Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

THE UNIVERSITY OF QUEENSLAND Bachelor of Engineering Thesis

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

Student Name: Declan SCOTT Course Code: CHEE4060 Supervisors: Dr. Simon Smart, Dr. David Harris Submission date: 16 June 2014

A thesis submitted in partial fulfilment of the requirements of the Bachelor of Engineering degree in Chemical Engineering

UQ Engineering Faculty of Engineering, Architecture and Information Technology

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

Abstract Technology for the production of syngas from coal and its subsequent conversion to synthetic crude oil has existed for a number of years with somewhat limited commercial implementation. This indirect coal liquefaction process was analysed to provide a preliminary assessment of its economic feasibility in Australia. Analysis was enabled through Aspen Plus simulation of coal preparation, gasification, syngas conditioning, Fischer-Tropsch (FT) synthesis and syncrude refining processes. This model was run with four Australian thermal coals as feedstock, specifically selected as their individual characteristics represent a wide range of coal types and because they have previously been subject to pilot scale gasification trials. Simulation results showed that coal type significantly affects the volume of FT liquid fuels produced, with production ranging from 21,000bbl/day to 43,000bbl/day. High carbon content coals yielded the highest fuel volumes and lowest relative CO2 production, although this is still significant at 3.9 kg CO2/kg fuel produced. Economic analysis of the modelled process was facilitated by simulation data, commodity price predictions and capital cost estimates for a reference CTL plant of similar size. Analysis showed that, for all the scenarios considered, the project offered a large, negative net present value and low internal rate of return ranging from 2.3% to 7.4%. This indicates that the process under consideration is not viable under the current market conditions in Australia. Breakeven prices required for crude oil to yield a viable process varied between $120$160/bbl depending on the coal feedstock. Process economics were shown to be most sensitive to product prices received, therefore crude oil prices would significantly affect the project viability. Results also showed that high carbon content coals offer the lowest breakeven fuel prices and highest returns, as the price premium for their higher heating value is more than offset by the additional fuel production achieved. However, none of the coals trialled offered a viable Australian CTL process and this appears to remain the case unless sustained increases in crude oil prices are experienced.

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

Acknowledgements I would like to thank Dr. David Harris and his team at CSIRO’s Queensland Centre for Advanced Technology (QCAT) for their valuable advice and assistance. Especially in selecting suitable coals to assess in this study and for sharing their extensive knowledge from previous coal gasification trials. Thanks also go to Dr. Nikolai Kinaev and Dr. Chris Greig from the UQ Energy Initiative for their practical advice from a wealth of industry experience. I would also like to thank Dr. Thomas Kreutz from Princeton University’s Environmental Institute for his guidance in developing an effective process model of the coal to liquids process based on his past experience in the area. Last but certainly not least, thanks go to my supervisor Dr. Simon Smart for giving me direction and keeping me on track throughout this project.

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

Table of Contents 1 Introduction ...................................................................................................................................... 1 1.1 Background:............................................................................................................................................... 1

1.2 Research Goal ............................................................................................................................................ 2

1.3 Project Scope ............................................................................................................................................. 2

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

2.1 Coal Liquefaction Pathways: ............................................................................................................... 2 2.1.1 Direct Coal Liquefaction ................................................................................................................................................. 3 2.1.2 Indirect Coal Liquefaction.............................................................................................................................................. 4

2.2 Gasification Technology ........................................................................................................................ 4 2.2.1 2.2.2 2.2.3 2.2.4

Moving-Bed Gasifiers:...................................................................................................................................................... 5 Circulated Fluidized-Bed Gasifiers: ........................................................................................................................... 5 Entrained Flow Gasifiers ................................................................................................................................................ 6 Gasifier Applications: ....................................................................................................................................................... 6

2.3 Syngas Conversion Technology .......................................................................................................... 6 2.3.1 Methanol to Gasoline ....................................................................................................................................................... 7 2.3.2 Fischer-Tropsch Synthesis: ........................................................................................................................................... 8

2.4 Additional Unit Operations:................................................................................................................. 9 2.4.1 2.4.2 2.4.3 2.4.4 2.4.5

Coal Preparation ................................................................................................................................................................ 9 Air Separation Unit ........................................................................................................................................................... 9 Water Gas Shift ................................................................................................................................................................ 10 Acid Gas Removal............................................................................................................................................................ 10 Syncrude Refining .......................................................................................................................................................... 11

2.5 Commercial Operations ...................................................................................................................... 11 2.6 Previous Techno-economic Studies ................................................................................................ 12

2.7 Summary & Research Gaps ................................................................................................................ 13

3 Model Development ..................................................................................................................... 14 3.1 Reference Case ........................................................................................................................................ 14 3.2 Model Assumptions .............................................................................................................................. 14

3.3 Model Summary ..................................................................................................................................... 14 3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.3.6 3.3.7 3.3.8

Coal Feedstock: ................................................................................................................................................................ 15 Coal Preparation: ............................................................................................................................................................ 15 Gasification:....................................................................................................................................................................... 16 Syngas Quench: ................................................................................................................................................................ 16 Water Gas Shift: ............................................................................................................................................................... 17 Acid Gas Removal: .......................................................................................................................................................... 17 Fischer-Tropsch Synthesis: ........................................................................................................................................ 18 Syncrude Refinery: ......................................................................................................................................................... 18

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 4 Model Verification ........................................................................................................................ 20 4.1 Gasification: ............................................................................................................................................. 20

4.2 Fischer-Tropsch Synthesis: ................................................................................................................ 21

4.3 Overall Model: ......................................................................................................................................... 22

5 Results: ............................................................................................................................................. 23 6 Cost Estimation .............................................................................................................................. 24 6.1 Capital Cost Estimation: ...................................................................................................................... 24 6.1.1 6.1.2 6.1.3 6.1.4

Direct Cost Estimates:................................................................................................................................................... 24 Indirect Cost Estimates: ............................................................................................................................................... 24 Exclusions: ......................................................................................................................................................................... 25 Capital Cost Conversion: .............................................................................................................................................. 25

6.2 Operating Costs: ..................................................................................................................................... 26

6.2.1 Fixed Operating Costs:.................................................................................................................................................. 26 6.2.2 Variable Operating Costs: ............................................................................................................................................ 26

7 Economic Analysis ........................................................................................................................ 30 7.1 Sensitivity Analysis: .............................................................................................................................. 33

7.2 Coal Selection:......................................................................................................................................... 34

8 Conclusions ..................................................................................................................................... 35 9 References ....................................................................................................................................... 36 10 Appendices ..................................................................................................................................... 39 10.1 Appendix A: Aspen Plus Model Components: ............................................................................ 39

10.2 Appendix B: Model Development Details: ................................................................................... 40

10.3 Appendix C: FT Synthesis Product Distribution: ....................................................................... 49 10.4 Appendix D: Fixed Operating Costs: ............................................................................................... 50

10.5 Appendix E: Price Estimations: ........................................................................................................ 51

10.6 Appendix F: Example Economic Calculations – CRC701 Scenario: .................................... 53

10.7 Appendix G: Sensitivity Analysis Results: .................................................................................... 54

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

List of Tables: Table 1: Breakeven Crude Oil Prices from Previous Technoeconomic ICL Studies ..................... 12 Table 2: Composition of Five Coal Feedstocks Used .................................................................... 15 Table 3: Average operating conditions from pilot scale gasification studies (Roberts, 2011) ....... 16 Table 4: Comparison of Model Gasifier with Reference Data ....................................................... 20 Table 5: Key Results from CTL Model run with 5 different coals................................................. 23 Table 6: Capital Cost Estimates for Five Operating Scenarios ...................................................... 25 Table 7: Key results from Economic Analysis of 5 Scenarios ....................................................... 31 Table 8: Cost of Oxygen Supply for Each Scenario....................................................................... 52

List of Figures: Figure 1: Commercial examples of different gasifier types (NETL, 2014) ..................................... 5 Figure 2: Gasifier suitability for different coal ranks (Kinaev, 2014) .............................................. 6 Figure 3: Potential Syngas Upgrading Pathways. Reprinted from Spath, 2003 .............................. 7 Figure 4: Fischer-Tropsch Reactor Designs (Spath and Dayton, 2008)........................................... 9 Figure 5: Block Flow Diagram of 20,000bbl/day ICL Reference Case (Drover, 2008) ................ 14 Figure 6: Comparison of Syncrude Compositon - Model Data vs Typical Co-LTFT Process ....... 21 Figure 7: Modelled Syncrude Composition vs ASF Distributions ................................................. 22 Figure 8: Dated Brent, Gasoline & Diesel Prices over the past 10 years (AIP, 2014) ................... 28 Figure 9: Average Refined Product Price vs Crude Oil Price since 2009 ...................................... 29 Figure 10: Contribution of Plant Costs towards Breakeven FT Product Prices ............................. 32 Figure 11: Effect of Percentage Change in Key Prices on Projects Internal Rate of Return ......... 33 Figure 12: Breakeven FT Product Price vs Effective Carbon Content of Coal.............................. 34

1

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

1 Introduction 1.1

Background:

Technologies for the conversion of coal and natural gas to liquid petroleum fuels have been available for over 80 years. However, the availability of relatively cheap crude oil supplies has resulted in these technologies remaining economically unattractive under most conditions. Implementation has therefore been limited except in extreme cases, often involving the prospect of oil sanctions (Lowenberg, 1992). However, significant increases in the price of crude oil of over 400% over the past 20 years have revived interest in liquid fuel production from alternative feed stocks and there are currently a number of projects under development worldwide (The World Bank, 2014). At this point in time, none of these projects under development are located within Australia. This is despite Australia being the world’s 4th largest coal producer with the 5th largest proven reserves, whilst having only the 38th largest oil reserves in comparison. Australia’s relatively small oil reserves, high cost of manufacturing, relatively small domestic market and the availability of low cost imports have led to the closure of 4 of the existing 8 domestic refinery’s over the past decade. This situation has no negative consequences under normal conditions however; it does place a high reliance on imports of refined products from Asian refiners. Any disruptions to this supply chain resulting from political issues or even severe weather events could have a significant impact on Australia’s transport sector. In addition to this, the inability to produce sufficient volumes of jet fuel domestically raises concerns over national security. Combination of these factors presents potentially favourable economic and political drivers for the production of liquid fuels from alternative feed stocks. Considering predicted volatility in future natural gas prices as a number of large export projects come online, and Australia’s large coal resources, a coal feedstock could be an attractive option (Wood and Carter, 2013). However, projects of this type generally exhibit significant economies of scale and the downside of this is the requirement for significant capital expenditure to build a large plant of economically viable capacity. Recent cost pressures on resource and industrial projects within Australia place significant risk on undertaking a project of this magnitude (BCA, 2013). To address these issues it is necessary to determine under what conditions a Coal to Liquid Fuels (CTL) plant might be economically viable within Australia. This study is the first step in conducting a robust techno-economic assessment of CTL plant viability in Australia, and involves the construction of a high-level tehno-economic model of an indirect coal liquefaction (ICL) process based on existing and accepted technology. This model is then used to evaluate the process for a number of different Australian coals, which have been subject to previous coal gasification trials (Roberts et al., 2011). The economic viability of the process is assessed under various conditions and the sensitivity of this to a number of factors is investigated.

2

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 1.2

Research Goal

The aim of this study is to conduct a preliminary assessment of the economic viability of jet fuel production through currently available indirect coal liquefaction technology for a number of different Australian coals. 1.3

Project Scope

Scope of the project includes development of a high-level simulation of an indirect coal liquefaction process, based on a generic plant configuration that has been considered in several commercial studies (Drover, 2008). Modelling covers production of refined transportation fuels from a raw coal feedstock and focuses on the key unit operations of coal gasification, syngas cleaning, Fischer-Tropsch conversion and syncrude refining. Analysis of the process for various coal feedstocks, which are relevant in an Australian context, is also included in the project scope. The specific coals were selected with assistance from CSIRO’s Queensland Centre for Advanced Technology to represent broader categories of coal available within Australia. Finally, a preliminary economic assessment of the process for each coal type will be presented. Detailed capital cost estimates are excluded from the scope as these are estimated from the capital cost of the reference case upon which the model was based. However, preliminary estimation of operating costs based on plant data as well as predicted fuel and feedstock prices are included. Analysis of the sensitivity of the process to future prices and capital cost assumptions is also presented.

2 Literature Review Although alternatives are briefly discussed, this literature review focuses on indirect coal liquefaction processes involving coal gasification and syngas upgrading. A number of available gasification technologies are compared and two key syngas upgrading pathways for liquid fuels production are presented. Additional unit operations required for an indirect coal liquefaction process are also identified and alternative technologies discussed. Finally, the state of existing ICL facilities is assessed and a brief summary of previous techno-economic studies is presented. 2.1

Coal Liquefaction Pathways:

Coal liquefaction is a broad term describing the conversion of solid coal to liquid fuels or chemicals. More specifically, it involves the production of liquids by chemically altering the coal structure to increase the hydrogen to carbon ratio (Kaneko et al., 2000). There are two main pathways that are used to achieve this conversion industrially; direct coal liquefaction and indirect coal liquefaction.

3

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 2.1.1 Direct Coal Liquefaction

Direct coal liquefaction reacts coal under elevated temperature and pressure with a solvent to rupture the bonds within high molecular weight organic molecules in the coal structure (Li and Fan, 2008). This produces a liquid product with a molecular weight similar to that of the fuels being produced that is highly aromatic, indicating a high composition of benzene derivatives (Dadyburjor et al., 2000). The solvent is usually a coal-derived, heavy aromatic material that dissolves the coals organic structure and provides the hydrogen required for the reactions taking place. The first of these reactions is the thermal cracking of large organic molecules to smaller free-radical fragment molecules. Hydrogenation of these free-radical molecules is then required to produce liquid products through the addition of hydrogen to eliminate the free radical (Dadyburjor et al., 2000). By definition, the free radical fragments contain an unpaired valence electron making them highly reactive and able to react with other fragments to form solid coke if insufficient hydrogen is provided (Liu et al., 2010). This basic understanding of the DCL process is generally accepted, although the exact reaction mechanisms are highly complex and not yet fully understood. Liquids produced from the DCL process are then separated according to their solubility, distilled into the desired fractions and then hydrotreated to produce the final liquid fuel products (Kaneko et al., 2000). There are a number of commercially available processes including the Exxon Donor Solvent Process, the H-Coal Process, the Shenhua and the NEDOL process, which was operated in the Latrobe Valley, Australia in the 1980s (Han and Chang, 2008). Direct coal liquefaction processes have been improved significantly since research interest intensified in the 1970s, with oil yields increasing from 44% to 70% (Liu et al., 2010). They currently offer the highest efficiency coal conversion process and a number of plants are being developed in China and worldwide. However, there are a number of disadvantages to the DCL process that limit its application. Although a high aromatic content is beneficial for octane properties, there are a number of potent mutagenic and carcinogenic aromatic liquids produced that pose significant health concerns and complicate waste disposal (Williams and Larson, 2003). Waste disposal is further complicated by high concentrations of salts in wastewater, which would require a different treatment strategy to the organic compounds (Kinaev, 2014). The high aromatic content also means that the fuel produced has a very low cetane rating and the production of diesel or jet fuel is not feasible from DCL products (Williams and Larson, 2003). Most DCL processes also require an external hydrogen source that is usually obtained from natural gas, thus making them undesirable if gas prices were to increase significantly in the future. Finally, DCL processes are also very cost-intensive and the modern, more efficient technologies are only just being proven on a commercial scale. As such, DCL processes were not considered any further for this study.

