Expanded Capacity Investment Scheme (CIS) – Design Paper

AUSTELA response to Capacity Investment Scheme consultation on its Design Implementation Paper
Overview
AUSTELA welcomes the Capacity Investment Scheme and the government’s focus on addressing capacity challenges in the coming years, particularly of the long-duration dispatchable capacity that will be critical in keeping the lights on as coal plants are retired over the next decade.

The underlying principle expressed in the Implementation paper is that projects will be assessed via a cost benefit analysis, not simply the lowest bid on floor price for energy. This is a good principle, but it will be important to give developers certainty as to the metrics to be used.

The key merit criteria articulated in Section 4.4, being the contribution to; a) System reliability; b)
Delivery of renewable energy and c) Additional system benefits, are entirely appropriate.

Concentrating Solar Thermal Power (CSP) systems are recognized as excellent contributors to each of these areas of concern. They incorporate long duration energy storage (typically around 15 hours) as a matter of course. They collect solar energy using low-cost mirror fields so add additional renewable generation, rather than simply storing electricity already generated. Additionally, through their steam turbine driven synchronous generators, they contribute all the system benefits that a dispatchable generator with inertia brings in the same way the gas turbines have done traditionally.

A recent study commissioned by the Australian Solar Thermal Research Institute and completed by
Fichtner Engineering and ITP Thermal (see attached report1) show this combined generation and storage capacity can offer the lowest LCOE for long duration dispatchability.

While we acknowledge much of the detail will be mapped out in individual tenders, there are issues that must be fixed in the Implementation Design Paper if the CIS is to attract long-duration dispatchable capacity into the National Electricity Market.

1. As they stand, the two different CISA contract types ‘Generation’ and ‘Clean Dispatchable’ do
not appear to adequately recognise the full benefits of technologies like CSP, which offers
generation and dispatchable capacity in one system, as well as inherent system strength
benefits.
The paper is unclear as to whether a CSP project would be considered under the Generation
or Clean Dispatchable category, this appears to indicate that the team have not yet
considered it explicitly as a key technology option.
While the Paper acknowledges hybrid projects, it states that they will bid into ‘Generation’
contracts. It is not clear that CSP fits the current definition of hybrid projects. Section 4.6
implies that Energy from Waste and Biomass projects would be considered as ‘Clean
Dispatchable’. CSP is not mentioned but shares similar characteristics to those technologies.

1
Kretschmann J, Lovegrove K, Klump F, Zapata J and Puppe M. The Australian Concentrating Solar Thermal
Value Proposition - Dispatchable Power Generation, Process Heat and Green Fuels. Prepared by Fichtner and
ITP for the Australian Solar Thermal Research Institute. October 2023.

1
As such, the current contracts risk missing out on the benefits and low-cost electricity
offered by CSP, which are ideally suited to meet the Merit criteria outlined on page 28.
We therefore propose addressing this problem by creating a new Hybrid CISA that recognises
a single system can be both generator and dispatchable capacity. Such a hybrid CISA could
follow from allowing such a project to simultaneously bid into Generation and Clean
Dispatchable tenders with one response.
2. The Paper does not yet define how different durations will be valued or rated according to
their usefulness i.e. would a project with 8hrs dispatchable capacity be twice as valuable as a
project with 4hrs? AUSTELA is keen to input into the rating mechanism as it is developed and
we believe a clear, transparent and consistent mechanism will be critical to attract
investment and strong responses to tenders.
3. With the above in place, the CIS should ensure that tenders are structured to ensure that a
significant share of contracts are awarded for dispatchable capacity with 8+hrs of storage as
soon as possible. While system needs in the short-term can be met by short- duration battery
storage, the strongest and most cost effective electricity system will have a mix of
technologies, with different capabilities, and it is important we see investment in long-
duration dispatchable capacity sooner rather than later, so that supply chains can be
developed efficiently.

On face value, CSP perfectly meets the merit criteria the CIS (p.28) is looking for. It offers system reliability, delivery of renewable energy (during the day and overnight) and additional system benefits through system strength and restart services. It is a key part of a least cost emissions free electricity system, it also has the potential to support local communities thanks to the ongoing jobs it requires. A greater share of project value stays in the region, and in Australia compared to imported components.

By including our proposed changes, the CIS can help Australia realise these benefits and derisk the energy transition through the deployment of long-duration dispatchable CSP alongside other technologies, from PV and wind to BESS and pumped hydro.

Background on AUSTELA and CSP
AUSTELA is an industry association, representing companies that are involved in solar thermal power generation. Its members includes Australian companies and Australian subsidiaries of international companies.

Solar thermal technologies take their energy from the sun and - unlike solar PV - store the energy as heat rather than converting it to electrons immediately. Concentrating Solar thermal Power (CSP) generation systems operate with an array of mirrors that concentrate the sun’s heat and store it in a medium (typically molten nitrate/potassium salts). The heat – to around 600˚Celsius – can be stored for many days and used to drive a steam turbine which produces electricity via a synchronous generator whenever needed, day or night. Typical systems have tanks with enough salt to run the power block in the absence of sun for 15 hours or more. CSP systems can also provide clean industrial process heat which typically displaces gas-powered heat in manufacturing.

While the renewable energy sector has been dominated by wind and PV, the imperative for reliable dispatchable renewable generation – to balance variable generation – makes CSP an ideal technology option to include in electricity grids. While relatively small in uptake to date, CSP has a 30-year track record and currently around 6.5GWe of installed capacity in more than 100 utility scale plants around

2
the world. Spain is the past leader in utility-scale CSP and China is currently building 28 CSP projects.
A recent key example is the Noor Energy project in Dubai, a 700MW CSP project hybridised with
250MW of solar PV.

