a problem for the operator,

as the released component is inert and does not comprise an environmental issue to be dealt with.

46

3.2.2.4 Comparing and Choosing the Liquefaction Technologies

With emergent but not yet significant adoption of small-scale LNG terminals throughout the globe,

literature regarding BOG liquefaction systems for these installations is still scarce and has not been widely

explored yet [101]. Nonetheless, and taking into account the size of the terminal at study, the most suitable NG

re-liquefaction cycle will present analogous characteristics to the re-liquefaction systems operating on small LNG

carrier vessels (for which more literature exists), as these also face BOG management issues during operation

and present similar capacities. For these heavy-duty maritime means of transportation, one of the BOG

management possibilities lies in the fact that auxiliary or main equipment such as steam turbines or recent diesel

or dual fuel engines have the ability to burn BOG. Alternatively, the gas can be re-liquefied back to the cargo

tanks [102].

Despite having different outputs, large-scale and small-scale LNG terminals share the same necessity – to

guarantee the best performance (and, consequently, energy consumption) possible for the system, meaning that

the choice of the most appropriate re-liquefaction cycle is of key importance. In any case, when operating a

small-scale facility, compactness and investment on equipment often play a more decisive role than the overall

efficiency of the equipment, as the BOG rate on small facilities is not expected to be very large.

Figure 36 - Qualitative Comparison of Different NG Liquefaction Technologies

As Mokhatab et al. [103] presented in their work, one example of this scale-problem is the propane

precooled mixed refrigerant system (C3MR), that while very efficient and typically applied to large-scale facilities,

represents a huge and unaffordable investment for a small-scale installation. However, it is important to refer

that a close link exists between the energy efficiency of the process and the OPEX of the installation [104], which

will result in a higher OPEX (operational expenditure) for any facility using expander technology. Figure 36 is

based and adapted from the graphical analysis of Mak et al. [91] and Castillo & Dorao [105] and is systematized

by Table 9, by comparing the different liquefaction technologies and their key characteristics.

47

Table 9 - Comparison of Different LNG Liquefaction Technologies [91].

Cycle Cascade / C3-MRC MRC Expander

Efficiency High Moderate / High Low

Complexity / Cost High Moderate Low

Heat Exchanger Area Low High Low

Flexibility High Moderate High

As it is possible to infer from Table 9, the expander technology offers two big advantages when compared

with cascade and MR cycles: low complexity and high flexibility of the installation, while keeping a small plot

area. Also, the complexity of the system is proportional to the initial investment on the system acquisition and

installation. Nevertheless, it is important, once again, to highlight the lower efficiency and higher energy

consumption of this technology, especially of the single expander cycle, that can reach values up to twice the

energy consumption of an MR cycle [105].

It is known that one of the issues that influences the most the expander cycle efficiency is the high

irreversible losses in the heat exchangers caused by the large temperature difference of the refrigerant and the

natural gas [95]. Being Trafaria Terminal a small-scale receiving terminal, and assuming a good management is

performed, the feed gas will not present a temperature difference from the cooling fluid that results in very high

entropy in the heat exchangers due to temperature differences [99]. This fact, together with the low investment

associated when compared with the two other technologies presented, points the gas expander cycle as the

adequate for BOG re-liquefaction for that terminal. Additionally, using nitrogen as refrigerant fluid diminishes

the potential fire hazards in the terminal, as the refrigerant is inert and it is not necessary to store hydrocarbon

fluids; also venting in an emergency situation does not constitute an environmental question, and emergency

flaring is not necessary [91].

Taking into account all the presented pros and cons, and presenting low complexity and high flexibility,

together with a lower initial investment, the liquefaction technology that will be studied for recovering BOG on

the Trafaria Terminal shall be the expander cycle using nitrogen as refrigerant, which characteristics are

deepened in the following section.

3.2.3 The Chosen Liquefaction Technology - N2 Expander

Cycle

The expander (or turbo-expander) liquefaction cycle, is based on the

reversed-Brayton cycle, as illustrated in Figure 37. While less efficient than

the other explored options, it is considered the most adequate solution for

BOG re-liquefaction when performing an analysis of cost versus benefit.

Anyhow, and as referred by Mokhatab et al. [91], these installations are

limited to 1–2 MTPA (million ton per annum) of single train capacity.

The reversed-Brayton cycle presents a refrigerator system consisting

Figure 37 – Reversed-Brayton Cycle

48

of a compressor unit, a heat exchanger assembly, an intercooler unit, and an expansion turbine sub-system [106],

composed of isobaric and isentropic processes. In order to safely produce work in the expander, it is important

that the refrigerant used should remain in gaseous phase [92], using a refrigerant such as nitrogen. The turbo-

expander is used as the main component to remove energy from the refrigerant, the nitrogen gas stream. To

increase the efficiency of this cycle, a lower expander outlet temperature is required, which might result in the

formation of liquid [107].

Analysing the processes of Figure 37, the cycle occurs as follow: in process 1-2 the compressor compresses

nitrogen and the fluid achieves high temperature and pressure; in process 2-3, the fluid exchanges heat with the

colder environment in an intercooler, resulting in a decrease of its temperature; in process 3-4 the fluid is

expanded in the turbine, causing a decrease of pressure and temperature; in process 4-1, the nitrogen passes

through the evaporator (heat exchanger), absorbing heat from the cold refrigerated fluid (LNG, in this case),

which then changes back to the conditions of state 1.

As in Trafaria Terminal other fuels (hydrocarbons) are stored and other industrial operations are performed

often, it is advised that the refrigerant fluid is nitrogen which, as an inert component, can mitigate the hazards

risk in such a sensitive installation. Moreover, one major advantage of using nitrogen as refrigerant is that by

attaining such low temperatures still in gaseous state, there are no mal-distributions in the heat exchangers,

improving the heat transfer of the system. Additionally, the liquefaction unit is very compact and the

refrigeration fluid can run on a closed loop [91], this way resulting in the advantage of not requiring refrigerant

storage nor makeup. The basic configuration of a single turbo-expander cycle is shown in Figure 38.

Figure 38 – Single Turbo-Expander Cycle

Many research projects have been accomplished on how to improve the efficiency of this liquefaction

technology. Based on the reverse-Brayton cycle, the expander cycle can be improved by adding a propane

precooled cycle [108], which cooling curve can match more efficiently the end of the LNG cooling curve than

nitrogen. The disadvantage of applying this improvement is the fact that the system safety (based on

hydrocarbon-free operation) is jeopardized by using propane, therefore compromising this benefit. For this

49

reason, this improvement will not be studied, as it can compromise the safety of the LNG installation and,

consequently, of the terminal. Other researchers also suggest pre-cooling with carbon dioxide [109], which could

mitigate the safety issue by using an inert gas as pre-cooling refrigerant.

Other option, despite entailing more capital costs and requiring more heat exchanger surfaces [91], is to

add a second turbo-expander to the cycle, reducing the temperature difference in the cold end of nitrogen and

LNG curve. The cycle with this configuration is called dual turbo-expander cycle and is shown in Figure 39.

Figure 39 - Dual Turbo-Expander Cycle

This configuration, dual expander cycle, comprises an interesting possibility to consider when choosing the

most appropriate cycle for Trafaria Terminal, as one of the major concerns of the terminal operator is to achieve

the intended results (liquefaction) with less equipment implementation costs associated and, preferentially,

more efficiently.

He & Ju [108] accomplished an analysis of different configurations of the turbo-expander LNG liquefaction

cycle in order to attain the maximum efficiency possible for expansion cycles. In their work, the thermodynamic

models obtained are computed and assessed according to the following equations:

i) Compressors

Compressors have the function of pressurizing the refrigerant (nitrogen) in the expansion cycle, generating

exergy losses caused by friction and heat losses in the process. The energy balance (equation (15)) is calculated

as follows:

�̇�𝑐 = �̇�𝑁2

(ℎ𝑜,𝑖𝑠𝑒𝑛𝑡𝑟𝑜𝑝𝑖𝑐 − ℎ𝑖,𝑐)

𝜂𝑐

= �̇�𝑁2(ℎ𝑜,𝑐 − ℎ𝑖,𝑐), (15)

where �̇�𝑐 is the work performed by the compressor, 𝜂𝑐 is the isentropic efficiency of the compressor, �̇�𝑁2 is the

mass of nitrogen, and ℎ𝑜,𝑐 and ℎ𝑖,𝑐 are respectively the outlet and inlet enthalpies in the compressor.

50

ii) Expanders (Turbines)

Turbines execute the depressurization of the refrigerant with the purpose of lowering its temperature.

Simultaneously, these components transmit mechanical power to the compressor by means of a connecting

shaft, driving it. The energy balance for expanders (equation (16)) is calculated as follows:

�̇�𝑒 = �̇�𝑁2 (ℎ𝑜,𝑖𝑠𝑒𝑛𝑡𝑟𝑜𝑝𝑖𝑐 − ℎ𝑖,𝑒) 𝜂

𝑒= �̇�𝑁2

(ℎ𝑜,𝑒 − ℎ𝑖,𝑒), (16)

where, �̇�𝑒 is the work performed by the expander, 𝜂𝑒 is the isentropic efficiency of the expander, �̇�𝑁2 is the

mass of nitrogen, and ℎ𝑜,𝑒 and ℎ𝑖,𝑒 are respectively the outlet and inlet enthalpies in the expander. Note that,

according to equation (16), �̇�𝑒 is negative.

iii) Intercooler (Heat Exchanger)

This heat exchanger is used to lower the discharge temperature of the fluid at the outlet of the compressors,

so that the work to the booster compressors becomes smaller. The energy balance for heat exchangers (equation

(17)) is calculated as follows:

�̇�𝑖 = �̇�𝑁2(ℎ𝑜,𝑖 − ℎ𝑖,𝑖) (17)

where �̇�𝑖 is the heat transferred on the intercooler, �̇�𝑁2 is the mass of nitrogen, and ℎ𝑜,𝑖 and ℎ𝑖,𝑖 are respectively

the outlet and inlet enthalpies in the condenser.

iv) LNG Heat Exchangers / Cycle evaporator

This equipment constitutes the heart of the liquefaction process. Here, the heat is removed from the

refrigerated fluid to the refrigerant, which absorbs it. In the present case, natural gas delivers heat to the

nitrogen, this way achieving liquid state. The energy balance for LNG heat exchangers (equation (18)) is calculated

as follows:

�̇�ℎ𝑒𝑥 = ∑ �̇�𝑁2,𝑛(ℎ𝑜,ℎ𝑥𝑛 − ℎ𝑖,ℎ𝑥𝑛)

𝑧

𝑛=1

(𝑛 = 1, 2, … , 𝑧), (18)

where �̇�ℎ𝑒𝑥 is the heat transferred within the heat exchangers, �̇�𝑁2,𝑛 is the mass of nitrogen in each heat

exchanger, 𝑧 is the number of heat exchangers on the system, and ℎ𝑜,ℎ𝑥𝑛 and ℎ𝑖,ℎ𝑥𝑛 are respectively the outlet

and inlet enthalpies in each heat exchanger.

v) Valves

Valves are considered isenthalpic equipment, with virtually no work interaction with the fluid, nor heat

transfer to the surroundings. The energy balance for valves (equation (19)) is calculated as follows,

ℎ𝑜,𝑣 = ℎ𝑖,𝑣 , (19)

where ℎ𝑜,𝑣 and ℎ𝑖,𝑣 are respectively the outlet and inlet enthalpies in the valve.

Kwak et al. [110] use the following equation to compute the specific power consumption in the liquefaction

cycle (equation (20)),

51

𝑊𝑠𝑝𝑐 = ∑�̇�𝐶

�̇�𝐵𝑂𝐺

, (20)

where 𝑊𝑠𝑝𝑐 is the specific power consumption, �̇�𝑐 is the work performed by the compressors, and �̇�𝐵𝑂𝐺 is the

mass of BOG.

These equations should be applied to the expander cycle configurations studied, in order to conclude what

is the most suitable technology for Trafaria Terminal, and guarantee the best relation of cost and efficiency

possible.

3.3 Contributions of the Thesis

Despite the fact that LNG terminals have been running effectively for more than 50 years in industry, each

case is a specific one and the technologies of systems available and of insulation materials are always evolving.

This means this terminal has its own specificities requiring a dedicated study for this case scenario, accounting

for all the constraints and advantages of the site, and reach the best solution possible. Being so, the contribution

of the thesis intends to be the particular and dedicated study of this terminal, for which is required that the best

solutions in terms of energy efficiency and overall energy conversion performance are studied and developed.

From the performed bibliographic review, the importance of a good planning and holistic knowledge of the

facilities was evidenced as a major factor contributing to the correct functioning of the future terminal, as this

will have to be planned from the very beginning and also be adapted to the existing facilities and operations that

are already performed in the terminal.

The present thesis will aim at performing the foundation study for the operation of a small-scale terminal,

and deliver feasible solutions for the owner of the project to implement the whole LNG system successfully and

informed of the options, constraints, pros and cons. Moreover, it intends to work out what options exist for

terminals with similar characteristics and simulate the best solutions found, in order to characterize the

insulation and features of the LNG tank, determine BOG production during operation, select the most adequate

liquefaction solution and explore other options to manage BOG production in the terminal.

For this effect, Chapter 4 will focus on the methodology followed to reach the results that will be presented

on Chapter 5, alongside with other solutions found and aspects to consider on the further actions when

implementing the project.

52

4. Modelling and Simulations Methodology

This chapter is dedicated to the study of different options for BOG management on the Trafaria terminal. It

is intended to identify which are the most adequate method and system to perform the needed operation with

the highest energy efficiency possible.

Modelling and simulation started thriving when, in the mid-50's of the 20th century, competition led

industries to start modifying their plants in order to achieve better performances, substitute equipment for more

profitable machinery and improve their automatic control methods [111]. Following this methodology, the

opportunity to anticipate systems was developed, by creating analytical models, saving time, effort and costs,

helping industry to strive for optimal system designs.

Modelling and simulation allow for the identification of the major characteristics of a system such as the

one under study. At this stage of concept, modelling and simulation allow the pre-knowledge necessary to take

decisions in terms of project design. The accuracy of the model used for simulation is bounded by the data

available at the time, and time itself. The present model is a compromise between these two factors, and the

results are in line with other studies published elsewhere.

As the idea of creating a universal model with all the imaginable items of the system is quite unrealistic, the

model produced should be attainable, as simple as possible, and address a few specific and of most importance

parts of the system being studied [112]. Being so, the models used in this work address specifically the BOG

production and management as the themes of greatest interest and importance, especially in terms of energy

efficiency.

4.1 Operation Conditions

In this section, the most relevant factors and conditions of operation are expected to be identified. It is

intended that the factors with greatest influence on the LNG system, and, therefore, on the BOG production and

management, are identified and accounted for later. Being so, this section presents the starting

conditions/characteristics of the terminal and of the fuel to store, also approaching the climate conditions under

which the system is expected to operate.

4.1.1 LNG Properties

Portugal is usually supplied with natural gas from Algeria. As so, Table 10 presents the composition of the

LNG expected to be used in Trafaria Terminal, assuming the same average characteristics as the LNG supplied to

Sines LNG terminal (REN Atlântico). This information is relevant for the study of the thermodynamic behaviour

of the fluid during liquefaction and gasification, as well as for the determination of the BOG rate.

53

Table 10 – LNG Composition Expected in Trafaria Terminal [113].

Components % Methane CH4 90.10

Ethane C2H6 7.10

Propane C3H8 2.20

i-Butane C4H10 0.30

Butane C₄H₁₀ 0.18

Nitrogen N2 0.12

As a simplification of the model, and due to the complexity and unpredictability of the composition of natural

gas, the thermodynamic characteristics of methane are used to approach the characteristics of natural gas, since

it is the hydrocarbon component with far higher concentration. These values are presented in Table 11, below:

Table 11 – Methane Properties [114].

Liquid Phase Gaseous Phase

Boiling Point [°C] -161.48 -

Latent Heat of Vaporization [kJ/kg] 510.83 -

Density [kg/m3] 422.36 1.8116 @ boiling point

0.7173 @ 0 °C

Specific Volume [m3/kg] 0.0024 0.552 @ boiling point

1.394 @ 0 °C

Lower Heating Value* - 39.05 MJ/Nm3

10.85 kWh/Nm3

* At reference conditions (referred to as normal conditions) of pref = 1.01325 bar(a) and Tref = 273.15 K (0⁰C).

4.1.2 Climate Characterization

The territorial occupation framework of the surroundings of the terminal was identified in section 2.2.1. In

the present section, the climate characteristics of the site are introduced.

Ambient temperature, radiation, and soil depth temperature are very important to calculate the tank BOG

production rate, as they influence directly the heat ingress into the tank through the dome, walls and bottom.

Thus, with the purpose of using reliable and accurate data of air temperature and radiation at the terminal site,

data of one full year was collected. This information was gathered from the nearest meteorological station to

the site (Station of Monte de Caparica), resorting to SNIRH (Nacional System of Hydric Resources Information),

which provides historical meteorological data of all the monitored meteorological stations in Portugal [115].

The region of Almada is on a climatic transition region, presenting Mediterranean characteristics with a

strong Atlantic Ocean influence [116]. In order to retrieve a representative period of time for each season of the

year, the data selected was one month of each season, with daily and hourly discrimination. The mean values for

each hour of the day of a whole month were then obtained so that a typical day for each season could be

represented. The last complete year possible to collect was from spring of 2016 to winter of 2017.

The data collected for each season of the reference period can be found in Annex V and is summarized in

Figure 40 and Figure 41:

54

Figure 40 - Ambient Temperature during the typical days of

each season- Reference Period

Figure 41 – Solar Radiation during the typical days of each season - Reference Period

Also a very important parameter to calculate the heat ingress through the walls and bottom of the tank is

the soil temperature. Despite the fact that soil temperature records do not exist for great depth such as the in-

ground tank will have, it is known that temperature does not vary significantly after the first few meters;

temperature in greater depths actually converges to a value near the average annual value [117].

Knowing the soil temperature is virtually insensitive to the diurnal cycle of the air temperature over a depth

of 1 meter [118], with the aim of using a value that can be representative of the soil temperature over the year,

the mean value of soil temperature at 10 cm of depth for 3 decades was taken as reference value for the

temperature of the soil [119]. This is illustrated in Figure 42, which demonstrates the temperature of the soil

does not suffer a very significant fluctuation during the year. The mean temperature computed for the soil

temperature at 10 cm depth is 16.18 ⁰C.

Figure 42 - Soil Temperature - Reference Period

4.1.3 Tank Operation Modes

According to the planned features of the terminal, the operation of the tank is expected to remain mainly

on holding mode, at least at the beginning. Meanwhile, it is supposed to work as a strategic backup for any vessel

that might come to the Port of Lisbon and is in need of refuelling. In any case, the LNG terminal will be subjected

to three modes of operation, which will be studied, and present the following characteristics:

Holding Mode - On this mode, the LNG remains stationary, stored in the tank. It is considered that the heat

ingress is solely the heat that enters the tank through the bottom, walls and dome, and only the BOG

0

5

10

15

20

25

30

00

:00

01

:00

02

:00

03

:00

04

:00

05

:00

06

:00

07

:00

08

:00

09

:00

10

:00

11

:00

12

:00

13

:00

14

:00

15

:00

16

:00

17

:00

18

:00

19

:00

20

:00

21

:00

22

:00

23

:00

°C Ambient Temperature

Winter Spring Summer Autumn

0

200

400

600

800

1000

00

:00

01

:00

02

:00

03

:00

04

:00

05

:00

06

:00

07

:00

08

:00

09

:00

10

:00

11

:00

12

:00

13

:00

14

:00

15

:00

16

:00

17

:00

18

:00

19

:00

20

:00

21

:00

22

:00

23

:00

W/m2 Solar Radiation

Winter Spring Summer Autumn

14,2

25

17,5

80

10

20

30

Spring Summer Autumn Winter

Soil

Tem

per

atu

re (

⁰C) Soil Temperature at 10cm depth

55

recovery system will be in operation.

Unloading a Vessel / Receiving LNG - While the heat ingress to the tank is constant, this mode has two

more variables that become important to include: the heat ingress through the piping system from the tank

to the ship, and also the heat transferred from the pumping system.

Loading a Vessel / Sending out LNG - In addition to what happens while unloading vessels, loading a vessel

implies that, while LNG is sent to the ship, BOG takes the place of the LNG in the tank, in order to equalize

the pressures in the tanks. This way, while receiving LNG from the terminal, the ship sends the

correspondent volume of gas up to the tank, called “displacement volume”.

The different modes of operation will also be studied for the best and worst scenarios of weather

conditions: the best scenario will be a typical winter day with no wind, whereas the worst scenario is

correspondent to a summer day, very windy. In the worst case scenario that is analysed, the heat that enters the

tank through the roof is boosted by largest quantities of radiation and high convection due to the wind effect.