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

4

2.1.2 Indirect Coal Liquefaction Indirect coal liquefaction involves the gasification of coal to a mixture of hydrogen and carbon monoxide, known as synthesis gas or syngas, which is then catalytically converted to liquid products (Kaneko et al., 2000). Coal gasification is the partial oxidation of coal in a reducing environment to produce syngas, which retains a large percentage of the coals heating value (Higman, 2008). Gasification occurs through a number of stages as the coal is heated within the gasifier. Drying occurs when the coal first enters the gasifier and moisture is driven out at temperatures of approximately 105 °C. Devolatilisation or pyrolysis then occurs from 500-900 °C where bonds between aromatic clusters in the coal structure are broken to form low molecular weight particles which vapourise and escape the coal structure (Shadle et al., 2000). The remaining highly porous solid product is known as char and would usually be combusted in an oxygen atmosphere. However, as the oxygen feed is restricted, carbon within the coal is only partially combusted according to the Reaction 1 below: R1)

(1+λ)C + O2 ⟶ 2λ CO + (1-λ)CO2

𝛥𝛥𝑚298𝐾 = 172.5𝜆 − 393.5 𝑘𝑘/𝑚𝑚𝑚

The variable λ varies from 0 to 1 depending on gasifier conditions and determines whether carbon is fully oxidised to CO2 or partially oxidised to CO. This exothermic reaction occurs in the combustion zone at the coal feed point and rapidly consumes the available oxygen whilst also providing the heat required for the endothermic, reversible gasification (R2-3) and methanation (R4) reactions below: R2)

R3) R4)

C + CO2 ⇌ 2CO

𝛥𝛥𝑚298𝐾 = 172.5 𝑘𝑘/𝑚𝑚𝑚

C + 2H2 ⇌ CH4

𝛥𝛥𝑚298𝐾 = −74.8 𝑘𝑘/𝑚𝑚𝑚

C + H2 O ⇌ CO + H2

𝛥𝛥𝑚298𝐾 = 131.3 𝑘𝑘/𝑚𝑚𝑚

In addition to this, the water-gas shift reaction also occurs within the gasifier as shown below: R5)

CO + H2 O ⇌ CO2 + H2

𝛥𝛥𝑚298𝐾 = −41.2 𝑘𝑘/𝑚𝑚𝑚

Reactions 1 to 5 describe the main reactions occurring within the gasifier. In addition to this, inorganic components within the coal (heteroatoms) also react under the reducing gasifier conditions. Sulphur reacts to form hydrogen sulphide (H2S) and carbonyl sulphide (COS), nitrogen reacts to form ammonia (NH3) and this can then react further to form hydrogen cyanide (HCN) (Kunze and Spliethoff, 2011). These pollutants must be removed in downstream syngas clean-up steps before further processing. Numerous commercial technologies have been developed for coal gasification and the main categories are discussed below. 2.2

Gasification Technology

Although a large number of commercial gasifiers are available, these all fall into three general categories; Moving-bed (aka Fixed-bed) gasifiers, Fluidized-bed gasifiers and Entrained-flow gasifiers (Simbeck, 1993). Commercial examples of each of these gasifier types are shown in Figure 1 on the following page.

5

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process Fixed Bed e.g. British Gas Lurgi

Circulating Fluidised Bed e.g. KBR Trig

Entrained Flow Dry Fed (e.g. Shell) Slurry Fed (e.g. GE)

Figure 1: Commercial examples of different gasifier types (NETL, 2014)

2.2.1 Moving-Bed Gasifiers: Moving-bed gasifiers contain a bed of coal which moves slowly downward under gravity whilst it is gasified by a counter current oxidant blast (Drover, 2008). Coal enters at the top of the reactor where it is dried and cools the exiting syngas. It then moves through the devolatisation and combustion zones, forming syngas whilst dry ash exits at the bottom of the reactor (Simbeck, 1993). Although the oxygen consumption is low and the Cold Gas Efficiency (CGE) of these reactors is relatively high, the product syngas contains methane as well as tars and oils which must be removed before further processing, reducing the overall efficiency (Drover, 2008). Commercially available moving-bed gasifiers include the SasolLurgi dry bottom gasifier and the BGL slagging gasifier (Higman, 2008) 2.2.2 Circulated Fluidized-Bed Gasifiers: Coal is generally fed in the side of a fluidised-bed gasifier whilst a counter-current oxidant flow from below ensures good back-mixing of the feed coal particles with those already undergoing gasification (Simbeck, 1993). This back-mixing means that some partially reacted coal is inevitably removed with the ash, reducing the carbon conversion (Higman, 2008). Syngas leaving the top of the reactor contains some entrained coal particles which are recovered via cyclone and returned to the reactor (Drover, 2008). The bed remains at a fairly uniform temperature which must remain below the ash fusion temperature of the coal, as melting ash will impact on the fluidisation ability. Fluidised bed reactors are characterised by moderate temperature and oxygen demands as well as low carbon conversion (Bell et al., 2011). Commercially available fluidised-bed gasifiers include the High-Temperature Winkler (HTW) and KBR TRIG processes (Kaneko et al., 2000).

6

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 2.2.3 Entrained Flow Gasifiers

Entrained-flow gasifiers operate with small feed particles below 100 microns in co-current flow with the oxidant blast (Drover, 2008). High temperatures above 1200C, and therefore high oxidant flow, are required in order to achieve sufficient conversion during the short residence time in the reactor. This high temperature results in a slagging operation for gasifier of this type, meaning that ash from coal is melted to form slag (Rezaiyan, 2005). These gasifiers have high oxidant requirements but produce a relatively clean syngas with a high sensible heat content. In addition to this, they are available in significantly larger capacity’s then other gasifiers and are the most common choice for applications downstream fuel production (Drover, 2008). 2.2.4 Gasifier Applications: Selection of a particular gasifier is also largely influenced by the coal feedstock to be processed. As shown in Figure 2, different gasifier types are more suited to certain coal ranks than others.

Figure 2: Gasifier suitability for different coal ranks (Kinaev, 2014)

Figure 2 indicates that dry fed entrained flow gasifiers are suited to a wide range of coal types, from sub-bituminous through to anthracitic coals. As such, selection of an entrained flow gasifier may be beneficial for a study aiming to compare a wide range of coal types. 2.3

Syngas Conversion Technology

Once the raw syngas product from the selected gasifier has been cleaned of contaminants it can be converted into a wide range of products through a number of different pathways. This includes the production of synthetic natural gas, ethanol production through fermentation, liquid hydrocarbon fuel production through Fischer-Tropsch synthesis and the production of a wide variety of fuels and chemicals through a methanol intermediate (Talmadge, 2013). The two most relevant pathways for liquid fuel production, Fischer-Tropsch synthesis and methanol to gasoline (MTG), are indicated in Figure 3 and discussed in the following section.

7

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

Figure 3: Potential Syngas Upgrading Pathways. Reprinted from Spath, 2003

2.3.1 Methanol to Gasoline The first step in the Methanol to Gasoline (MTG) process is the production of methanol from clean syngas. This is achieved through the hydrogenation of CO and CO2 as shown below: R6)

R7)

CO + 2H2 ⟶ CH3 OH

CO2 + 3H2 ⟶ CH3 OH + H2 O

𝛥𝛥𝑚298𝐾 = −90.8 𝑘𝑘/𝑚𝑚𝑚

𝛥𝛥𝑚298𝐾 = 49.6 𝑘𝑘/𝑚𝑚𝑚

This process was commercialised in 1923, operating at 340°C, 30-50MPa with a zinc chromium oxide catalyst (Pletcher, 1982). Since then various improvements have been made to lower the reactor pressure and improve conversion through the ICI and Lurgi processes. Methanol can be used as a standalone fuel however, in response to crude oil price rises in the 1970’s, Mobil developed the methanol to gasoline process to produce conventional transportation fuel (Dadyburjor et al., 2000). In this process, methanol is dehydrated over a zeolite ZSM-5 catalyst to dimethyl ether and then further dehydrated to hydrocarbons in the C2 – C10 range as shown below (Kaneko et al., 2000). R8)

2CH3 OH ⇋ CH3 OCH3 + H2 O

R10)

Light Alkenes ⟶ Heavier Alkenes

R9)

CH3 OCH3 ⟶ (CH2 )2 + H2 O

The overall heat of reaction for this complex sequence of reaction ranges from 1.5-1.75 kJ/g methanol depending on the specific product distribution. This process was implemented in New Zealand in 1980 with a natural gas feed for syngas generation and subsequent production of 2700m3/d of gasoline. However, the gasoline producing section of the plant was shut down when crude oil prices dropped and it became more attractive to produce methanol (Spivey, 2014). As the name suggests, the MTG process is used for the production of gasoline not low octane fuels such as diesel or jet fuel. As this study focuses on the production of jet fuel, the MTG is not considered any further here.

8

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 2.3.2 Fischer-Tropsch Synthesis:

Franz Fischer and Hans Tropsch first reported the use of iron catalysts to produce liquid hydrocarbons from synthesis gas in Germany, 1923. Originally termed the Synthol process, this is now widely known as the Fischer-Tropsch process after its founders (Spath, 2003). The mechanism for this process is a polymerisation reaction where chemisorbed methyl species are formed by the dissociation of CO and the stepwise addition of H2, which is summarised below. R11)

CO + 2H2 ⟶ --(CH2 )-- + H2 O

𝛥𝛥𝑚227𝐶 = −165 𝑘𝑘/𝑚𝑚𝑚

In addition to this, the formation of alcohols, carbonyls, carboxylic acids and water-gas shift reactions also occurs during FT Synthesis (de Klerk, 2011). FT synthesis is characterized by the large amount of heat released from the exothermic polymerization reactions as well as the production of a very wide range of hydrocarbon products (C1 – C100+) (Spath, 2003). However, as the probability of chain growth and chain termination is independent of chain length, the selectivity of hydrocarbon products can be predicted according to the AndersonSchulz-Flory (ASF) statistical distribution as follows (Dry, 2002): Wn =n.(1- α)2.αn-1 Where Wn is the weight fraction of hydrocarbons containing n carbon atoms and α is the chain growth probability, which can be determined experimentally. Although the ASF provides a reasonably accurate description of FT products, there are two main deviations that should be accounted for, these are a higher than predicted methane selectivity and a higher than predicted C2 selectivity (de Klerk, 2011). Chain growth probabilities, and therefore product distributions, are very temperature dependent. The production of gasoline and light olefins is facilitated through high temperature FT processes operating between 330-350 °C. Low temperature processes operating between 220 – 250 °C are employed for the production of waxes and diesel (Kaneko et al., 2000). Heat removal and temperature control is one of the primary considerations in FT reactor design. This has led to a number of different designs, with the four main reactor designs that have been used commercially in the past depicted in Figure 4 on the following page.

9

Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

a)

b)

c)

d)

Figure 4: Fischer-Tropsch Reactor Designs (Spath and Dayton, 2008)

The circulating (CFB) and fixed (SAS) fluidized bed reactors, shown in Figure 4a and 4b, are both high-temperature reactors which have been used commercially by SASOL (Kaneko et al., 2000). The multi-tubular fixed bed and fixed slurry bed are both low temperature reactors. In the slurry bed reactor, syngas is bubbled through a suspension of catalyst particles in a FT wax product, providing excellent catalyst contact. This design offers better temperature control, lower catalyst loading, and significantly lower catalyst attrition rates compared to fluidized bed reactors. Slurry beds also cost 75% less than the more complex, multi-tubular beds but have only recently been commercialized. If maximum diesel and jet fuel production is desired, slurry bed reactors with a cobalt catalyst are the optimum technology (Kreutz et al., 2008). 2.4

Additional Unit Operations: 2.4.1 Coal Preparation

Coal preparation requirements are dependent on the coal feedstock and the gasifier technology implemented. Coal is generally stockpiled to allow for supply disruptions to rail transportation or mining activities. This run-of-mine coal is then fed to a crusher to reduce the maximum size to 5cm (Austin, 1991). If an entrained flow gasifier is to be used, particle diameter must be further reduced to below 100 microns with a ball mill, rod mill or similar (Wills, 2006). Dry fed gasifiers require coal moisture content between 2 and 12%, which can be achieved by heating the coal slightly with a mixture of air, nitrogen and combustion gases. Slurry fed gasifiers do not require coal drying, with crushed coal instead being mixed with treated water and fine slag particles before being fed into a wet rod mill for grinding (Drover, 2008). 2.4.2 Air Separation Unit Entrained flow gasifiers almost exclusively use oxygen instead of air as their oxidant for the gasification process. This necessitates the provision of an air separation unit (ASU) unless an alternative arrangement can be made to source oxygen from a supplier. Costs associated with this unit are significant, with the ASU estimated to account for 10-15% of total capital costs and significant power requirements for indirect coal liquefaction processes (Higman, 2008).

10 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process Cryogenic ASUs are the dominant technology for air separation and involve air compression, pre-purification and drying, liquefaction and then distillation into O2 and N2 fractions. This is recognized as a reliable process, an important factor for gasification plants which are severely impacted by disruptions to oxygen supply. Cryogenic ASUs exhibit significant economies of scale, making them most cost-effective technology for producing large volumes of oxygen, and can also achieve an oxygen purity over 99% (Smith and Klosek, 2001). Pressure swing adsorption (PSA) offers an alternative for oxygen production and relies on the ability of some materials, such as zeolites, to preferentially adsorb nitrogen. Oxygen purity is approximately 93-95% and capital costs increase more rapidly with scale up than cryogenic plants, making PSA units uneconomic for large volumes (Ruthven et al., 1993). Membrane technologies can also be used for oxygen production and are classified into two distinct areas. Conventional polymeric membranes separate oxygen through a partial pressure driving force over a selective material. Unit capacities are limited to 20 TPD and provide an oxygen purity of only 40%. Ion transport membranes (ITMs) are solid inorganic crystal structures that allow the selective transport of oxygen ions at high temperatures (Higman, 2008). This is still a developing process but the need for high temperature operation offers significant potential for integration with gasification processes. 2.4.3 Water Gas Shift Syngas from gasification must be treated in order to obtain the correct H2:CO ratio required by the chosen Fischer-Tropsch reactor. This is achieved through the water-gas shift reaction (R5), which can be conducted downstream of syngas cleaning or directly after the gasifier. Water gas shift of raw syngas, also known as a sour shift, is enabled through the use of a cobalt-molybdenum catalyst, which not only tolerates sulphur in the feed but requires it to remain active (Higman, 2008). Sour shifts are generally conducted after a water quench of the syngas in a scrubber, which supplies syngas with sufficient water to conduct the shift reaction and negates the need for steam addition. As the shift reaction results in an equilibrium H2 to CO ratio exceeding that required for FT synthesis, a portion of the syngas bypasses the reactor to control this ratio to the desired level (Drover, 2008). Conventional water gas shift processes can be used for de-sulphurised syngas and may be applied at high temperature (300-500°C) with an iron oxide catalyst or low temperature (200270°C) with a copper-zinc catalyst (Drover, 2008). The exothermic nature of the shift reaction means that the equilibrium conversion to hydrogen is favoured at low temperatures. Reactors are often run in series, with an initial high temperature process to promote reaction kinetics and a subsequent low temperature process to improve conversion. 2.4.4 Acid Gas Removal Acid gas removal refers to the removal of sulphur compounds and other contaminants from the syngas to prevent catalyst poisoning in the downstream FT reactor. Potential separation processes include selective absorption in a physical or chemical solvent, adsorption on a solid or diffusion through a semi-permeable membrane (Kohl and Nielsen, 1997). Solid adsorbents

11 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process are not cost-effective for removing the high percentage of impurities in syngas from gasification and liquid solvents, which can be more easily regenerated, are preferred (de Klerk, 2011). Physical solvents incur a higher upfront capital cost but lower ongoing operating costs than chemical solvents, which require significant energy for solvent regeneration. As such, the vast majority of coal to liquid plants utilise physical solvents for syngas cleaning, in particular the Rectisol process. This process uses a pure refrigerated methanol solvent to selectively absorb and remove H2S, CO2 and other contaminants in a series of absorption and regeneration columns. Products from the Rectisol process are a clean syngas stream, concentrated CO2 stream and an H2S rich stream. This H2S stream cannot be discharged but can be used to recover elemental sulphur through the Claus process. This involves the partial combustion of H2S to SO2 followed by the reaction of this with remaining H2S to form water and elemental sulphur, which can be sold as a by-product (Higman, 2008). A portion of the high-purity CO2 can be recycled to the coal preparation plant for drying and coal transportation purposes whilst the remainder can be compressed and transported to a CO2 sequestration site if this is available (Drover, 2008) 2.4.5 Syncrude Refining Syncrude produced from FT synthesis differs from conventional crude oil in a number of ways, thus requiring a unique refinery design. Distinguishing characteristics of syncrude include high alcohol content, high alkene (olefin) content, low concentration of aromatic or cyclic compounds and the absence of contaminant compounds containing nitrogen or sulphur (de Klerk, 2011). Conventional refining technologies can be adapted to match these characteristics and allow for efficient upgrading of syncrude from FT synthesis. Syncrude is typically fractionated via distillation with C9 and heavier components then fed to a hydrotreater to convert alcohols and alkenes to alkane hydrocarbons. Long-chain waxes greater than C15 are typically fed to a hydrocracker to reduce the carbon chain to a length more desirable for final products. A wide variety of technologies exist to upgrade lighter hydrocarbon fractions including alkylation, oligomerisation and aromatisation (Furimsky, 2010). Product blending from different units is required to ensure that the final products meet the wide range of specifications imposed on transport fuels. Individual refinery designs can vary greatly according to the syncrude composition from different synthesis reactors and the final products being targeted. Various commercial designs, as well as proposed designs for different synthesis reactors and target products are outside the scope of this report but the reader is referred to the detailed discussion by de Klerk, 2011 for further information if required. 2.5

Commercial Operations

South African Synthetic Oil Ltd (SASOL) has operated commercial coal-to-liquids plants in South Africa since 1956. The original plant has since been converted to exclusively produce chemicals however an additional two plants have been built since then. In addition to this SASOL have a gas to liquids (GTL) demonstration plant in Oryx, Qatar (Kaneko et al., 2000).