AUSTELA regularly comments on the mix of future renewable energy technologies, energy system design and market rules and incentives. We brief ministers, senior departmental officeholders and regulators and continue to be available to the CIS team as it designs this critical mechanism for the energy transition.

Responses to specific questions
Cadence of tender schedule and products

Q: We welcome feedback on the proposed scheduling approach, including views on alternative options:

Would your organisation benefit more from a 6-monthly cycle of simultaneous tenders for both generation and clean dispatchable products, or would an alternating 12-monthly cycle (consisting of one tender every six months?

A: We favour a 6 month cycle of simultaneous tenders for both generation and clean dispatchable rather than an alternating 12 month cycle.

As they stand, the two different CISA contract types ‘Generation’ and ‘Clean Dispatchable’ do not appear to adequately recognize the full benefits of technologies like CSP, which offers generation and dispatchable capacity in one system, as well as inherent system strength benefits.

The paper is unclear as to whether a CSP project would be considered under the Generation or Clean
Dispatchable category, this appears to indicate that the team have not yet considered it explicitly as a key technology option.

In its current form, simultaneous tenders could potentially lead to more bids from CSP projects, as, thanks to its inherent generation and storage capacity, a CSP project could bid for both Generation and Dispatchable Capacity CISAs.

CSP’s strength is in its ability to provide new Genaration and Dispatchable Capacity, as well as offering system services. On its own, CSP is likely to lose out to lower-cost forms of generation or dispatchable capacity. This would see the NEM miss out on a technology that offers renewable generation and storage capacity 24/7, as well as the lowest LCOE over long-durations, according to modelling by Fichtner and ITP Thermal.

We therefore propose addressing this problem by creating a new Hybrid CISA that recognises a single system can be both generator and dispatchable capacity. Such a hybrid CISA could follow from allowing such a project to simultaneously bid into Generation and Clean Dispatchable tenders with one response. Thus the cycle of simultaneous tenders is a better starting point to facilitate such an approach.

Incentives for participation in the contracts market

Q: We welcome feedback on risks to contract market liquidity and whether the design elements outlined in this section are sufficient to preserve incentives to participate in the contracts market.

A: These seem reasonable. Austela has no strong views on this point.

3
Generation support mechanism

Q: Would the proposed Eligible Wholesale Contract requirements present a significant barrier to your organisation participating in the wholesale contracts market with a generation project with a CISA?

A: These seem reasonable. Austela has no strong views on this point.

Q: Would the proposed negative price provisions present a significant barrier to any renewable capacity business model considered by your organisation? Could these provisions have any negative impact on project NEM bidding behaviour?

A: These seem reasonable. Austela has no strong views on this point.

Special Purpose Vehicle requirement

Q: Would the proposed Special Purpose Vehicle requirement present a major barrier to your organisation’s business model for renewable capacity and clean dispatchable capacity projects?

A: This sounds reasonable, most CSP project globally are structured in this way, so there is no reason it would be a barrier for project proponents.

Alternative options to preserve incentives for generators to participate in wholesale contracts markets

Q: We welcome feedback on the alternative options to preserve incentives to participate in wholesale contracts markets, including:
• Whether an option structure would be of value for the generation CISA
• Views on the inclusion of Eligible Wholesale Contract revenue into the net revenue
calculation vis-à-vis the volumetric exclusion of Eligible Wholesale Contract revenue
• Views on the potential requirement for the Project Operator to physically deliver any Green
Products to the Australian Government

A: These seem reasonable. Austela has no strong views on these points.

Merit and eligibility criteria

Q: We welcome feedback on the proposed eligibility and merit criteria.

A: In regards eligibility we note that:

• The definitions of eligible technologies are to mirror the definitions under the Renewable A
Energy (Electricity) Act and agree that this is appropriate.
• A minimum registered capacity of 30MW is proposed. This is appropriate and is in line with
the minimum size that is likely to be contemplated by developers of CSP plants.
• The requirement that land tenure would need to be established and grid connection being
progressed seems reasonable.
• Timing and delivery dates seem reasonable as they discussed. We would warn against any
further changes to that thinking that lead to shorter time frames that could be realistically be
delivered by a CSP project for example. It should be noted that longer duration energy
systems in general take longer to build and commission than battery projects, so timing
requirements should always be considered from the context of potential perverse incentives.

4
• Participation in other schemes such as CEFC funding or ARENA grants are not expected to be
counted as revenue support. This is correct and important. Technologies like CSP that do not
yet have a track record in Australia are going to require such support in a synergistic manner.

In regards merit assessment we note that:

The underlying principle expressed in the Implementation paper is that projects will be assessed via a cost benefit analysis, not simply the lowest bid on floor price for energy. This is a good principle, but it will be important to give developers certainty as to the metrics to be used.

The key merit criteria articulated in Section 4.4, being the contribution to ; a) System Reliability; b)
Delivery of renewable energy and c)Additional system benefits, are entirely appropriate.

Concentrating Solar Thermal Power (CSP) systems are recognized as excellent contributors to each of these areas of concern. They incorporate long duration energy storage (typically around 15 hours) as a matter of course. They collect solar energy using low cost mirror fields so add additional renewable generation, rather than simply storing electricity already generated. Additionally, through their steam turbine driven synchronous generators, they contribute all the system benefits that a dispatchable generator with inertia brings in the same way the gas turbines have done traditionally.