As seen in section 4.1.2, the soil temperature does not oscillate much during the year; for this reason, the

soil temperature will be considered constant during the year. For simulation purposes, only the heat that enters

through the dome of the tank will be varying according to the annual temperature variation, and the heat that

enters the walls and bottom of the tank will be considered constant.

4.2 Designing the Tank

When it comes to design a tank and define its characteristics, it is important to state the assumptions and

simplifications made. As the present work intends to study energy use of the LNG terminal facility, structural

design and resistance, despite being a crucial section of a full and complete tank project, will not be detailed

herein. Hence, the following design and construction features were not calculated: wind load, hydrostatic tests

and seismic forces on shell; soil settlement and tank foundations; roof draining system; and major components

and fittings (deck penetration, piping, anchors and gauging system).

Due to the difficulty of finding particular rules and standards for the whole project and construction of an

in-ground tank, the characteristics of the tank planned for the terminal are based in similar industrial projects

and examples found in the literature. Also the insulation materials and its features were collected and adapted

from those examples, in order to rely on trustful and implemented technology; once again, it is important to

refer that each company (mostly Japanese for in-ground tanks projects) has its own methodologies to implement

such a project, using different approaches to reach a similar result.

This chapter will be based, among other references, in the examples presented in Annex IV - Tank Projects

and Examples - and Annex IX – Tank Insulation Examples.

4.2.1 Initial Design Conditions and Intended Characteristics of the Tank

The first and more important characteristics of the tank to start with are its physical ones, namely, the type

56

of tank and the designed volume. As described in section 3.1.2, the chosen tank will be built in-ground; in order

to be able to receive the full cargo of a small LNG carrier ship, the volume of the tank shall be of 30.000 m3, the

average expected capacity of the bigger vessel that can be accommodated by the terminal jetty water depth.

As a starting point, consulting the examples found in the dedicated literature and addressing the available

space in the location chosen to build the in-ground tank, it was decided that the internal diameter of the tank

should be 40 meters. The decision of the diameter was a compromise between diameter and depth, considering

also the soil stability on site. To meet the same capacity, a smaller diameter would require deeper foundations

(and more investment to do so) and a larger diameter would reduce the available space at ground level; thus, an

internal diameter of 40 meters meets the expectations of the owner of the terminal, while also in accordance

with the examples of other projects found for similar capacities.

With three features defined – type of tank, volume, and internal diameter – the following characteristics of

the tank were calculated and are presented in Table 12:

Table 12 – LNG Tank Specifications.

Feature Value

Maximum Capacity [m3] 33 333

Rated Capacity [m3] 30 000

Internal Diameter [m] 40.00

Working Height [m] 23.87

Rated Height [m] 26.53

Heel Height [m] 1.33

Bottom Area [m2] 1 256.64

Wall Area [m2] 3 333.33

Operating Temperature [⁰C] - 164

Operating Pressure [MPa] 0.107

Temperature and pressure were defined according to several examples of other industrial tank projects.

According to API RP 2350 [120], relevant operation levels in the tank are defined as:

Normal Capacity (Normal Fill Level): this level is comprised below the rated capacity, and is a level

predefined by the terminal operator in order to alert and allow for time to stop the flow before the fill level

reaches the safe fill level avoiding the latter to be passed. This level will not be defined in this work as it will

depend on specific safety measures still to define by the operator.

Rated Capacity (Safe Fill Level): this level is usually 90% of the maximum capacity of the tank for tanks with

internal floating roofs or decks.

𝑅𝑎𝑡𝑒𝑑 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 = 0.9 × 𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 = 30 000 𝑚3 (21)

Maximum Capacity (Overfill Level): it is correspondent to the level of product in a tank at which an overflow

and spilling out will happen in case any additional product is added to the tank.

𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 =𝑅𝑎𝑡𝑒𝑑 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦

0.9= 33 333 𝑚3 (22)

Heel Level (Minimum Level): it is correspondent to the volume usually left in the tank before bunkering,

and it is the minimum volume that remains in the tank in order to keep it cold. As a thumb rule, 5 % of the

57

volume of the tank can be assumed for heel level [121].

𝐻𝑒𝑒𝑙 𝐿𝑒𝑣𝑒𝑙 = 0.05 × 𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 = 1 667 𝑚3 (23)

After defining this operation levels, the next step is to define the dome characteristics.

This feature has, as seen before in other sections of the present work, a methodology of construction

specific of every company that projects and builds such tanks; there is no legislation or standards that define

specifically how the dome of an LNG tank should be built or the dimensions it should have. For these reasons,

and in order to attain plausible and meaningful results, the dimensions of the dome were determined by

following examples and projects of existing LNG tanks (see Annex IV).

The final project of the tank is illustrated in Figure 43, and can also be found in Annex VIII.

Figure 43 - Tank Project Scheme with Main Dimensions and Levels

4.2.2 Choosing the Insulation

An appropriate insulation is a key element on cryogenic storage tanks construction and will critically

influence the daily amount of BOG. The insulation chosen determines the hold period in case of supply rupture

and allows the regulation of temperatures in the storage tank, but has a strong impact on the use of energy of

the facility.

The insulation structure was designed to meet the daily BOG rate limit (up to a calculated percentage value

of the tank working capacity), and shall endure the best case scenario and the worst case scenario possible, over

a period of 24 hours, guaranteeing that the designed BOG rate will not be overcome. To do so, several examples

of materials, their thicknesses, and characteristics were taken into account and used to determine the insulation

of the bottom, walls and roof of the tank, having as basis examples of industrial LNG projects and technologies

researched.

58

4.2.2.1 Bottom

This section of the tank has to present structural properties capable of enduring cryogenic temperatures

and the forces that will occur inside the tank, while providing the necessary insulation.

The bottom of the tank has an outer layer of concrete, with a layer of 9% nickel steel, and then another

layer of concrete. On top, 4 layers of asphalt and foam glass exist to provide impermeability and insulation to the

bottom. LNG tanks usually have a heating system installed on the ground below to guarantee the water present

on the soil will not freeze and affect the foundations of the tank. This system, however, was not contemplated

in the calculations as it does not influence significantly the heat ingress through the bottom.

Figure 44 illustrates the chosen layers of the tank bottom, based on industrial LNG projects consulted (see

Annex IX).

Figure 44 - Bottom of the tank – Composition (layers)

4.2.2.2 Walls

Similarly to the bottom, the walls have to simultaneously provide structural resistance and insulation to

maintain the LNG temperature. Being so, and once again following industrial examples (see Annex IX), in direct

contact with LNG is a layer of 9% nickel steel, concrete for structural resistance, a thick layer of expanded perlite

to provide the necessary insulation, and once again concrete as the outer layer. The composition of the walls of

the tank is illustrated in Figure 45.

Figure 45 – Walls of the tank – Constitution (layers)

59

4.2.2.3 Roof

In this analysed section it is included the dome and also the suspended deck, which function is to support

the insulation material; the deck is made of aluminium and supports a layer of foam glass to provide insulation.

The dome structure, outer layer, is made of concrete. Figure 46 illustrates the composition of the suspended

deck and dome.

Figure 46 - Dome (Roof) of the tank – Constitution (layers)

4.3 Boil-off Production

In a small tank such as the one planned for Trafaria Terminal and destined to withstand long periods on

holding mode, BOG production is an issue that can severely influence the profitability of the operation of the

terminal. As a starting point of the simulations, the maximum daily BOG rate will be found in this section. The

objective of first defining this parameter is to allow to choose and design the insulation of the tank in order to

meet the obtained value and mark out the remaining calculations.

4.3.1 Defining the Maximum Boil-Off Gas Rate

In the literature, the maximum admitted value for BOG seems to differ from case to case and there is no

consensus regarding standard values, knowing, logically, that the lower the daily BOG, the less will be the energy

losses with re-liquefaction or any other BOG management methods applied. Yet, some authors indicate the value

of 0.05% of the liquid volume as the daily maximum allowable [88] [122]. BOG remains a key issue regarding

economic and technical reasons [122], as its rate also influences safety (pressure increase inside of the tank) and

the total cost of the LNG plant, which is planned and built according to the calculated re-liquefaction needs.

It is also known that the heat input to LNG tanks depends on the structure of the tank, lines and vessel

discharge conditions. Especially small tanks like the present case-study exchange more heat with the

surroundings due to their large surface area to volume ratio.

In order to assume a maximum allowed daily value of BOG for the tank in study, the calculation of the value

was performed using the results stated by Adom et al. [122], that presented a percentage of BOG per day for

four different tank capacities (200 000 m3, 180 000 m3, 160 000 m3 and 140 000 m3) and different percentages

60

of methane. Figure 47 shows the values taken from the mentioned reference:

Figure 47 - Daily percentage of BOG for different tank capacities and LNG methane content [122].

With the aim of retrieving a reliable set of values for a 30 000 m3 tank, an LNG composition 90.1% methane

content was assumed.

From the linear regression equations it was possible to find that the slope for all the lines corresponds to

0.0003. Knowing that the methane content of the Algerian gas (the one Portugal receives) is about 90.1% and

that the pretended volume is 30 000 m3, the new value for y-intercept was calculated. The y-intercept values

were plotted and the equation was best fitted with a linear tendency line, achieving a good correlation factor (R2

= 0.9642), as it is possible to see in Figure 48:

Figure 48 – Linear Fitting of Ordinate Interception Values.

From the fitted equation presented in Figure 48, it was possible to calculate the ordinate interception value

for a tank capacity of 30 000 m3, allowing to attain the following equation for the maximum daily BOG (equation

(24)):

𝑦 = 0.0003𝑥 + 0.03964 (24)

where 𝑦 is the BOG daily rate, and 𝑥 is the percentage of methane of the LNG.

Applying to the equation above (equation (24)) the percentage of methane of 90.1%, the maximum daily

percentage of BOG was defined as:

𝐵𝑂𝐺𝑑𝑎𝑖𝑙𝑦 𝑚𝑎𝑥𝑖𝑚𝑢𝑚 = 0.0667 %

This BOG rate shall be compared with the value of 0.05% proposed in the literature, and also with a lower

value - 0.039% - approximately half the extrapolated value and considered interesting for comparison by the

project members, in order to identify the difference of investment in insulation, and understand what changes

can be performed to the BOG management systems studied.

y = 0,0003x + 0,0166R² = 0,9918

y = 0,0003x + 0,0127R² = 0,9909

y = 0,0003x + 0,0102R² = 0,9883

y = 0,0003x + 0,0037R² = 0,9909

0,0250

0,0350

0,0450

0,0550

80 85 90 95 100

BO

G v

ol%

/day

% Methane

BOG rate vs % Methane

140 000 m3 160 000 m3 180 000 m3 200 000 m3

y = -0,000206x + 0,045820R² = 0,964235

0,000

0,005

0,010

0,015

0,020

120 140 160 180 200 220

b

Capacity (thousand m3)

Y-intercept Values

61

4.3.2 Heat Input to the Tank

To determine the total heat input to the tank and define the insulation required to meet a daily BOG rate

equal or smaller than the attained result, it was necessary to assess the total heat entering the tank and,

consequently, find the actual BOG rate, through an iterative method, until the insulation was adjusted to the

maximum defined BOG rate, independent of the climate conditions throughout the year.

Being so, the calculation of the total heat input was performed by analysing three different sections of the

system: bottom, walls and roof. The total heat input value was then computed by adding the heat that enters

the tank by each one of the three analysed sections.

Therefore, the total heat input to the tank is given by equation (25):

𝑞𝑡𝑎𝑛𝑘 = 𝑞𝑏𝑜𝑡𝑡𝑜𝑚 + 𝑞𝑤𝑎𝑙𝑙𝑠 + 𝑞𝑑𝑜𝑚𝑒 (25)

In order to account for the climate differences during the year, the total heat input was studied for two

considered scenarios:

Best Case Scenario: winter (lowest temperatures), windless conditions (natural convection).

Worst Case Scenario: summer (highest temperatures), windy conditions (forced convection).

4.3.2.1 Heat Input through the Bottom

To calculate the heat input through the bottom of the tank, it was analysed as a regular infinite plane wall,

which allows the application of the Fourier Law of heat conduction for one dimension. Thus, knowing that the

tank operation temperature is -164°C and the soil temperature is practically constant during the year (16.18°C)

[118], the wall-face temperatures were defined.

As the bottom is constituted by different materials (destined to provide structural resistance and insulation)

with different characteristics, this system will be analysed as a multilayer wall.

Having the thicknesses and thermal conductivity of each material layer, and also the area of the bottom of

the tank, the thermal resistance of each layer was calculated according to equation (26), below, considering solely

conduction, where L is the thickness of the material, k is the thermal conductivity and A is the surface area of the

layer:

𝑇ℎ𝑒𝑟𝑚𝑎𝑙 𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒 =𝑇ℎ𝑖𝑐𝑘𝑛𝑒𝑠𝑠

𝐶𝑜𝑛𝑑𝑢𝑐𝑡𝑖𝑣𝑖𝑡𝑦.𝐴𝑟𝑒𝑎=

𝐿

𝑘.𝐴 [K/W] (26)

It is possible to attain the total heat by applying the following equations to the several layers of material:

𝑇∞1 − 𝑇1 = 𝑞𝑐𝑜𝑛𝑑𝐿1

𝑘1.𝐴

𝑇1 − 𝑇2 = 𝑞𝑐𝑜𝑛𝑑

𝐿2

𝑘2. 𝐴

𝑇2 − 𝑇∞2 = 𝑞𝑐𝑜𝑛𝑑

𝐿3

𝑘3. 𝐴

(27)

(28)

(29)

where, 𝑇∞1 is the temperature of the coldest medium (LNG), 𝑇∞2 the temperature of the warmest medium

62

(soil), and 𝑇1 and 𝑇2 the temperatures of the insulation layers between.

These equations lead then to equation (30), which allows to compute the total heat input through the

bottom of the tank, 𝑞𝑏𝑜𝑡𝑡𝑜𝑚, where 𝑇𝑆𝑂𝐼𝐿 and 𝑇𝐿𝑁𝐺 are respectively the temperatures of the soil and LNG:

𝑞𝑏𝑜𝑡𝑡𝑜𝑚 =𝑇𝑆𝑂𝐼𝐿−𝑇𝐿𝑁𝐺

∑ 𝑇ℎ𝑒𝑟𝑚𝑎𝑙 𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒𝑠 [ 𝑊 ] (30)

Knowing the total heat input, it is then possible to calculate the surface temperature of each layer of the

bottom, using equation (31), where T1 is the temperature of the warmest surface and T2 is the temperature of

the coldest surface of each material:

𝑇1 = 𝑇2 + 𝑞𝐿

𝑘.𝐴 [ 𝐾 ] (31)

4.3.2.2 Heat Input through the Walls

Following a similar approach to the heat input through the bottom, to calculate the heat input through the

walls of the tank, this section is analysed as one-dimensional heat flow through multiple cylinder sections,

allowing to apply the Fourier Law of heat conduction for the several layers that compose the walls. Thus, similarly

to the approach used to calculate the heat input through the bottom of the tank, knowing that the tank operation

temperature is -164°C and the soil temperature is practically constant during the year (16.18°C), the wall-face

temperatures were defined.

Having the height of the tank, the radius, and the thermal conductivity of each material layer, the thermal

resistance of each layer was calculated according to equation (32), below, considering solely conduction, where

L is the height of the tank of the material, k is the thermal conductivity, 𝑅1 is the internal radius of the layer and

𝑅2 the external radius of the layer:

𝑇ℎ𝑒𝑟𝑚𝑎𝑙 𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒 =ln(𝑅2 𝑅1⁄ )

2𝜋.𝑘.𝐿 [ K/W ] (32)

It is possible to attain the total heat applying the following equations:

𝑇∞1 − 𝑇1 = 𝑞𝑐𝑜𝑛𝑑ln(𝑅2 𝑅1⁄ )

2𝜋.𝑘1.𝐿 ,

𝑇1 − 𝑇2 = 𝑞𝑐𝑜𝑛𝑑 ln(𝑅3 𝑅2⁄ )

2𝜋.𝑘2.𝐿 ,

𝑇2 − 𝑇∞2 = 𝑞𝑐𝑜𝑛𝑑 ln(𝑅3 𝑅4⁄ )

2𝜋.𝑘3.𝐿,

(33)

(34)

(35)

where, 𝑇∞1 is the temperature of the coldest medium (LNG), 𝑇∞2 the temperature of the warmest medium

(soil), and 𝑇1 and 𝑇2 the temperatures of the insulation layers between.

Leading then to the equation that allows to compute the total heat input, 𝑞𝑤𝑎𝑙𝑙𝑠, with equation (36) where

𝑇𝑆𝑂𝐼𝐿 and 𝑇𝐿𝑁𝐺 are the temperatures of the soil and LNG:

𝑞𝑤𝑎𝑙𝑙𝑠 =𝑇𝑆𝑂𝐼𝐿−𝑇𝐿𝑁𝐺

∑ 𝑇ℎ𝑒𝑟𝑚𝑎𝑙 𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒𝑠 [ 𝑊 ] (36)

Knowing the total heat input, it is then possible to calculate the surface temperature of each layer of the

wall, using equation (37), where 𝑇1 is the temperature of the warmest surface and 𝑇2 is the temperature of the

coldest surface of each material:

𝑇1 = 𝑇2 +ln(𝑅2 𝑅1⁄ )

2𝜋.𝑘.𝐿 [ 𝐾 ] (37)

63

4.3.2.3 Heat Input through the Roof

As for the calculation of heat input through the roof, besides accounting for the ambient temperature it is

also necessary to account for the solar radiation incident on the outer surface of the roof. The determination of

the heat input through the roof is the influenced by climacteric conditions in an in-ground tank, as temperature,

radiation and also wind conditions can affect the results.

To perform this calculation and compute the heat input in this complex interface, one may make recourse

to the resistances of insulation and structure materials, and the absorptivity of the white paint with which the

dome is painted; heat input sources, incident radiation, ambient temperature and associated convection are

analysed. Knowing the physical characteristics of the dome (roof), and according to Figure 46, an equivalent heat

transfer diagram was made in order to define the calculations to determine the heat input through the roof of

the tank, as displayed in Figure 49, below:

Figure 49 - Equivalent Heat Transfer Diagram – Dome

where 𝑞′′𝑠𝑜𝑙𝑎𝑟 is the solar irradiation, 𝑞1 is the heat transferred by convection, ℎ𝑒𝑥𝑡 is the convection

coefficient considered, αpaint is the absorptivity of the white paint, 𝑞2 is the heat rate that enters through the

layer of concrete, 𝑞3 is the heat that enters through the walls and bottom of the tank, and 𝑞𝑙𝑎𝑡𝑒𝑛𝑡 is the latent

heat (phase change of LNG) inside the tank; 𝑇𝑎𝑖𝑟 is the ambient temperature, 𝑇1 is the surface temperature, 𝑇2

the temperature between the dome and the insulation deck and 𝑇3 the temperature inside the tank. 𝑅𝑐𝑜𝑛𝑐𝑟𝑒𝑡𝑒

and 𝑅𝐴𝑙+𝐹𝑜𝑎𝑚 are the thermal resistances (conductivity) of the concrete layer and the deck (foam glass and

aluminium).

To simplify the calculations, the convection resistance between the concrete layer (inner interface) and the

insulation deck (upper interface) is discarded, as equal temperatures of those surrounding walls were assumed

and, therefore, no natural convection exists; it is then considered that the natural gas temperature is uniform in

both interfaces, as shown in Figure 50. Also, as a simplification, the radiation from the dome to its surroundings

is not considered.