12 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process Through their operations SASOL have made significant improvements to FT reactor technology, specifically through scale-up of the conventional fluidised bed reactor and development of the SAS fixed bed reactor. Shell is another major oil company that has made significant developments in FT technology, mainly through their GTL operations. They currently operate the commercial Pearl GTL plant in Qatar and a demonstration plant in Bintulu, Malaysia (Drover, 2008). Developments made at these plants are still applicable to CTL operations as once the syngas has been produced and conditioned to the required specifications, the same FT synthesis technology can be used in both GTL and CTL operations. In addition to these existing operations, there are also a number of CTL plants currently under development or coming online in China. These include the Yinchuan and Yankunag Yulin CTL plants which are currently under construction (Higman, 2013) 2.6

Previous Techno-economic Studies

Although not in an Australian context, a number of recent techno-economic studies of fuel production from indirect coal liquefaction have been undertaken. Table 1 summarises the breakeven crude oil prices found by these studies to make the ICL process viable. All breakeven prices are based on a 12% discount rate and have been converted to 2014 AUD based on U.S. CPI data and the Q1 2014 exchange rate. It should be noted that some studies refer to the price of West Texas Intermediate (WTI) crude prices whilst others refer to Dated Brent crude prices. Table 1: Breakeven Crude Oil Prices from Previous Technoeconomic ICL Studies

Study

Location

DOE (2007a) Mei (2004) Baker & O'Brien (2007) DOE (2007b) Vliet et al. (2009) Hatch (2008)

Illinois, USA Virginia, USA USA Alaska, USA Europe Alaska, USA

Breakeven Crude Price (USD/bbl) $43.00 $47.00 $55.00 $64.00 $88.00 $138.00

Breakeven Crude Price (’14 AUD/bbl) $54.00 $64.00 $69.00 $80.00 $117.00 $166.00

From Table 1 it can be seen that there is a very wide range of predicted crude prices, from $54 to $166/bbl, required to make the ICL process viable. This can be explained in part by the time when the study was conducted, the difference in location for each study and the different assumptions made during analysis. A serious cost escalation for power and process plants also occurred between 2004 and 2009 (IHS, 2013). It should also be noted that some of the studies are commercial engineering studies such as the Hatch Fairbanks report whilst others are academic papers (Mei, 2004). It is likely that predicted prices from academic papers might fail to account for the full range of costs involved with the project. The large disparity between the various estimates means that there is no one price that can be used as an indication of whether the process might be viable within Australia. It is necessary to develop independent estimates which are more applicable under Australian market conditions in order to determine what circumstances might make the process economically viable.

13 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 2.7

Summary & Research Gaps

Indirect coal liquefaction processes are comprised of a number of different well-established technologies. Although these technologies have existed for a number of years, there are relatively few commercial scale ICL processes due to economic constraints and uncertainty. As a result, there is limited information on the integration of gasification, Fischer-Tropsch synthesis and related units into a full-scale facility. Research continues to address various separate issues related to individual technologies; however some of the main research gaps are related to the process as a whole. This includes analysis of the market conditions required for a CTL process to be economically viable on a long-term basis. Recent work has addressed this in various studies however limited work has been done on this area in an Australian context. As such, this study looks to address this research gap through the preliminary assessment of an Australian indirect CTL project

14 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

3 Model Development 3.1

Reference Case

To assess the viability of an indirect CTL plant, a model was developed using Aspen Plus® process simulation software based on a reference case. The reference case used was an indirect liquefaction process design developed in a conceptual study by engineering consultant Hatch. This design incorporated the production of 20,000bbl/day of liquid fuels with additional power generation through an indirect coal gasification and FT synthesis process. The study was conducted in 2008 for a proposed development in Fairbanks, Alaska (Drover, 2008). This was selected as the reference case as it presented well-established and commercially available technologies in a widely accepted process configuration. A simplified overview of this process is shown in the block flow diagram below. Air

Cryogenic ASU

O2

CO2

Syncrude Refinery

Sulfur

N2

Coal Feed

FT Products

Syncrude

Coal Preparation

Dried Coal

Gasification Slag

Raw Syngas

Water-Gas Shift

AGR

Clean Syngas

Syngas Bypass

LTFT Synthesis Fuel Gas

Steam

Power Generation

Power

Figure 5: Block Flow Diagram of 20,000bbl/day ICL Reference Case (Drover, 2008)

3.2

Model Assumptions

A number of assumptions were made to enable the simulation of the process depicted in Figure 5. From the outset, the cryogenic ASU and power generation units were excluded from the model, as they were not the main focus of this study. It was assumed that oxygen and electric power was supplied ‘over the fence’ from a third party supplier, which is a common commercial arrangement for many plants. Modelling the conversion of H2S to elemental sulphur through the Claus process was also excluded from the scope. Detailed heat integration of the process was not conducted and it was assumed that the significant heat generation from gasification and FT synthesis reactions was sufficient to meet the heating requirements in the refinery and around the plant. Previous studies indicate that the plant is likely to require additional cooling capacity (Drover, 2008). 3.3

Model Summary

A steady-state model of the process depicted in Figure 1, adjusted for the exclusions listed above, was developed in Aspen Plus®. Boundary conditions for the model were inputs of runof-mine (ROM) coal with varying characteristics and pre-separated oxygen and nitrogen, with outputs of refined FT liquid fuels (Jet Fuel and Gasoline), CH4 as a by-product, concentrated CO2 and H2S streams and untreated water. The Peng-Robinson cubic equation of state (EOS) was used throughout the model, with the exception of solids handling applications, as this is

15 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process recommended by Aspen Plus® for gas processing and refining applications. A detailed description of model implementation and unit operating conditions can be found in Appendix B. A brief summary of the modelling approach for key unit operations is given below. 3.3.1 Coal Feedstock: The model CTL process was run with five different coal feedstocks to compare the impact of coal type on operation. These coals included the sub-bituminous Healy coal upon which the proposed Hatch CTL plant was based and four different Australian thermal coals. These four coals have previously undergone extensive laboratory analysis and pilot scale testing in an entrained flow gasifier (Roberts et al., 2011). They vary in rank from sub-bituminous to semianthracite and were selected for this initial work to cover a wide range of coal types. As such, they are also well suited to this study as experimental gasification data is available and they provide a good range of Australian coals over which the CTL process can be assessed. The characteristics of these four coals and the sub-bituminous Healy coal are detailed in Table 2. Table 2: Composition of Five Coal Feedstocks Used

Coal Code Proximate Analysis (wt%, dry) Moisture Fixed Carbon Volatile Matter Ash Ultimate Analysis (wt. % dry) Ash C H N Cl S O Ash fusion temperatures (°C)

CRC701

CRC702

CRC703

CRC704

Healy

17.4 53.8 41.3 5.0

2.4 53.2 35.8 11.1

1.3 82.8 8.1 9.1

10.3 44.4 44.7 10.9

28.3 37.3 43.3 19.3

5.0 70.2 4.5 1.4 0.0 1.1 17.8 1310

11.1 74.1 4.8 1.7 0.0 1.5 6.8 1310

9.1 76.9 3.0 8.1 0.0 0.6 2.2 >1600

10.9 70.2 5.2 0.9 0.0 0.3 12.5 1440

19.3 58.1 4.2 0.7 0.0 0.3 17.3 -

3.3.2 Coal Preparation: A ROM coal feed rate of 650 TPH for all scenarios was specified based on the Hatch reference case for a 20,000 bbl/day plant (Drover, 2008). Coal was specified as a nonconventional solid with particle size distribution and individual coals were characterised through the specification of proximate and ultimate analysis data listed in Table 2. This data is used by Aspen to calculate the coal enthalpy and density according to the respective HCOALGEN and DCOALIGT property models. ROM coal was fed through a single roll crusher and generic grinding model with screen and internal recycle to reduce the maximum particle diameter to 100 microns. Coal was combined with nitrogen upstream of a stoichiometric (RStoic) reactor and Separator, which were used to model the drying of coal by altering the proximate analysis moisture content to 5%. This dry, ground coal was then suitable for gasification.

16 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 3.3.3 Gasification: Dried coal is still specified as a non-conventional solid and must first be decomposed into its individual components, as specified by the ultimate analysis data. This is achieved through an RStoic reactor, which converts coal into conventional components recognised by Aspen for phase and chemical equilibria calculations. Decomposed coal components, in addition to a 99.5% pure O2 stream and high-pressure steam are then fed into the gasifier. No specific technology provider was selected for the gasifier and it is instead modelled as a generic entrained flow gasifier with gas quench. This is achieved through an RGibbs equilibrium reactor model, which minimises the Gibb’s Free Energy of the products formed. Modelling on the basis of thermodynamic equilibrium is acceptable due to the high operating temperature of entrained flow gasifiers (Higman, 2008). This equilibrium model implements the gasification reactions (R1- R5) listed in Section 2.1.2 to convert coal components into a mixture of syngas, CO2 and other contaminants. Close control of oxidant feed to the gasifier is essential to reach the temperatures required for operation whilst at the same time avoiding over-oxidation of coal. Required operating temperatures for entrained flow gasifiers typically range from 1400-1600 °C and the lower temperature limits are constrained by the ash fusion temperature of the coal, which must be exceeded to ensure proper slagging operation. Optimum operating temperature involves a compromise between minimising oxygen consumption and maximising carbon conversion, whilst at the same time satisfying slagging requirements. The specific operating temperature required for each coal was based off the pilot scale studies conducted by CSIRO (Roberts et al., 2011). A number of trials were run for each of the four coals and average operating temperature, carbon conversion, O:C ratio and cold gas efficiency (CGE) from these trials are shown in Table 3. Table 3: Average operating conditions from pilot scale gasification studies (Roberts, 2011)

Coal

CRC701 CRC702 CRC703 CRC704

Temperature (°C)

Conversion (%)

O:C Ratio (mol:mol)

CGE (%)

1558 1479 1441 1707

97.3 99.0 95.1 95.4

1.2 1.4 1.3 1.4

73.4 71.0 71.2 62.0

Based off the data in Table 3, a design specification was implemented to achieve the required operating temperature for each coal by varying the oxygen flow rate. The carbon conversion for each coal type was specified in the initial RStoic reactor, where coal is decomposed into conventional components recognised by Aspen. The heat duty of the gasifier was set at 3% of the higher heating value (HHV) of the input coal, as this is typical of commercial entrained flow gasifiers (Higman, 2008). 3.3.4 Syngas Quench: Rawhot syngas cannot be used directly for heat recovery due to the presence of entrained molten ash. Therefore, a direct gas quench system is employed, where cool syngas is recycled

17 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process and injected into the raw syngas stream to solidify the molten ash. Aspen Plus® does not model the melting and solidification of ash however, the quench recycle is implemented and its rate varied until a syngas temperature of 900°C, required to ensure ash solidification, is achieved. Quenched syngas is then further cooled to 250°C for heat recovery in a simplified heat exchanger. Ash and unconverted coal are removed through a simple separator model before splitting the syngas into a product stream and a recycle stream, which is compressed and combined with the incoming hot syngas. This specific quench configuration is similar to that employed in the Shell SCGP process, although many variations exist depending on the technology provider. 3.3.5 Water Gas Shift: A sour water gas shift reactor, required to increase the H2:CO ratio, is implemented directly after the syngas quench in keeping with the reference case design. As a cobalt catalyst is to be used for FT synthesis, a final H2:CO ratio of 1.9-2.1 is required (Drover, 2008). This is achieved by first splitting the incoming cool syngas stream, with a certain fraction bypassing the reactor entirely whilst the remainder is fed into a pre-heater. This counter-current heat exchanger model utilises heat contained in the shifted syngas from the exothermic shift reaction to preheat the incoming syngas to the 277°C temperature required for reactor operation (Drover, 2008). Preheated syngas is then fed into the WGS reactor along with additional steam, which is required to ensure equilibrium is reached and to prevent carbon formation on the catalyst. Above reaction temperatures of 230°C, solid carbon formation on catalysts will be minimised by keeping the H2O/CO ratio in the feed above 2 (Abdulhamid, 2007). Based on the H2O and CO content in the syngas feed, a calculator block is used to determine the steam feed rate required to achieve a ratio of 2.1, thus incorporating an operating safety margin. The WGS reactor is simulated with an REquil reactor model, which calculates the equilibrium condition for the water gas shift reaction (R5) according to the temperature approach method. It is not sufficient to simply achieve the required H2:CO ratio for FT synthesis at the WGS reactor outlet, as this will be impacted by the downstream acid gas removal and pressure swing adsorption processes. Therefore, a design specification is implemented that varies the bypass rate around the WGS reactor to achieve the required H2:CO ratio of 2 in the FT reactor feed stream. 3.3.6 Acid Gas Removal: A detailed model of the Rectisol process for acid gas removal was not developed, as this is a well-established process and simulating it was not the main objective of this study. Instead, separation efficiencies were based off those achieved in previous detailed simulations of the process obtained through correspondence with Princeton University. This gave a CO2 removal of 93.4%, 100% H2S removal to the sulphur-rich stream and recovery in the clean syngas of 96.8% of CO and 99.7% of H2. This was simulated using a simple component separator model in Aspen Plus®. Future improvements to the model when detailed heat integration is

18 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process performed would include consideration of the power requirements to be expected from the Rectisol process based on literature values. Pressure swing adsorption is required to separate a small percentage of hydrogen from the clean syngas for use in hydrotreating and hydrocracking refinery units. Again, a simple component separator has been use to model the PSA process. The amount of hydrogen removed is set by a design specification, which ensures that there is a very low concentration of hydrogen in the hydrocracker effluent to prevent excess H2 removal from the syngas. 3.3.7 Fischer-Tropsch Synthesis: A low temperature Fischer-Tropsch reactor with cobalt catalyst (Co-LTFT) is implemented, as this is the optimum technology for targeting diesel and jet fuel production. As described in Section 2.3.2, the product yields from Fischer-Tropsch reactors can be described by an ASF distribution. Bertoncini et al, 2009 characterised this distribution for low temperature cobaltbased reactors by analysing the product composition from a FT pilot plant through gas chromatography. This work provides all necessary parameters to characterise the product distribution, including chain growth probabilities (α) for alkanes, alkenes and alcohols, overall weight fractions for each hydrocarbon family and individual selectivity’s for methane and ethane. Based off this data the Fischer-Tropsch reaction was simulated using an RYield reactor model. A FORTRAN code routine was required to calculate the mass fraction of each component in the product based on the ASF distribution parameters and from this, determine the fractional yields that must be specified for each component in the RYield reactor. Syncrude product was fed into a flash drum model to simulate the phase separation between FT wax and hot vapour, which would occur within the reactor in reality. This hot vapour of unconverted syngas and lighter syncrude compounds is cooled by preheating the incoming syngas and then fed into a 3-phase separator model to produce a hot condensate, lighter vapour stream and reaction water stream containing traces of syncrude. These syncrude traces are stripped from the aqueous product and combined with the light vapour stream. This combined vapour stream is further separated into a cold condensate syncrude product, tail gas stream and additional reaction water stream. The wax stream, hot condensate and cold condensate are the three sources of syncrude for the refinery. 3.3.8 Syncrude Refinery: A detailed refinery model was not developed and product yields were instead based on a refinery design for jet fuel production from Co-LTFT syncrude proposed by de Klerk, 2011. Product distribution between jet fuel, gasoline and fuel gas was based on this design however, in order for product fuels to have appropriate composition, simulation of certain refinery units was required. Syncrude contains a significant percentage of alcohols and alkenes, which are not suitable for final fuel products. Hydrotreating of heavier syncrude components (C9+) is required to saturate these hydrocarbons with hydrogen and produce alkanes. This process was simulated with an RStoic reactor model, which implemented the reactions R12 an R13 as shown.