A recent study commissioned by the Australian Solar Thermal Research Institute and completed by
Fichtner Engineering and ITP Thermal (see attached report) show this combined generation and storage capacity can offer the lowest LCOE for long duration dispatchability.

It is noted that the three merit criteria will be considered both in Stage A and Stage B.

The approach to Stage A appears to be appropriate, however we would advocate that the three merit criteria be given a high weighting, and not be overshadowed by the other areas of interest raised in section 4.4.2.

In regards Stage B, Financial Value assessment, we note again that the intention is that projects will be assessed via a cost benefit analysis, not simply the lowest bid on floor price for energy. While supporting this principle, we argue that greater certainty and transparency is needed for this aspect.
Developers of CSP projects (and other dispatchable technologies) can configure future projects with varying durations of energy storage and varying sizes of solar fields compared to the power block.
They need some clear signals of how a project is to be measured for its impact on the value assessment in order to design the most attractive project.

The Paper does not yet define how different durations will be valued or rated according to their usefulness i.e. would a project with 8hrs dispatchable capacity be twice as valuable as a project with
4hrs? AUSTELA is keen to input into the rating mechanism as it is developed and we believe a clear, transparent and consistent mechanism will be critical to attract investment and strong responses to tenders.

Participation of hybrid projects

Q: Would the proposed approach enable the better participation of hybrid projects in CIS tenders?

More work needs to be done to define hybrid projects and to recognise technologies that inherently offer generation and dispatchable capacity.

5
As they stand, the two different CISA contract types ‘Generation’ and ‘Clean Dispatchable’ do not appear to adequately recognize the full benefits of technologies like CSP, which offers generation and dispatchable capacity in one system, as well as inherent system strength benefits.

The paper is unclear as to whether a CSP project would be considered under the Generation or Clean
Dispatchable category, this appears to indicate that the team have not yet considered it explicitly as a key technology option. While the Paper acknowledges hybrid projects, it states that they will bid into
‘Generation’ contracts. It is also not clear that CSP fits the current definition of hybrid projects.
Section 4.6 implies that Energy from Waste and Biomass projects would be considered as Clean
Dispatchable. CSP is not mentioned but shares similar characteristics to those technologies.

As such, the current contracts risk missing out on the benefits and low-cost electricity offered by CSP, which are ideally suited to meet the Merit criteria outlined on page 28.

CSP’s strength is in its ability to provide new Generation and Dispatchable Capacity, as well as offering system services. On its own, CSP is likely to lose out in a head to head floor price competition with wind and PV. It would also loose in a floor price competition with pure electricity storage options if the extra value of its added generation is not recognized. This would see the NEM miss out on a technology that offers renewable generation and storage capacity 24/7, as well as the lowest
LCOE over long-durations, according to modelling by Fichtner and ITP Thermal.

Furthermore, to maximise value CSP projects globally are often co-located with PV. This combination is used in Dubai at Noor Energy 1 and achieved a record low CSP LCOE at the time of US$7c per kwh.

Q: Would your organisation consider bidding for separate clean dispatchable capacity and generation
CISA for the components of a hybrid? Would the proposed schedule that includes simultaneous clean dispatchable capacity and generation tenders (detailed in section 1.1.3) support this option?

A: As noted above, CSP projects inherently provide both new generation and clean dispatachable capacity. As such CSP project developers would indeed be expected to bid for provision of both if this were presented as a clear possibility.

We therefore propose addressing this problem by creating a new Hybrid CISA that recognises a single system can be both generator and dispatchable capacity. Such a hybrid CISA could follow from allowing such a project to simultaneously bid into Generation and Clean Dispatchable tenders with one response. Thus the cycle of simultaneous tenders is a better starting point to facilitate such an approach.

Simultaneous tenders could potentially lead to more bids from CSP projects, as, thanks to its inherent generation and storage capacity, a CSP project could bid for both Generation and
Dispatchable Capacity CISAs.

6

The Australian
Concentrating Solar
Thermal Value
Proposition
Dispatchable Power
Generation, Process Heat
and Green Fuels

Prepared for
Contact

Fichtner Australia Pty Ltd

Level 28, 31 Market Street
Sydney NSW 2000
Australia
www.fichtner.com.au

Johannes Kretschmann
+61 419 999 300
+49 (151) 4020 2271
Johannes.Kretschmann@fichtner.de
info@fichtner.com.au

Fichtner GmbH & Co. KG

S774Doc-676563417-363 / v0.2 2 CST Value Proposition
Document Approval

Name Signature Position Date

Prepared by: Dr. Florian Klumpp MD Fichtner Australia 06.10.2023
Dr. Keith Lovegrove MD ITP Thermal
Jose Zapata Principal ITP Thermal
Michael Puppe CST PM Fichtner GmbH

Checked by: Johannes PD Fichtner GmbH 08.10.2023
Kretschmann

Approved by: Johannes PD Fichtner GmbH 08.10.2023
Kretschmann

S774Doc-676563417-363 / v0.2 3 CST Value Proposition
Document Revision Record

Rev. Date Details of Fichtner Doc Ref. Prepared by Checked Approved
revision by by

0 23.06.2023 Working S774Doc-676563417-299 KLP, KLG, KRT KRT
Draft

1 03.10.2023 Final S774Doc-676563417-337 KLP, KLG, JZ KRT KRT
Draft MPU

2 08.10.2023 Final Report S774Doc-676563417-363 KLG KRT KRT
(web)

S774Doc-676563417-363 / v0.2 4 CST Value Proposition
Disclaimer

The content of this document is intended for the exclusive use of Fichtner’s client and other contractually agreed recipients. It may only be made available in whole or in part to third parties with the client’s consent and on a non-reliance basis. Fichtner is not liable to third parties for the completeness and accuracy of the information provided therein.