Figure 50 - Scheme of the Temperatures considered for the Heat Ingress Calculation in the Dome

64

From Figure 49 and Figure 50 it is possible to infer the following equations ((38) and (39)):

𝑞𝐷𝑜𝑚𝑒 = 𝑞2 = 𝑞1 + 𝑞𝑠𝑜𝑙𝑎𝑟

′′ . 𝐴𝐷𝑜𝑚𝑒 . 𝛼 =𝑇1 − 𝑇3

𝑅𝑐𝑜𝑛𝑐𝑟𝑒𝑡𝑒 + 𝑅𝐴𝑙+𝐹𝑜𝑎𝑚

, (38)

𝑞1 =

𝑇𝑎𝑖𝑟 − 𝑇1

1ℎ𝑒𝑥𝑡 . 𝐴𝑑𝑜𝑚𝑒

(39)

With equations (38) and (39) it is possible to obtain equation (40):

𝑇1 − 𝑇3

𝑅𝑐𝑜𝑛𝑐𝑟𝑒𝑡𝑒+𝑅𝐴𝑙+𝐹𝑜𝑎𝑚

= 𝑞𝑠𝑜𝑙𝑎𝑟′′ . 𝐴𝐷𝑜𝑚𝑒 . 𝛼 +

𝑇𝑎𝑖𝑟−𝑇1

1𝐴𝐷𝑜𝑚𝑒 . ℎ𝑒𝑥𝑡

(40)

Then, from equation (40), it is possible to attain equation (41) in order to find 𝑇1:

𝑇1 =

𝐴𝐷𝑜𝑚𝑒 . (𝑅𝑐𝑜𝑛𝑐𝑟𝑒𝑡𝑒 + 𝑅𝐴𝑙+𝐹𝑜𝑎𝑚)(ℎ𝑒𝑥𝑡 . 𝑇𝑎𝑖𝑟 + 𝑞𝑠𝑜𝑙𝑎𝑟′′ . 𝛼) + 𝑇3

𝐴𝐷𝑜𝑚𝑒(𝑅𝑐𝑜𝑛𝑐𝑟𝑒𝑡𝑒+𝑅𝐴𝑙+𝐹𝑜𝑎𝑚). ℎ𝑒𝑥𝑡 + 1 (41)

Knowing 𝑇1, it is possible to find 𝑞2, the total heat input through the dome referring to equation (38), and

thus finding the heat input to the tank through the dome. In order to study different effects of temperature and

wind, the simulations were performed for different seasons of the year, and assuming convection coefficients

that simulate a day with no wind and a very windy day.

Taking into account the work of Avraham Shitzer [123] regarding the correlation between wind speed (m/s)

and the convective heat transfer coefficient (W/m2.K) performed by Tikuisis and Osczevski [124], the typical

values of the convection heat transfer coefficients for natural and forced convection stated by Moran et al. [125],

and the IPMA classification of wind velocity [126], the coefficients used in the study of the effect of the wind are

10 W/m2.K for calm wind (approximately at 1.34 m/s), and 90 W/m2.K for strong wind (approximately at 50 m/s).

4.3.3 Heat Input through Piping and Pumping System

Although only applicable while loading or unloading the tank, there is a substantial heat input through

circulation and loading pipelines, between the tank and the vessels at the jetty.

To be able to calculate the heat that enters the LNG system through piping and pumps, it was necessary to

determine the length and diameter of the pipelines, the thickness of the piping insulation and its thermal

conductivity, and the LNG/BOG flow rate. Also, the LNG, BOG, and air properties were considered in this

calculations.

Following a similar methodology to the one used by Wordu and Peterside [127], the heat ingress through

pipelines was computed using the following equations: equations (42) to (45) calculate the velocity of the LNG in

the pipelines, and Prandtl (𝑃𝑟), Reynolds (𝑅𝑒) and Nusselt (𝑁𝑢) numbers, used to calculate the heat transfer

coefficient (ℎ0), as inputs in equation (46).

𝑃𝑟 =𝐶𝑝𝑎𝑖𝑟

µ𝑎𝑖𝑟

𝑘𝑎𝑖𝑟, (42)

𝑣 =

𝑄𝑓𝑖𝑙𝑙𝑖𝑛𝑔 × 𝐴

3600,

(43)

65

𝑅𝑒 =𝑣 × 𝐷 × 𝜌𝑎𝑖𝑟

µ𝑎𝑖𝑟

, (44)

𝑁𝑢 = 0,3 +

0,62𝑅𝑒12𝑃𝑟1/3

[1 + (0,4𝑃𝑟

)

23

]

14

[1 + (𝑅𝑒

282000)

58

]

45

(45)

where 𝐶𝑝𝑎𝑖𝑟, µ𝑎𝑖𝑟 , and 𝑘𝑎𝑖𝑟 are respectively the specific heat (J/kg.K), viscosity (kg/m.s) and thermal

conductivity (W/m.K) of the ambient air; 𝑣 is the velocity of the gas (m/s), 𝑄𝑓𝑖𝑙𝑙𝑖𝑛𝑔 is the filling volumetric rate of

the tank or ship (m3/h), and 𝐴 is the area of the pipeline (m2); 𝐷 is the diameter of the pipeline (m), 𝜌𝑎𝑖𝑟 is the

density of the ambient air (kg/m3).

Prandtl number characterizes the thermal and momentum boundary layers, representing the relation

between momentum diffusion and heat diffusion; Reynolds number characterizes the flow regime, through the

relation between inertia and viscous forces; Nusselt number represents the relation between the heat

transferred by convection and by conduction in a fluid subjected to a temperature gradient. An increase in

Reynolds number results in an increase of the Nusselt number, which will increase the heat transferred by

convection, and consequently the convection coefficient (ℎ0).

Equation (46) was used to compute the heat transfer coefficient:

𝑁𝑈𝐷 =

ℎ0𝐷

𝑘𝑎𝑖𝑟

↔ ℎ0 =𝑁𝑈𝐷 × 𝑘𝑎𝑖𝑟

𝐷 (46)

Finally, the heat input through pipelines is given by equation (47):

𝑄𝑝𝑖𝑝𝑒𝑙𝑖𝑛𝑒𝑠 =

2 𝜋𝐿[𝑇∞ − 𝑇𝐿𝑁𝐺]

1ℎ0𝑟0

+𝑙𝑛 (

𝑟𝑒

𝑟𝑖)

𝑘𝑖𝑛𝑠

(47)

where 𝑄𝑝𝑖𝑝𝑒𝑙𝑖𝑛𝑒𝑠 is the heat transferred to the LNG through the pipelines (W), 𝐿 is the length of the pipelines

(m), 𝑇∞ is the temperature of the ambient air (⁰C), 𝑇𝐿𝑁𝐺 is the temperature of the LNG (⁰C), 𝑟𝑖 is the internal

radius of the pipelines (m), 𝑟𝑒 is the external radius of the pipelines (𝑟𝑖 plus the thickness of insulation) (m), and

𝑘𝑖𝑛𝑠 is the thermal conductivity of the insulation (W/m.K).

The BOG production rate due to heat ingress through pipelines is given by equation (48):

𝐵𝑂𝐺𝑝𝑖𝑝𝑒𝑙𝑖𝑛𝑒𝑠 = 3,6 × 𝑄𝑝𝑖𝑝𝑒𝑙𝑖𝑛𝑒𝑠 (

𝛽

𝜆) (48)

where 𝛽 is a safety factor (5%), 𝜆 is the latent heat of the LNG (kJ/kg), and 𝐵𝑂𝐺𝑝𝑖𝑝𝑒𝑙𝑖𝑛𝑒𝑠 has as units kg h-1.

As for the heat ingress through the pumping system (loading/unloading pumps), the heat transfer to the

fluid is given by equation (49):

𝑄𝑝𝑢𝑚𝑝𝑖𝑛𝑔 =

𝑚𝑝𝑢𝑚𝑝𝑠𝑔𝐻

3600(

1

𝜂𝑝

− 1) (49)

where 𝑄𝑝𝑢𝑚𝑝𝑖𝑛𝑔 is the heat transferred to the LNG by the pumping system, 𝑚𝑝𝑢𝑚𝑝𝑠 is the pumping capacity

of each pump (kg/h), 𝑔 is the acceleration of gravity, 𝐻 is the pump head of the LNG pumps, and 𝜂𝑝 is the

efficiency of the pumps.

The BOG production rate due to heat ingress through the piping system is given by equation (50):

𝐵𝑂𝐺𝑝𝑢𝑚𝑝𝑠 = 3,6𝑄𝑝𝑢𝑚𝑝𝑖𝑛𝑔

𝜆 (50)

66

It is also important to refer that while loading a vessel, to maintain the pressure in both systems, the

terminal will receive from the vessel the same volume of natural gas that is sent of LNG to the vessel. Similarly,

when loading the tank, the same volume of gas is transferred from the tank to the vessel; e.g., if the tank receives

10.000m3 of LNG, the ship will receive from the terminal 10.000m3 of natural gas to occupy the displaced volume.

It is called vapour displacement and it is not processed by the BOG recovery unit, as this volume is necessary at

its gaseous state. Hence, this volume was not accounted for when dimensioning the BOG re-liquefaction system.

4.4 Simulation Scenarios

4.4.1 BOG Production

According to the model of BOG rate applied to the case-study of Trafaria terminal (see Section 4.3.1), the

BOG daily production is calculated as 0.0667% of the maximum volume of the tank. As a comparison, the

calculations will also be performed for the value usually described in the literature as the typical BOG rate (0.05%)

and a lower value (0.039%), for comparison, of the maximum volume of the tank [88] [122] – in order to assess

the differences between the model, what is usually mentioned as common, and a lower value of design BOG.

The advantages and disadvantages of designing the tank for each BOG production will be assessed, namely

the differences of holding period, the initial investment in insulation, and the impact in the BOG management

measures studied. The scenarios will be defined as:

1) BOGrate = 0.039 % - best scenario - winter season, windless conditions.

2) BOGrate = 0.039 % - worst scenario - summer season, very windy conditions.





4.4.2 BOG Management

Knowing the constraints of the terminal and the legal impossibility of commercializing natural gas in gaseous

form, the present work explores three scenarios of BOG management: a re-liquefaction system using a simple

turbo expander cycle, the installation of a cogeneration plant close to the LNG storage, and the emergency

system (forbidden except in case of emergency). This section intends to explain the methodology applied to the

BOG management system and how to analyse its performance.

4.4.2.1 Single Expander Re-Liquefaction System

Due to the size of the LNG terminal and storage facilities, only the hypothesis of a single expander re-

liquefaction system will be studied, as the expected BOG rate is very small and the dual turbo-expander cycle

implies a superior initial investment.

In order to determine the capacity of the re-liquefaction system and the quantity of refrigerant fluid (N2),

the Mollier diagrams of methane (see Annex X) and nitrogen (see Annex XI) were used with the aim of

determining the enthalpy, pressure and temperature changes of both fluids during the liquefaction process (as

67

the major component of LNG, the Mollier diagram of methane was used to study the behaviour of the BOG, as

an approximation). The cycles studied are depicted in Figure 51, which represents the nitrogen cycle (refrigerant),

and the BOG cycle, collected from the tank in gaseous form, and returned to the tank, liquefied, after transferring

heat to the refrigerant.

Figure 51 - Single Expander Nitrogen Cycle

i) Determining the Refrigeration Power Necessary

The determination of the refrigeration power required for the liquefaction unit will depend on the temperature

and pressure at which BOG is collected and enters the liquefaction system, as well as the temperature and

pressure the LNG is to be delivered to the tank; this information allows to determine the enthalpies at the

beginning (ℎ𝐵𝑂𝐺) and end (ℎ𝐿𝑁𝐺) of the liquefaction cycle, and to determine the heat necessary to remove from

the BOG in order to condense to LNG (ℎ𝑟𝑒𝑓). This is defined by using equation (51):

ℎ𝑟𝑒𝑓 = ℎ𝐵𝑂𝐺 − ℎ𝐿𝑁𝐺 (51)

Knowing the BOG mass flow rate it is then possible to calculate the power of the refrigeration/liquefaction

cycle (𝑃𝑟𝑒𝑓) by using equation (52):

𝑃𝑟𝑒𝑓 = �̇�𝐵𝑂𝐺 × ℎ𝑟𝑒𝑓 (52)

Equation (51) determines the heat that will have to be absorbed by the refrigeration fluid, nitrogen, so as

to achieve the cooling capacity required, given by equation (52).

ii) Determining the Liquefaction Cycle Stages

Knowing the required power of the liquefaction cycle, the heat that has to be removed from the BOG by

the refrigeration fluid, and knowing how the liquefaction cycle is (see Figure 51), resourcing to the Mollier

diagram of nitrogen (see Annex XI) it is possible to obtain the pressure, temperature and enthalpy correspondent

to each stage of the cycle.

The liquefaction system considered and illustrated in Figure 51- nitrogen expander cycle –, operates as

follows: the refrigerant is compressed in two stages with intercooling (states 1 to 5), then passes through a heat

exchanger, being pre-cooled by a colder nitrogen stream (states 5 to 6); it is then expanded (states 6 to 7), and

finally absorbs heat from the BOG (liquefying and sub-cooling it) in two counter-current heat exchangers (states

7 to 1), returning back to its original state (state 1). It is considered that this cycle operates in a closed loop, and

with minimal losses during operation.

68

According to the examples from the works of Gerdsmeyer and Isalski [128], A. Vorkapić et al. [129], and

Gómez et al. [130], the nitrogen pressure at the inlet of the first compressor (state 1) shall be between 0.8 and

1 MPa and the pressure at the expander inlet shall be between 3.5 and 6.0 MPa. The efficiency of the turbo-

compressors shall be considered 0.8. Knowing the pressure ratio of the cycle (𝑝𝑅) given by equation (53), and

through equation (54), the intermediate pressure (𝑝𝑖), pressure of states 2 and 3, can be calculated as follows:

𝑝𝑅 =𝑝2

𝑝1

, (53)

Where 𝑝1 is the inlet pressure, and 𝑝2 is the discharge pressure. To guarantee that the work done by each

compressor is minimal and divided evenly, equation (54) is applied:

𝑝𝑅 =𝑝𝑖

𝑝1

×𝑝2

𝑝𝑖

⇒ 𝑝𝑖 = √𝑝1 𝑝2 (54)

The outlet temperature of the expander shall be lowered to a temperature between -168⁰C to -180⁰C, in

order to achieve sub-cooled LNG. The subcooled LNG shall be delivered to tank at -165⁰C.

With the required refrigeration power of the liquefaction system (𝑃𝑟𝑒𝑓), the mass flow rate of refrigerant

shall be computed using equation (55):

�̇�𝑁2=

𝑃𝑟𝑒𝑓

(ℎ1 − ℎ7) + (ℎ6 − ℎ5) , (55)

where �̇�𝑁2 is the mass flow rate of nitrogen, ℎ1 and ℎ7 are the enthalpies at states 1 and 7, respectively,

and ℎ5 and ℎ6 are the enthalpies at states 5 and 6, respectively.

The total work of the compressors (�̇�𝑐) is expressed by equation (56), while the total heat dissipated in the

intercoolers (𝑄𝑖) is expressed by equation (57):

�̇�𝑐 = �̇�𝑁2[(ℎ2 − ℎ1) + (ℎ4 − ℎ3)], (56)

�̇�𝑖 = �̇�𝑁2[(ℎ3 − ℎ2) + (ℎ5 − ℎ4) + (ℎ6 − ℎ5)] (57)

The total heat dissipated from the nitrogen to the surroundings shall be calculated using equation (58):

�̇�𝑑𝑖𝑠𝑠 = �̇�𝑖 + �̇�ℎ𝑒 (58)

where �̇�ℎ𝑒 is the heat absorbed by the refrigerant stream in the heat exchangers.

The liquefaction cycle configuration was studied according to the values described in this section, and for

different BOG inlet temperatures (-155⁰C, -140⁰C, -125⁰C and -110⁰C) in order to study the energy performance

of the liquefaction system at different temperatures (e.g. during the year seasons).

4.4.2.2 Cogeneration

The study of the cogeneration scenarios starts by having the data of the different BOG rate scenarios and

considering the BOG mass flow as the feed gas for the cogeneration plant (at reference conditions – 𝑇𝑟𝑒𝑓 =

273.15 𝐾, 𝑝𝑟𝑒𝑓 = 1.01325 𝑏𝑎𝑟(𝑎)). This way, and assuming the electric efficiency as 40% and the thermal efficiency

as 35%, it is possible to calculate several parameters associated with the implementation of a cogeneration.

Through equation (59) the consumption of natural gas, assumed as an ideal gas, is calculated for reference

conditions:

𝑄𝑟𝑒𝑓 = 𝑄 ×𝑇𝑟𝑒𝑓

𝑇×

𝑝

𝑝𝑟𝑒𝑓

(59)

Where 𝑄 is the volumetric rate consumption of natural gas at a given pressure, 𝑝, and temperature, 𝑇.

69

Equation (60) gives the power of the cogeneration,

𝑃𝑐𝑜𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛 = 𝑄𝑟𝑒𝑓 ×𝐿𝐻𝑉

3600 (60)

where 𝑃𝑐𝑜𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛 is the power of the cogeneration (in kW), 𝑄𝑟𝑒𝑓 is the consumption of natural gas (in

Nm3/h), and 𝐿𝐻𝑉 is the lower heating value in kJ/Nm3.

Equations (61) and (62) are, respectively, the electric and thermal power of the cogeneration:

𝑃𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐 = 𝑃𝑐𝑜𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛 × 𝐸𝜂 (61)

𝑃𝑡ℎ𝑒𝑟𝑚𝑎𝑙 = 𝑃𝑐𝑜𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛 × 𝐻𝜂 (62)

where 𝑃𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐 and 𝐸𝜂 are the electrical power (in kW) and efficiency, and 𝑃𝑡ℎ𝑒𝑟𝑚𝑎𝑙 and 𝐻𝜂 are the thermal

power (in kW) and efficiency.

Following the guidelines found in the Portuguese and European legislation dedicated to this subject

(Decree-Law n.23/2010 [131] and EU directive 2004/8/CE [132]), the Primary Energy Saving (𝑃𝐸𝑆), during a

certain period of time, that results from the cogeneration operation can be calculated as shown in equation (63):

𝑃𝐸𝑆 =𝐻𝐶𝐻𝑃

𝑅𝑒𝑓 𝐻𝜂+

𝐸𝐶𝐻𝑃

𝑅𝑒𝑓 𝐸𝜂− FTotal (63)

where 𝐻𝐶𝐻𝑃 and 𝐸𝐶𝐻𝑃 are the thermal and electrical energy produced in cogeneration, 𝑅𝑒𝑓 𝐻𝜂 and 𝑅𝑒𝑓 𝐸𝜂

are, respectively, the considered reference values of thermal and electrical efficiencies for the combustion of

natural gas as fuel, and FTotal is the total amount of fuel consumed during the analysed period.

Also as a reference, it is possible to calculate the total CO2 emissions avoided during the operation in

cogeneration, applying equation (64):

(𝐴. 𝐸. 𝐶𝑂2)𝐶𝐻𝑃 =𝑃𝐸𝑆

𝐸𝐶𝐻𝑃

× (𝐸. 𝐶𝑂2) (64)

where 𝐸. 𝐶𝑂2 is the CO2 emission factor calculated for natural gas [133], correspondent to 201,96

gCO2/kWh, and (𝐴. 𝐸. 𝐶𝑂2)𝐶𝐻𝑃 are the avoided emissions of CO2 by producing electricity in cogeneration.

4.4.2.3 Emergency System

The emergency system, a flare, shall be only used in case no other option to handle the BOG is available,

and as last resource. Flaring represents a great loss for the operator, and can endanger the profitability of the

terminal. It is estimated that this system shall only be used in case the BOG re-liquefaction system, cogeneration,

or other recovery system is damaged or under maintenance, and in case a phenomenon like the rollover effect

occurs.

One advantage of in-ground tanks is the fact that due to the very stable temperatures around the tank walls

and bottom, it is highly unlikely that the rollover effect occurs. Despite being unlikely, the cogeneration and BOG

recovery cannot be designed accounting for possible rollover effects to happen (these events can increase the

BOG 10 to 30 times – see section 3.1.4), and in the event of a rollover phenomenon, it is considered that all the

exceeding BOG (that the recovery systems cannot process) is lost.

70

5. Results and Discussion

This chapter covers the results of simulations and the assessment of the BOG management options

presented, analysing its energy efficiency and feasibility for the specific case of the Trafaria Terminal.

5.1 Heat Input

5.1.1 Designed Heat Ingress to the Tank According to the values determined for the BOG rate (the value of the model (0.067%), the typical value

from the literature (0.050%) and a lower BOG value (0.039%) that can be considered of the same order of

magnitude), the energy ingress through each area of the tank was calculated, and the volumes of BOG and LNG,

and mass of natural gas were computed and are presented in Table 13. As described in chapter 4, the winter

scenarios are considered virtually windless, and the summer scenarios, very windy, in order to depict,

respectively, the best and worst scenarios.