19 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 𝑅12)

𝑅13)

𝐶𝑥 𝐻2𝑥 + 𝐻2 ⟶ 𝐶𝑥 𝐻2𝑥+2

𝐶𝑥 𝐻2𝑥−1 𝑂𝑂 + 𝐻2 ⟶ 𝐶𝑥 𝐻2𝑥+2 + 𝐻2 𝑂

In addition to this, LTFT syncrude contains a large amount of long-chain hydrocarbon waxes that must be cracked to produce hydrocarbons of a more suitable length for transportation fuels. Simulation of this process was based off data on Shell’s severe hydrocracking technology developed specifically for the treatment of heavy syncrude components (Eilers, 1990). Results from this work included the mass fraction distribution for different carbon length chains in the product from a hydrocracking reactor. This information was used in Aspen Plus® to calculate the various yields for each hydrocarbon chain required to achieve the specified mass fraction. FORTRAN code to achieve this was adapted from that provided by Princeton University, allowing hydrocracking to be simulated with an RYield reactor model. Carbon chains heavier than C15 in the product are recycled to the reactor to undergo additional hydrocracking. In reality, hydrotreating and hydrocracking reactions can be accomplished in a single reactor by feeding the lighter components lower down in the reactor to allow hydrotreating but avoid cracking. Lighter hydrocarbon components in syncrude are upgraded through oligomerisation processes. Propane and propene can be combined through a catalytic polymerisation process to form longer chain hydrocarbons more suitable for transport fuels. Butene and iso-butane can also be reacted through an alkylation process to form heavier hydrocarbons. Hydrocarbon isomers were not included in the simulation so, for the purposes of this model, normal butane represents the iso-butane in this reaction. The polymerisation and alkylation reactions implemented are shown below (R14 and R15 respectively) and were both simulated with RStoic reactor models. 𝑅14)

𝑅15)

𝐶3 𝐻8 + 𝐶3 𝐻6 ⟶ 𝐶6 𝐻14

𝑖𝑖4 𝐻10 + 𝐶4 𝐻8 ⟶ 𝐶8 𝐻18

𝛥𝛥𝑚298𝐾 = −82.7 𝑘𝑘/𝑚𝑚𝑚

𝛥𝛥𝑚298𝐾 = −82.8 𝑘𝑘/𝑚𝑚𝑚

Average chain length hydrocarbons from hydrotreating, hydrocracking and straight run syncrude would typically be fed to an aromatisation process to produce an aromatic liquid. This has good octane properties for gasoline blending, whilst unconverted material is blended for jet fuel or sent to fuel gas. Again, to prevent the number of specified components from becoming excessive, aromatic compounds such as benzene and toluene are not included in the model and the aromatisation reactions are not modelled. Instead, recognising that the products from this process are distributed between fuel gas, jet fuel and gasoline, the exact split for each product is varied to achieve the final product yields specified in the LTFT refinery design by de Klerk, 2011. In this way, although not all refining processes have been modelled in detail, the distribution between the final products is accurate which is the main requirement for further economic evaluation.

20 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

4 Model Verification Basic model verification is required to confirm the validity of the results before undertaking any further analysis. Accurate modelling of the gasifier is of paramount importance as operating in an over or under-oxidised zone will significantly impact on syngas composition and affect all downstream units. 4.1

Gasification:

To verify gasifier operation, results were compared with reference data for the performance of a typical Australian bituminous coal in a dry-feed, entrained flow gasifier given in Higman, 2008. The model was run with CRC702, a typical bituminous coal, as feedstock and a comparison of the raw syngas composition and key operational parameters for the modelled gasifier with the reference data is shown in Table 4. Table 4: Comparison of Model Gasifier with Reference Data

Raw Syngas (Mole %) H2 CO CO2 N2 Ar H2S 3 Nm CO+H2/t coal Nm3 O2/ Nm3 CO+H2 kg steam/Nm3 CO +H2

Reference Data 33.0% 62.0% 2.0% 1.0% 1.0% 0.50% 2100 0.26 0.07

Model Data 31.2% 65.3% 2.1% 0.7% 0.1% 0.47% 2116 0.31 0.05

Absolute Error -1.8% 3.3% 0.1% -0.3% -0.9% -0.03% 16.17 0.05 -0.02

Percent Error -5% 5% 3% -32% -87% -6% 1% 18% -28%

As shown in Table 4, the syngas composition from the model closely matches the reference data with only a slightly lower H2 to CO ratio. As in the reference case the CO2 fraction in the syngas is low, indicating that the gasifier is not operating in an over-oxidised zone. Nitrogen and argon fractions are significantly lower for the model data, which would be due to a higher assumed purity for the oxidant used. However, the 99.5% pure O2 used is obtainable through a cryogenic ASU system and the lower inert gas compositions in the modelled syngas will not have any significant effect on downstream operations. Oxidant feed is higher and steam feed lower for the modelled gasifier, indicating that a greater proportion of the required oxygen is being sourced from oxidant rather than steam. The lower steam feed can also explain the slightly lower H2:CO ratio in the model syngas. Steam feed was set as a constant percentage of coal feed rate according to reported literature values and oxidant feed has been varied to achieve the required operating temperature (Kunze and Spliethoff, 2011). Optimisation of the H2:CO ratio by varying both O2 and steam feed rates is possible but is outside the scope of this study. Although improvements to the gasifier model are possible, its correlation with reference data is satisfactory for the preliminary study being undertaken here.

21 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 4.2

Fischer-Tropsch Synthesis:

Confirmation that the model accurately describes the syncrude product distribution expected from FT synthesis is also important for model verification. Typical syncrude compositions from industrial low temperature FT reactors with cobalt catalysts (Co-LTFT) are provided in de Klerk, 2011. Comparison of this data with the modelled syncrude composition is shown in Figure 6. 50%

Reference

Syncrude Composition (Wt%)1

45%

Model

40% 35% 30% 25% 20% 15% 10% 5% 0% C1

C2

C3-4

C5-10 C11-22

Alkanes

C22+

C2

C3-4

C5-10 C11-22

Alkenes

Figure 6: Comparison of Syncrude Compositon - Model Data vs Typical Co-LTFT Process

It should be noted that syncrude composition is reported as weight percent of FT products only, excluding inert gases and water gas shift products. Figure 6 shows that the model produces a lighter syncrude compared to the reference data, where a higher percentage of waxy hydrocarbons (C22+) are produced. The differences are significant enough to warrant further investigation. FT synthesis was modelled based on the experimental data of Bertoncini et al, 2009 which determined ASF distributions to describe syncrude composition based on product distribution from a pilot scale Co-LTFT reactor. As seen in Appendix C, the ASF distributions obtained were shown to accurately describe the syncrude composition obtained from the Co-LTFT reactor used. To confirm that these distributions had been successfully implemented in the Aspen Plus® model, the model syncrude composition obtained was plotted against the original ASF distributions for the three hydrocarbon families as shown in Figure 7 on the following page.

22 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

Syncrude Composition (Wt%)

3.00%

ASF (alkane) ASF (alkene) ASF (alcohol) Alkane wt% Alcohol wt% Alkene wt%

2.50% 2.00% 1.50% 1.00% 0.50% 0.00% 0

5

10

15

Carbon Number

20

25

30

Figure 7: Modelled Syncrude Composition vs ASF Distributions

As expected, the syncrude composition from the model exactly matches the ASF distributions, with the notable exceptions of C2H4 and CH4. These are well known not to follow an ASF distribution and have been modelled based on the individual component selectivity’s. Confirming that the model follows the respective ASF distributions indicates that the discrepancy shown in Figure 6 is not due to an error in model implementation and can instead be attributed to a difference in operation between the pilot scale FT reactor and the reference case given in de Klerk, 2011. Although both are Co-LTFT processes, syncrude composition can still be significantly affected by reactor technology, operation and level of catalyst deactivation. Experimental work by Bertoncini et al was conducted with a CSTR slurry reactor at 230 °C and 20 bar pressure whilst no reactor technology or operating conditions were specified in de Klerk, 2011. Different reactor types or operating conditions may be a significant factor in the different product distributions. It is also possible that the ASF distributions obtained from the pilot scale reactor are not directly applicable when scaled up to a commercial scale reactor. Future improvements to the FT synthesis model can be made by implementing ASF distributions that have been derived from the syncrude composition of commercial scale FT reactors. 4.3

Overall Model:

As a high-level check on the overall CTL model, the Healy coal upon which the Hatch Fairbanks CTL study was based was run as feedstock. Production of refined FT liquids from the model was 21,076 bbl/day compared to a total production of 20,021 bbl/day for the Hatch reference case, a 5% discrepancy. Product distribution was not comparable as the CTL model targeted the production of jet fuel and gasoline only whilst the Hatch reference plant produced diesel in addition to these. The different refinery designs may be a small factor leading to the discrepancy in total production between the two cases. Once it had been established that the model described the process with reasonable accuracy, further analysis could be undertaken on the results obtained.

23 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

5 Results: The verified model was run with the five different coals listed in Table 2 and the key results from each case are shown in Table 5. Model operating conditions were kept constant in each case except for the coal properties (composition and grindability) and gasification conditions. Gasifier temperature was specified based on the CSIRO pilot studies and varies for each coal based on their specific slagging requirements as shown in Table 3. The carbon conversion for each coal type that was obtained at these temperatures also varies for each case. The oxidant flow to the gasifier is varied to obtain the required temperature and the steam flow is set based on the carbon content of the incoming coal, therefore these parameters both change between the cases. Expected carbon conversion of the Healy coal was not known and was assumed equal to the conversion of the most similar coal from the pilot studies, CRC701.

Production

Gasification

Coal Prep

Table 5: Key Results from CTL Model run with 5 different coals

Parameter Coal Code Coal Feed Grindability Crusher/Mill Power Carbon Conversion Gasifier Temperature Gasifier Heat Duty O:C Ratio Oxidant Flow Steam Flow Syncrude Jet Fuel Gasoline Total FT Product Fuel Gas CH4 Produced CO2 Emitted

Units tonne/hr HGI MW % C MW mol/mol tonne/hr tonne/hr tonne/hr bbl/day bbl/day bbl/day tonne/hr GJ/hr tonne/hr

CRC701 CRC702 651 651 27 45 24.8 15.7 97.3 99 1550 1480 121 148 1.14 1.02 424 524 50 61 157 212 19,484 26,373 10,301 13,942 29,785 40,315 5.19 7.02 762 1039 711 814

Scenario CRC703 CRC704 651 651 74 76 10.0 9.8 95.1 95.4 1440 1710 148 132 0.98 1.19 479 504 115 53 207 170 25,792 21,109 13,635 11,160 39,427 32,269 6.87 5.62 1037 824 827 749

Healy 651 45 15.7 97 1550 74 1.22 320 36 111 13,787 7,289 21,076 3.67 538 516

As seen in Table 5, the volume of refined FT liquids produced varies significantly between each case, ranging from 21,076 to 40,315 bbl/day. This can be explained by different feedstock characteristics, with a wide variety of coal ranks used. The Healy coal is a subbituminous coal characterised by high moisture and ash content, and as such results in the lowest production of FT liquids. CRC702 and CRC703 are both high carbon, low moisture and ash content coals and as such produce significantly higher volumes of refined FT liquids. The O:C ratio required for the gasification of these coals is lower than the other cases however, the total amount of oxidant required is higher due to the higher carbon content. It should be noted that a significant amount of concentrated CO2 from the AGR process is produced for each case. The majority of this CO2 is produced in the WGS process, where CO is shifted to CO2 and H2 to obtain the required H2:CO ratio. This CO2 production is a

24 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process significant environmental concern for the process; however the fact that it is a concentrated stream (96 mol%) is beneficial if CO2 sequestration or enhanced oil recovery options were to be pursued. Given the current political uncertainty in Australia regarding the taxation of CO2 emissions and the regulatory challenges facing the introduction of carbon sequestration, these options have not been investigated further in this study. However, appropriate handling of CO2 emissions would be required for the development of a sustainable coal to liquids project. The model results listed in Table 5 formed the basis for estimation of operating costs and revenue for the process. This was required as part of the economic analysis conducted to determine the rate of return and breakeven product prices for the CTL plant under each scenario.

6 Cost Estimation An economic analysis of the modelled CTL process was conducted to provide preliminary estimates of the viability of this process within Australia. This required estimation of plant capital costs for each scenario as well as operational costs based on the model data. 6.1

Capital Cost Estimation:

Capital cost estimates for the process were based off the order-of-magnitude estimate developed in the Hatch Fairbanks study for the 20,000 bbl/day reference plant. This was prepared using a combination of quoted, estimated and factored costs for various equipment items and as such the accuracy for the overall estimate is quoted as ±40% (Drover, 2008). The estimate includes both direct and indirect costs, with the basis of the estimates for these two areas listed below: 6.1.1 Direct Cost Estimates: Direct costs for each plant area include: 1. The design, procurement, delivery, installation and commissioning of all main process units, utilities and civil works. 2. The design, procurement and erection of a control room and all plant buildings. 3. Site fencing and security during construction. 6.1.2 Indirect Cost Estimates: Indirect cost estimates for the project include: 1. Engineering, Procurement & Construction Management (EPCM) fees at 9% of direct costs. 2. License fees to all licensors. 3. Project contingency of 15% 4. Location factor of 15% over Gulf Coast priced equipment due to Alaskan location.

25 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 6.1.3 Exclusions: The capital cost estimate for the plant excluded the following items: 1. Cost of acquiring land on which to build the plant. 2. Any roads, railways or pipelines required outside the plant boundaries. 3. Financing costs, working capital allowance and insurance. 6.1.4 Capital Cost Conversion: The total capital cost estimate for the 20,00 bbl/day plant, including direct and indirect costs and excluding items listed above was $4,146 million in 2008 US dollars. This was converted to a capital cost estimate for each of the five scenarios analysed in this study through the following process. 1. Removal of direct costs for the air separation unit and integrated power plant as these are both excluded from the developed model. 2. Application of a capacity factor to direct costs for all units downstream of the gasifier based on the relative amount of FT liquids produced compared to the reference case. 3. Re-calculation of EPCM and contingency costs at the same level of 9% and 15% of Total Direct Costs respectively. Conversion to Feb. 2014 USD based on the Chemical Engineering Plant Cost Index (CEPCI). 4. Conversion to AUD based on the average Q1 2014 exchange rate.

The resulting capital cost estimates for the five different scenarios are shown in Table 6. Table 6: Capital Cost Estimates for Five Operating Scenarios

Capacity Factor DIRECT COSTS (Million USD) Feedstock Preparation Air Separation Island Gasifier Island Gas-Shift Island* Gas Purification Island* FT Synthesis & Refinery* Power Plant Utilities* Offsites Total Direct Costs INDIRECT COSTS (Million USD) Engineering & Supervision License Fees Project Contingency Total Indirect Costs

Reference 1.0

CRC701 1.41

CRC702 1.91

CRC703 1.87

CRC704 1.53

Healy 1.0

$61 $386 $1,131 $112 $308 $372 $352 $368 $81 $3,170

$61 $1,131 $158 $436 $526 $519 $81 $2,911

$61 $1,131 $213 $590 $712 $703 $81 $3,491

$61 $1,131 $209 $577 $697 $688 $81 $3,442

$61 $1,131 $171 $472 $570 $563 $81 $3,048

$61 $1,131 $112 $308 $372 $368 $81 $2,432

$285 $150 $541 $976

$262 $150 $499 $911

$314 $150 $593 $1,057

$310 $150 $585 $1,045

$274 $150 $521 $945

$219 $150 $420 $789

26 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process Total (Million '08 USD)

CEPCI (2008 Ave.) CEPCI (Feb 2014) CEPCI Ratio Q1 '14 Exchange Rate PLANT COST (Million '14 AUD)

$4,146 575.4 575.0 0.999 1.09 $4,527

$3,822 575.4 575.0 0.999 1.09 $4,173

$4,548 575.4 575.0 0.999 1.09 $4,966

$4,487 575.4 575.0 0.999 1.09 $4,899

$3,993 575.4 575.0 0.999 1.09 $4,360

$3,221 575.4 575.0 0.999 1.09 $3,517

Note: Items marked with * are the units which have been adjusted according to the capacity factor.