S774Doc-676563417-363 / v0.2 5 CST Value Proposition
About this report:

ASTRI commissioned Fichtner Australia Pty Ltd - part of the Fichtner Group - and ITP Thermal Pty Ltd - part of the ITP Energised Group to assess the role and value of CST technologies and systems within
Australia’s future energy landscape,

Founded in 1922, the Fichtner Group is one of the leading independent engineering consultancy firms in the world. Being active in CST since the late 1970s and having been involved in 140+ CST projects,
Fichtner has acquired an unrivalled wealth of experience in CST. Fichtner has been involved in multiple
CST projects in Australia, including site selection and feasibility studies, design planning, tendering and technical advisory services. The Fichtner has worked on projects across Australia since 1979 and due to the growing demand, Fichtner Australia was founded in 2021. Fichtner Australia is a trusted partner in all types of engineering, infrastructure, and consulting projects, with a strong focus also on renewable energy, including CST.

ITP Thermal Pty Ltd was established in 2016 as a new company within the ITP Energised group, with a mandate to lead solar thermal projects globally. In doing so it accesses staff and resources in the other
ITP Energised group companies. The ITP Energised Group, formed in 1981, is a specialist renewable energy, energy efficiency and carbon markets consulting company. The group has offices and projects throughout the world. IT Power (Australia) was established in 2003 and has undertaken a wide range of projects.

Cite as:

Kretschmann J, Lovegrove K, Klump F, Zapata J and Puppe M. The Australian Concentrating Solar Thermal
Value Proposition - Dispatchable Power Generation, Process Heat and Green Fuels. Prepared by Fichtner and ITP for the Australian Solar Thermal Research Institute. October 2023.

S774Doc-676563417-363 / v0.2 6 CST Value Proposition
Table of Contents

Executive Summary ................................................................................................................................................................. 20

1 Introduction ..................................................................................................................................................................... 28

2 CST Technologies: Technology Options, Developments and Deployments .......................................... 31

2.1 CST Technology Overview ............................................................................................................................ 31

2.2 Australia’s Solar Resource ............................................................................................................................. 34

2.3 International Deployment ............................................................................................................................. 36

2.4 Developments in Australia ............................................................................................................................ 43

3 Value of CST ..................................................................................................................................................................... 46

3.1 Competitive Medium-Duration to Long Intraday Energy Storage ............................................... 47

3.2 Firm Capacity ...................................................................................................................................................... 49

3.3 Grid Benefits ....................................................................................................................................................... 53

3.4 Combined Heat and Power .......................................................................................................................... 54

3.5 Socio-Economic Benefits ............................................................................................................................... 55

3.6 Diversification of Supply Options............................................................................................................... 56

4 CST Cost ............................................................................................................................................................................ 57

4.1 Today’s CST Cost .............................................................................................................................................. 57

4.2 Future CST Cost Developments .................................................................................................................. 60

4.3 GenCost Comparison ...................................................................................................................................... 61

5 Grid Connected Power Generation ......................................................................................................................... 64

5.1 Key Findings........................................................................................................................................................ 64

5.2 Introduction ........................................................................................................................................................ 65

5.3 Predicted CST Uptake ..................................................................................................................................... 66

5.3.1 Approximate analysis for SWIS ................................................................................................................... 69

5.3.2 Combined uptake ............................................................................................................................................. 70

5.4 Quantifying CST Value .................................................................................................................................... 71

5.5 CST’s Role in the System ............................................................................................................................... 74

5.6 Sensitivity to Configuration, Cost, and other Factors ......................................................................... 79

S774Doc-676563417-363 / v0.2 7 CST Value Proposition
6 Remote Area Power Generation .............................................................................................................................. 82

6.1 Key Findings........................................................................................................................................................ 82

6.2 Introduction ........................................................................................................................................................ 83

6.3 Reference Sites .................................................................................................................................................. 84

6.4 Newman Reference Location ....................................................................................................................... 85

6.5 Mount Isa Reference Location ..................................................................................................................... 88

6.6 Market Size.......................................................................................................................................................... 89

7 Industrial Process Heat ................................................................................................................................................ 92

7.1 Key Findings........................................................................................................................................................ 92

7.2 Introduction ........................................................................................................................................................ 93

7.3 Mid Temperature Medium Size End-Use Case ..................................................................................... 93

7.4 Comparative Assessment with PV-eTES .................................................................................................. 97

7.5 Impact of Project Size and Operating Temperature ........................................................................... 99

7.6 Market Size........................................................................................................................................................ 100

8 Green Fuels..................................................................................................................................................................... 104

8.1 Key Findings...................................................................................................................................................... 104

8.2 Introduction ...................................................................................................................................................... 105

8.3 Energy Requirements and End Use Cases ............................................................................................ 105

8.3.1 Overview............................................................................................................................................................. 105

8.3.2 Hydrogen ........................................................................................................................................................... 106

8.3.3 Methanol ............................................................................................................................................................ 110

8.3.4 Ammonia............................................................................................................................................................ 111

8.4 Techno-economic Optimisation of Green Fuel Plants ..................................................................... 113

8.4.1 Methodology .................................................................................................................................................... 114

8.4.2 Benefits of CST for green fuel production ............................................................................................ 116

8.4.3 Economic impacts of CST ............................................................................................................................ 121

8.5 Market Size........................................................................................................................................................ 127