Table 13 - BOG Scenarios and Results

BOG RATE

Season Area of the

tank

Heat [kW]

Heat [MJ/h]

BOG mass flow rate

[kg/h]

BOG mass flow rate [kg/day]

LNG Volume [m3/h]

LNG Volume

[m3/day]

BOG Volume

[m3/day]

BOG mass flow rate

[kg/s]

0.039 %

Winter Best

Scenario

Bottom 9.51 34.23 67.01 1 608.23 0.16 3.81 887.74 0.019

Walls 9.14 32.90 64.40 1 545.55 0.15 3.66 853.14 0.018

Roof 10.67 38.40 75.17 1 804.12 0.18 4.27 995.87 0.021

TOTAL 29.31 105.53 206.58 4 957.90 0.49 11.74 2 736.75 0.057

Summer

Bottom 9.51 34.23 67.01 1 608.23 0.16 3.81 887.74 0.019

Walls 9.14 32.90 64.40 1 545.55 0.15 3.66 853.14 0.018

Roof 11.46 41.26 80.77 1 938.39 0.19 4.59 1 069.99 0.022

TOTAL 30.11 108.38 212.17 5 092.18 0.50 12.06 2 810.87 0.059

0.050 %

Winter

Bottom 9.51 34.23 67.01 1 608.23 0.16 3.81 887.74 0.019

Walls 9.14 32.90 64.40 1 545.55 0.15 3.66 853.14 0.018

Roof 17.89 64.41 126.10 3 026.28 0.30 7.17 1 670.50 0.035

TOTAL 36.54 131.54 257.50 6 180.07 0.61 14.63 3 411.39 0.072

Summer

Bottom 9.51 34.23 67.01 1 608.23 0.16 3.81 887.74 0.019

Walls 9.14 32.90 64.40 1 545.55 0.15 3.66 853.14 0.018

Roof 19.27 69.38 135.82 3 259.71 0.32 7.72 1 799.36 0.038

TOTAL 37.92 136.51 267.23 6 413.50 0.63 15.18 3 540.24 0.074

0.067 %

Winter

Bottom 9.51 34.23 67.01 1 608.23 0.16 3.81 887.74 0.019 Walls 9.14 32.90 64.40 1 545.55 0.15 3.66 853.14 0.018 Roof 28.30 101.88 199.43 4 786.35 0.47 11.33 2 642.06 0.055

TOTAL 46.95 169.00 330.84 7 940.13 0.78 18.80 4 382.94 0.092

Summer Worst

Scenario

Bottom 9.51 34.23 67.01 1 608.23 0.16 3.81 887.74 0.019 Walls 9.14 32.90 64.40 1 545.55 0.15 3.66 853.14 0.018 Roof 30.59 110.14 215.60 5 174.48 0.51 12.25 2 856.30 0.060

TOTAL 49.24 177.26 347.01 8 328.27 0.82 19.72 4 597.19 0.096

The worst scenario, the simulation of a BOG rate of 0.067%, in a windy summer day, presents an increase

of mass flow rate of 40.1% when compared with the best scenario. Figure 52 to Figure 54 illustrate the BOG mass

flow rate for the simulated scenarios, in winter and summer.

Figure 52 - Daily Mass Flow Rate -

BOG=0.039%


BOG=0.050%


BOG=0.067%

190

200

210

220

00:

00

02:

00

04:

00

06:

00

08:

00

10:

00

12:

00

14:

00

16:

00

18:

00

20:

00

22:

00

kg/hBoil Off

(BOG = 0,039%)

BOG=0.0399%, WinterBOG=0.0399%, Summer

240

250

260

270

280

00

:00

02

:00

04

:00

06

:00

08

:00

10

:00

12

:00

14

:00

16

:00

18

:00

20

:00

22

:00

kg/hBoil Off

(BOG = 0,050%)


300

320

340

360

380

00

:00

02

:00

04

:00

06

:00

08

:00

10

:00

12

:00

14

:00

16

:00

18

:00

20

:00

22

:00

kg/hBoil Off

(BOG = 0,067%)


71

Figure 55 to Figure 57 illustrate the volume of LNG that evaporates per hour during the day, while Figure

58 to Figure 60 show the volume of BOG that is produced in the same period, during summer and winter.

Figure 55 - LNG Volume evaporated -

BOG=0.0399%


BOG=0.050%


BOG=0.067%

Figure 58 - BOG Volume - BOG=0.039%



The effect of the daily ambient temperature, incident radiation and forced convection is noticeable in Figure

52 to Figure 60, as well as the effect, within each BOG rate scenario, of the summer and winter seasons. For a

BOG rate of 0.039% the maximum expected BOG is 214.1 kg of BOG per hour, resulting in approximately 5092.2

kg per day; for a BOG rate of 0.067% the maximum expected BOG is 352.1 kg of BOG per hour, resulting in

approximately 8328.3 kg per day. The summer BOG production for BOG rates of 0.039 and 0.067% is,

respectively, approximately 3 and 5% higher than the winter BOG production in each case, resulting in the need

for processing, respectively 134 and 388 more kg of BOG per day. For a BOG rate of 0.067%, it is expected that

the BOG rate results in a maximum production of BOG 35.9% higher than for a BOG rate of 0.039%, highlighting

the effect of a poorer insulation.

5.1.2 Designed Heat Ingress through Piping and Pumping System

The heat ingress through piping and pumping system was analysed for summer and winter conditions to

assess the effect of the ambient temperature. Recalling Chapter 2, the length of the run down lines (loading and

unloading line) used in the calculations is 220m, and have a filling rate of 1000m3/h; its diameter is 0.41m, the

insulation has a thickness of 0.23m, and a thermal conductivity of 0.04 W/m.K, following the typical dimensions

presented on the work of Wordu and Peterside [127]. Two LNG pumps were considered in the calculations of

the heat ingress through the pumping system; the pump head of the LNG pumps is 41m, and the considered

efficiency is 0.8. The summer and winter temperatures used are the average for each season in the period

analysed, respectively, 8⁰C and 25⁰C. The LNG temperature is -161⁰C. A safety factor (design margin) of 5% was

used in the BOG calculations.

Table 14 presents the computed values for the heat ingress in these systems.

0,47

0,48

0,49

0,50

0,51

01

:00

03

:00

05

:00

07

:00

09

:00

11

:00

13

:00

15

:00

17

:00

19

:00

21

:00

23

:00

m3/hBoil Off - Volume LNG

(BOG = 0,039%)


0,58

0,60

0,62

0,64

0,66

01

:00

03

:00

05

:00

07

:00

09

:00

11

:00

13

:00

15

:00

17

:00

19

:00

21

:00

23

:00


(BOG = 0,050%)


0,70

0,75

0,80

0,85

01

:00

03

:00

05

:00

07

:00

09

:00

11

:00

13

:00

15

:00

17

:00

19

:00

21

:00

23

:00


(BOG = 0,067%)


110112114116118120

00:

00

02:

00

04:

00

06:

00

08:

00

10:

00

12:

00

14:

00

16:

00

18:

00

20:

00

22:

00

m3/hBoil Off - Volume NG

(BOG = 0,039%)


136

140

144

148

152

00:

00

02:

00

04:

00

06:

00

08:

00

10:

00

12:

00

14:

00

16:

00

18:

00

20:

00

22:

00


(BOG = 0,050%)


160

170

180

190

200

00:

00

02:

00

04:

00

06:

00

08:

00

10:

00

12:

00

14:

00

16:

00

18:

00

20:

00

22:

00


(BOG = 0,067%)


72

Table 14 – Heat Ingress in the LNG System through Piping and Pumping System

Season LNG

Velocity [m/s]

Prandtl number

Reynolds number

Nusselt number

Heat Transfer Coefficient [W/m2.K]

Heat Ingress Through

Pipelines [kW]

Heat Ingress through Pumping

System [kW]

Winter 2,10 0,71 79 871,20 185,25 12,20

112,32 23,59

Summer 123,62

The volume of BOG, evaporated LNG, and mass of natural gas were computed for the piping and pumping

systems and are presented in Table 15.

Table 15 – BOG Scenarios and Results - Piping and Pumping System

Season q [kW] q [kW]

w/ safety margin

q [MJ/h] BOG mass flow rate

[kg/h]

BOG mass flow rate [kg/day]

LNG Volume [m3/h]

LNG Volume [m3/day]

BOG Volume [m3/day]

BOG mass flow rate [kg/s]

Winter 135,92 142,71 513,76 1 005,74 24 137,65 2,27 54,43 13 323,94 0,279

Summer 147,21 154,57 556,47 1 089,34 26 144,20 2,46 58,95 14 431,55 0,303

5.1.3 Total Heat Ingress With the calculation of heat ingress through the different tank areas, piping, and pumping system, the total

heat ingress for each operation mode was calculated and is presented in Table 16.

Table 16 – Heat Ingress in the LNG System during each Operation Mode

Tank [kW] Piping [kW] Pumping System [kW] TOTAL [kW]

Operation Mode Best

Scenario Worst

Scenario Best

Scenario Worst

Scenario Best

Scenario Worst

Scenario Best

Scenario Worst

Scenario

Holding Mode (BOG = 0.039%) 29,31 30,11 - - - - 29,31 30,11

Loading / Unloading 29,31 30,11 112,32 123,62 23,59 23,59 165,23 177,32

Holding Mode (BOG = 0.05%) 36,54 37,92 - - - - 36,54 37,92


Holding Mode (BOG = 0.067%) 46,95 49,24 - - - - 46,95 49,24


It is important to notice that while loading a vessel, the terminal will have to receive the displacement

volume delivered by the vessel. This volume shall not be processed on the BOG management system, as it is

necessary to maintain the pressure in the storage tank, although there is always a certain volume generated by

the increment in temperature all along the loading and unloading lines.

5.1.4 Insulation Costs

To accomplish the designed BOG rate, the insulation had to be modified in order to adjust the maximum

heat ingress to 0.039 %, 0.050 %, and 0.067% per day. The adjustment was performed on the dome insulation,

as this is the section of the tank through which more heat ingresses to its interior. Hence, the thickness of foam

glass insulation in the deck was optimized in order to assure that, in the worst case scenario of each BOG rate,

the design BOG rate is not exceeded.

The insulation costs for the different BOG rates studied are presented in Table 17. The specifications of each

material are as presented in section 3.1.3.

Table 17 - Tank Insulation Costs

BOG Rate 0.039% 0.050% 0.067%

MATERIAL Cost per m3

[€] Volume

[m3] Total Cost

[€] Cost per m3

[€] Volume

[m3] Total Cost

[€] Cost per m3

[€] Volume

[m3] Total Cost

[€]

Concrete 79.93 4 919.99 393 254.61 79.93 4 919.99 393 254.61 79.93 4 919.99 393 254.61

9% Nickel Steel 125.18 29.24 3 659.52 125.18 29.24 3 659.52 125.18 29.24 3 659.52

Sand 9.96 125.66 1 251.60 9.96 125.66 1 251.60 9.96 125.66 1 251.60

Felt/Asphalt 1.88 15.08 28.36 1.88 15.08 28.36 1.88 15.08 28.36

Foam Glass 888.46 1 143.54 1 015 990.92 888.46 923.63 820 608.05 888.46 804.25 714 543.06

Expanded Perlite 1 544.37 7 051.67 10 890 405.34 1 544.37 7 051.67 10 890 405.34 1 544.37 7 051.67 10 890 405.34

Aluminium 43.21 150.80 6 516.05 43.21 150.80 6 516.05 43.21 150.80 6 516.05

TOTAL Cost [€] 12 311 106.38 12 115 723.51 12 009 658.52

73

The initial investment difference between the best and worse insulation is approximately of 301 448 €,

being this the estimated cost of having a smaller BOG rate. According to a basis of estimate for LNG storage tank

cost analysis [134], it is also necessary to account for other relevant components: the labour component

corresponds to 2.5% of the cost of the tank, the LNG tank foundations are estimated to cost around 15 million

euros, and the civil works, piping, electrical and structural systems, and infrastructures construction are

estimated to be around 25% of the tank cost. To these systems, it is necessary to add the cost of the cryogenic

pumps and fittings on the tank, as well as the BOG management system.

5.1.5 Holding Mode Operation

Operation on holding mode – not loading or unloading LNG - can be a significant part of the operation of

an LNG terminal. Table 18 illustrates the holding mode capacity of each BOG rate configuration chosen.

Table 18 – Holding Mode Capacity

BOG Rate Days in Holding Mode Years in Holding Mode 0.039 %/day 2506 6.86 0.050 %/day 2000 5.48 0.067 %/day 1500 4.11

These values represent a hypothetical situation and compare the impact of the BOG rate in the duration of

the holding mode, until all the LNG changes phase and evaporates; these are only valid assuming that no rollover

phenomenon occurs, nor is any BOG returned to the tank as LNG.

5.2 BOG Management

5.2.1 Single Expander Re-Liquefaction System

The single expander nitrogen BOG re-liquefaction system (see sections 3.2.3 and 4.4.2.1) was designed

according to the information available in the literature and industrial systems. According to these guidelines, two

liquefaction cycles were simulated to achieve the required liquefaction of BOG. The two hypothesis were studied

for the set resulting of the sum of each worst scenario of the tank BOG rates considered, with the worst scenario

of BOG production of piping and pumping system, plus a 5% design margin to guarantee the liquefaction system

is capable of processing the maximum expected BOG production. The hypotheses were simulated for four

different BOG inlet temperatures, to study the performance of the systems and obtain a well-defined tendency.

For the first cycle, hereafter designated as Cycle A, the pressure of the nitrogen at the inlet of the first

compressor is 0.8 MPa, the calculated intermediate pressure is 1.79 MPa, and the pressure at the inlet of the

expander is 4 MPa. Cycle B, designation given to the alternative cycle hereafter, intends to assess the effect of

expanding to a lower pressure (0.4 MPa) and increasing, this way, the heat the refrigerant can absorb from the

BOG. Both cycles start compression at -168⁰C, allowing the nitrogen temperature to be always smaller than the

BOG temperature, and that the minimum temperature difference between the refrigerant and the BOG is 2⁰C;

the compression work in both cycles takes place by compressing superheated nitrogen, to guarantee that no

liquid nitrogen is present at compressors inlet. As, at this stage, the exact pipes layout and the distance to the

re-liquefaction system are not known in detail, the BOG temperature is variable and, as an approximation, its

pressure will be considered equal to the tank operating pressure, 0.107 MPa. The LNG shall be delivered slightly

subcooled to the tank, at -165⁰C.

Figure 61 illustrates both simulated cycles in the Mollier diagram that can also be found in Annex XII.

74

Figure 61 – Liquefaction Cycles A (in blue) and B (in orange)

Table 19 presents the simulation results of both simulated cycles, for four different temperatures, and

according to equations (15) to (20) and (51) to (58) – see sections 3.2.3 and 4.4.2.1:

Table 19 – Liquefaction Cycles Simulation Results

TBOG inlet BOG RATE =

Cycle A Cycle B 0.039% 0.050% 0.067% 0.039% 0.050% 0.067%

- 155 ⁰C

�̇�𝑩𝑶𝑮 [kg/s] 0.362 0.377 0.400 0.362 0.377 0.400

𝒉𝑩𝑶𝑮 [kJ/kg] 525 525 525 525 525 525

𝑷𝑹𝒆𝒇 [kW] 197.3 205.4 218.2 197.3 205.4 218.2

𝑷𝑹𝒆𝒇𝒎𝒂𝒓𝒈𝒊𝒏 [kW] 207.2 215.6 229.1 207.2 215.6 229.1

�̇�𝑪𝒐𝒎𝒑 [kW] 259.0 269.6 286.4 197.8 205.9 218.8

�̇�𝑵𝟐 [kg/s] 4.144 4.313 4.583 2.878 2.995 3.182

- 140 ⁰C

�̇�𝑩𝑶𝑮 [kg/s] 0.362 0.377 0.400 0.362 0.377 0.400

𝒉𝑩𝑶𝑮 [kJ/kg] 555 555 555 555 555 555

𝑷𝑹𝒆𝒇 [kW] 208.2 216.7 230.2 208.2 216.7 230.2

𝑷𝑹𝒆𝒇𝒎𝒂𝒓𝒈𝒊𝒏 [kW] 218.6 227.5 241.7 218.6 227.5 241.7

�̇�𝑪𝒐𝒎𝒑 [kW] 273.2 284.4 302.2 208.7 217.2 230.8

�̇�𝑵𝟐 [kg/s] 4.372 4.550 4.835 3.036 3.160 3.358

- 125 ⁰C

�̇�𝑩𝑶𝑮 [kg/s] 0.362 0.377 0.400 0.362 0.377 0.400

𝒉𝑩𝑶𝑮 [kJ/kg] 590 590 590 590 590 590

𝑷𝑹𝒆𝒇 [kW] 220.9 229.9 244.2 220.9 229.9 244.2

𝑷𝑹𝒆𝒇𝒎𝒂𝒓𝒈𝒊𝒏 [kW] 231.9 241.4 256.5 231.9 241.4 256.5

�̇�𝑪𝒐𝒎𝒑 [kW] 289.9 301.7 320.6 221.4 230.5 244.9

�̇�𝑵𝟐 [kg/s] 4.638 4.827 5.129 3.221 3.352 3.562

- 110 ⁰C

�̇�𝑩𝑶𝑮 [kg/s] 0.362 0.377 0.400 0.362 0.377 0.400

𝒉𝑩𝑶𝑮 [kJ/kg] 620 620 620 620 620 620

𝑷𝑹𝒆𝒇 [kW] 231.7 241.2 256.3 231.7 241.2 256.3

𝑷𝑹𝒆𝒇𝒎𝒂𝒓𝒈𝒊𝒏 [kW] 243.3 253.2 269.1 243.3 253.2 269.1

�̇�𝑪𝒐𝒎𝒑 [kW] 304.1 316.5 336.3 232.3 241.8 256.9

�̇�𝑵𝟐 [kg/s] 4.866 5.065 5.381 3.379 3.517 3.737

As expected, the mass flow rate of nitrogen required to absorb the heat from the BOG increases with its

temperature and with its mass flow rate. Cycle A requires a flow rate of nitrogen approximately 31% bigger than

Cycle B, which requires more compression work to achieve the same liquefaction power. Also, as it is possible to

infer from the linear regression equations presented in Figure 62, the amount of nitrogen required by Cycle A is

proportionally more than the amount required by Cycle B, when the BOG temperature is increased.

75

Figure 62 - Evolution of the Flow Rate of Refrigerant Required Vs BOG inlet Temperature

The nitrogen cycle calculations of each analysed cycle, for the different BOG rates and temperatures are

presented in Annex XIV. The results of performance assessment of each cycle are presented in Table 20.

Table 20 – Liquefaction Cycles Performance Assessment

TBOG inlet Cycle A Cycle B

kWh/ton €/ton kg/kWh kWh/ton €/ton kg/kWh

- 155 ⁰C 198.71 20.31 5.03 151.78 15.51 6.59

- 140 ⁰C 209.64 21.42 4.77 160.14 16.37 6.24

- 125 ⁰C 222.40 22.73 4.50 169.89 17.36 5.89

- 110 ⁰C 233.33 23.85 4.29 178.24 18.22 5.61

Results obtained indicate Cycle B as the most efficient one, presenting a significantly smaller specific energy

consumption to perform the liquefaction of the BOG, of approximately less 24%. The reference cost of electrical

energy was consulted from an energy supplier, for a consumer receiving energy in medium voltage, and its value

is 0.1022€/kWh. Figure 63 illustrates the energy performance and associated cost of the liquefaction cycles.

Figure 63 – Specific Consumption and Cost per Ton of BOG Processed

It is important to refer that, regarding the energy consumption values computed, the circulation pumps,

lubrication pumps, and other inherent auxiliary equipment were not contemplated.

Table 21 and Figure 64 present the values of the COP for the cycles, as well as the real COP (COPactual) values.

The COP value was calculated using equation (12), and it does not consider the work performed by the turbine

as the driver of the second compressor. The KPI COPactual was calculated due to the specific characteristic of this

liquefaction cycle, by having a turbo-expander; it was calculated dividing the absorbed heat from the BOG by the

work performed by compressor 1, assuming that the work performed by the second compressor is recovered by

the work produced by the expander. It is worthwhile to note that the use of a turbo-expander contributes

drastically to the efficiency of the cycle by providing the whole power absorbed by the second stage of

compression. A perfect matching of these two turbomachines is essential to reach high COP values.

y = 0,2691x + 4,3092

y = 0,1869x + 2,9925

y = 0,2432x + 3,8969

y = 0,169x + 2,706

2,00

3,00

4,00

5,00

6,00

-155⁰C -140⁰C -125⁰C -110⁰CNit

roge

n F

low

Rat

e [k

g/s]

Temperature

Evolution of the Flow Rate of Refrigerant Required Vs BOG inlet Temperature

CYCLE A, BOGr=0.067%

CYCLE B, BOGr=0.067%

CYCLE A, BOGr=0.05%

CYCLE B, BOGr=0.05%

CYCLE A, BOGr=0.039%

CYCLE B, BOGr=0.039%

0,0

1,0

2,0

3,0

4,0

5,0

6,0

0,00

50,00

100,00

150,00

200,00

250,00

-155ºC -140ºC -125ºC -110ºC

Co

st p

er t

on

of

BO

G li

qu

efie

d

[€/t

on

]

Ener

gy p

er t

on

of

BO

G li

qu

efie

d

[kW

h/t

on

]

Temperature

Specific Consumption and Cost per Ton of BOG Processed

CYCLE A, kWh/ton

CYCLE B, kWh/ton

CYCLE A, €/ton

CYCLE B, €/ton

76

Table 21 – Coefficients of Performance

Cycle A Cycle B

COP 0,800 1,047

COPactual 1,600 1,646

Figure 64 - Coefficients of Performance - Cycle Comparison

The COP values of Cycle B are higher than those of Cycle A, as expected; the actual COP shows the actual

potential of each cycle, reflecting the work recovered from the expander.