Table 6 shows that removal of the air separation and power generation units significantly reduces the direct costs of the project. However, this is partly offset and in some case exceeded by the increase in direct costs resulting from the increased capacity required for downstream units with higher quality coals. Adjustment for the CEPCI has minimal impact on the capital cost in this case. No location factor has been applied to the capital cost as a 15% location factor and 40% labour productivity factor was applied in the original cost estimate from Hatch. Multiple sources list a location factor for Fairbanks, Alaska that is higher than for Australia (Richardson, 2010, DOD, 2014). However, the location factors listed in these sources for Alaska relative to the Gulf Coast are well in excess of the 15% implemented by Hatch. It is not the objective of this study to try and improve the initial capital cost estimates and they are taken as correct for an Alaskan location, hence a location factor for Australia is not applied. However, it appears that the location factor of 15% applied for the initial estimate may have been optimistic. As such, the capital cost estimates presented here are considered conceptual and likely to have an accuracy range of at least -10% / +50% (Greig, 2014). 6.2

Operating Costs:

Operating cost estimates for the different scenarios are comprised of fixed operating expenditure, based on estimates from the Hatch Fairbanks study, and variable operating costs determined by model data and predicted commodity prices. 6.2.1 Fixed Operating Costs: An order-of-magnitude estimate for the fixed operating expenditure associated with the 20,000 bbl/day reference case is provided in the Hatch Fairbanks report (Drover, 2008). This includes maintenance, chemical and catalyst costs for individual plant areas as well as sales, general, administrative and labour expenses for the facility as a whole. A similar procedure to that used for capital expenditure was used to determine fixed operating cost estimates for each scenario based on the reference case data. Costs associated with the air separation and power generation units were removed, downstream unit costs were adjusted according to the relevant capacity factor and all costs were converted from USD to AUD. This process is summarised in Appendix D, which gives the resulting fixed operating cost estimates for each scenario. 6.2.2 Variable Operating Costs: Model results provided the coal and utility feed rates and associated production rates of liquid fuels for each scenario, thus allowing expected variable operating costs and sales revenue to be determined. As the process would not be operating at modelled rates throughout the year, a

27 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process capacity factor, defined as the fraction of maximum operation per year, is applied. A capacity factor of 0.85 is assumed based on a previous techno-economic study of a similar CTL process (Mantripragada, 2008). This factor is applied to all model input and production rates. In order to determine the variable costs and revenues expected from the model operating data, estimated commodity prices are required for the life of the plant. It is recognised that these prices can be inherently unpredictable however, all efforts were made to source conservative predictions from reputable sources. In addition to this, a sensitivity analysis was conducted as discussed in Section 7.1 to determine the impact that a variation in these prices would have on the process. The basis upon which individual prices were estimated is detailed below. All estimations are for real prices (exclusive of inflation) in 2014 AUD. Coal Prices: Predicted prices for a typical Australian thermal coal were obtained from The World Bank’s “Commodity Markets Outlook – January 2014” report (Baffes and Cosic, 2014). This provides predicted prices until 2025 in real 2010 U.S. dollars. Predictions were converted to 2014 USD according to the historical U.S. Consumer Price Index (CPI) and then converted to AUD according to the average Q1 2014 exchange rate (US BLS, 2014). These prices are for a standard thermal coal with a higher heating value (HHV) of 26.8 GJ/tonne. As is common practice when estimating coal prices, this was converted to a price for each of the five coal feedstocks used based on the relative magnitude of the coals HHV. The higher heating value of each coal was calculated based on their proximate and ultimate analysis. Calculations can be seen in Appendix E1 and provided price estimates for each of the 5 coals until 2025. In the absence of any extended predictions, these prices were assumed to stabilise at the 2025 values. Electricity Prices: Through consultation with the University of Queensland Energy Initiative it was determined that the CTL process being modelled was unlikely to receive wholesale electricity prices as the power demand of the process was not high enough. However, the plant would be likely to receive a discount on retail prices and, based on prior experience, the electricity price to be incurred in 2015 was estimated at $145/MWh under an optimistic scenario. This price is expected to grow by 0.3% in real terms based on a national economic forecast commissioned by the Australian Energy Market Operator (AEMO) (NIEIR, 2013). Natural Gas Prices: Natural gas is produced as a minor by-product of FT synthesis and the revenue from this byproduct must be accounted for. The Australian Energy Regulator lists the average Q1 2014 gas price as $4.38/GJ and it was assumed that this increases throughout the project at a rate of 1% per year as predicted in the NIEIR national economic forecast (NIEIR, 2013).

28 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process FT Product Prices: Forecasts of the future prices for the FT fuels produced (jet fuel and gasoline) could not be sourced for the Australian market. However, predicted prices for Dated Brent Crude Oil, which is being increasingly used as a pricing benchmark in the Asia-Pacific region, are available until 2025 (Platts, 2011, Baffes and Cosic, 2014). Although not a perfect relationship, the retail prices for refined products are closely linked to the price of Dated Brent Crude. In order to develop predictions for refined product prices, historical prices of gasoline and diesel were compared with historical prices of Dated Brent crude oil. Prices for gasoline and diesel are given as terminal gate prices by the Australian Petroleum Institute (AIP), which include a fuel excise tax and goods & services tax (GST). The current fuel excise tax on both gasoline and diesel in Australia is $0.38143/L and a GST of 10% is applied on top of this (ATO, 2014). These tax components were removed to provide an estimate for the average refinery price received for diesel and gasoline, neglecting the transportation cost component included in terminal gate prices. Figure 8 shows this average price, as well as individual TGPs and Dated Brent crude oil prices since 2004. 290

Diesel (TGP) Dated Brent

Gasoline (TGP) Ave. TGP (excl. tax)

Price (AUD/bbl)

240

190

140

90

40

2004

2005

2006

2007

2008

2008

Year

2009

2010

2011

2012

2013

Figure 8: Dated Brent, Gasoline & Diesel Prices over the past 10 years (AIP, 2014)

In order to develop a relationship for the average refinery price received, this price was plotted as a function of crude oil prices since 2009. Earlier data was neglected to exclude the effects of the 2008 oil crisis. A linear regression of average refinery price vs crude oil price can be seen in Figure 9 on the following page.

29 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 160

Refined Product Price (AUD/bbl)

150 140 130

y = 1.0367x + 22.82 R² = 0.8856

120 110 100 90 80 60

70

80

110 90 100 Crude Oil Price (AUD/bbl)

120

130

Figure 9: Average Refined Product Price vs Crude Oil Price since 2009

The linear regression in Figure 9 gives the following relationship between refined product prices excluding taxes and Dated Brent crude oil prices. 𝐴𝐴𝐴𝐴𝐴𝐴𝐴 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑃𝑃𝑃𝑃𝑃 = 1.04 × 𝐷𝐷𝐷𝐷𝐷 𝐵𝑟𝑟𝑟𝑟 𝐶𝐶𝐶𝐶𝐶 𝑃𝑃𝑃𝑃𝑃 + $22.82

The slope of the linear regression is close to 1, indicating a fairly constant refinery margin as would be expected. This linear relationship was used to develop predictions for average refined product prices based on predicted Dated Brent Crude prices from the World Bank (Baffes and Cosic, 2014). These crude oil price predictions, provided until 2025 in real 2010 USD, were adjusted to 2014 USD according to historical U.S. CPI data and converted to AUD according to the average Q1 2014 exchange rate. Expected refined product prices could then be calculated according to the relationship developed with all calculations shown in Appendix E2. It was assumed that all FT products produced by the CTL process received this same average refinery price. This is a close approximation for diesel and gasoline, as seen in Figure 8 and historical Australian jet fuel prices were not available for comparison in this study. In the absence of price forecasts beyond 2025, it was assumed that all prices stabilised at the 2025 value going forward. Oxygen Costs: The cost of oxygen was based on the usual commercial arrangement indicated by Praxair, a leading provider of cryogenic ASUs, when operating an external plant for over-the fence supply The cryogenic ASU would be owned and operated by Praxair whilst the CTL plant would pay a fixed monthly facility fee and a variable cost depending on the volume of oxygen produced (Wilcox, 2005). The fixed fee is in the range of 2.75% of the ASU capital cost per month, which covers maintenance, repair and back up supply costs. The ASU capital cost is estimated at $27 million per 1000 short TPD of capacity in 2004 USD. This can then be scaled to the capacity required for each of the five scenarios according to the following formula (Wilcox, 2005). 𝑠ℎ𝑜𝑜𝑜 𝑇𝑇𝑇 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 0.6 𝐴𝐴𝐴 𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐶𝐶𝐶𝐶 = $27 𝑚𝑚𝑚𝑚𝑚𝑚𝑚 × � � 1000 𝑠ℎ𝑜𝑜𝑜 𝑇𝑇𝑇

30 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process The total ASU capital cost was adjusted to 2014 dollars according to the relative CEPCI values and then converted into AUD based on the Q1 2014 exchange rate. A location factor of 1.3 was applied to account for construction in Australia compared to a U.S. location (Richardson, 2010). This capital cost estimate can be used to calculate the fixed fee paid to the oxygen supplier, which is 2.75% of the total capital cost per month. The variable price charged for oxygen production is based on the electricity consumption of the process. For gasifier applications, the power consumption for the ASU is estimated at 16 kWh per 1000 cubic feet (CF) of oxygen produced at normal temperature and pressure (NTP) (Wilcox, 2005). The oxygen requirements for each scenario are known from the process model and electricity prices have been predicted over the life of the plant. From this, the variable costs associated with the ASU can be calculated as follows. 𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 𝐶𝐶𝐶𝐶 𝑝𝑝𝑝 𝑦𝑦𝑦𝑦 =

16 𝑘𝑘ℎ 0.145 𝐴𝐴𝐴 × 𝐶𝐶𝐶 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 × 1000 𝐶𝐶 1 𝑘𝑘ℎ

Where CFY required is the cubic feet per year of oxygen required according to the process model, adjusted by the assumed plant capacity factor of 85%. Calculations of these variable costs and fixed facility fees for each scenario can be seen Appendix E3 and allow the total cost of oxygen for the CTL plant to be determined. Price predictions for all relevant inputs and products of the CTL allowed variable operating costs and revenue to be calculated for the process based on model data. This, in conjunction with capital cost estimates discussed in Section 6.1 allow for a detailed economic analysis of the process under each of the five scenarios.

7 Economic Analysis An economic analysis of the modelled CTL process was conducted to determine the breakeven FT product price and internal rate of return for each of the five scenarios. A number of assumptions regarding operating and financial conditions were required to achieve this and these are detailed below. 1. 30 year operating life for the plant. 2. Plant capacity factor of 85%. (Process runs at 85% capacity on average during the year) 3. Construction is completed over 4 years and all capital expenditure is allocated as follows over these 4 years: 10%, 20%, 50%, 20%. (Drover, 2008). 4. Double-declining balance depreciation method used to depreciate plant over its 30year life. Plant has no salvage value. 5. Corporate tax rate of 30%. 6. Discount rate of 12%. 7. Neglecting all financing costs (eg. interest repayments). These would be significant for such a large capital expenditure and neglecting these means that total cost estimates will be optimistic.

31 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 8. Working capital required is neglected. 9. Sales price (exclusive of GST and fuel excise tax) is equal for jet fuel and gasoline produced. 10. Future commodity prices are assumed to follow the predictions detailed in section 6.2.2. Based on these assumptions it was possible to calculate the yearly Earnings Before Interest and Tax (EBIT) for each scenario, including fixed and variable operating costs, revenue from fuel and natural gas sales and a depreciation tax shield. Corporate tax of 30% of EBIT was applied to calculate yearly net profit. Being a non-cash flow, depreciation expenses were added back to this net profit to determine the yearly free cash flow (FCF) for the project. These FCFs were discounted back to 2015 dollars using the 12% discount rate for the project. The sum of these discounted FCFs gave the net present value (NPV) for the project under each scenario. The discount rate required to achieve zero NPV, also known as the internal rate of return (IRR) for the project, was then calculated. In addition to calculating the projects IRR, the breakeven price for FT products was also calculated. This was done by neglecting the previous price predictions for FT products and instead assuming a constant price throughout the life of the project. This constant price was varied to find the breakeven price that gave a NPV of zero with a 12% discount rate. The Equivalent Crude Oil Price (ECOP) at this breakeven price for FT products was then calculated according to the relationship developed in Section 6.2.2. These calculations were performed for each of the five scenarios and example calculations for CRC701 can be seen in Appendix F. A summary of the key results for each scenario can be seen in Table 7 on the following page. Table 7: Key results from Economic Analysis of 5 Scenarios

Scenario 2015 Coal Price FT Product Net Present Value Internal Rate of Return Breakeven FT Price Gasoline Sales Price* Jet Fuel Sales Price * ECOP

Units $/tonne bbl/day 106 $ % $/bbl $/L $/L $/bbl

CRC701 81.28 29,785 -$1,670 5.0% $165 $1.56 $1.25 $137

CRC702 109.54 40,315 -$1,393 7.4% $147 $1.44 $1.13 $120

CRC703 116.03 39,427 -$1,435 7.2% $149 $1.45 $1.14 $122

CRC704 94.55 32,269 -$1,747 5.0% $163 $1.55 $1.24 $135

Healy 59.12 21,076 -$1,816 2.3% $189 $1.73 $1.41 $160

*

Note that gasoline and jet fuel sales price are calculated from the breakeven FT product price with the fuel excise tax and GST included. Current fuel excise taxes in Australia are $0.38143/L for gasoline and $0.09835/L for kerosene when used as fuel in an aircraft (i.e. jet fuel) (ATO, 2014). Goods & Services Tax (GST) is an additional 10% on top of the fuel excise.

Table 7 shows that, for all the scenarios considered, development of the CTL process yields a large negative net present. As such the internal rate of return (IRR) for each scenario is less than the required discount rate of 12%. This indicates that the CTL process considered is not currently viable in Australia at the assumed conditions and future prices. Table 7 also shows

32 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process the breakeven FT product price that would be required over the life of the project in order to achieve zero NPV at a discount rate of 12%. These prices can be broken down into their individual levelised cost categories, as shown in Figure 10 below. 200

Breakeven FT Product Price ($/bbl)

180 160 140

Electricity

120

Oxygen Cost

100

Fixed Opex

80

Coal Cost

60

Capital Cost

40

Current Price

20 0 CRC701

CRC702

CRC703 Coal Type

CRC704

Healy

Figure 10: Contribution of Plant Costs towards Breakeven FT Product Prices

Figure 10 shows that capital expenditure and coal feedstock costs are the main components making up the breakeven price for FT fuels produced. It is also shows that the relative contribution of capital costs increases for lower rank coals (CRC701 and Healy) as the same gasifier and coal preparation equipment is required for lower fuel production. The equivalent crude oil price required to achieve a market price for FT fuels equal to their breakeven price for each scenario is also given in Table 7. This ranges from $120-160/bbl depending on the scenario, and this entire range is above the current Dated Brent crude oil price of AUD116.12/bbl as at 06/06/2014 (Bloomberg, 2014). The market price for refined fuels that would be expected from this crude oil price is indicated on Figure 10. Although this current market price is not far from the breakeven prices at the lower end of the range, the price would have to remain above this level for the entire life of the plant in order for the project to remain viable. The effect of changing prices and other key variables on the process economics was investigated further through a sensitivity analysis.

33 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 7.1

Sensitivity Analysis:

A sensitivity analysis was conducted by independently varying key inputs to the economic model by ±30% of their base value and determining the effect on the IRR for each scenario. The key parameters varied were capital expenditure and fixed operating costs as well as the prices for coal, FT products, electricity, natural gas and oxygen supply. As discussed in Section 6.1, capital costs may vary by more than 30%, however all parameters were varied by the same percentage for consistency. Sensitivity analysis results for the four main scenarios can be seen in Figure 11. 15%

15%

CRC701

13%

9%

9%

7%

7% IRR (%)

11%

IRR (%)

11%

5% 3%

5% 3%

1%

1%

-1%

-1%

-3%

-3%

-5% -30%

-20%

-10%

15%

0%

% Change

10%

20%

-5% -30%

30%

9%

7%

7% IRR (%)

9% IRR (%)

11%

5% 3%

-1%

-1%

-3%

-3% 0%

% Change

10%

20%

30%

10%

20%

30%

10%

20%

30%

CRC704

3% 1%

-10%

0%

% Change

5%

1%

-20%

-10%

13%

11%

-5% -30%

-20%

15%

CRC703

13%

CRC702

13%

-5% -30%

-20%

-10%

0%

% Change

Figure 11: Effect of Percentage Change in Key Prices on Projects Internal Rate of Return

As seen in Figure 11, the internal rate of return for the project under all four scenarios is clearly most sensitive to changes in FT product prices over all other variables. Total capital expenditure and coal prices also have a marked effect on the projects IRR but not of the same magnitude as FT product prices. The impact of all other variables investigated is largely insignificant compared to these three major effects. All numerical results from the sensitivity analysis and graphical results for the Healy coal can be found in Appendix G. The high sensitivity to FT product prices reinforces the importance that crude oil prices have on the viability of the CTL process. Although breakeven FT product prices have been identified and the crude oil prices required to achieve these has been estimated, waiting for these prices to be reached is not enough to ensure the viability of the process. Their needs to be a reasonable degree of certainty that crude oil prices will remain above these levels over

34 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process the life of the project. As seen in Figure 11, small changes to these prices and the resulting price for refined products will have a very significant impact on the viability of the CTL process. 7.2

Coal Selection:

If market conditions were to change and improve the viability of the CTL process, selection of the most appropriate coal feedstock is an important consideration. The coals considered in this study range from sub-bituminous to semi-anthracite and as such vary widely in carbon content and higher heating value (HHV). Table 7 shows that running CRC702 or CRC703 coals as feedstock gives the lowest breakeven FT product price and the highest rate of return for the project. These are both high carbon content, high heating value coals. Figure 12 below shows the breakeven price for all five coals as a function of their effective carbon content. Effective carbon content is simply the total carbon content of the coal multiplied by the gasifier carbon conversion.

Breakeven FT Price ($/bbl)

200 190 180

y = -127.41x + 239.17 R² = 0.9892

170 160 150 140 40%

45%

60% 65% 50% 55% Effective Carbon Content of Coal (%)

70%

75%

Figure 12: Breakeven FT Product Price vs Effective Carbon Content of Coal

Figure 12 shows that as the effective carbon content of the coal feedstock increases, the process can produce FT products at a lower breakeven price. This is to be expected as higher carbon content coals will produce a larger volume of liquid products if effective gasification is achieved. However, these high carbon content coals are also generally more expensive due to their higher heating value. Figure 8 suggests that for a coal-to-liquids process, it is preferable to use a more expensive, high carbon content coal as feedstock to produce greater volumes of fuel rather than a cheaper, lower carbon content coal. Under current market conditions however, none of the coal feedstocks appear to offer a viable coal-to-liquids process for Australia. This study suggests that if these conditions were to change, high carbon content coals should be strongly considered as feedstocks for the CTL process. Appropriate coal selection can significantly improve the economics of a CTL process and could prove the difference in ensuring the viability of the process if market conditions were to change in the future.