8.5.1 Green fuel market estimates ...................................................................................................................... 127

8.5.2 Potential for CST in green fuels ................................................................................................................ 129

9 Conclusion ...................................................................................................................................................................... 131

S774Doc-676563417-363 / v0.2 8 CST Value Proposition
10 References ...................................................................................................................................................................... 132

Appendix A CST Technologies ................................................................................................................. 134

Appendix B OpenCEM Overview ............................................................................................................ 148

Appendix C Sensitivity Analysis of Uptake in grid connected Systems ................................... 163

Appendix D Green Fuels Data .................................................................................................................. 172

S774Doc-676563417-363 / v0.2 9 CST Value Proposition
List of Tables

Table 2-1: Gen 3 road map pathways, targets, and challenges (NREL, 2021) ........................................... 34
Table 2-2: Summary of major CST milestones worldwide ................................................................................. 37
Table 2-3: Global CST capacity (GW) deployed in relevant IEA and IRENA scenarios towards net
zero .................................................................................................................................................................... 41
Table 3-1: Capacity credits by energy source for 2023-24 as per the WEM ESOO (AEMO, 2022a) . 52
Table 4-1: CST reference configuration .................................................................................................................... 58
Table 4-2: Subsystem specific cost and total cost for the CST reference configuration (NSW
medium) ........................................................................................................................................................... 59
Table 4-3: Example CST cost for nighttime dispatch configurations (2023 / VIC Low) ......................... 59
Table 4-4: Example CST cost for day and nighttime configurations (2023 / VIC Low) .......................... 60
Table 5-1: Summary of results to NEM 2050 CST uptake from various assumption changes. ........... 80
Table 6-1: Reference configurations for each technology ................................................................................ 83
Table 6-2: Remote area reference sites ..................................................................................................................... 84
Table 7-1: Reference sites for process heat assessment .................................................................................... 93
Table 7-2: Size distribution of sites using heat in the range < 500°C ....................................................... 103
Table 8-1: Key properties for the green fuels considered in the study at ambient conditions ....... 105
Table 8-2: An overview of the operating conditions for the selected electrolyser technologies ... 110
Table 8-3: Estimated green fuel market volumes for 2050 ............................................................................ 127

Appendix

Table B-1: Technology options included in modelling .................................................................................... 149
Table B 2: Comparison of ISP2022 and OpenCEM installed capacities for 2050. ................................. 157
Table B-3: Comparison of net demand and total utility generation predicted by OpenCEM .......... 159
Table B-4: AEMO draft cost assumptions for ISP 2024 compared to ISP 2022...................................... 160

Table C- 1: Impact of CST uptake scenarios on emissions and system NPC ............................................ 168

Table D-1: A general overview of the methanol synthesis process, including DAC, normalised to 1
kg of methanol (Bellotti, Rivarolo, & Magistri, 2020) .................................................................. 172
Table D-2: An overview of the ammonia synthesis process, including the ASU, with values
normalised to 1 kg of ammonia (Gilbert, Alexander, Thornley, & Brammer, 2013; Smith,
Hill, & Torrente-Murciano, 2019) ........................................................................................................ 173
Table D-3: Summary of the KPIs of the electrolyser technologies considered in this project. ......... 174
Table D-4: Key advantages and disadvantages of the three electrolyser technologies considered in
this project. .................................................................................................................................................. 175

S774Doc-676563417-363 / v0.2 10 CST Value Proposition
List of Figures

Figure 1-1: Power generation principle using CST with thermal energy storage ...................................... 28
Figure 1-2: Use-cases considered for CST in Australia ......................................................................................... 30
Figure 2-1: Overview of solar power plant options ............................................................................................... 32
Figure 2-2: Global and Australian direct normal irradiation (SolarGIS, 2023) ............................................. 35
Figure 2-3: Australian direct normal irradiation distribution by land area (SolarGIS, 2023) .................. 35
Figure 2-4: Overview of global CST deployment as of today (Q1 2023) ....................................................... 37
Figure 2-5: Global installed CST capacity by country ............................................................................................ 39
Figure 2-6: Global installed CST capacity by technology .................................................................................... 39
Figure 2-7: Sundrop Farms greenhouses and concentrating solar power plant (courtesy of Aalborg
CSP) .................................................................................................................................................................... 43
Figure 2-8: Vast Solar’s CST demonstration power station at Jemalong, New South Wales ................ 44
Figure 3-1: Key services and benefits of CST ............................................................................................................ 46
Figure 3-2: LCoS for different storage capacities (full-load-hours - FLH) ..................................................... 48
Figure 3-3: LCoS results, medium-duration storage (8-hour storage duration, 230 annual cycles)
(CSIRO, 2023a) ............................................................................................................................................... 48
Figure 3-4: LCoS results, long intraday storage (24-hour storage duration, 117 annual cycles)
(CSIRO, 2023a) ............................................................................................................................................... 49
Figure 3-5: Comparative example of CST-PV hybrid and combined cycle (CCGT) providing firm
capacity............................................................................................................................................................. 51
Figure 3-6: Indicative firm capacity of CST plants across the eastern states (Rutovitz et al., 2013) ... 53
Figure 4-1: Considered CST cost reduction scenario in comparison to other reduction scenarios ... 61
Figure 4-2: 2023 GenCost LCoE for various technologies compared with CST .......................................... 62
Figure 4-3: 2030 GenCost LCoE for various technologies compared with CST .......................................... 62
Figure 4-4: Capital costs of storage technologies using GenCost2023 methods and assumptions
with CST corrected. ...................................................................................................................................... 63
Figure 5-1: Transmission system coverage compared to direct normal irradiation ................................. 65
Figure 5-2: Comparison of dispatchable renewable LCoEs using AEMO 2023 draft cost data for
ISP2024 with addition of new CST cost model. (12 hour duration using ISP Draft 2024
Data 6.5%WACC) .......................................................................................................................................... 67
Figure 5-3: OpenCEM results for capacity (top) and dispatch (bottom) in the NEM, for 20h, SM 3.88,
Fichtner cost for CST 35%CAGR , other costs and assumptions from the draft ISP 2023
costs in the Step Change scenario ........................................................................................................ 68
Figure 5-4: OpenCEM results for capacity (top) and dispatch (bottom) from SWIS basic model for
20h, SM 3.88, Fichtner cost for CST....................................................................................................... 70
Figure 5-5: OpenCEM results for CST capacity uptake in the SWIS and NEM ............................................ 70
Figure 5-6: CST value compared to CAPEX and lifetime total cost for the NEM. ...................................... 72
Figure 5-7: CST value compared to CAPEX and lifetime total cost for Western Australia SWIS. ........ 73
Figure 5-8: Accumulated cost and value in the NEM and SWIS combined ................................................. 74
Figure 5-9: Winter week 2050 dispatch for 20 hour SM 3.88 CST system .................................................... 75
Figure 5-10: Summer week 2050 dispatch for 20 hour SM 3.88 CST system ................................................ 76
Figure 5-11: Normalised average hourly CST dispatch for South West NSW (SWNSW) zone ............... 77