Notice that, with such a significant load variation (between modes of operation of the plant), it is advised

that the turbomachinery components have adequate characteristics to guarantee the system can process

different BOG flow rates with optimized efficiency.

The capacity control of the re-liquefaction system may be achieved using a pressostat, by the increase of

the pressure of the tank, which starts the re-liquefaction system, and through an LNG valve; when this valve is

opened, cold LNG enters the tank, lowering the temperature inside, and, therefore, its pressure - see Annex XV.

Moreover, it is necessary that the temperature of the tank and of the re-liquefied BOG is monitored

carefully, to avoid the vapour in the storage tank to become sub-cooled; overcooling may result in the occurrence

of vacuum inside the tank, potentially damaging the tank structure [91].

5.2.2 Cogeneration

Taking the LHV of BOG as 39.05MJ/Nm3 (10.85kWh/Nm3), and after converting the produced BOG rate to

Nm3, it was possible to calculate for each scenario, the power of a potential cogeneration installation in the

Terminal. Assuming the thermal efficiency of the engine as 35% and the electrical efficiency as 40%, it was

possible to calculate several characteristics of this possible installation, as summarized in Table 22.

Table 22 – Cogeneration Prospect Scenarios

Scenarios BOG rate = 0.039 %/day BOG rate = 0.050 %/day BOG rate = 0.067 %/day

Best (Winter)

Worst (Summer)

Best (Winter)

Worst (Summer)

Best (Winter)

Worst (Summer)

BOG Rate [m3/h] 114.03 117.12 142.14 147.51 182.62 191.55

BOG Rate [Nm3/h] 121.91 125.21 151.96 157.70 195.24 204.78

Power of the Cogeneration [MW]

1.322 1.358 1.648 1.711 2.118 2.221

Electric Power [kW] 528.95 543.28 659.34 684.25 847.12 888.53

Thermal Power [kW] 462.83 475.37 576.92 598.72 741.23 777.46

Produced Electricity per month [MWh]

380.84 391.16 474.73 492.66 609.93 639.74

Produced Heat per month [MWh]

333.24 342.26 415.38 431.07 533.69 559.77

PES [MWh] 187.293 192.366 233.462 242.281 299.952 314.614

Avoided CO2 Emissions per month [Ton CO2 eq.]

37.83 38.85 47.15 48.93 60.58 63.54

Avoided CO2 Emissions [gCO2/kWh]

99.32

The different BOG scenarios enable the operation of a small/medium capacity cogeneration (1.3MW –

2.2MW), allowing the production of electricity on site, for self-consumption or to sell to the electric network,

while producing heat that can be sold to the adjacent industries or also be used internally.

Cogeneration is a method that allows energy conversion with reduced environmental impacts when

compared with other forms of producing energy through fossil fuels. The avoided CO2 emissions were calculated

in order to perceive the positive impact of such an installation, and a monthly estimated reduction of 38 to 64

tons of CO2 is expected. The avoided CO2 emissions per kWh are constant in all analysed scenarios as its

0,0000,5001,0001,5002,000

COP COP real

COP vs COPactual

Cycle A

Cycle B

77

calculation is based on the thermal and electric efficiencies, assumed equal in all cases.

It is also important to refer that natural gas should be compressed before the engine inlet (to around 5-

7bar, according to manufacturers) to ensure the correct fuel injection, and also the auxiliary systems such as

pumps and an eventual chiller for the production of chilled water, require additional energy consumption.

5.2.3 Emergency System

As described in section 4.4.2.3, despite being necessary, the flare is to be the last resource to manage BOG.

Table 23 shows the profit loss associated with burning the BOG produced in each considered scenario during a

certain period, as well as if a rollover phenomenon happens.

Table 23 – Flare Profit Loss

BOG Rate Scenario Profit Loss

€/h €/day €/month

0.039 %/day Best (Winter) 26.46 635.10 19 052.88

Worst (Summer) 27.18 652.30 19 568.90

0.050 %/day Best (Winter) 32.99 791.65 23 749.58

Worst (Summer) 34.23 821.56 24 646.65

0.067 %/day Best (Winter) 42.38 1 017.11 30 513.41

Worst (Summer) 44.45 1 066.83 32 004.97

€/h €/ 2h €/ 3h

Rollover Phenomenon

10 x BOG rate 264 – 342 529 – 684 793 – 1026

30 x BOG rate 793 – 1026 1587 – 2 053 2381 - 3080

The price of the LNG used was 7$/MMBTU [135], which is equivalent to 0.02€/kWh. Assuming a LHV of

10.85kWh/Nm3, the price of 1 Nm3 of BOG is 0.2171€. The presented ranges of value for the rollover

phenomenon are the best and worst case under study. For this event three scenarios were considered according

to the study on the behaviour of existing LNG installations during rollover performed by GIIGNL [136]: one, two

and three hours of excessive BOG, in the proportion of 10 to 30 times the expected value under normal

operation.

With the relevant differences found between the two analysed BOG rates, the insulation costs, and the

holding mode capacity of each scenario, it is important that the organization assesses more deeply and with

more detailed information in project phase, the cost vs benefit of choosing one BOG rate option or another.

Either way, the BOG management should be adequate to the chosen option and be as efficient as possible.

Simulations were performed to study a re-liquefaction system and a cogeneration plant burning BOG as fuel.

Regarding the re-liquefaction options simulated, Cycle B presents a COP value that is 24% higher than that

of Cycle A, while the real COP value is only 3% higher, accounting for the work produced by the turbine to drive

the second compressor. Also, the specific energy consumption of Cycle B is 24% smaller than that of Cycle A,

which represents a significant amount of energy saved while performing the same liquefaction work; these values

indicate Cycle B as the most energy efficient cycle, thus the most adequate to perform the BOG re-liquefaction.

With good results and without a great impact on the LNG storage in holding mode, the cogeneration plant

might also be a good option to recover BOG as, despite the inherent inefficiencies of any internal combustion

engine, the thermal and electrical energy produced can still be used for the benefit of the terminal, while avoiding

CO2 emissions – approximately 38 to 64 CO2 tons per month. The cogeneration can be a good method to process

the holding mode BOG, while having as backup (and capable of processing that and the BOG produced on loading

and unloading events) a re-liquefaction system, based on the most energy efficient cycle simulated - Cycle B.

78

6. Concluding Remarks

6.1 Conclusions

With the growth of primary energy demand, LNG is expected to represent an increasingly larger role as a

fuel - projections indicate that the LNG trade will have an annual growth of 3.9% until 2035. This progressively

bigger demand has boosted the development of LNG supply chains, and Portugal, due to its significant maritime

traffic, rises as a promising future for the development of a new SSLNG supply terminal at the entrance of the

port of Lisbon (Trafaria).

The several processes that comprise a regular regasification terminal (jetty, storage, regasification and boil-

off gas handling) are deeply linked, and the planning of such facilities and processes involving a profound

knowledge of the characteristics that each component can assume, as these are restrained by each other and by

the intrinsic characteristics of the terminal. Knowing the characteristics of the terminal of Trafaria and

considering regulatory political restrictions it is not possible to solve the BOG issue sending it to the pipelines of

the natural gas network; therefore two options are left as valid: either the BOG is used as fuel on an adjacent

facility or it is re-liquefied and stored back to the tank in its liquid state, awaiting to be supplied to the customers.

BOG is quite dependent on the tank location. One advantage found in operating an in-ground tank is that,

while the heat ingress is performed mainly through the dome and the remaining areas of the tank have a very

constant and predictable heat ingress, this type of tank construction minimizes the heat ingress into the tank,

resulting in small variations of the BOG production during the day. This type of tank also presents structural

advantages, especially important in the Trafaria terminal, which is located on a hillside in a steep slope. For a

tank with the same insulation characteristics, in spite of having a higher initial investment due to the need of

building deeper foundations, an in-ground tank has the advantage of having a reduced heat ingress, as the effect

of convection and irradiation are minimized; this will also result in BOG management solutions with smaller

power and consequent energy consumption, as the BOG production is smaller, paying-off throughout the project

life.

After determining the production rate of BOG, the insulation of the in-ground tank was conceived in order

to assure the heat ingress would not overcome the defined rate. The used model allowed to define the daily BOG

rate as 0.067%, which was then compared with the value 0.050% found in the literature, and a lower value for

comparison – 0.039%. The heat ingress in the LNG system was calculated for different operation modes, and for

different seasons of the year, namely the ingress through the tank structure, and through pipelines and pumping

system during loading and unloading processes.

As for the loading and unloading processes, these produce significantly more BOG, despite the fact that

these actions do not happen very often (depending on the frequency of the terminal bunkering processes),

contributing to the BOG production through the heat ingress into pipelines and pumping system. Independently

the BOG management method chosen, the method is to be prepared to process the total amount of BOG

79

production in the LNG system, from the tank, piping, and pumping system.

The addressed methods for BOG management include a system of BOG re-liquefaction and a cogeneration.

The BOG re-liquefaction system studied and applied to the present SSLNG case-study is a nitrogen single

expander cycle, as this technology was found the most adequate to the type and size of the terminal. Both

options were studied for the different seasons and the two BOG rates considered.

The re-liquefaction cycle was studied with two different cycle configurations, namely two different inlet

pressures at the first compression stage. The specific energy consumption and COP were calculated for both

cycles, pointing out Cycle B, with lower inlet pressure at the first compression stage, as the most efficient cycle.

Although the present study indicates some small-scale re-liquefaction plant issues, namely, specific needs

in terms of BOG and re-liquefaction control system, the re-liquefaction system can be further improved, namely

by adding a pre-cooling cycle to the BOG stream or studying other configurations of the cycle such as the dual

expander cycle, although more expensive in terms of investment.

Cogeneration has also been proved an alternative, since the BOG produced is sufficient to power a small

cogeneration (between 1.3 MW - scenario of lower BOG production - and 2.2 MW - scenario of higher BOG

production). In addition, cogeneration enables primary energy savings to be made while producing electricity, as

well as avoiding GHG emissions (CO2). It shall be a decision of the operator of the terminal to decide whether the

electric and thermal energy produced can be consumed in the terminal or exported to any facility nearby the

terminal.

One possibility of using the heat produced by a cogeneration in the terminal is replacing the electric

resistances tracing in the bottom of the tank by a piping system on which heated fluid, capable of preventing the

frost heave, circulates; nonetheless, the feasibility of such opportunity requires further study.

It is also important to understand whether the investment in a cogeneration plant shall be of interest to the

organization or if, on the contrary, the installation of a sole BOG re-liquefaction system is preferable; by choosing

the re-liquefaction system, it is possible to maintain a largest amount of LNG available, considering the potential

long periods of operation in holding mode, thus guaranteeing the availability of the maximum LNG load possible

at all times. However, and knowing that the tank can endure long periods of time without losing a very significant

volume of LNG, cogeneration might be an interesting solution for the terminal BOG management.

Still, it is important that, even if a cogeneration plant is installed in the terminal, the re-liquefaction system

is still present as backup, and can process the BOG from the tank and from the loading and unloading lines. This

should reassure that even if the cogeneration plant fails or stops for maintenance, there is an efficient and

reliable way of managing the BOG. In any case, the use of the flare should be avoided, as it is only an emergency

system as it does represent a large profit loss to the terminal owner, as described in this work.

80

6.2 Future Work

The present thesis was performed as a preliminary work for a possible SSLNG terminal project at Trafaria

(Lisbon port). It serves to identify the different available technologies that constitute one such SSLNG terminal.

This study identified scenarios for the chosen location and capacities regarding the energy consumption related

to the energy-consuming processes that an SSLNG receiving terminal entails, addressing the issue of the BOG

management.

This work may be used as a starting ground of the final SSLNG project at Trafaria, as it points out specific

requirements of the BOG management system. Whether the project will move forward or not, this work is

intended to help the project engineers with guidelines and a dedicated study of the conditions of the terminal.

It is also intended to deliver the OZ Energia company information concerning what to expect from this SSLNG

terminal in the studied location, in particular regarding the energy behaviour of the whole system.

Regarding the tank, it might be interesting to simulate the tank design as a completely underground tank,

i.e., the dome totally covered by earth, to assess the effect of protecting the tank from the direct effect of

convection and irradiation, studying the respective energy efficiency improvement.

As future work regarding the re-liquefaction cycle, the calculations and methodology applied in this thesis

may be used to study equipment supplier solutions and respective costs, i.e., a realistic project budget. In order

to better understand the operational aspects of this SSLNG terminal, the present simulations should be

complemented by performing detailed dynamic simulations regarding the loading and unloading processes of

the tank, in order to perceive, conceive, and optimize the adequate solutions for the terminal. Necessarily, the

project must follow the natural engineering project spiral.

81

References [1] Direcção Geral de Energia e Geologia, “Energia em Portugal: 2014,” DGEG, Lisboa, 2016.

[2] J. Amador, “Produção e Consumo de Energia em Portugal: Factos Estilizados,” Boletim Económico, Banco de Portugal, p. 72, 2010.

[3] P. M. F. Rocha, Estratégia Nacional para a Energia em Portugal em 2020 - Eixo da Eficiência Energética (MSc thesis), Faculdade de Ciências Sociais e Humanas, 2013.

[4] Resolução do Conselho de Ministros n.º 29/2010, “Estratégia Nacional para a Energia,” Diário da República, 1.ª série — N.º 73, pp. 1289-1296, 15 April 2010.

[5] Oxford Institute for Energy Studies, “The Outlook for Natural Gas Demand in Europe,” OIES Papers, June 2014.

[6] International Energy Agency; OECD, “Energy Policies of IEA Countries - Portugal,” 2016 Review, pp. 15-17, 2016.

[7] S. Fankhausera, C. Gennaiolia and M. Collins, “The political economy of passing climate change legislation: Evidence from a survey,” Global Environmental Change, vol. 35, pp. 52-61, 14 August 2015.

[8] N. Rivers and M. Jaccard, “Combining top-down and bottom-up approaches to energy-economy modeling using discrete choice methods.,” Energy Journal, vol. 26 (1), pp. 83 - 106, 2005.

[9] M. Abdelaal and A. Hegab, “Combustion and emission characteristics of a natural gas-fueled diesel engine with EGR,” Energy Conversion and Management, vol. 64, pp. 301-312, 2012.

[10] BP, “BP Energy Outlook 2035 (Report),” BP, 2015.

[11] J. T. Jensen, “The LNG Revolution,” Energy Journal of the International Association for Energy Economics, vol. 21 (2), pp. 1-45, 2003.

[12] S. Kumar, H.-T. Kwon, K.-H. Choi, W. Lim, J. H. Cho and K. Tak, “LNG: An eco-friendly cryogenic fuel for sustainable development,” Applied Energy, vol. 88, no. 12, pp. 4264-4273, 21 June 2011.

[13] F. Burel, R. Taccani and N. Zuliani, “Improving sustainability of maritime transport through utilization of Liquefied Natural Gas (LNG) for propulsion,” Energy, vol. 57, pp. 412-420, 30 April 2013.

[14] International Maritime Organization, “Third IMO GHG Study 2014: Executive Summary and Final Report,” Micropress Printers, Suffolk, UK, 2015.

[15] S. Brynolf, E. Fridell and K. Andersson, “Environmental assessment of marine fuels: liquefied natural gas, liquefied biogas, methanol and bio-methanol,” Journal of Cleaner Production, vol. 74, pp. 86-95, 15 March 2014.

[16] E. Dodge, “Growth of LNG Fuel in Maritime Shipping,” 20 January 2014. [Online]. Available: http://www.theenergycollective.com/ed-dodge/329406/growth-lng-fuel-maritime-shipping.

[17] DNV, “Marpol 73/78 Annex VI - Regulations for the Prevention of Air Pollution from Ships: Technical and Operational implications (Technical Guide),” Det Norske Veritas, Høvik, Norway, 2009.

[18] Danish Maritime Authority, “North European LNG Infrastructure Project - a Feasibility Study for an LNG Filling Station Infrastructure and test of Recomendations (Report),” DMA, Copenhagen, 2012.

[19] M. Acciaro, “Real option analysis for environmental compliance: LNG and emission control areas,” Transportation Research Part D: Transport and Environment, vol. 28, pp. 41-50, 2013.

[20] Marintek, “Emissions Factors for CH4, NOx, particulates and black carbon for domestic shipping in Norway.,” Marintek Report, 2010.

[21] International Gas Union, “Life Cycle Assessment of LNG,” IGU, WCG Paris 2015, Paris, France, 2015.

[22] S. Rüster and A. Neumann, “Economics of the LNG Value Chain and Corporate Strategies: An Empirical Analysis of the Determinants of Vertical Integration,” in 26th USAEE International Conference, Ann Arbor, Michigan, U.S., 2006.

[23] British Chamber of Commerce, “LNG 50 - A Celebration Of The First Commercial Shipment Of LNG: A Brief Story of LNG,” BCCS, Singapore, 2014.

[24] J. P. Beale, “The Facts About Lng (document: RPT-06903-01),” CH-IV International, Millersville; Houston, 2006.

[25] A. Bernatik, P. Senovsky and M. Pitt, “LNG as a potential alternative fuel e Safety and security of storage facilities,” Journal of Loss Prevention in the Process Industries, vol. 24(1), pp. 19-24, 10 August 2010.

[26] H. S. Tira, J. Herreros, A. Tsolakis and M. L. Wyszynski, “Characteristics of LPG-diesel dual fuelled engine operated with rapeseed methyl ester and gas-to-liquid diesel fuels,” Energy, vol. 47, pp. 620-629, 2012.

[27] Diffen, “Diffen,” 2012. [Online]. Available: https://www.diffen.com/difference/Diesel_vs_Petrol. [Accessed 11 March 2018].

[28] U.S. Department of Energy, “Liquefied Natural Gas Research - Report to Congress,” United States Department of Energy, Washington, DC 20585, 2012.

[29] M. M. Foss, “LNG Safety and Security,” University of Texas at Austin, Sugar Land, Texas, 2006.

[30] Energy Charter Secretariat, “LNG and Natural Gas Quality Standards,” Energy Charter Secretariat, Brussels, Belgium, 2004.

82

[31] Bristol, “The Wobbe Index and Natural Gas Interchangeability (Application Data Document 1660AD-5a),” Emerson, 2007.

[32] World Energy Council, “World Energy Perspective: Cost of Energy Technologies,” World Energy Council, London, 2013.

[33] United States Energy Information Administration, “Annual Energy Outlook 2015 with projections for 2040,” Office of Integrated and International Energy Analysis, U.S. Department of Energy, Washington DC, 2015.

[34] OPEC, “Annual Report 2014,” Organization of the Petroleum Exporting Countries, Austria, 2014.

[35] I. Tamura, T. Tanaka, T. Kagajo, S. Kuwabara, T. Yoshioka, T. Nagata, K. Kurahashi and H. Ishitani, “Life cycle CO2 analysis of LNG and city gas,” Applied Energy, vol. 68, pp. 301-319, 2001.

[36] N. Zhanga and N. Lior, “A novel near-zero CO2 emission thermal cycle with LNG cryogenic exergy utilization,” Energy, vol. 31, p. 1666–1679, 2006.

[37] A. Feller, D. Shunk and T. Callarman, “Value Chains Versus Supply Chains,” BP Trends, March 2006.

[38] C. v. Hirschhausen, A. Neumann, S. Ruester and D. Auerswald, “Advice on the Opportunity to Set up an Action Plan for the Promotion of LNG Chain Investments - Economic, Market, and Financial Point of View (Study for the European Commission, DG-TREN),” Dresden University of Technology, Dresden, 2008.

[39] S. Dorigoni, C. Graziano and F. Pontoni, “Can LNG increase competitiveness in the natural gas market?,” Energy Policy, vol. 38, no. 12, pp. 7653-7664, 5 August 2010.

[40] Đ. Dobrota, B. Lalić and I. Komar, “Problem of Boil - off in LNG Supply Chain,” Transactions On Maritime Science - Regular Papers, vol. 02, pp. 91-100, February 2013.

[41] S. Cornot-Gandolphe, “LNG Cost Reductions And Flexibility In LNG Trade Add To Security Of Gas Supply,” ENERGY PRICES & TAXES, 1st Quarter 2005 - International Energy Agency, pp. xxix - xxxvi, 1st quarter 2005.