35 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

8 Conclusions An indirect coal liquefaction (ICL) process for the production of transportation fuels from coal was modelled to provide a preliminary assessment of its viability within Australia. This techno-economic model was based on a commercially proposed coal-to-liquids project in Fairbanks, Alaska that was produced by engineering consultant Hatch. Process modelling was conducted with Aspen Plus® simulation software and focuses on the coal preparation, gasification, syngas conditioning, Fischer-Tropsch synthesis and syncrude refining processes. Once verified, the model was run with four different Australian thermal coals as feedstock to assess the impact of coal type on the process. These coals range in rank from sub-bituminous to semi-anthracite and were selected as they represent a wide range of different coals and have been subject to previous pilot scale gasification trials (Roberts et al., 2011). Simulation results showed that coal type has a significant impact on the volume of FT liquids produced, ranging from 21,000bbl/day to 43,000bbl/day. Under all scenarios considered significant amounts of CO2 were produced, ranging from 3.9 - 4.8kg per kg liquid fuel produced. Higher carbon content coals (CRC702 and 703) yielded the largest volumes of liquid fuels and had the lowest relative CO2 production. The high concentration CO2 stream produced is well suited for sequestration or enhanced oil recovery applications however these have not been considered here due to issues surrounding their implementation. Future studies should focus on determining the impact of these options on the economics of the CTL process, which can be done by adjusting the model presented here. Without this component, the process is unlikely to ever receive governmental or community support. The economics of the current CTL process were assessed here and the project was shown to yield a large, negative net present value with an internal rate of return ranging from 2.3 to 7.4% for the different scenarios, well below the discount rate of 12%. This indicates that, under the assumptions made in this study, an indirect coal liquefaction process is not currently viable within Australia under the current market conditions. The breakeven crude oil price required for the process to be viable was found to vary from $120-$160/bbl depending on the scenario. The lower end of this range is not far from the current market price of $116/bbl, however prices would have to remain elevated throughout the life of the project for it to be viable. Sensitivity analysis showed that CTL process economics are most sensitive to a change in product prices and therefore fluctuations in crude oil prices can significantly affect the viability of the process. If such a change were to improve the prospects for the CTL process in the future, results showed that high carbon content coals offer the most economically attractive feedstock. The premium paid for the higher heating value of the coal is more than offset by the increased liquid fuel production and these coals offer a significantly lower breakeven sales price. However, at this point in time none of the coals trialled offered a viable CTL process under current market conditions and it appears this will remain the case unless crude oil prices were to undergo sustained increases in the future.

36 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

9 References ABDULHAMID, H. 2007. Report on Catalytic WGS Reaction Studies. Clean Hydrogen-rich Synthesis Gas ChrisGas. AIP 2014. TERMINAL GATE PRICES (WHOLESALE). Canberra: Australian Institute of Petroleum. ATO. 2014. Excise Tariff Working Pages [Online]. Available:

http://law.ato.gov.au/atolaw/view.htm?docid=PAC/BL030002/1 [Accessed May 2014]. AUSTIN, L. 1991. Size Reduction. Coal Preparation. Pennsylvania: SME. BAFFES, J. & COSIC, D. 2014. Commodity Markets Outlook - January 2014. The World Bank. BAKER & O'BRIEN 2007. High Quality Competitive Distillate Fuels from Coal-to-Liquids Processing. NPRA Annual Meeting. BCA 2013. Securing Investment in Australia's Future. Business Council of Australia. BELL, D., TOWLER, B. & FAN, M. 2011. Coal Gasification and its Applications, Oxford, Elsevier. BLOOMBERG. 2014. Energy & Oil Prices [Online]. Available: http://www.bloomberg.com/energy/ [Accessed June 2014]. DADYBURJOR, D. B., LIU, Z. & DAVIS, B. H. 2000. Coal Liquefaction. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons, Inc. DE KLERK, A. 2011. Fischer-Tropsch Refining, Germany, Wiley-VCH. DOD 2014. DOD AREA COST FACTORS (ACF). Department of Defence. DOE 2007a. Baseline Technical and Economic Assessment of a Commercial Scale Fischer-Tropsch Liquids Facility: Final Report. Department of Energy. DOE 2007b. Alaska Coal Gasification Feasibility Studies – Healy Coal-to-Liquids Plant: Final Report. Department of Energy. DROVER, B. 2008. Fairbanks Economic Development Corporation - Coal-to-Liquids FEL1 Study. Hatch. DRY, M. 2002. The Fischer-Tropsch Process: 1950 - 2000. Catalysis Today. FURIMSKY, E. 2010. Catalysis in the Refining of Fischer-Tropsch Syncrude. Royal Society of Chemistry. GREIG, C. 11/06/2014 2014. RE: Review of Capital Cost Estimations with UQ Energy Initiative. Type to SCOTT, D. HAN, S. & CHANG, C. 2008. Synthetic Liquid Fuels. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons. HIGMAN, C. 2008. Gasification, Burlington, Elsevier. HIGMAN, C. 2013. State of the Gasification Industry – The Updated Worldwide Gasification Database. Schwalbach, Germany: Higman Consulting GmbH. IHS. 2013. Power Capital Cost Index [Online]. Available:

http://www.ihs.com/info/cera/ihsindexes/index.aspx [Accessed 06/05/2014. KANEKO, T., DERBYSHIRE, F., MAKINO, E., GRAY, D., TAMURA, M. & LI, K. 2000. Coal Liquefaction. Ullmann's Encyclopedia of Industrial Chemistry. Wiley-VCH Verlag GmbH & Co. KGaA.

37 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process KINAEV, N. 2014. RE: CTL Undergraduate Thesis Meeting. KOHL, A. & NIELSEN, R. 1997. Gas Purification, Houston, Gulf Publishing Company. KREUTZ, T., LARSON, E., LIU, G. & WILLIAMS, R. 2008. Fischer-Tropsch Fuels from Coal and Biomass. Pittsburgh: Princeton Environmental Institute. KUNZE, C. & SPLIETHOFF, H. 2011. Modelling, comparison and operation experiences of entrained flow gasifier. Energy Conversion and Management, 52, 2135-2141. LI, F. & FAN, L.-S. 2008. Clean Coal Conversion Processes - Progress and Challenges. Energy & Environmental Science. LIU, Z., SHI, S. & LI, Y. 2010. Coal liquefaction technologies—Development in China and challenges in chemical reaction engineering. Chemical Engineering Science, 65, 12-17. LOWENBERG, A. 1992. The Origins and Demise of South African Apartheid: A Public Choice Analysis, Michigan. MANTRIPRAGADA, H. 2008. CO2 reduction potential of coal-to-liquids (CTL) plants. Department of Engineering and Public Policy: Carnegie Mellon University. MEI, H. 2004. Market Risk Analysis of Coal Liquefaction. Master of Science, West Virginia University. NETL. 2014. Commercial Gasifiers [Online]. Available:

http://www.netl.doe.gov/research/coal/energy-systems/gasification/gasifipedia/typesgasifiers [Accessed 06/05/2014. NIEIR 2013. National and state economic forecasts to 2034-35 consistent with AEMO scenarios 2012. Victoria: National Institute of Economic and Industry Research. PLATTS 2011. Dated Brent: The Pricing Benchmark for Asia-Pacific Sweet Crude Oil. Market Issues: Oil. PLETCHER, D. 1982. Indirect Coal Liquefaction Processes. Environmental Protection Agency. REZAIYAN, J. 2005. Gasification Technologies: A Primer for Engineers and Scientists, Boca Raton, Taylor & Francis Group. RICHARDSON 2010. Richardson International Construction Factors Manual. Nevada: Richardson Products. ROBERTS, D., HARRIS, D., TREMEL, A. & ILYUSHECHKIN, A. 2011. Linking laboratory data with pilot scale entrained flow gasification performance. Fuel Processing Technology. RUTHVEN, D., FAROOQ, S. & KNAEBEL, K. 1993. Pressure Swing Adsoprtion, VCH New York. SHADLE, L. J., BERRY, D. A. & SYAMLAL, M. 2000. Coal Conversion Processes, Gasification. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons, Inc. SIMBECK, D. 1993. Coal Gasification Guidebook: Status, Application, and Technologies. California: Electric Power Research Institute. SMITH, A. R. & KLOSEK, J. 2001. A review of air separation technologies and their integration with energy conversion processes. Fuel Processing Technology, 70, 115-134. SPATH, P. 2003. Technical and Economic Assessment of Synthesis Gas to Fuels and Chemicals with Emphasis on the Potential for Biomass-Derived Syngas. Springfield: National Renewable Energy Laboratory. SPATH, P. & DAYTON, D. 2008. BIOPPRODUCTS FROM SYNGAS NREL.

38 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process TALMADGE, M. 2013. Syngas Upgrading to Hydrocarbon Fuels - Technology Pathway . Denver: National Renewable Energy Laboratory. THE WORLD BANK. 2014. World Databank - Global Economic Monitor (GEM) Commodities [Online]. Available:

http://databank.worldbank.org/data/views/variableselection/selectvariables.aspx?sourc e=global-economic-monitor-(gem)-commodities [Accessed May 2014]. US BLS 2014. Consumer Price Index for All Urban Consumers. In: UNITED STATES DEPARTMENT OF LABOUR (ed.). VLIET, O. P. R., FAAIJ, A. P. C. & TURKENBURG, W. C. 2009. Fischer–Tropsch diesel production in a well-to-wheel perspective: A carbon, energy flow and cost analysis. Energy Conversion and Management, 50, 855-876. WILCOX, W. R. 2005. Oxygen Costs [Online]. [Accessed May 2014]. WILLIAMS, R. & LARSON, E. 2003. A comparison of direct and indirect liquefaction technologies for making fluid fuels from coal. Energy for Sustainable Development. WILLS, B. 2006. Wills' Mineral Processing Technology, Elsevier Science & Technology Books. WOOD, T. & CARTER, L. 2013. Getting gas right: Australia's energy challenge. Grattan Institute.

39 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

10 Appendices 10.1 Appendix A: Aspen Plus Model Components: Table A1: CTL Aspen Plus Model Components

ID H2 CO CO2 H2S COS H2O O2 N2 AR CH4 CL2 HCL HCN NH3 S C S-S COAL COALASH C2H6 C3H8 C4H10 C5H12 C6H14 C7H16 C8H18 C9H20 C10H22 C11H24 C12H26

Type Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Solid Solid Nonconventional Nonconventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional

Component name HYDROGEN CARBON-MONOXIDE CARBON-DIOXIDE HYDROGEN-SULFIDE CARBONYL-SULFIDE WATER OXYGEN NITROGEN ARGON METHANE CHLORINE HYDROGEN-CHLORIDE HYDROGEN-CYANIDE AMMONIA SULFUR CARBON-GRAPHITE SULFUR COAL COALASH ETHANE PROPANE N-BUTANE N-PENTANE N-HEXANE N-HEPTANE N-OCTANE N-NONANE N-DECANE N-UNDECANE N-DODECANE

ID C13H28 C14H30 C15H32 C16H34 C17H36 C18H38 C19H40 C20H42 C30H62 C2H4 C3H6 C4H8 C5H10 C6H12 C7H14 C8H16 C9H18 C10H20 C20H40 CH3OH C2H5OH C3H7OH C4H9OH C5H11OH C6H13OH C7H15OH C8H17OH C9H19OH C10H21OH C20H41OH

Type Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional Conventional

Component name N-TRIDECANE N-TETRADECANE N-PENTADECANE N-HEXADECANE N-HEPTADECANE N-OCTADECANE N-NONADECANE N-EICOSANE N-TRIACONTANE ETHYLENE PROPYLENE 1-BUTENE 1-PENTENE 1-HEXENE 1-HEPTENE 1-OCTENE 1-NONENE 1-DECENE 1-EICOSENE METHANOL ETHANOL 1-PROPANOL N-BUTANOL 1-PENTANOL 1-HEXANOL 1-HEPTANOL 1-OCTANOL 1-NONANOL 1-DECANOL 1-EICOSANOL

40 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process

10.2 Appendix B: Model Development Details: 10.2.1 B1: Overall Process Flowsheet:

Figure A1: Overall Process Flowsheet for AspenPlus CTL Model

Figure A1 depicts the overall layout of the CTL model developed in AspenPlus. Plant areas are divided into individual hierarchies to simplify the layout and also allow different stream classes (e.g. Conventional or non-conventional solids) to be used in different plant areas. The layout and model development for individual plant areas is detailed through the remainder of Appendix A.

41 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.2.2 B2: Coal Preparation Hierarchy:

Figure A2: Coal Preparation Simulation Flowsheet Table A2: Coal Preparation Model Specifications

Block Name (Aspen Unit) COALTYPE (Selector)

Block Parameters

Description

Inlet Stream ID: CRC701 Multiplication Factor: 180.7

Allows the selection of 1 of the 5 coal feedstocks for the process.

Specify’s amount of N2 from ASU required for coal transportation drying. (AspenTech, 2009)

CRUSHER (Crusher) MIXER B-MILL (Crusher) SCREEN DRY-REAC (RStoic)

N2 = 5 x Coal Feed (kg/s) Crusher Type: Single Roll Max. Particle D. 0.0508m Pressure: 0 bar Dist. Function: RRSB D63: 70 microns Cut Size: 100 microns Pressure = 1atm, Duty = 0W Rxn: Coal -> 1/18 H2O

WATER (Calculator)

H2ODRY = 5 %, Conv = (H2OIN-H2ODRY)/(100-

Calculates conversion required in DRY-REAC to reduce moisture content of outlet coal to 5%.

COALMULT (Multiplier) N2FEED (Calculator)

Multiply’s mass flowrate of coal by a specified factor. Allows all coals to be specified as 1kg/s and only the multiplication factor needs to be varied when changing the feed rate.

Models power requirement to reduce particle size of ROM coal to a maximum diameter of 2 inches (0.0508m) (Drover, 2008) Combines crushed, oversized grinded coal and nitrogen gas for transportation/drying. Determines power requirement to grind crushed coal and achieve a PSD with 63% of particles with diameter less than 70 microns. Recycles oversize coal particles (>100 microns) for grinding. Converts some moisture content of coal specified in Proxanal to free water to simulate the drying process. Coefficient = 1/MWH2O as Aspen assigns coal a MW of 1g/mol.

H2ODRY)

H2O-SEP (Flash2)

-

Separates dried coal from free water and nitrogen transportation gas.

42 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.2.3 B3: Gasifier Hierarchy:

Figure A3: Gasifier Simulation Flowsheet Table A3: Gasification Model Specifications

Block Name NC-CHNG (ClChng) DCOMP (RStoic) CDCOMP (Calculator)

Block Parameters

Changes solids Sub-Stream type from Non-conventional with Particle Size Distribution (NCPSD) to Non-Conventional (NC).

COALNC -> H2O + O2 + N2 + Csolid + AshNC + S-Ssolid + CL2+ H2 Conversion = 100% DCOMP Conv = 0.973 T(DCOMP) = Tin

Decomposes coal into conventional components that can be used by Aspen in phase and chemical equilibria calculations. Conversion overwritten by calculator block

O2 = Dry x Ultanal(O)/(100 x MWO2)

GASFR (RGibbs) GAS-FEED (Calculator)

GASFTEMP (Design-Spec)

Description

None

Pressure = 30 bar Products: CO2, CO, CH4, H2O, H2, H2S, COS, HCl, NH3, HCN, N2, O2, Ar

H2O = C x 0.09 O = C x O:C Ratio O2FEED=(O - H2O – Ocoal x 2)/2 OXIDANT = O2FEED / O2FRAC Temp = 1550°C Vary: O:C Ratio from 0.6-1.5

Calculates stoichiometric coefficients for each of the components coal is decomposed into based on the ultimate analysis composition of the coal. Calculation for O2 shown as example. Note: Dry = 1 – Mass fraction of H2O in coal. Determines the thermodynamic equilibrium for gasification reactions by minimising the Gibb’s Free Energy of products specified. Heat Duty specified by HHV calculator block and Temperature achieved by varying O2 feed with GASFTEMP design-spec. Calculates steam requirements according to H2O:C ratio = 0.09 mol/mol (Kunze and Spliethoff, 2011). Calculates O2 gas required in feed as total oxygen required less oxygen in steam and coal. Total oxidant required calculated according to the oxygen purity of 99.5%. O:C Ratio is varied by GASFTEMP design-spec. Varies O:C ratio to achieve the required temperature in the gasifier. O:C ratio used by GAS-FEED calculator to determine the flow of oxidant required.