S774Doc-676563417-363 / v0.2 11 CST Value Proposition
Figure 5-12: Frequency distribution of CST dispatch compared to PV ............................................................ 78
Figure 5-13: Frequency distribution of CST dispatch compared to wind dispatch level. .......................... 78
Figure 5-14: Frequency distribution of CST dispatch in the NEM compared to residual demand level.
............................................................................................................................................................................. 79
Figure 6-1: Average wind capacity factor for Newman (left) and Mount Isa (right) ................................. 85
Figure 6-2: LCoE based on renewable energy share without consideration of back-up cost (Newman
- 2023) ............................................................................................................................................................... 85
Figure 6-3: LCoE based on different back-up generation cost (Newman - 2023) ..................................... 86
Figure 6-4: LCoE based on different back-up generation cost (Newman - 2030) ..................................... 86
Figure 6-5: Annual generation split for 90% RE share (Newman - 2023) ..................................................... 87
Figure 6-6: Annual average generation profile for 90% RE share (Newman - 2023) ............................... 88
Figure 6-7: LCoE based on renewable energy share (Mt. Isa - 2023) ............................................................. 89
Figure 6-8: LCoE based on different back-up generation cost (Mt. Isa - 2023) ......................................... 89
Figure 6-9: Overview of distribution of mining energy demand and correlation with DNI potential
(adapted from ITP, 2019) ........................................................................................................................... 90
Figure 7-1: LCoH and solar shares for three investigated sites and different solar field sizes ............. 94
Figure 7-2: Combined LCoH for all sites for different solar shares and fuel cost (2023 cost) .............. 95
Figure 7-3: Combined LCoH for all sites for different solar shares and fuel cost (2030 cost) .............. 96
Figure 7-4: LCoH for all sites for CST-TES and PV-ETES (solar only LCoH)................................................... 98
Figure 7-5: LCoH of CST for various DNI levels and process heat consumption (200°C level) ............ 99
Figure 7-6: LCoH of CST for various temperature and process heat consumption (DNI 2,000
kWh/m2/a) ................................................................................................................................................... 100
Figure 7-7: Locations and intensity of energy use and DNI level in Australia 2019-20 ....................... 101
Figure 7-8: Industrial heat use in Australia 2016-17, by sector and temperature of use (ITP, 2019)102
Figure 8-1: PEM-electrolyser - a high-level overview of the mass and energy requirements for green
hydrogen production ............................................................................................................................... 109
Figure 8-2: SOEC-electrolyser - a high-level overview of the mass and energy requirements for
green hydrogen production .................................................................................................................. 109
Figure 8-3: A high-level overview of the mass and energy requirements for the production of
renewable methanol ................................................................................................................................ 111
Figure 8-4: A high-level overview of the mass and energy requirements for the production of green
ammonia ....................................................................................................................................................... 113
Figure 8-5: Energy system model of the Fichtner H2-Optimizer ................................................................... 114
Figure 8-6: Overview of the methodology for the techno-economic optimisation performed........ 114
Figure 8-7: Distribution of heat generated by the CST plant in Case CH 3OH +CST .............................. 116
Figure 8-8: Sorted annual duration curve of heat utilisation from the CST plant in Case CH 3OH +CST
.......................................................................................................................................................................... 117
Figure 8-9: Sources of thermal energy for methanol production in the case with and without CST117
Figure 8-10: Sorted duration curves of the capacity factor of all electricity generators in Case NH3
+CST ............................................................................................................................................................... 118
Figure 8-11: Installed capacity of all electricity sources ...................................................................................... 119
Figure 8-12: Curtailed electricity from all energy sources for each case in the optimisation .............. 120
Figure 8-13: Sorted annual duration curve of the hydrogen production in Case NH 3 +CST ............... 121
Figure 8-14: Hydrogen Storage State of Charge over the year in Case NH3 +CST .................................. 121