[42] S. Kumar, H.-T. Kwon, K.-H. Choi, J. H. Cho, W. Lim and I. Moon, “Current status and future projections o fLNG demand and supplies: A global prospective,” Energy Policy, vol. 39, no. 7, pp. 4097-4104, 13 May 2011.

[43] Redes Energéticas Nacionais, “Technical Data 2016: National Natural Gas System (Technical Report),” Redes Energéticas Nacionais, Lisboa, 2016.

[44] REN, “RNTIAT – Existências e Reservas Operacionais (Period: 01-10-2015 to 30-09-2016) (Report),” RNTIAT, Redes Energéticas Nacionais, Lisboa, 2015.

[45] Marine Traffic, “Marine Traffic - Density Map - 2017,” 2018. [Online]. Available: https://www.marinetraffic.com/en/ais/home/centerx:3.0/centery:-2.6/zoom:2. [Accessed 04 March 2018].

[46] Porto de Lisboa, “Porto de Lisboa / Instalações Portuárias / Terminais de Carga,” 1 January 2016. [Online]. Available: http://www.portodelisboa.pt/portal/page/portal/PORTAL_PORTO_LISBOA/PORTO_LISBOA/INSTALACOES_PORTUARIAS/TERMINAIS_CARGA.

[47] Porto de Lisboa, “Publicação Estatística,” Terminais Portuários e Logística, 2016.

[48] J. Antunes, “O Shipping, o Ambiente e a Poluição Atmosférica da Costa Portuguesa,” Revista de Marinha N980, July/August 2014.

[49] Worley Parsons, “CCS Learning from the LNG Sector: A Report for the Global CCS Institute,” Global Carbon Capture and Storage Institute, Melbourne, 2013.

[50] Google, “Google Maps,” 2017. [Online]. Available: https://www.google.pt/maps/@38.6740096,-9.2287522,675a,35y,180h/data=!3m1!1e3. [Accessed January 2017].

[51] Roche, “LNG receiving Terminal on the Saint-Laurent: Pre-Feasibility of the Jetty Component of the Project - LNG Carrier ships characteristics,” Gaz Metro – Enbridge – Gaz de France, Saint-Laurent, 2004.

[52] Instituto Hidrográfico da Marinha Portuguesa, “Instituto Hidrográfico da Marinha Portuguesa / Simoc,” 2 January 2016. [Online]. Available: http://www.hidrografico.pt/simoc.php.

[53] G. Lloyd, “Rules for Classification and Construction: Ship Technology - Seagoing Ships: 6) Liquefied Gas Carriers,” Germanischer Lloyd SE, Hamburg, 2008.

[54] DGPM, “National Ocean Strategy 2013-2020,” DGPM, Direcção-Geral de Política do Mar - Governo de Portugal, 2013.

[55] Município de Almada, “SIGMA - Sistema de Informação Geográfica do Município de Almada,” 2017. [Online]. Available: http://websig.smasalmada.pt/websicas/framesetup.asp. [Accessed May 2017].

[56] A.Takagi and K.Maruyama, “Advanced construction technologies for LNG in-ground storage tanks,” Tunnelling and Underground Space Technology, vol. 7, pp. 347-353, 1992.

[57] Magalog, “The general design of a LNG-terminal,” European Comission: Executive Agency for Competitiveness and Innovation, 2011.

[58] BSI, “BS EN 1473:2007 - Installation and equipment for liquefied natural gas - Design of onshore installations,” British Standard, 28 February 2007.

83

[59] REN Atlântico, “Ship to Shore Sines Port and LNG Jetty Data (4th Edition) (Technical Report),” REN Atlântico, Terminal de LNG, Sines, Portugal, 2013.

[60] SIGTTO, “Site Selection and Design for LNG Ports and Jetties - With views on Risk Limitation during Port navigation and cargo operations,” Worldwide Marine Technology, Wales, United Kingdom, 2000.

[61] H. Sugiyama, H. Mochizuki and E. Kanao, “Caracteristiques de la Conception et du Fonctionnement d’un Terminal de GNL de Sodeshi,” NKK Corporation; Shizuoka Gas Corporation; Shimizu LNG Corporation.

[62] P. Quist, “LNG Jetties as a standard product (MSc. Thesis),” Ballast Nedam Engineering, Delft University of Technology, Delft, 2001.

[63] C. Andrieu, “Small Scale LNG Import Terminal: Not as Simple as a Reduced One,” Tractebel Engineering, Houston, Texas, USA, 2013.

[64] M. D. Tusiani and G. Shearer, LNG: A Nontechnical Guide, Tulsa, Oklahoma: Pennwell, 2007.

[65] American Petroleum Institute, “Oil-and-Natural-Gas-Overview/Exploration-and-Production/Natural-Gas/LNG-Storage,” 28 December 2015. [Online]. Available: http://www.api.org/Oil-and-Natural-Gas-Overview/Exploration-and-Production/Natural-Gas/LNG-Storage.

[66] B. Long and B. Garner, Guide to Storage Tanks & Equipment, Bury St Edmunds and London UK: Professional Engineering Publishing, 2004.

[67] M. Tanabe and A. Miyake, “Approach enhancing inherent safety application in onshore LNG plant design,” Journal of Loss Prevention in the Process Industries, vol. 25, no. 5, pp. 809-819, 12 April 2012.

[68] A. Ezzarhouni, “Opportunities in Capex Saving For Onshore LNG Storage Tanks,” in 7th Annual LNG Tech Global Summit, Rotterdam, 2012.

[69] Center for Energy Economics, “Offshore LNG Receiving Terminals - Guide to Commercial Frameworks for LNG in North America (Technical Guide),” Center for Energy Economics; Bureau of Economic Geology; The University of Texas, Austin, 2006.

[70] EIA, “EIA / Today in Energy,” 27 April 2015. [Online]. Available: https://www.eia.gov/todayinenergy/detail.cfm?id=20972. [Accessed 2 January 2016].

[71] J. E. Fesmire and S. D. Augustynowicz, “Cryogenic Thermal Insulation Systems - 16th Thermal and Fluids Analysis Workshop,” Orlando, Florida, 2005.

[72] Voestalpine Grobblech GmbH, Nickel Steel Plates for LNG and LPG industry (Technical Catalog), Austria: Voestalpine, 2011.

[73] F. J. Powell and S. L. Matthews, Thermal Insulation: Materials and Systems, vol. ASTM Special Technical Publication 992, Dallas: ASTM, 1984.

[74] Pittsburgh Corning, “SPECIFICATION GUIDELINES FOR FOAMGLAS® INSULATION (Technical Brochure),” Pittsburgh Corning, Pittsburgh, 2004.

[75] Pittsburgh Corning, Foamglas® High Load Bearing Cellular Glass Insulation (Technical Brochure), United States of America: Pittsburgh Corning, 2017.

[76] Perlite Institute, “Why Perlite Works - Genesis in Fire (Informative Brochure),” Perlite Institute, Harrisburg, 2009.

[77] Perlite Institute, “Perlite for Non-Evacuated Cryogenic and Low Temperature Service (Informative Brochure),” Perlite Institute, Harrisburg, 2013.

[78] CTS, Suspended Aluminum LNG Decks - Lightweight Alternative to Welden Nickel Decks (Technical Catalog), Netherlands: CTS Netherlands B.V., 2017.

[79] L. Lue, Chemical Thermodynamics, Leo Lue & Ventus Publishing ApS, 2009.

[80] IELE, “LNG Frequently Asked Questions (Report),” University of Houston Law Center - Institute for Energy, Law & Enterprise, Houston, 2005.

[81] R. Sedlaczek, “Diploma Thesis: Boil-Off In Large-and Small-Scale LNG Chains,” Faculty of Engineering Science and Technology - Department of Petroleum Engineering and Applied Geophysics, Trondheim, 2008.

[82] M. M. F. Hasan, A. M. Zheng and I. A. Karimi, “Minimizing Boil-Off Losses in Liquefied Natural Gas Transportation,” Industrial & Engineering Chemistry Research, vol. 48, no. 21, p. 9571–9580, 28 January 2009.

[83] Y. Li, Z. Li and W. Wang, “Simulating on rollover phenomenon in LNG storage tanks and determination of the rollover threshold,” Journal of Loss Prevention in the Process Industries, vol. 37, pp. 132-142, 25 July 2015.

[84] A. Acton and V. Meerbeke, “Rollover in LNG storage e an industry view.,” in LNG8 Conference, LA, USA, 1972.

[85] British Petrol and International Gas Union, “Guidebook to Gas Interchangeability and Gas Quality,” IGU, 2011.

[86] GL, “Study on Standards and Rules for Bunkering of Gas-Fuelled Ships - Final Report for the European Maritime Safety Agency,” Germanischer Lloyd, Lisbon, Portugal, 2012.

[87] C. Park, K. Song, S. Lee, Y. Lim and C. Han, “Retrofit design of a boil-off gas handling process in liquefied natural gas receiving terminals,” Energy, vol. 44, no. 1, pp. 69-78, 23 February 2012.

84

[88] E. Querol, B. Gonzalez-Regueral, J. García-Torrent and M. J. García-Martínez, “Boil off gas (BOG) management in Spanish liquid natural gas (LNG) terminals,” Applied Energy, vol. 87, no. 11, p. 3384–3392, 24 April 2010.

[89] IPCC, “IPCC Fourth Assessment Report: Climate Change 2007,” International Panel on Climate Change, 2007.

[90] S. Huang, J. Hartono and P. Shah, “BOG recovery from long jetties during LNG ship-loading,” in 15th International Conference & Exhibition on Liquefied Natural Gas, Barcelona, Spain, 2007.

[91] J. Y. Mak, S. Mokhatab, J. V. Valappil and D. A. Wood, Handbook of Liquefied Natural Gas, First ed., Oxford: Elsevier, 2014.

[92] H.-M. Chang, “A thermodynamic review of cryogenic refrigeration cycles for liquefaction of natural gas,” Cryogenics, vol. 72, no. 2, pp. 127-147, 5 October 2015.

[93] NASA, “Figure of Merit Characteristics Compared to Engineering Parameters,” National Aeronautics and Space Administration, Marshall Space Flight Center - MSFC, Alabama, 2010.

[94] Y. A. Çengel and M. A. Boles, Termodinâmica, Third ed., McGraw-Hill, 2001.

[95] A. Trigilio, A. Bouza and S. D. Scipio, “Modelling and Simulation of Natural Gas Liquefaction Process,” in Advances in Natural Gas Technology, Intech, 2012, pp. 213 - 234.

[96] J.-i. Yoon, H.-s. Lee, S.-t. Oh, S.-g. Lee and K.-h. Choi, “Characteristics of Cascade and C3MR Cycle on Natural Gas Liquefaction Process,” International Journal of Chemical and Molecular Engineering, vol. 3, p. No.11, 2009.

[97] Chart Energy & Chemicals, Brazed Aluminum Heat Exchangers, La Crosse, Wisconsin, USA: Chart.

[98] International Gas Union, “IGU World LNG Report,” IGU, Barcelona, Spain, 2017.

[99] T. B. He and Y. Ju, “Performance improvement of nitrogen expansion liquefaction process for small-scale LNG plant,” Cryogenics, vol. 61, pp. 111-119, 2014.

[100] A. J. Finn, G. Johnson and T. Tomlinson, “Developments in natural gas liquefaction,” Hydrocarbon Processing, vol. 78, no. 4, pp. 47-59, March 1999.

[101] T.-V. Nguyen, E. D. Rothuizen, W. B. Markussen and B. Elmegaard, “Thermodynamic comparison of three small-scale gas liquefaction systems,” Applied Thermal Engineering, vol. 128, pp. 712-724, 2018.

[102] H. Sayyaadi and M. Babaelahi, “Thermoeconomic optimization of a cryogenic refrigeration cycle for re-liquefaction of the LNG boil-off gas,” Internation Journal of Refrigeration, vol. 33, no. 6, pp. 1197-1207, 2010.

[103] S. Mokhatab, W. A.Poe and J. G. Speight, Handbook of Natural Gas Transmission and Processing, 1st ed., Oxford: Elsevier, 2006.

[104] B. Austbø and T. Gundersen, “Optimization of a single expander LNG process,” Energy Procedia, vol. 64, pp. 63-72, 2015.

[105] L. Castillo and C. Dorao, “Influence of the plot area in an economical analysis for selecting small scale LNG technologies for remote gas production,” Journal of Natural Gas Science and Engineering, vol. 2, no. 6, pp. 302-309, 2010.

[106] A. R. Jha, Cyogenic Technology and Applications, Oxford, UK: Butterworth-Heinemann (Elsevier), 2006.

[107] J. W. Moon, Y. P. Lee, Y. W. Jin, E. S. Hong and H. M. Chang, “Cryogenic Refrigeration Cycle for Re-Liquefaction of LNG Boil-Off Gas,” in Cryocoolers 14, Boulder, Colorado, 2007.

[108] T. He and Y. Ju, “Optimal synthesis of expansion liquefaction cycle for distributed-scale LNG (liquefied natural gas) plant,” Energy, vol. 88, pp. 268-280, 2015.

[109] Z. Yuan, M. Cui, Y. Xie and C. Li, “Design and analysis of a small-scale natural gas liquefaction process adopting single nitrogen expansion with carbon dioxide pre-cooling,” Applied Thermal Engineering, vol. 64, pp. 139-146, 2014.

[110] D.-H. Kwak, J.-H. Heo, S.-H. Park, S.-J. Seo and J.-K. Kim, “Energy-efficient design and optimization of boil-off gas (BOG) Re-liquefaction process for LNG-fuelled ship,” Energy, vol. 148, pp. 915-919, 2018.

[111] W. L. Luyben, Process Modelling, Simulation and Control for Chemical Engineers, McGraw-Hill International, 1996.

[112] C. W. d. Silva, Modeling and Control of Engineering Systems, CRC Press - Taylor & Francis Group, 2009.

[113] REN Atlântico, “Regras Técnicas de Uso do Terminal de Gás Natural Liquefeito de Sines (Technical Guide),” REN Atlântico, Terminal de GNL, S.A., 2013.

[114] Air Liquide, “Gas Encyclopedia,” Air Liquide, 2018. [Online]. Available: https://encyclopedia.airliquide.com/methane. [Accessed 12 February 2018].

[115] Agência Portuguesa do Ambiente, “SNIRH - Sistema Nacional de Informação de Recursos Hídricos,” 2017. [Online]. Available: http://snirh.apambiente.pt/index.php?idMain=2&idItem=1&objCover=920123704&objSite=920685506. [Accessed 12 Outubro 2017].

[116] Município de Almada, “Plano Municipal de Defessa da Floresta contra Incêndios (Report),” CMA, Almada, 2011.

[117] M. Badache, P. Eslami-Nejad, M. Ouzzane, Z. Aidoun and L. Lamarche, “A new modeling approach for improved ground temperature profile determination,” Renewable Energy, vol. 85, pp. 436-444, 2016.

85

[118] G. Florides and S. Kalogirou, “Annual Ground Temperature Measurements at Various Depths (technical paper),” Higher Technical Institute, Nicosia, Cyprus, 2005.

[119] Instituto Português do Mar e da Atmosfera, “Boletim meteorológico para a agricultura (Periodic Report),” IPMA, 2016/2017.

[120] American Petroleum Institute, “API RECOMMENDED PRACTICE 2350 - Overfill Protection for Storage Tanks in Petroleum Facilities,” American Petroleum Institute, Washington, D.C., 2005.

[121] ABS, “LNG Bunkering: Technical and Operational Advisory,” Houston, TX, USA, 2017.

[122] E. Adom, S. Z. Islam and X. Ji, “Modelling of Boil-Off Gas in LNG Tanks: A Case Study,” International Journal of Engineering and Technology, vol. 2, no. 4, pp. 292-296, 2010.

[123] A. Shitzer, “Wind-chill-equivalent temperatures: regarding the impact due to the variability of the environmental convective heat transfer coefficient,” International Journal of Biometeorology, vol. 50, pp. 224-232, 2006.

[124] P. Tikuisis and R. J. Osczevski, “Dynamic model of facial cooling,” Journal of Applied Meteorology, vol. 12, p. 1241–1246, 2002.

[125] M. J.Moran, H. N. Shapiro, D. D. Boettner and M. B. Bailey, Fundamentals of Engineering Thermodynamics, 7th ed., United States of America: John Wiley & Sons,Inc., 2011.

[126] IPMA, “IPMA,” 2018. [Online]. Available: https://www.ipma.pt/pt/educativa/faq/meteorologia/previsao/faqdetail.html?f=/pt/educativa/faq/meteorologia/previsao/faq_0032.html. [Accessed 27 April 2018].

[127] A. Wordu and B. Peterside, “Estimation of Boil-off-Gas BOG from Refrigerated Vessels in Liquefied Natural Gas Plant,” International Journal of Engineering & Technology, vol. 3, no. 1, pp. 44-49, 2013.

[128] K-D.Gerdsmeyer and W.H.Isalski, “ON-BOARD RELIQUEFACTION FOR LNG SHIPS,” Tractebel Gas Engineering, 2005.

[129] A. Vorkapić, P. Kralj and D. Bernečić, “Ship systems for natural gas liquefaction,” Multidisciplinary Scientific Journal of Maritime Research, vol. 30, pp. 105-112, 2016.

[130] J. R. Gómez, M. R. Gómez, R. F. Garcia and A. D. M. Catoira, “On board LNG reliquefaction technology: a comparative study,” Polish Maritime Reasearch, vol. 21, pp. 77-88, 2014.

[131] Ministério da Economia, da Inovação e do Desenvolvimento, “Decreto-Lei n.º 23/2010,” Diário da República, 1ªSérie, no. 59, pp. 934-946, 25 Março 2010.

[132] “Directiva 2004/8/CE do Parlamento Europeu e do Conselho,” Jornal Oficial da União Europeia, no. Promoção da cogeração com base na procura de calor útil no mercado interno da energia, 11 Fevereiro 2004.

[133] International Panel for Climate Change, “IPCC Guidelines for National Greenhouse Gas Inventories, Vol.2: Stationary Combustion,” IPCC, 2006.

[134] Michael Baker Jr., Inc, “BASIS OF ESTIMATE: Interior Gas Utility, Fairbanks Gas Distribution, Advancement Project - Task 3: LNG Storage Tank Cost Analysis,” Michael Baker Jr., Inc., Anchorage, AK, 2013.

[135] S&P Global, “S&P Global Platts - Natural Gas - LNG,” 2018. [Online]. Available: https://www.platts.com/commodity/natural-gas/natural-gas-lng. [Accessed 20 April 2018].

[136] International Group of Liquefied Natural Gas Importers, “Rollover in LNG Storage Tanks,” GIIGNL, Neuilly-sur-Seine (France), 2015.

[137] Gas Infrastructure Europe, “GIE LNG Map - information by entry point (Informative Map),” GIE, 2018.

[138] REN - Redes Energéticas Nacionais, “Natural Gas National Transport Grid Map 2017 (Informative Map),” REN, 2017.

[139] Toho Gas Co., Ltd. and Mitsubishi Heavy Industries, Ltd., “Construction Of The World’s Largest Lng In-Ground Storage Tank (Project Report),” THG and MHI, Nagoya, Japan; Yokohama, Japan, 2000.

[140] ARUP, “Gas and LNG Storage | The Future of Modular LNG Tanks (Project Report),” Arup, 2017.

[141] H. Lun, F. Filippone, D. C. Roger and M. Poser, “Design and Construction Aspects of Post-Tensioned LNG Storage Tanks in Europe and Australasia,” in Concrete Industry Conference, New Zealand, 2006.

[142] Ishikawajima-Harima Heavy Industries Co., Ltd. and Kajima Corporation, “Design and Erection of Temporary Steel Roof for Under-Ground LNG Storage Tank,” in 13th International Conference and Exhibition on Liquefied Natural Gas, Korea, 2001.

[143] Sangaku Maths, “Sangakoo Maths,” 2017. [Online]. Available: https://www.sangakoo.com/en/unit/the-spherical-dome-surface-area-and-volume. [Accessed 18 October 2017].

[144] M. F. M Fahmy, H. I. Nabih and T. A. El-Rasoul, “Optimization and comparative analysis of LNG regasification processes,” Energy, vol. 91, pp. 371-385, 14 August 2015.

[145] International Gas Union, “IGU world LNG report - 1st Edition,” IGU, 2014.

[146] J. Mak, “Integration of LNG Regasification with Refinery and Power Generation”. Santa Ana, CA (US) Patent US 8316665 B2, 27 November 2012.

86

[147] H. Dong, L. Zhao, S. Zhang, A. Wang and J. Cai, “Using cryogenic exergy of liquefied natural gas for electricity production with the Stirling cycle,” Energy, vol. 63, pp. 10-18, 20 October 2013.