43 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.2.4 B4: Syngas Quench Hierarchy:

Figure A4: Syngas Quench Simulation Flowsheet

Table A4: Syngas Quench Model Specifications

Block Name SG-MIX (Mixer) SG-QUENC (Design-Spec) HX-1 Filter (SSplit) SG-SPLIT S-TO-MIX (ClChng) Q-BLOWER (Compr)

Block Parameters Pressure drop = 0 bar SG-2 Temp = 900°C Vary: SG-SPLIT Split Fraction T = 250°C, P = 0 bar Stream: SLAG

Description Combines raw, hot syngas from gasifier with cooled quench syngas stream to reduce temperature and (in reality) solidify molten ash although this is not modelled. Varies split fraction of cooled syngas stream to achieve the temperature of 900°C which would be required to solidify molten ash. Reduces quenched SG temp to 250°C. Would be used for heat recovery in reality. Removes all Coal ash and unconverted solid carbon to SLAG waste stream.

CISOLID & NC Split Fraction = 1

Pressure drop = 0 bar None

Splits portion of cool syngas for quench stream. Split fraction determined by SG-QUENC. Converts Sub-Stream class from solids to MIXED.

Type: Polytropic Discharge P = 39.6 bar Polytropic η = 0.88 Mechanical η = 0.92

Compressed recycled syngas stream to the required pressure (≈ 39.6 bar) in order to quench hot incoming syngas stream from gasifier (Kreutz, 2012).

44 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.2.5 B5: Water Gas Shift Hierarchy:

Figure A5: Water Gas Shift Simulation Flowsheet

Table A5: Water Gas Shift Model Specifications

Block Name SPLIT (Splitter) SG-SPLI (Design Spec) PREHEAT (HeatX) WGSSTEAM (Calculator) WGS (REquil) MIX (Mixer)

Block Parameters BYPASS Split = 0.5 Pressure drop = 0 bar CLNGAS H2:CO ratio = 2.0 Vary: Bypass Split Fraction Flow Direction = Counter-current Cold stream outlet T = 277°C Steam = 2.1COHOT-SG - H2OHOT-SG RXN: CO + H2O -> CO2 + H2 Duty = 0 MW, Pressure = 0 bar. Pressure drop = 0 bar

Description Splits a portion of incoming syngas to bypass the WGS reactor and achieve the required H2:CO ratio. Split fraction determined by SG-SPLI design-spec found in main flowsheet. Varies the split fraction of the bypass stream around the WGS reactor to achieve the required H2:CO ratio in the CLNGAS stream entering the FT reactor. Uses heat from shifted syngas to preheat feed stream to a temperature of 277°C required for the WGS reaction (Drover, 2008). Ensures sufficient steam is added to the WGS reactor to avoid carbon formation on catalyst. Minimum required H2O:CO ratio required is 2 (Abdulhamid, 2007) Calculates the thermodynamic equilibrium for the water-gas shift reaction to simulate the shift of CO to H2. Combines syngas which has bypassed the WGS reactor with shifted syngas which has been cooled by pre-heating the feed stream.

45 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.2.6 B6: AGR Hierarchy:

Figure A6: Acid Gas Removal Simulation Flowsheet

Table A6: Acid Gas Removal Model Specifications

Block Name SG-HX (Heater) KNOCKOUT (Flash2)

RECTISOL (Sep2)

PSA (Sep2) H2-REQ (Design-Spec)

Block Parameters Temperature = 95°C Pressure drop = 0 bar Temperature = 25°C Pressure drop = 0 bar CLNGAS Split Fractions: H2 = 99.7%, CO = 96.7%, CH4 = 96.8%, H2S = 0.01% CO2-Rich Stream Split Fractions: CO2 = 93.4%, CO = 3.2%, CH4 = 3.2%, H2S = 0%. H2 Stream Component Flows: H2 = 0.17 kmol/s. All others = 0 kmol/s Excess H2 = 0.01 kmol/s Vary: PSA H2 Split

Description Cool syngas down after exothermic shift reaction to required temperature of 95°C (Sudiro, 2009). Significant heat available for recovery but heat integration not modelled here. Knockout drum to separate syngas from entrained water from the WGS reactor as this will freeze in the Rectisol process. Model Rectisol process as a simple separator with recoveries of H2, CO and CH4 as well as separation efficiency’s of CO2, H2S and COS estimated from a detailed simulation of the Rectisol process (Kreutz, 2012)

Models Pressure Swing Adsorption process assuming that hydrogen can be removed from the syngas stream with no other impurities. Exact amount of hydrogen removed is overwritten by H2-REQ design spec based on refinery requirements. Varies the amount of H2 removed from syngas by PSA in order to get minimal excess H2 remaining in the outlet from the hydrocracker in the REFINERY hierarchy.

46 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.2.7 B7: LTFT Hierarchy

Figure A7: FT Synthesis Simulation Flowsheet

Table A7: FT Synthesis Model Specifications

Block Name PREHEAT (HeatX) FT (RYield)

Block Parameters Flow: Counter-current Cold Stream Outlet T = 164°C Temperature = 230°C. Pressure drop = 0 bar

C-FT (Calculator)

αalkane= 0.91, αalcohol= 0.85, αalkene= 0.74 xalkane = 0.78, xalkene = 0.12, xalc = 0.03, xCH4 = 0.08, xC2H4 = 0.001, Conv = 90%

FLASH (Flash2) FLASH2 (Flash3) H2OSTRIP (Sep2)

Temperature = 230°C Pressure = 25 bar Temperature = 140°C Pressure = 25 bar CLN-H2O Split Fractions: H2O = 1, All others = 0.

Description Utilises heat in hot recycled syncrude vapours to preheat incoming syngas to 164°C (Drover, 2008). Models syncrude production from syngas at 230°C. Mass yields for all syncrude components are calculated by C-FT Calculator block. Calculates mass yields of all syncrude components (alkanes, alkenes and alcohols) based on the individual ASF distributions derived for each hydrocarbon family by Bertoncini et al. Simulates phase separation of syncrude waxes from hot vapours and unconverted syngas that would occur within the FT reactor in reality. 3-phase knockout drum to separate heavy syncrude components (HOT-COND) from vapours (VAP-2) and aqueous product (RXN-H2O). Removes syncrude components from aqueous product stream (assuming 100% separation) and combines them with syncrude vapours.

47 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process VAP-MIX (Mixer)

SC-FRAC (Sep) H2O-MIX (Mixer)

Pressure = 25 bar Phases: Vapour-Liquid

Combines syncrude vapours from FLASH2 with syncrude components stripped from aqueous product.

WATER Split Fractions: H2O = 1, All others = 0 TAILGAS Split Fractions: C1-C2, CO, H2 & all contaminants = 1, All others = 0. Pressure = 25 bar Phases: Vapour-Liquid

Separates all water and light components (C1-C2) as well as unconverted CO, H2 and contaminants (H2S, HCN, NH3) from syncrude product. Light components sent to tail gas and water combined with stripped aqueous product.

10.2.8 B8: Refinery Hierarchy:

Figure A8: Syncrude Refinery Simulation Flowsheet

Combines clean aqueous product with traces of water removed from syncrude in SC-FRAC.

48 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process Table A8: Syncrude Refinery Model Specifications

Block Name MAINFRAC (Sep2) WAX-MIX HT-MIX HT (RStoic) HT-FRAC (Sep) HC-REC (Mixer) HC (RYield) C-HC (Calculator) HC-FRAC (Sep) AROMAT (Hierarchy) GAS-SPL (Design-Spec) FUEL-SPL (Design Spec) OLIG (Hierarchy)

Block Parameters Stream C3-C5 Split Fractions: C3-C5 = 1, All others = 0. Stream C6-C8 Split Fractions: C6-C8 = 1, All others = 0. TAILGAS Split Fractions: H2, CO, CO2, H2S, H2O, CH4 = 0 Pressure drop: 0 bar Pressure: 10 bar Temperature = 330°C, Pressure = 25 bar, Conversion = 1 Alkenes: CxH2x + H2 -> CxH2x+2 Alcohols: CxH2x-1OH + H2 -> CxH2x+2 + H2O HC-FEED Split Fractions: H2, C15+ components = 1, All others = 0 Pressure drop = 0 bar

Description Simplified model of the main distillation column to separate syncrude components into various fractions for further processing according to a proposed LTFT Jet Fuel refinery (de Klerk, 2011). Combines MAINFRAC C9+ material & wax from LTFT FLASH Combines Hydrotreater feed with H2 from PSA unit. Models hydrotreating of all alkenes and alcohols according to general reactions shown to the left to produce alkanes (and H2O for alcohols). 100% conversion assumed for all.

Separates out C15+ material requiring hydrocracking. In reality H2 would be in overheads and require recovery with additional units. Combines bottoms stream from HT-FRAC with recycled C15+ material exiting HC as this requires further cracking. Temperature = 330°C Simulates hydrocracking of long-chain hydrocarbons according to the mass yields calculated by C-HC. Pressure drop = 0 bar. See FORTRAN code in Aspen Plus file. Calculates HC mass yields based off Shell’s hydrocracking technology developed for heavy syncrude waxes (Eilers, 1990). C3-C8 Split Fractions: C9+ components = 0, All others =1 Separates C15+ material from HC product so this can be recycled for additional hydrocracking. C9-14 sent to Jet Blending. C15+ Split Fractions: C15+ components = 1, All others = 1 In proposed refinery design, aromatisation produces aromatic liquid for gasoline blending with unconverted material sent to jet blending or fuel gas. Process not modelled but distribution between gasoline, jet fuel and fuel gas varied by GAS-SPL & FUELSPL design-specs to achieve the overall product distribution for the refinery specified by de Klerk, 2011. Refinery Product Distribution (mass %): Jet Fuel = 64.6%, Gasoline = 32.1%, Fuel Gas = 3.3% GASOLINE/(FUELGAS + GASOLINE + JET) =0.321 Varies the split fraction of gasoline from AROMAT splitter in order to achieve product distribution for gasoline of 32.1%. Vary: AROMAT.SPLIT Gasoline split fraction FUELGAS/(FUELGAS + GASOLINE + JET) = 0.033 Varies the split fraction of fuel gas from AROMAT splitter in order to achieve product distribution for fuel gas of 3.3%. Vary: AROMAT.SPLIT Fuel Gas split fraction Block: POLY (RStoic), T = 550°C, P = 10 bar. Upgrading of light hydrocarbons through polymerisation (reacting propane and propene) and alkylation (reacting isoRXN: C3H8 + C3H6 -> C6H14 butane and butane) to form longer chain hydrocarbons more Block: ALKY (RStoic), T = 550°C, P = 10 bar. suitable to fuels. Alkylate sent to gasoline blending and RXN: C4H10 + C4H8 -> C8H18 polygasoline sent to jet fuel blending.

49 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.3 Appendix C: FT Synthesis Product Distribution: The syncrude product distribution from FT synthesis was modelled on the basis of experimental results from Bertoncini et al, 2011. These results were obtained by analysing the syncrude composition from a Co-LTFT reactor and determining an Anderson-Schulz-Flory (ASF) distribution to characterise the product distribution for each hydrocarbon family. This was done by plotting Log (Wn/n) vs n, where Wn is the weight fraction of a hydrocarbon chain of length n, as shown in Figure A9.

Figure A9: Determination of ASF distributions for Co-LTFT Products (Bertoncini et al, 2011)

Alpha (α) values characterising the three ASF distributions were determined through Figure A9. It can be seen from the high coefficient of determination values that these distributions closely describe the actual product distribution. These α values and additional data from Bertoncini et al, 2011 required to model the FT synthesis process can be seen in Table A9 below. Table A9: ASF Distribution Parameters

Parameter Value α (alkane) 0.91 α (alcohol) 0.85 α (alkene) 0.74 Alkane (wt%) 0.78 Alkene wt% 0.12 Alcohol (wt%) 0.027 CH4 (wt%) 0.082 C2H4 (wt%) 0.001 The weight fraction for an individual component can be calculated from the ASF distribution as follows. This is then multiplied by the total weight fraction for that hydrocarbon family to get the weight fraction for that component in the entire syncrude product. Wn =n.(1- α)2.αn-1

50 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.4 Appendix D: Fixed Operating Costs: Table A10 lists the details of estimated fixed operating expenses for the CTL plant under each of the 5 scenarios. These were adjusted from original estimates in the Hatch Fairbanks CTL report and details of this adjustment can be found in Section 6.2.1. Table A10: Estimated Fixed Operating Expenses for 5 Scenarios

Capacity Factor Fixed O&M ('000 USD/yr) Coal Feed Coal Drying Air Separation Unit (ASU) Gasifier & Steam Handling Syngas Shift & Conversion Syngas Purification Fischer-Tropsch Synthesis Upgrading General Facilities Power Generation Operating Labor Sales, General & Admin Total Q1 '14 Exchange Rate Total ('000 AUD/yr)

Reference 1.00

CRC701

CRC702

CRC703

CRC704

HATCH

1.41

1.91

1.87

1.53

1.00

156,950 1,213 4,707 38,202 10,070 8,221 25,813 6,814 5,300 8,186 33,850 30,000 329,326 1.09 359,854

1,213 38,202 14,231 11,618 36,480 9,630 5,300 33,850 30,000 180,524 1.09 197,258

1,213 38,202 19,263 15,726 49,377 13,034 5,300 33,850 30,000 205,965 1.09 225,058

1,213 38,202 18,838 15,379 48,289 12,747 5,300 33,850 30,000 203,818 1.09 222,712

1,213 38,202 15,418 12,587 39,522 10,433 5,300 33,850 30,000 186,526 1.09 203,816

1,213 38,202 10,070 8,221 25,813 6,814 5,300 33,850 30,001 159,484 1.09 174,268

51 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.5 Appendix E: Price Estimations: 10.5.1 E1: Predicted Coal Prices: Future coal prices were predicted from the World Banks prediction for the price of a standard Australian thermal coal until 2025 (Baffes and Cosic, 2014). This was converted to prices for each of the 5 coals considered according to the relative heating values (HHV) of the coals. Additional data to convert from 2010 USD to 2014 AUD was as follows. • • •

US CPI (2010): 218.06 US CPI (Q1 '14): 237.07 AUD:USD (Q1 '14): 1.09

Based of this data predicted coal prices were calculated and can be seen in Table A11. Table A11: Predicted Coal Prices until 2025 for 5 Coal Types

Coal Type World Bank World Bank World Bank

CRC701 CRC702 CRC703 CRC704 Healy

HHV (GJ/t) 26.4 " " 22.4 30.2 31.9 26.0 16.3

Units

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

'10 USD/t

82.6 89.8 97.9 83.1 95.1 115.2 113.7 70.2

82.3 89.5 97.5 82.8 94.7 114.8 113.3 70.0

82.1 89.3 97.3 82.6 94.5 114.5 113.1 69.8

81.7 88.8 96.8 82.2 94.0 113.9 112.5 69.5

81.4 88.5 96.5 81.9 93.7 113.5 112.1 69.2

80.9 88.0 95.9 81.4 93.1 112.8 111.4 68.8

80.5 87.5 95.4 81.0 92.6 112.3 110.8 68.4

80.0 87.0 94.8 80.5 92.1 111.6 110.2 68.0

79.6 86.5 94.3 80.1 91.6 111.0 109.6 67.7

79.1 86.0 93.7 79.6 91.0 110.3 108.9 67.3

78.6 85.5 93.1 79.1 90.5 109.6 108.2 66.8

'14 USD/t '14 AUD/t '14 AUD/t '14 AUD/t '14 AUD/t '14 AUD/t '14 AUD/t

10.5.2 E2: FT Product Prices: Future FT Product prices were estimated based on World Bank predictions for crude oil prices and the relationship between the crude oil price and the average refinery price received for liquid fuels. This relationship was developed from historical prices, as described in section 6.2.2, and is given as follows: 𝐴𝐴𝐴𝐴𝐴𝐴𝐴 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑃𝑃𝑃𝑃𝑃 = 1.04 × 𝐷𝐷𝐷𝐷𝐷 𝐵𝐵𝐵𝐵𝐵 𝐶𝐶𝐶𝐶𝐶 𝑃𝑃𝑃𝑃𝑃 + $22.82

From this, expected prices until 2012 were calculated and can be seen in Table A 12. Table A 12: Estimated FT Product Prices based of World Bank Predictions for Crude Oil Prices until 2025

Product

Units

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

Dated Brent Crude

'10 USD/bbl

Dated Brent Crude

'14 USD/bbl

Dated Brent Crude

'14 AUD/bbl

91.6 99.6 108.6

89.3 97.1 105.8

87.7 95.3 103.9

86.1 93.6 102.0

84.6 92.0 100.3

83.1 90.3 98.5

81.7 88.8 96.8

80.2 87.2 95.0

78.8 85.7 93.4

77.4 84.1 91.7

76.0 82.6 90.1

FT Product Price

'14 AUD/bbl

135.4

132.5

130.6

128.6

126.8

124.9

123.2

121.3

119.6

117.9

116.2

52 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.5.3 E3: Oxygen Supply Costs: Over-the-fence supply costs were calculated according to the following procedure which is typical of commercial arrangements implemented by Praxair (Wilcox, 2005). 1. Determine oxygen requirements from process model in short tons per day (TPD) and cubic feet per year at NTP (CFY). 2. Calculate the ASU capital cost by scaling according to the capacity required, adjusting for the change in CEPCI and the USD:AUD exchange rate and including a Location Factor of 1.19 for construction in Australia compared to a US location (Richardson, 2012). 𝐴𝐴𝐴 𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐶𝐶𝐶𝐶 = 𝑈𝑈𝑈 27 𝑚𝑚𝑚𝑚𝑚𝑚𝑚 × �

𝑠ℎ𝑜𝑜𝑜 𝑇𝑇𝑇 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 0.6 𝐶𝐶𝐶𝐶𝐶2014 1.09 𝐴𝑈𝑈 � × × × 𝐿𝐿 𝐶𝐶𝐶𝐶𝐶2004 1000 𝑠ℎ𝑜𝑜𝑜 𝑇𝑇𝑇 1 𝑈𝑈𝑈

3. Calculate the annual facility fee: AFF = 12 x 0.0275 x Capital Cost. 4. Calculate the variable cost paid to the oxygen supplier based on estimated power consumption of the plant.