S774Doc-676563417-363 / v0.2 12 CST Value Proposition
Figure 8-15: Overall investment costs of all cases ................................................................................................. 123
Figure 8-16: Change of required installed capacity when using PEM or SOEC in combination with CST
.......................................................................................................................................................................... 124
Figure 8-17: Levelised cost of hydrogen when using SOEC in combination with a CST plant ............ 125
Figure 8-18: Relative change of levelised cost of the final product................................................................ 126
Figure 8-19: Total Market Potential for CST in 2050 with varying shares of sites suitable for CST and
varying market shares of CST ............................................................................................................... 130

Appendix

Figure A-1: Schematic of parabolic trough power plant with storage system ......................................... 136
Figure A-2: Typical layout of a medium scale solar field ................................................................................... 137
Figure A-3: Parabolic trough collector development (source Flabeg) ......................................................... 138

Figure B-1: NEM zones used by AEMO and replicated in OpenCEM (source AEMO) ........................... 148
Figure B-2: NEM transmission between planning regions from AEMO ...................................................... 154
Figure B-3: Capacity expansion results by technology from AEMO ISP 2022 (top) and OpenCEM
using AEMO assumptions (bottom) ................................................................................................... 156
Figure B 4: Dispatched energy results by technology from AEMO ISP 2022 (top) and OpenCEM
using AEMO assumptions (bottom). .................................................................................................. 158
Figure B 5: Comparison of dispatchable renewable LCOEs using cost data from ISP 2022............... 160
Figure B-6: Comparison of dispatchable renewable LCOEs using AEMO draft cost data for ISP2024
with addition of new CST cost model ............................................................................................... 162

Figure C-1: Comparison of dispatchable renewable LCoEs using AEMO draft cost data for ISP2024
with addition of new CST cost model ............................................................................................... 163
Figure C-2: Variation of Solar Multiple for CST with 20hour storage duration ........................................ 164
Figure C-3: Change in Capacity (left) and dispatched energy (right), ISP2022 vs Draft ISP2023
assumptions................................................................................................................................................. 165
Figure C-4: Change in Capacity and dispatched energy from introducing new cost model vs draft
ISP2024 .......................................................................................................................................................... 165
Figure C-5: Change in Capacity and dispatched energy from a 10% CST CAPEX reduction 20h SM
3.88. ................................................................................................................................................................. 166
Figure C-6: Change in Capacity and dispatched energy from a 10% CST CAPEX increase 20h SM 3.88
.......................................................................................................................................................................... 166
Figure C-7: Change in Capacity and dispatched energy from increasing battery costs by 10% ...... 166
Figure C-8: Capacity trajectories to 4GW by 2050 as a function of CAGR. ................................................ 167
Figure C-9: Change in Capacity and dispatched energy fixed vs optimal dispatch strategy ............. 169
Figure C-10: Change in Capacity (left) and dispatched energy (right) removing Snowy 2.0, CST at 20h
SM 3.88 .......................................................................................................................................................... 169
Figure C-11: Change in Capacity and dispatched energy when coordinated DER batteries are not
forced in. ....................................................................................................................................................... 170

S774Doc-676563417-363 / v0.2 13 CST Value Proposition
Figure C-12: Change in Capacity (left) and dispatched energy (right) when coordinated DER batteries
are not forced in, but emissions are held constant. .................................................................... 170
Figure C-13: Change in Capacity (left) and dispatched energy (right) from zero emissions by 2040
trajectory ....................................................................................................................................................... 171

Figure D-1: The relative positioning of the three types of electrolyser technologies with respect to
technological maturity and physical equipment size. ................................................................. 176

S774Doc-676563417-363 / v0.2 14 CST Value Proposition
List of Abbreviations

Abbreviation Description

A-CAES Advanced Adiabatic Compressed Air Energy Storage

AEMO Australian Energy Market Operator

ALK Alkaline Water Electrolysis

ANU Australian National University

APS Announced Pledges Scenario

ARENA Australian Renewable Energy Agency

ASTRI Australian Solar Thermal Research Institute

ASU Air Separation Unit

BESS Battery Energy Storage System

BNEF Bloomberg New Energy Finance

BOM Bureau of Meteorology (Australia)