[148] J. Mak, “Configurations and Methods for Offshore LNG Regasification and Heating Value Conditioning”. Patent US 20100126187 A1, 2010.

[149] API Energy, “Liquefied Natural Gas (LNG) Operations: Consistent Methodology for Estimating the Greenhouse Gas Emissions,” American Petroleum Institute, Washington DC, 2015.

[150] B. Eisentrout, S. Wintercorn and B. Weber, “Study focuses on six LNG regassification systems,” LNG Journal, pp. 21-22, July/August 2006.

[151] D. Patel, J. Mak, D. Rivera and J. Angtuaco, “LNG Vaporizer Selection Based On Site Ambient Conditions,” International Gas Union (IGU); Gas Technology Institute (GTI); and International Institute of Refrigeration (IIR), 2013.

[152] T. Dendy and R. Nanda, “Utilization of Atmospheric Heat Exchangers in LNG Vaporization Processes: A comparison of systems and methods (Technical Report),” American Institute of Chemical Engineers (AIChE), USA, 2008.

[153] CH·IV, “Oregon LNG Import Terminal: Vaporizer Alternatives Study,” CH·IV International, Baltimore; Houston, 2007.

[154] R. Agarwal, “LNG Regasification — Technology Evaluation And Cold Energy Utilisation (Technical Report),” Queensland University of Technology, Australia, 2013.

[155] Y. W. Chin, “Cycle analysis on LNG Boil-off Gas Reliquefaction Plant,” Journal of the Korea Institute of Applied Superconductivity and Cryogenics, vol. 8, pp. 34-38, 2006.

[156] V. S. Bisht, “Thermodynamic Analysis of Kapitza Cycle based on Nitrogen Liquefaction,” IOSR Journal of Engineering (IOSRJEN), vol. 4, no. 5, pp. 38-44, May 2014.

[157] IHI, “LNG Storage Solutions: A Key Consideration and Element in LNG Terminal Operation,” IHI, 2015.

[158] Chevron Energy Technology Company, “COMPOSITE CONCRETE CRYOGENIC TANK (C3T): A PRECAST CONCRETE ALTERNATIVE FOR LNG STORAGE (Technical Report),” CETC, 2016.

[159] ASHRAE, ASHRAE Handbook—Fundamentals (SI), ASHRAE, 2009.

[160] Intergovernment Panel on Climate Change, “Fourth Assessment Report: Climate Change,” WMO; UNEP, 2007.

[161] REN, “REN / LNG Terminal,” 2 January 2016. [Online]. Available: https://www.ign.ren.pt/en/terminal-de-gnl3.

[162] SIGTTO, “LNG Transfer Arms and Manifold Draining, Purging and Disconnection Procedure,” Witherby Seamanship International, London , 2012.

[163] British Standards, “BS EN 1473:2007 - Installation and Equipment for Liquefied Natural Gas - Design of Onshore Installations,” Standards Policy and Strategy Committee, 2007.

[164] Ê. C. d. Costa, Refrigeração, São Paulo, Brasil: Edgard Blücher, 1986.

[165] T. M. Flynn, Cryogenic Engineering: Second Edition Revised and Expanded, New York: Marcel Dekker, 2005.

[166] A. Benito, “Accurate determination of LNG quality unloaded in Receiving Terminals: An Innovative Approach,” GERG Academic Network Event, Brussels, Belgium , 2009.

[167] Orçamentos 2009-2017, “Orçamentos e Orçamentação na Construção Civil,” 2014. [Online]. Available: http://orcamentos.eu/precos-de-betao-pronto/. [Accessed 21 September 2017].

[168] Alibaba Group, “Alibaba,” 2017. [Online]. Available: https://www.alibaba.com/product-detail/Constructions-Sand_117792245.html?spm=a2700.7724838.2017115.104.7d299801xNSsdC. [Accessed 21 September 2017].

[169] Alibaba Group, 2017. [Online]. Available: https://www.alibaba.com/product-detail/Nickel-Steel-Plates-9-Inox-Stainless_60545095394.html?spm=a2700.7724857.main07.21.4e9e903cCp6LP&s=p. [Accessed 21 09 2017].

[170] Alibaba Group, “Alibaba,” 2017. [Online]. Available: https://www.alibaba.com/product-detail/Bitumen-Asphalt-Roofing-Felt-For-Waterproof_60617179669.html?spm=a2700.7724838.2017115.1.23861b7cKyvS5A. [Accessed 21 September 2017].

[171] Alibaba Group, “Alibaba,” 2017. [Online]. Available: https://www.alibaba.com/product-detail/Foam-Glass-Board-for-Oil-Platform_60389703830.html?spm=a2700.7724838.2017115.1.156e29e9WyyJzI. [Accessed 21 September 2017].

[172] U.S. Geological Survey, “Mineral Commodity Summaries,” U.S.A., 2016.

[173] Alibaba Group, “Alibaba,” 2017. [Online]. Available: https://www.alibaba.com/product-detail/5383-5454-5456-aluminum-alloy-plain_1657854430.html?spm=a2700.7724838.2017115.65.23861b7cKyvS5A. [Accessed 21 September 2017].

[174] A. Shitzer, “Wind-chill-equivalent temperatures: regarding the impact due to the variability of the environmental convective heat transfer coefficient,” International Journal of Biometeorology, vol. 50, p. 224–232, 2005.

[175] J. M. Pfotenhauer, Refrigeration & Liquefaction (Technical Report), Madison: University of Wisconsin, 2010.

[176] G. Chicco and P. Mancarella, “Distributed multi-generation: a comprehensive view,” Renewable and Sustainable Energy Reviews, vol. 13, p. 535–551, 2007.

87

[177] A. Arteconi, D. Brandoni, D. Evangelista and F. Polonara, “Life cycle green house gas analysis of LNG as a heavy vehicle fuel in Europe,” Applied Energy , vol. 87, p. 2005–2013, 13 November 2009.

[178] X. Wang and M. J. Economides, “Purposefully built underground natural gas storage,” Journal of Natural Gas Science and Engineering, vol. 9, pp. 130 - 137, 11 June 2012.

[179] U. Lee, K. Kim and C. Han, “Design and optimization of multi-component organic Rankine cycle using liquefied natural gas cryogenic exergy,” Energy, vol. 77, p. 520–532, 12 September 2014.

[180] D. W. Plachta and M. C. Guzik, “Cryogenic Boil-Off Reduction System,” Cryogenics, vol. 60, pp. 62-67, 16 December 2013.

[181] Y. Li, X. Chena and M.-H. Cheinb, “Flexible and cost-effective optimization of BOG (boil-off gas) recondensation process at LNG receiving terminals,” Chemical Engineering Research and Design, vol. 90, no. 10, pp. 1500-1505, 25 January 2012.

88

ANNEXES

89

I. Gas Infrastructure: Europe’s LNG Map of 2018

Figure 65 – Europe’s LNG Map of 2018 [137]

90

II. Gas Infrastructure: Portugal

Figure 66 – Portuguese Gas Infrastructure [138]

91

III. Trafaria’s Terminal – Plant and General View

Figure 67 – Europe’s LNG Map of 2018

92

IV. Tank Projects and Examples – Determining the Dome Height and Characteristics

The design and methodology for the construction of LNG tanks varies largely with the techniques used by

each constructor, and are also disclosure-sensitive due to the industrial secrecy associated. Also, it is difficult to

find standard design procedures specifically for the design of LNG in-ground tanks, as this type of tanks is the

least common to build. As a method to overcome this difficulties, the available tank projects and drawings were

analysed in order to reach a conclusion especially regarding the ratios used for the design of the height of the

dome and radius of the spherical cap.

Below, it is possible to find the project drawings consulted and their dimensions:

Figure 68 – Tank Project 1 - In-ground Tank, 200 000 m3 [139]

Figure 69 – Tank Project 2 – Above ground Tank, 15 000 m3

[140]

Figure 70 – Tank Project 3 – Above ground Tank, 40 000 m3 [140]

93

Figure 71 – Tank Project 4 – Above ground Tank, 40 000 m3 [141]

Figure 72 – Tank Project 5 - In-ground Tank, 200 000 m3 [142]

Table 24 – Synthesis of the Characteristics of the Examples found in the Literature and Industrial Projects.

Drawing Tank Type

Capacity Total

Height Dome Height

Tank Internal

Diameter

Tank External

Diameter

Dome Radius

Dome Height /

Total Height

Dome Height /

Diameter

Maximum Pressure

m3 m m m m m % % bar

1 In-ground 200 000 60.5 13.9 74 75.6 60.5 0.230 0.188 0.24

2 Above

Ground 15 000 27.9 3.15 30 - - 0.113 0.105 -

3 Above

Ground 40 000 39.38 5.4 37.5 40 40 0.137 0.144 0.19

4 Above

Ground 196 107 49.56 10.74 80.2 81.8 80.2 0.217 0.134 -

5 In-ground 200 000 63.6 14.4 72 - 89.137 0.226 0.200 0.19

94

Table 25 – Conclusions Regarding the Dome Height for different types of tanks, with two different approaches.

CONCLUSIONS Dome height corresponds to __ %

of the total tank height Dome height corresponds to __ %

of the tank diameter

For above ground tanks projects 15.56 % 12.76 %

For in-ground tanks projects 22.81 % 19.39 %

For all the tanks projects 18.46 % 15.42 %

As the in-ground tanks present different values of Dome Height/Total Height and Dome Height/Tank

Diameter than above-ground tanks, in order to define the dome height for the present tank project, only the in-

ground tanks dimensions were used to calculate the final dimensions.

Knowing that (equation (65)):

ℎ𝑇𝑜𝑡𝑎𝑙 = ℎ𝐷𝑜𝑚𝑒 + ℎ𝐶𝑖𝑙𝑖𝑛𝑑𝑒𝑟 , (65)

then, the height of the dome is given by the following equation (66):

ℎ𝐷𝑜𝑚𝑒 = ___ % × (ℎ𝐷𝑜𝑚𝑒 + ℎ𝐶𝑖𝑙𝑖𝑛𝑑𝑒𝑟) (66)

Taking ℎ𝐶𝑖𝑙𝑖𝑛𝑑𝑒𝑟 = 29.84 𝑚,

Then the dome height will assume the following values:

Calculating the approximation by Total Height: ℎ𝐷𝑜𝑚𝑒 = 8.82 𝑚

Calculating the approximation by Diameter: ℎ𝐷𝑜𝑚𝑒 = 7.76 𝑚

Using the Diameter approximation, and knowing that the projects that were studied to perform this

approximation certainly accounted for the structural resistance needed to stand the dome and to stand

mechanical stresses the structure might suffer, this method will reassure that these factors will also be accounted

for on the actual tank, as the structural simulation will not be performed in this study.

Otherwise, if the Total Height approximation was to be chosen, the dome height would not be influenced

by the diameter, which could mean that regardless of the diameter and for a fixed Total height, the height of the

dome would be the same, not accounting for the extra stress on the structure that a higher diameter causes.

After this analysis, the method chosen is the approximation by Diameter.

Assuming finally, then: ℎ𝐷𝑜𝑚𝑒 = 7.76 𝑚.

Knowing the height of the dome, it was then possible to specify its area, the radius of the circumference in

which the dome arc is inscribed, and the volume of the dome.

Equation (67) is used to calculate the radius of the circumference in which the arc of the dome is inscribed

(𝑅𝐷𝑜𝑚𝑒) [143]:

95

𝑅𝐷𝑜𝑚𝑒 =𝑅𝑇𝑎𝑛𝑘

2 + ℎ𝐷𝑜𝑚𝑒2

2ℎ𝐷𝑜𝑚𝑒

(67)

where 𝑅𝑡𝑎𝑛𝑘 is the radius of the tank, and ℎ𝐷𝑜𝑚𝑒 is the height of the dome.

Figure X illustrates the spherical cap (dome) and the considered variables to calculate the several parameters of

the dome of the tank.

Figure 73 - Spherical Cap Illustration [143]

Equations (68) and (69) allow to calculate the area (𝐴𝐷𝑜𝑚𝑒) and volume (𝑉𝐷𝑜𝑚𝑒) of the dome [143]:

𝐴𝐷𝑜𝑚𝑒 = 2 𝜋 𝑅𝑑𝑜𝑚𝑒 ℎ𝑑𝑜𝑚𝑒 (68)

𝑉𝐷𝑜𝑚𝑒 =𝜋. ℎ𝑑𝑜𝑚𝑒

2

3(3𝑅𝑑𝑜𝑚𝑒 − ℎ𝑑𝑜𝑚𝑒) (69)

Then, the dome of the tank assumes the following characteristics:

Table 26 - Dome Characteristics

Dome

Heigth Diameter Radius (circle)

Volume Area

m m m m3 m2

7.76 40 29.66 28.35 1 445.66

96

V. Radiation and Temperature Data Collected

Table 27 – Meteorological Data Collected – Monte de Caparica Station

Spring - April 2016 Summer - July 2016 Autumn - October 2016 Winter - January 2017

Mean

Radiation Mean

Temperature Mean

Radiation Mean

Temperature Mean

Radiation Mean

Temperature Mean

Radiation Mean

Temperature

W/m2 °C W/m2 °C W/m2 °C W/m2 °C

00:00 0 12,74 0 20,5 0 16,7 0 10,15

01:00 0 12,59 0 20,0 0 16,3 0 9,81

02:00 0 12,40 0 19,5 0 16,1 0 9,51

03:00 0 12,07 0 19,2 0 15,7 0 9,44

04:00 0 11,87 0 18,9 0 15,6 0 9,29

05:00 0 11,81 0 18,5 0 15,5 0 9,14

06:00 22,2 11,62 26,5 18,4 0 15,2 0 8,89

07:00 48,3 12,04 99,5 19,5 0 15,1 0 8,68

08:00 161,6 13,09 262,6 21,4 40,2 15,9 0 8,51

09:00 296,4 14,12 448,5 22,9 143,1 17,2 11,48 8,53

10:00 441,0 15,01 597,0 24,2 284,4 18,6 94,16 9,54

11:00 535,2 15,67 739,3 25,6 383,2 19,7 214,94 10,47

12:00 636,5 16,32 801,3 26,4 444,3 20,6 287,03 11,3

13:00 664,5 16,83 840,5 27,0 461,4 21,2 312,87 12,01

14:00 643,9 17,12 830,3 27,3 427,0 21,6 293,42 12,79

15:00 580,2 17,35 753,5 27,3 345,0 21,7 236,35 13,16

16:00 475,4 17,03 643,0 27,2 238,2 21,4 161,77 13,39

17:00 329,1 16,39 500,8 26,8 102,0 20,6 62,94 13,07

18:00 168,6 15,61 321,1 26,1 20,5 19,4 0 12,09

19:00 50,6 14,61 148,5 24,9 0 18,4 0 11,27

20:00 22,7 13,73 35,0 23,1 0 18,0 0 10,77

21:00 0 13,33 0 21,9 0 17,6 0 10,62

22:00 0 13,23 0 21,5 0 17,4 0 10,52

23:00 0 13,12 0 21,2 0 17,1 0 10,48

97

VI. LNG Regasification

Despite the fact that the total costs of LNG technology have decreased substantially on the past twenty

years due to the improvement of the liquefaction process, the regasification technologies and systems have not

suffered any considerable improvements [144]. As so, this process requires the existence of vaporizers, the

devices that power regasification at LNG receiving terminals. The optimum choice of an LNG vaporisation system

is established by the terminal location, its environmental conditions, regulatory limitations and operability

considerations [145]. Planning of operational costs on an LNG terminal to be constructed should include and

account for the energy costs associated with regasifying LNG, as these are significant on the overall estimated

budget. One of the methods used to viable the regasification process and make it monetarily and technically

more efficient might pass through the use of the natural gas on other industrial processes, as a refrigerating fluid

[146] on power plants or co-generations next to the LNG facility, as LNG contains a considerable cold energy to

recover through electricity generation [147], allowing to perform this procedure in a symbiotic, integrated and

more efficient approach. Also the recovery using an ORC (organic Rankine cycle) system combined with electricity

generation plants has been referred as an opportunity to recover energy from this process by Lee, Kim and Han

(2014) on their research.

The LPG components existent on the natural gas may not be removed during the liquefaction process of

LNG which can result in a high heating value of the mix (high Wobbe index – consult section 1.2.2), such as

ethane, propane and butane (C2+) [144]. As the heating value is controlled onshore on each terminal by diluting

the LNG with nitrogen in order to attain the desired characteristics of natural gas supplied to the network [148].

Therefore, regasification processes that can seize the economic advantage of recovering these other valuable

gases are much needed and there is a need to enhance this processes and methods. It is then possible to conclude

that the optimal LNG regasification method is the one that assures the maximum amount of C2+ (ethane-plus

extraction) removed from LNG (if conceivable) and at the same meet the pipeline high heating value

specifications, if possible granting energy recovering.

Aware of the ideal regasification process characteristics and knowing that the extraction of ethane-plus is

an option applied in very few locations [149], the most common methods used for LNG regasification nowadays

comprise the Submerged Combustion Vaporizers (SCV), Open Rack Vaporizers (ORV), Ambient Air Vaporizer

(AAV) and Shell and Tube Vaporizers (STV). Among these it is also possible to find, in certain cases, LNG

regasification made through benefiting from waste heat from a co-located power plant or industry [150].

Submerged Combustion Vaporizers (SCV)

This type of vaporizer (see Figure 74) is designed to use the low pressure fuel gas from the BOG recovery

98

system. While the products of combustion pass

into a water bath to recovery their heat energy

that is contained in the flue gases, the LNG

passes through stainless steel tubes that are

submerged on the high temperature water,

which acts as the intermediate fluid that

transfers energy from the combustion process

to the LNG, vaporizing it [151]. The energy

consumption in this process happens as electric

energy is necessary for the combustion air blower and both the pumps for water and LNG. The exhaust gases

delivered the blower are later discharged to the atmosphere through an exhaust stack.

Open Rack Vaporizers (ORV)

Seawater is used on this type of

vaporizer (see Figure 75) as the

unique source of heat, being

supplied by seawater intake pumps

and to an overhead distribution

header, flowing over long thin

finned tube panels where LNG

circulates, and working as counter-

flow heat exchanger. As LNG circulates inside the panels, it is vaporised by the warmer seawater, which is cooled

during the process and then returned to an outfall.

Shell and Tube Vaporizers (STV)

This technology (see Figure 76) uses a

low temperature heat transfer fluid which can

be heated by seawater, ambient air or other

heat source as a fired heater, resulting in

different configurations as direct or indirect

heating is employed. In case of using

seawater, this again is cooled during the

process and delivered to an outfall. One

possible configuration of this equipment is to

have a closed loop heated water-glycol system to provide heat to vaporize the LNG through heat exchanger tubes

and a superheater. The STV technology with direct heat exchanger offers a good choice for space-constrained

sites, as this is the most compact system between the presented ones, but has the disadvantage of representing

the highest initial investment for an LNG vaporizer [152].

Figure 74 - Submerged Combustion Vaporizer (SCV)

Figure 75 - Open Rack Vaporizer (ORV)

Figure 76 - Shell and Tube Vaporiser (STV)

99

Ambient Air Vaporizer (AAV)

These vaporizers (see Figure 77)

consist, just as on the ORV technology, on a

heat exchanger that consists on a long,

finned and direct contact tubes that

aerodynamically facilitate the air draft

downwards. Water condensation and

melting ice can also be collected as it

represents a source of potable/service

water. In cases where the process in fan

assisted, this system is called FAV - Fan

Assisted (Ambient Air) Vaporization.

The AAV can also be used as a way to recover energy from a gas engine/turbine exhaust or some other fired

heater source, these sources provide a way to ensure natural is warmed at the required temperature on the

pipeline distribution network [153].

During planning, when the choice of the vaporizers is performed, it is important to have in mind that a key

parameter for the vaporization technology selection is the flexibility of the operation that is required and while

the efficiency of the equipment is expected to increase its performance at nominal rates than at lower ones,

once again being important not to oversize the vaporization equipment. Knowing this, the typical capacity per

unit of the ORV and ACV technologies do not suit the SSLNG terminals, as the nominal flow assume values that

are too low to obtain a worthy efficiency from the utilization of these kinds of equipment. Being so, the most

suited and commonly used equipment for a SSLNG terminal are the AAV units, which are passive type and have

the capacity of better matching the demands from this kind of terminals, reducing its energy costs near to zero,

as a free heat source is used [144].

Since direct and indirect heat transfer processes occur while the LNG regasification is performed, a

dissipation of cold energy takes place, wasting energy. To recover this energy several possibilities arise as viable,

such as air conditioning, cold storage warehousing, district cooling, production of dry ice and it may even be used

as a contribution to re-liquefy the boil-off gas [154].