5.

𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 𝐶𝐶𝐶𝐶 =

16 𝑘𝑘ℎ 0.145 𝐴𝐴𝐴 × (0.85 × 𝐶𝐶𝑌 𝑎𝑎 𝑚𝑚𝑚. 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐) × 1000 𝐶𝐶 1 𝑘𝑘ℎ

Determine the total annual cost: AFF + Variable Cost.

Results from these calculations are shown in Table 8. Table 8: Cost of Oxygen Supply for Each Scenario

Oxygen Required Oxygen Required Capital Cost Capital Cost Annual Fixed Fee Electricity Price Variable Cost Total Cost

Units stpd 1010 ft3/yr 106 '04 USD 106 '14 AUD 106 AUD/yr AUD/kWh 106 AUD/yr 106 AUD/yr

CRC701

CRC702

CRC703

CRC704

HATCH

11,221 49.3 $147 $175 $63 0.145 $106 $169

13,871 60.9 $167 $199 $72 0.145 $131 $203

10,493 46.1 $141 $168 $61 0.145 $99 $160

13,327 58.5 $163 $194 $70 0.145 $126 $196

8,473 37.2 $124 $148 $53 0.145 $80 $133

53 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.6 Appendix F: Example Economic Calculations – CRC701 Scenario: Time Period Year

Coal Price ($/t)

Plant Year

Commodity Prices FT Gas Electricity Product Price CapEx ($) ($/MWh) ($/bbl) ($/GJ)

Expenses

Revenue Fixed Opex

Oxygen Coal Cost Electricity Costs

Deprec.

Gas

Profit & FCF's

Fuel Revenue

EBIT ($)

Tax ($)

NPV & IRR

Net Profit

FCF ($)

Cumulative FCF

NPV (r =0.12)

2015

0

83.09

135.35

145.00

4.42 -4.17E+08 -

-

-

-

-

-

-

-

-

-

-4.17E+08

-4.17E+08

-4.17E+08

2016

1

82.79

132.53

145.44

4.47 -8.35E+08 -

-

-

-

-

-

-

-

-

-

-8.35E+08

-1.25E+09

-7.45E+08

2017

2

82.59

130.56

145.87

4.51 -2.09E+09 -

-

-

-

-

-

-

-

-

-

-2.09E+09

-3.34E+09

-1.66E+09

2018

3

82.19

128.60

146.31

4.56 -8.35E+08 -

-

-

-

-

-

-

-

-

-

-8.35E+08

-4.17E+09

-5.94E+08

2019

4

81.88

126.76

146.75

4.60 -

-2.78E+08 -1.61E+08 -3.98E+08

-2.71E+07 -1.97E+08 2.61E+07 1.17E+09 1.35E+08

-4.06E+07 9.48E+07

3.73E+08

-3.80E+09

2.37E+08

2020

5

81.38

124.91

147.19

4.65 -

-2.60E+08 -1.62E+08 -3.97E+08

-2.72E+07 -1.97E+08 2.64E+07 1.15E+09 1.38E+08

-4.15E+07 9.68E+07

3.56E+08

-3.44E+09

2.02E+08

2021

6

80.98

123.19

147.63

4.70 -

-2.42E+08 -1.62E+08 -3.94E+08

-2.72E+07 -1.97E+08 2.66E+07 1.14E+09 1.42E+08

-4.26E+07 9.94E+07

3.42E+08

-3.10E+09

1.73E+08

2022

7

80.48

121.35

148.07

4.74 -

-2.26E+08 -1.62E+08 -3.92E+08

-2.73E+07 -1.97E+08 2.69E+07 1.12E+09 1.43E+08

-4.29E+07 1.00E+08

3.26E+08

-2.78E+09

1.48E+08

2023

8

80.07

119.63

148.52

4.79 -

-2.11E+08 -1.63E+08 -3.90E+08

-2.74E+07 -1.97E+08 2.72E+07 1.11E+09 1.44E+08

-4.33E+07 1.01E+08

3.12E+08

-2.46E+09

1.26E+08

2024

9

79.57

117.91

148.96

4.84 -

-1.97E+08 -1.63E+08 -3.88E+08

-2.75E+07 -1.97E+08 2.74E+07 1.09E+09 1.44E+08

-4.33E+07 1.01E+08

2.98E+08

-2.17E+09

1.08E+08

2025

10

79.07

116.19

149.41

4.89 -

-1.84E+08 -1.63E+08 -3.85E+08

-2.76E+07 -1.97E+08 2.77E+07 1.07E+09 1.44E+08

-4.32E+07 1.01E+08

2.85E+08

-1.88E+09

9.17E+07

2026

11

79.07

116.19

149.86

4.94 -

-1.72E+08 -1.64E+08 -3.83E+08

-2.76E+07 -1.97E+08 2.80E+07 1.07E+09 1.59E+08

-4.76E+07 1.11E+08

2.83E+08

-1.60E+09

8.13E+07

2027

12

79.07

116.19

150.31

4.98 -

-1.60E+08 -1.64E+08 -3.83E+08

-2.77E+07 -1.97E+08 2.83E+07 1.07E+09 1.70E+08

-5.10E+07 1.19E+08

2.79E+08

-1.32E+09

7.17E+07

2028

13

79.07

116.19

150.76

5.03 -

-1.50E+08 -1.64E+08 -3.83E+08

-2.78E+07 -1.97E+08 2.85E+07 1.07E+09 1.81E+08

-5.42E+07 1.26E+08

2.76E+08

-1.04E+09

6.32E+07

2029

14

79.07

116.19

151.21

5.09 -

-1.40E+08 -1.64E+08 -3.83E+08

-2.79E+07 -1.97E+08 2.88E+07 1.07E+09 1.90E+08

-5.71E+07 1.33E+08

2.73E+08

-7.70E+08

5.58E+07

2030

15

79.07

116.19

151.66

5.14 -

-1.30E+08 -1.65E+08 -3.83E+08

-2.80E+07 -1.97E+08 2.91E+07 1.07E+09 2.00E+08

-5.99E+07 1.40E+08

2.70E+08

-5.00E+08

4.93E+07

2031

16

79.07

116.19

152.12

5.19 -

-1.22E+08 -1.65E+08 -3.83E+08

-2.81E+07 -1.97E+08 2.94E+07 1.07E+09 2.08E+08

-6.25E+07 1.46E+08

2.67E+08

-2.33E+08

4.36E+07

2032

17

79.07

116.19

152.58

5.24 -

-1.13E+08 -1.65E+08 -3.83E+08

-2.81E+07 -1.97E+08 2.97E+07 1.07E+09 2.16E+08

-6.49E+07 1.51E+08

2.65E+08

3.20E+07

3.86E+07

2033

18

79.07

116.19

153.03

5.29 -

-1.06E+08 -1.66E+08 -3.83E+08

-2.82E+07 -1.97E+08 3.00E+07 1.07E+09 2.24E+08

-6.71E+07 1.57E+08

2.62E+08

2.94E+08

3.41E+07

2034

19

79.07

116.19

153.49

5.34 -

-9.88E+07 -1.66E+08 -3.83E+08

-2.83E+07 -1.97E+08 3.03E+07 1.07E+09 2.31E+08

-6.92E+07 1.61E+08

2.60E+08

5.55E+08

3.02E+07

2035

20

79.07

116.19

153.95

5.40 -

-9.23E+07 -1.66E+08 -3.83E+08

-2.84E+07 -1.97E+08 3.06E+07 1.07E+09 2.37E+08

-7.11E+07 1.66E+08

2.58E+08

8.13E+08

2.68E+07

2036

21

79.07

116.19

154.41

5.45 -

-8.61E+07 -1.67E+08 -3.83E+08

-2.85E+07 -1.97E+08 3.09E+07 1.07E+09 2.43E+08

-7.30E+07 1.70E+08

2.56E+08

1.07E+09

2.37E+07

2037

22

79.07

116.19

154.88

5.51 -

-8.04E+07 -1.67E+08 -3.83E+08

-2.86E+07 -1.97E+08 3.12E+07 1.07E+09 2.49E+08

-7.47E+07 1.74E+08

2.55E+08

1.32E+09

2.10E+07

2038

23

79.07

116.19

155.34

5.56 -

-7.50E+07 -1.67E+08 -3.83E+08

-2.87E+07 -1.97E+08 3.15E+07 1.07E+09 2.54E+08

-7.62E+07 1.78E+08

2.53E+08

1.58E+09

1.87E+07

2039

24

79.07

116.19

155.81

5.62 -

-7.00E+07 -1.68E+08 -3.83E+08

-2.87E+07 -1.97E+08 3.19E+07 1.07E+09 2.59E+08

-7.77E+07 1.81E+08

2.51E+08

1.83E+09

1.66E+07

2040

25

79.07

116.19

156.28

5.67 -

-6.53E+07 -1.68E+08 -3.83E+08

-2.88E+07 -1.97E+08 3.22E+07 1.07E+09 2.64E+08

-7.91E+07 1.85E+08

2.50E+08

2.08E+09

1.47E+07

2041

26

79.07

116.19

156.74

5.73 -

-6.10E+07 -1.68E+08 -3.83E+08

-2.89E+07 -1.97E+08 3.25E+07 1.07E+09 2.68E+08

-8.04E+07 1.88E+08

2.49E+08

2.33E+09

1.31E+07

2042

27

79.07

116.19

157.21

5.79 -

-5.69E+07 -1.68E+08 -3.83E+08

-2.90E+07 -1.97E+08 3.28E+07 1.07E+09 2.72E+08

-8.16E+07 1.90E+08

2.47E+08

2.57E+09

1.16E+07

2043

28

79.07

116.19

157.69

5.85 -

-5.31E+07 -1.69E+08 -3.83E+08

-2.91E+07 -1.97E+08 3.31E+07 1.07E+09 2.76E+08

-8.27E+07 1.93E+08

2.46E+08

2.82E+09

1.03E+07

2044

29

79.07

116.19

158.16

5.90 -

-4.96E+07 -1.69E+08 -3.83E+08

-2.92E+07 -1.97E+08 3.35E+07 1.07E+09 2.79E+08

-8.37E+07 1.95E+08

2.45E+08

3.06E+09

9.16E+06

2045

30

79.07

116.19

158.63

5.96 -

-4.63E+07 -1.69E+08 -3.83E+08

-2.93E+07 -1.97E+08 3.38E+07 1.07E+09 2.82E+08

-8.47E+07 1.98E+08

2.44E+08

3.31E+09

31

79.07

116.19

159.11

6.02 -

-4.32E+07 -1.70E+08 -3.83E+08

-2.93E+07 -1.97E+08 3.41E+07 1.07E+09 2.85E+08

-8.56E+07 2.00E+08

32

79.07

116.19

159.59

6.08 -

-4.03E+07 -1.70E+08 -3.83E+08

-2.94E+07 -1.97E+08 3.45E+07 1.07E+09 2.88E+08

-8.64E+07 2.02E+08

2046 2047

2.43E+08 2.42E+08

3.55E+09 3.79E+09 NPV =

8.14E+06 7.24E+06 6.44E+06 -1.67E+09

54 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.7 Appendix G: Sensitivity Analysis Results: 10.7.1 G1: Numerical Results from Sensitivity Analysis: Table A13: IRR Sensitivity Analysis - CRC701

Percent Change -30% -20% -10% 0% 10% 20% 30%

Coal Price 7% 6.6% 5.8% 5.0% 4.2% 3.4% 2.4%

FT Product -5.8% -0.6% 2.6% 5.0% 7.1% 9.0% 10.8%

Internal Rate of Return Electricity Gas Oxygen 5.8% 4.9% 6.0% 5.6% 4.9% 5.7% 5.3% 5.0% 5.4% 5.0% 5.0% 5.0% 4.8% 5.1% 4.7% 4.5% 5.2% 4.3% 4.2% 5.2% 4.0%

Fixed Opex 6.2% 5.8% 5.4% 5.0% 4.6% 4.2% 3.8%

CapEx 7.9% 6.7% 5.8% 5.0% 4.4% 3.8% 3.3%

Table A14: IRR Sensitivity Analysis - CRC702

Percent Change -30% -20% -10% 0% 10% 20% 30%

Coal Price 10% 8.89% 8.16% 7.40% 6.60% 5.76% 4.88%

FT Product -1.7% 2.2% 5.0% 7.4% 9.5% 11.4% 13.2%

Internal Rate of Return Electricity Gas Oxygen 8.0% 7.2% 8.3% 7.8% 7.3% 8.0% 7.6% 7.3% 7.7% 7.4% 7.4% 7.4% 7.2% 7.5% 7.1% 7.0% 7.5% 6.8% 6.7% 7.6% 6.5%

Fixed Opex 8.4% 8.1% 7.7% 7.4% 7.1% 6.7% 6.4%

CapEx 10.7% 9.4% 8.3% 7.4% 6.6% 5.9% 5.4%

Table A15: IRR Sensitivity Analsysis - CRC703

Percent Change -30% -20% -10% 0% 10% 20% 30%

Coal Price 10% 8.79% 8.00% 7.17% 6.30% 5.39% 4.41%

FT Product -2.1% 1.9% 4.8% 7.2% 9.3% 11.2% 13.0%

Internal Rate of Return Electricity Gas Oxygen 7.6% 7.0% 7.9% 7.5% 7.1% 7.6% 7.3% 7.1% 7.4% 7.2% 7.2% 7.2% 7.0% 7.2% 6.9% 6.8% 7.3% 6.7% 6.7% 7.3% 6.4%

Fixed Opex 8.2% 7.8% 7.5% 7.2% 6.8% 6.5% 6.1%

CapEx 10.4% 9.1% 8.1% 7.2% 6.4% 5.7% 5.2%

Table A16: IRR Sensitivity Analysis - CRC704

Percent Change -30% -20% -10% 0% 10% 20% 30%

Coal Price 7% 6.71% 5.89% 5.03% 4.11% 3.13% 2.05%

FT Product -6.9% -0.9% 2.5% 5.0% 7.2% 9.1% 10.9%

Internal Rate of Return Electricity Gas Oxygen 5.8% 4.8% 6.1% 5.5% 4.9% 5.8% 5.3% 5.0% 5.4% 5.0% 5.0% 5.0% 4.8% 5.1% 4.6% 4.5% 5.2% 4.3% 4.2% 5.2% 3.9%

Fixed Opex 6.2% 5.8% 5.4% 5.0% 4.6% 4.2% 3.8%

CapEx 7.8% 6.7% 5.8% 5.0% 4.4% 3.8% 3.3%

55 Techno-Economic Assessment of an Australian Indirect Coal Liquefaction Process 10.7.2 Healy Coal IRR Sensivity Results Table A17: IRR Sensitivity Analysis - Healy Coal

Percent Change -30% -20% -10% 0% 10% 20% 30%

Coal Price 5% 3.92% 3.16% 2.34% 1.45% 0.48% -0.61%

FT Product -16.40% -4.2% -0.3% 2.3% 4.5% 6.3% 8.0%

Internal Rate of Return Electricity Gas Oxygen 3.2% 2.1% 3.5% 2.9% 2.2% 3.1% 2.6% 2.3% 2.7% 2.3% 2.3% 2.3% 2.0% 2.4% 1.9% 1.7% 2.5% 1.5% 1.4% 2.5% 1.1%

10% 8% 6% 4%

IRR (%)

2% 0%

-2% -4% -6% -8%

-10% -30%

-20% Coal Price

-10%

0% % Change FT Product

Figure A10: IRR Sensitivity - Healy Coal

10% Electricity

20%

30% Gas

Fixed Opex 3.8% 3.3% 2.9% 2.3% 1.8% 1.2% 0.6%

CapEx 4.6% 3.7% 3.0% 2.3% 1.8% 1.3% 0.9%