On the present study case, as seen on section 2.2.1, regasification will not take part on the terminal due to

the regulatory impossibility of commercializing natural gas in gaseous state. Although it represents less initial

investment (not necessary to acquire re-liquefaction equipment) for the terminal operator, this will result in the

need of a more efficient boil-off recovery system, as it cannot be recovered through the regasification process.

Being so, despite the fact that regasification is a core process on a standard LNG receiving terminal, it will not be

part of Trafaria Terminal, due to its impracticability.

Figure 77 - Ambient Air Vaporizer (AAV)

100

VII. Other Liquefaction Cycles

Not particularly relevant for this work, but still part of the research made of liquefaction cycles for natural

gas, some other liquefaction cycles are presented on this section, for reference.

i) Linde Cycle

The Linde cycle (see Figure 78) is capable of

expanding the compressed gas without producing

work. It adopts a multi-stage compression (usually

three), followed by cooling and expansion of the gas at

low temperature through a valve. One of the

possibilities to increase the efficiency of the cycle is by

means of a frigorific machine of ammonia (NH3).

This system works with the inversion temperature

above the compression temperature, using a

recuperative heat exchanger to pre-cool the high

pressure stream. This liquefier requires a source of

make-up gas, and the refrigerator absorbs heat, converting liquid to vapour at saturation temperature of low

pressure.

ii) Claude Cycle

The Claude cycle adopts de expansion of the compressed and cooled

gas with production of a mechanical work (mechanical expander). This

expansion can be performed on mechanical expanders such as a turbine

coupled directly on the turbo compressor (about 55% of the compressed

gas) and an expansion valve (about 45% of the expansion valve).

The JT (Joule-Thomson) expansion presented on both Figure 78 and

Figure 79 correspond to a throttling process through flow resistance, such

as a valve or porous plug [92]. Also on the images, C stands for a

compressor, AC for an aftercooler, HX for heat exchangers, JT for a valve,

and finally E for an expander.

This cycle is often used as a part of re-liquefaction plants of BOG alongside with a nitrogen cycle also based

on the Claude cycle [155].

Figure 79 - Claude Cycle [92]

Figure 78 - Linde-Hampson Cycle [175]

101

iii) Kapitza Cycle

Kapitza (1939) modified the basic

Claude cycle by eliminating the third heat

exchanger (low temperature heat

exchanger, see Figure 80). A rotary

expansion engine was also introduced

instead of the reciprocating expander,

increasing the efficiency of the system. The

throttling valve represents an isenthalpic

expansion, and is used to reduce the

pressure of the compressed air, in order to

produce and store the liquid [156].

This cycle can operate at relatively low pressure, lower than the critical pressure of nitrogen (34 bar).

As Moon et al. (2007) state on their comparison between the Claude and Kapitza cycles for BOG re-

liquefaction, the Kapitza refrigeration cycle can be more efficient than the Claude cycle in terms of plant

operability and cost.

Figure 80 - Kapitza Cycle [156]

102

VIII. Tank Project – Final Drawing

Figure 81 - Tank Project – Final Drawing

103

IX. Tank Insulation Examples

Figure 82 - Insulation Example 1 [157]

Figure 83- Insulation Example 2 [157]

104



105



106

X. Mollier Diagram - Methane

Figure 88 - Mollier Diagram – Methane [159]

107

XI. Mollier Diagram - Nitrogen

Figure 89 - Mollier Diagram – Nitrogen [159]

108

XII. Mollier Diagram – Liquefaction Cycles A and B

Figure 90 - Liquefaction Cycles A and B [159]

109

XIII. Mollier Diagram – Methane Cycle

Figure 91 - Methane Cycles [159]

110

XIV. Simulation Results - Liquefaction Cycles A and B

i) BOG inlet temperature = - 155⁰C

Cycle A – BOG rate = 0.039 %

�̇�𝑩𝑶𝑮 [kg/s] 0.362

𝑻𝑩𝑶𝑮 [K] 118.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 259.0

𝒉𝑩𝑶𝑮 [kJ/kg] 525.000 �̇�𝒊𝒏𝒕𝒆𝒓𝒄𝒐𝒐𝒍𝒆𝒓𝒔 [kW] -414.40

𝑻𝑳𝑵𝑮_𝑺𝒖𝒃𝒄𝒐𝒐𝒍𝒆𝒅 [K] 108.150 �̇�𝒉𝒆𝒂𝒕 𝒆𝒙𝒄𝒉𝒂𝒏𝒈𝒆𝒓𝒔 [kW] 207.20

𝒉𝑳𝑵𝑮_𝑺𝒖𝒃𝒄𝒐𝒐𝒍𝒆𝒅 [kJ/kg] -20.000 �̇�𝑵𝟐 [kg/s] 4.144

𝑷𝑹𝒆𝒇 [kW] 197.323

𝑷𝑹𝒆𝒇𝒎𝒂𝒓𝒈𝒊𝒏 [kW] 207.190

Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1

Stages Enthalpy Pressure Temperature

Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1

Expander Heatex HE 2

Heatex HE 3

kJ/kg MPa K kW kW kW kW kW kW kW kW

1 95 0.8 105 129.50 - - - - - - -

2 120 1.79 135 - -145.04 - - - - -

3 85 1.79 113 - 129.50 - - - - -

4 110 4.0 140 - - - -165.76 - - - -

5 70 4.0 136 - - - - -103.60 - - -

6 45 4.0 130 - - - - - -103.60 - -

7 20 0.8 100 - - - - - - 82.88 -

8 40 0.8 100 - - - - - - - 227.92


�̇�𝑩𝑶𝑮 [kg/s] 0.377

𝑻𝑩𝑶𝑮 [K] 118.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 269.55




𝑷𝑹𝒆𝒇 [kW] 205.370


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 134.78 - - - - - - -

2 120 1.79 135 - -150.948 - - - - -

3 85 1.79 113 - 134.78 - - - - -

4 110 4.0 140 - - - -172.51 - - - -

5 70 4.0 136 - - - - -107.82 - - -

6 45 4.0 130 - - - - - -107.82 - -

7 20 0.8 100 - - - - - - 86.26 -

8 40 0.8 100 - - - - - - - 237.20

111


�̇�𝑩𝑶𝑮 [kg/s] 0.400

𝑻𝑩𝑶𝑮 [K] 118.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 286.41




𝑷𝑹𝒆𝒇 [kW] 218.218


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 143.20 - - - - - - -

2 120 1.79 135 - -160.39 - - - - -

3 85 1.79 113 - 143.20 - - - - -

4 110 4.0 140 - - - -183.30 - - - -

5 70 4.0 136 - - - - -114.56 - - -

6 45 4.0 130 - - - - - -114.56 - -

7 20 0.8 100 - - - - - - 91.65 -

8 40 0.8 100 - - - - - - - 252.04

Cycle B – BOG rate = 0.039 %

�̇�𝑩𝑶𝑮 [kg/s] 0.362

𝑻𝑩𝑶𝑮 [K] 118.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 197.84




𝑷𝑹𝒆𝒇 [kW] 197.323


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 125.90 - - - - - - -

2 140 1.27 145 - -143.88 - - - - -

3 90 1.27 105 - 71.94 - - - - -

4 110 4.0 155 - - - -115.10 - - - -

5 70 4.0 135 - - - - -71.94 - - -

6 45 4.0 130 - - - - - -106.47 - -

7 8 0.4 92 - - - - - - 92.08 -

8 40 0.4 92 - - - - - - - 187.04

112


�̇�𝑩𝑶𝑮 [kg/s] 0.377

𝑻𝑩𝑶𝑮 [K] 118.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 205.91




𝑷𝑹𝒆𝒇 [kW] 205.370


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 125.90 - - - - - - -

2 140 1.27 145 - -143.88 - - - - -

3 90 1.27 105 - 71.94 - - - - -

4 110 4.0 155 - - - -115.10 - - - -

5 70 4.0 135 - - - - -71.94 - - -

6 45 4.0 130 - - - - - -106.47 - -

7 8 0.4 92 - - - - - - 92.08 -

8 40 0.4 92 - - - - - - - 187.04


�̇�𝑩𝑶𝑮 [kg/s] 0.400

𝑻𝑩𝑶𝑮 [K] 118.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 218.78




𝑷𝑹𝒆𝒇 [kW] 218.218


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 139.22 - - - - - - -

2 140 1.27 145 - -159.11 - - - - - -

3 90 1.27 105 - - 79.56 - - - - -

4 110 4.0 155 - - - -127.29 - - - -

5 70 4.0 135 - - - - -79.56 - - -

6 45 4.0 130 - - - - - -117.74 - -

7 8 0.4 92 - - - - - - 101.83 -

8 40 0.4 92 - - - - - - - 206.85

113

ii) BOG inlet temperature = - 140⁰C


�̇�𝑩𝑶𝑮 [kg/s] 0.362

𝑻𝑩𝑶𝑮 [K] 133.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 273.24




𝑷𝑹𝒆𝒇 [kW] 208.185


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 136.62 - - - - - - -

2 120 1.79 135 - -153.016 - - - - - -

3 85 1.79 113 - - 136.62 - - - - -

4 110 4.0 140 - - - -174.88 - - - -

5 70 4.0 136 - - - - -109.30 - - -

6 45 4.0 130 - - - - - -109.30 - -

7 20 0.8 100 - - - - - - 87.44 -

8 40 0.8 100 - - - - - - - 240.45


�̇�𝑩𝑶𝑮 [kg/s] 0.377

𝑻𝑩𝑶𝑮 [K] 133.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 284.39




𝑷𝑹𝒆𝒇 [kW] 216.675


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 142.19 - - - - - - -

2 120 1.79 135 - -159.257 - - - - -

3 85 1.79 113 - - 142.19 - - - - -

4 110 4.0 140 - - - -182.01 - - - -

5 70 4.0 136 - - - - -113.76 - - -

6 45 4.0 130 - - - - - -113.76 - -

7 20 0.8 100 - - - - - - 91.00 -

8 40 0.8 100 - - - - - - - 250.26

114


�̇�𝑩𝑶𝑮 [kg/s] 0.400

𝑻𝑩𝑶𝑮 [K] 133.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 302.18




𝑷𝑹𝒆𝒇 [kW] 230.230


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 151.09 - - - - - - -

2 120 1.79 135 - -169.22 - - - - - -

3 85 1.79 113 - - 151.09 - - - - -

4 110 4.0 140 - - - -193.39 - - - -

5 70 4.0 136 - - - - -120.87 - - -

6 45 4.0 130 - - - - - -120.87 - -

7 20 0.8 100 - - - - - - 96.70 -

8 40 0.8 100 - - - - - - - 265.92


�̇�𝑩𝑶𝑮 [kg/s] 0.362

𝑻𝑩𝑶𝑮 [K] 118.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 208.73




𝑷𝑹𝒆𝒇 [kW] 208.185


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 132.83 - - - - - - -

2 140 1.27 145 - -151.802 - - - - -

3 90 1.27 105 - - 75.90 - - - - -

4 110 4.0 155 - - - -121.44 - - - -

5 70 4.0 135 - - - - -75.90 - - -

6 45 4.0 130 - - - - - -112.33 - -

7 8 0.4 92 - - - - - - 97.15 -

8 40 0.4 92 - - - - - - - 197.34

115


�̇�𝑩𝑶𝑮 [kg/s] 0.377

𝑻𝑩𝑶𝑮 [K] 133.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 217.24




𝑷𝑹𝒆𝒇 [kW] 216.675


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 138.24 - - - - - - -

2 140 1.27 145 - -

157.992 - -

- - -

3 90 1.27 105 - - 79.00 - - - - -

4 110 4.0 155 - - - -126.39 - - - -

5 70 4.0 135 - - - - -79.00 - - -

6 45 4.0 130 - - - - - -116.91 - -

7 8 0.4 92 - - - - - - 101.11 -

8 40 0.4 92 - - - - - - - 205.39


�̇�𝑩𝑶𝑮 [kg/s] 0.400

𝑻𝑩𝑶𝑮 [K] 133.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 230.83




𝑷𝑹𝒆𝒇 [kW] 230.230


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 146.89 - - - - - - -

2 140 1.27 145 - -167.88 - - - - - -

3 90 1.27 105 - - 83.94 - - - - -

4 110 4.0 155 - - - -134.30 - - - -

5 70 4.0 135 - - - - -83.94 - - -

6 45 4.0 130 - - - - - -124.23 - -

7 8 0.4 92 - - - - - - 107.44 -

8 40 0.4 92 - - - - - - - 218.24

116

iii) BOG inlet temperature = - 125⁰C


�̇�𝑩𝑶𝑮 [kg/s] 0.362

𝑻𝑩𝑶𝑮 [K] 148.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 289.88




𝑷𝑹𝒆𝒇 [kW] 220.857


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 144.94 - - - - - - -

2 120 1.79 135 - -162.33 - - - - -

3 85 1.79 113 - - 144.94 - - - - -

4 110 4.0 140 - - - -185.52 - - - -

5 70 4.0 136 - - - - -115.95 - - -

6 45 4.0 130 - - - - - -115.95 - -

7 20 0.8 100 - - - - - - 92.76 -

8 40 0.8 100 - - - - - - - 255.09


�̇�𝑩𝑶𝑮 [kg/s] 0.377

𝑻𝑩𝑶𝑮 [K] 148.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 301.70




𝑷𝑹𝒆𝒇 [kW] 229.863


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 150.85 - - - - - - -

2 120 1.79 135 - -168.95 - - - - -

3 85 1.79 113 - - 150.85 - - - - -

4 110 4.0 140 - - - -193.09 - - - -

5 70 4.0 136 - - - - -120.68 - - -

6 45 4.0 130 - - - - - -120.68 - -

7 20 0.8 100 - - - - - - 96.54 -

8 40 0.8 100 - - - - - - - 265.49

117


�̇�𝑩𝑶𝑮 [kg/s] 0.400

𝑻𝑩𝑶𝑮 [K] 148.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 320.57




𝑷𝑹𝒆𝒇 [kW] 244.244


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 160.28 - - - - - - -

2 120 1.79 135 - -179.52 - - - - - -

3 85 1.79 113 - - 160.28 - - - - -

4 110 4.0 140 - - - -205.16 - - - -

5 70 4.0 136 - - - - -128.23 - - -

6 45 4.0 130 - - - - - -128.23 - -

7 20 0.8 100 - - - - - - 102.58 -

8 40 0.8 100 - - - - - - - 282.10


�̇�𝑩𝑶𝑮 [kg/s] 0.362

𝑻𝑩𝑶𝑮 [K] 148.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 221.43




𝑷𝑹𝒆𝒇 [kW] 220.857


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 140.91 - - - - - - -

2 140 1.27 145 - -161.042 - - - - -

3 90 1.27 105 - - 80.52 - - - - -

4 110 4.0 155 - - - -128.83 - - - -

5 70 4.0 135 - - - - -80.52 - - -

6 45 4.0 130 - - - - - -119.17 - -

7 8 0.4 92 - - - - - - 103.07 -

8 40 0.4 92 - - - - - - - 209.35

118


�̇�𝑩𝑶𝑮 [kg/s] 0.377

𝑻𝑩𝑶𝑮 [K] 148.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 230.46




𝑷𝑹𝒆𝒇 [kW] 229.863


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 146.66 - - - -

2 140 1.27 145 - -167.61 - -

3 90 1.27 105 - 83.80 - -

4 110 4.0 155 - - -134.09 -

5 70 4.0 135 - - - - -83.80

6 45 4.0 130 - - - - - -124.03

7 8 0.4 92 - - - - - 107.27

8 40 0.4 92 - - - - - 217.89


�̇�𝑩𝑶𝑮 [kg/s] 0.400

𝑻𝑩𝑶𝑮 [K] 148.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 244.87




𝑷𝑹𝒆𝒇 [kW] 244.244


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 155.83 - - - - - - -

2 140 1.27 145 - -178.09 - - - - - -

3 90 1.27 105 - - 89.05 - - - - -

4 110 4.0 155 - - - -142.47 - - - -

5 70 4.0 135 - - - - -89.05 - - -

6 45 4.0 130 - - - - - -131.79 - -

7 8 0.4 92 - - - - - - 113.98 -

8 40 0.4 92 - - - - - - - 231.52

119

iv) BOG inlet temperature = - 110⁰C


�̇�𝑩𝑶𝑮 [kg/s] 0.362

𝑻𝑩𝑶𝑮 [K] 163.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 304.13




𝑷𝑹𝒆𝒇 [kW] 231.719


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 152.07 - - - - - - -

2 120 1.79 135 - -170.31 - - - - -

3 85 1.79 113 - - 152.07 - - - - -

4 110 4.0 140 - - - -194.64 - - - -

5 70 4.0 136 - - - - -121.65 - - -

6 45 4.0 130 - - - - - -121.65 - -

7 20 0.8 100 - - - - - - 97.32 -

8 40 0.8 100 - - - - - - - 267.64


�̇�𝑩𝑶𝑮 [kg/s] 0.377

𝑻𝑩𝑶𝑮 [K] 163.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 316.53




𝑷𝑹𝒆𝒇 [kW] 241.168


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 158.27 - - - - - - -

2 120 1.79 135 - -177.26 - - - - -

3 85 1.79 113 - - 158.27 - - - - -

4 110 4.0 140 - - - -202.58 - - - -

5 70 4.0 136 - - - - -126.61 - - -

6 45 4.0 130 - - - - - -126.61 - -

7 20 0.8 100 - - - - - - 101.29 -

8 40 0.8 100 - - - - - - - 278.55

120


�̇�𝑩𝑶𝑮 [kg/s] 0.400

𝑻𝑩𝑶𝑮 [K] 163.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 336.34




𝑷𝑹𝒆𝒇 [kW] 256.256


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 95 0.8 105 168.17 - - - - - - -

2 120 1.79 135 - -188.35 - - - - - -

3 85 1.79 113 - - 168.17 - - - - -

4 110 4.0 140 - - - -215.25 - - - -

5 70 4.0 136 - - - - -134.53 - - -

6 45 4.0 130 - - - - - -134.53 - -

7 20 0.8 100 - - - - - - 107.63 -

8 40 0.8 100 - - - - - - - 295.98


�̇�𝑩𝑶𝑮 [kg/s] 0.362

𝑻𝑩𝑶𝑮 [K] 163.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 232.32




𝑷𝑹𝒆𝒇 [kW] 231.719


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 147.84 - - - - - - -

2 140 1.27 145 - -168.96 - - - - - -

3 90 1.27 105 - - 84.48 - - - - -

4 110 4.0 155 - - - -135.17 - - - -

5 70 4.0 135 - - - - -84.48 - - -

6 45 4.0 130 - - - - - -125.03 - -

7 8 0.4 92 - - - - - - 108.14 -

8 40 0.4 92 - - - - - - - 219.65

121


�̇�𝑩𝑶𝑮 [kg/s] 0.377

𝑻𝑩𝑶𝑮 [K] 163.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 241.80




𝑷𝑹𝒆𝒇 [kW] 241.168


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 153.87 - - - - - - -

2 140 1.27 145 - -175.85 - - - - -

3 90 1.27 105 - - 87.93 - - - - -

4 110 4.0 155 - - - -140.68 - - - -

5 70 4.0 135 - - - - -87.93 - - -

6 45 4.0 130 - - - - - -130.13 - -

7 8 0.4 92 - - - - - - 112.55 -

8 40 0.4 92 - - - - - - - 228.61


�̇�𝑩𝑶𝑮 [kg/s] 0.400

𝑻𝑩𝑶𝑮 [K] 163.150 𝜼𝑪𝒐𝒎𝒑 0.8

𝒑𝑩𝑶𝑮 [MPa] 0.107 �̇�𝑪𝒐𝒎𝒑 [kW] 256.92




𝑷𝑹𝒆𝒇 [kW] 256.256


Stages 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-1


Work Comp.1

Heatdiss Cond.1

Work Comp.2

Heatdiss Cond.2

Heatex HE 1


Heatex HE 3


1 105 0.4 105 163.49 - - - - - - -

2 140 1.27 145 - -186.85 - - - - -

3 90 1.27 105 - - 93.43 - - - - -

4 110 4.0 155 - - - -149.48 - - - -

5 70 4.0 135 - - - - -93.43 - - -

6 45 4.0 130 - - - - - -138.27 - -

7 8 0.4 92 - - - - - - 119.58 -

8 40 0.4 92 - - - - - - - 242.91

122

XV. Re-liquefaction System – Capacity Control Scheme

Figure 92 - Capacity Control Scheme of the Re-liquefaction System

Simulation and Design of a Boil-Off Gas Re-liquefaction ... · Simulation and Design of a Boil-Off...

Documents

Transcript of Simulation and Design of a Boil-Off Gas Re-liquefaction ... · Simulation and Design of a Boil-Off...