BOOT Build Own Operate Transfer

BOP Balance of Plant

CAGR Compound Annual Growth Rate

CAAGR Compound Average Annual Growth Rate

CAPEX Capital Expenditures

CC Capacity Credit

CCGT Combined Cycle Gas Turbine

CCS Carbon Capture and Storage

CEFC Clean Energy Finance Cooperation

CEPCI Chemical Engineering Plant Cost Index

CH3OH Methanol

CHP Combined Heat and Power

CO2 Carbon Dioxide

CSIRO Commonwealth Scientific and Industrial Research Organisation

COD Commercial Operation Date

CSP Concentrated Solar Power

S774Doc-676563417-363 / v0.2 15 CST Value Proposition
Abbreviation Description

CST Concentrated Solar Thermal

CSTA China Solar Thermal Alliance

DAC Direct Air Capture

DER Distributed Energy Resource

DEWA Dubai Electricity and Water Authority

DKIS Darwin Katherine Interconnected System

DNI Direct Normal Irradiation

DOE Department of Energy

DPP Deeper Decarbonisation Perspective

DSG Direct Steam Generation

DSP Demand Side Program

EC European Community

EPC Engineering, Procurement, and Construction

ESOO Electricity Statement of Opportunities

ESR Electric Storage Resource

ESS Energy storage system

ESTELA European Solar Thermal Electricity Association

eTES Electric Thermal Energy Storage

EU European Union

EV Electric Vehicle

FIT Feed-In Tariff

FID Final Investment Decision

FLH Full-Load-Hours

FOM Fixed Operations and Maintenance

FSC Field Supervisory Control

GenCost Generation Cost report by CSIRO

GDP Gross Domestic Product

GHI Global Horizontal Irradiation

H2 Hydrogen

S774Doc-676563417-363 / v0.2 16 CST Value Proposition
Abbreviation Description

HCE Heat Collection Element

HP High Pressure

HTF Heat Transfer Fluid

IASR Input Assumption Scenario Report

IATA International Air Transport Association

ICE Internal Combustion Engines

IEA International Energy Agency

IRENA International Renewable Energy Agency

IPP Independent Power Purchase

IRR Internal Rate of Return

ISP Integrated System Plan

KNO3 Potassium Nitrate

KPI Key Performance Indicator

LCoA Levelised Cost of Ammonia

LCoE Levelised Cost of Electricity

LCoH Levelised Cost of Heat

LCoH Levelised Cost of Hydrogen

LCoM Levelised Cost of Methanol

LCoP Levelised Cost of Product

LCoS Levelised Cost of Storage

Li-ion Lithium Ion

LOC Local Controller

LP Low Pressure

MENA Middle East and North Africa

MMBTU Million British Thermal Units

NaNO3 Sodium Nitrate

NECPs National Energy and Climate Plans

NEM National Electricity Market

NH3 Ammonia

S774Doc-676563417-363 / v0.2 17 CST Value Proposition
Abbreviation Description

NPC Net Present Cost

NREL National Renewable Energy Laboratory

NPI National Pollution Inventory

NPV Net Present Value

NSW New South Wales

NT Northern Territory

NWIS North West Interconnected System

NZE Net Zero Emissions

OCGT Open Cycle Gast Turbine

OEM Original Equipment Manufacturer

O&M Operations and Maintenance

OPEX Operational Expenditure

QLD Queensland

PEM Proton Exchange Membrane

PHES Pumped Hydro Energy Storage

PPA Power Purchase Agreement

PtH Power-to-Heat

PV Photovoltaics

Q Quarter

RCM Reserve Capacity Mechanism

R&D Research & Development

RE Renewable Energy

RET Renewable Energy Target

SA South Australia

SAF Sustainable Aviation Fuel

SAM System Advisor Model

SCA Solar Collector Assemblies

SCE Solar Collector Element

SDGs Sustainable Development Goals

S774Doc-676563417-363 / v0.2 18 CST Value Proposition
Abbreviation Description

SEGS Solar Electricity Generating System

SM Solar Multiple

SOEC Solid Oxide Electrolyser Cells

SSG Solar Steam Generators

SWIS Western Australian South West Interconnected System

TES Thermal Energy Storage

TRL Technology Readiness Level

UAE United Arab Emirates

UNSW University of New South Wales

U.S. United States

USGS United States Geological Survey

UTS University of Technology Sydney

VIC Victoria

VOM Variable Operations and Maintenance

VRE Variable Renewable Energy

WA Western Australia

WACC Weighted Average Cost of Capital

WEM Wholesale Electricity Market

WEO World Energy Outlook

WETO World Energy Transitions Outlook

S774Doc-676563417-363 / v0.2 19 CST Value Proposition
Executive Summary

Concentrated Solar Thermal (CST) uses mirrors to concentrate the sun’s rays to a small area (receiver) to produce medium to high temperature heat (from 150°C up to 1,000°C or beyond). The heat can be used immediately or stored in a Thermal Energy Storage (TES) system- for multiple hours or even days. The heat can be used directly as industrial process heat or further utilised to generate electric power or to drive chemical processes.

This report examines the values that CST can provide to Australia across grid connected and remote / off- grid power generation, industrial process heat and green fuels production.

CST
Linear Fresnel Parabolic Trough Solar Tower Parabolic Dish

Thermal Energy
Storage

Grid Connected Remote Area Process Heat Green Fuel
Power Power Production

Globally, power generation has, so far, been the most common use case for CST1, with more than 100 commercial plants in operation with a total installed capacity of over 6.5 GWe and around 4 GWe more under construction. The first commercialisation of CST started in 1984 in the U.S. (California) with nine CST plants put into operation, most of which operated for more than 30 years. The largest CST project to date is the Noor Energy 1 project in Dubai with a total CST capacity of 700 MW and storage capacities of up to
15 hours. This project has also offered the lowest CST tariff to date at US$ 73 /MWh (A$ 102/ MWh).

Compared to the power sector, CST deployment for industrial process heat is lagging behind. This reflects the overall lag in the decarbonisation of this important sector- accounting for around 28% of global final energy consumption. Today, there is only around 400 MWt of CST process heat capacity in operation, but the sector is gaining momentum with several medium to large-scale (> 1 GWt) projects under construction or development. Concerning the green fuel market there are so far no commercial projects in operation using CST for part of their power and heat supply. There are however first projects under development considering CST.

While CST has been deployed in multiple locations around the world and despite the favourable solar resource, there is so far no commercial CST plant in operation in Australia and the technology is still

1
For power generation, the technology is also often referred to as concentrated solar power (CSP)

S774Doc-676563417-363 / v0.2 20 CST Value Proposition
mistakenly viewed as an emerging technology by some stakeholders. Australia is among the regions with the world’s best Direct Normal Irradiation (DNI), relevant for CST. More than 85% of Australia’s land mass has DNI levels considered suitable for CST. While Australia offers regions with up to 2,900 kWh/m²/a, most of the globally installed CST capacity is at sites with lower DNI. For example, the site of the Noor
Energy 1 project offers only around 2,100 kWh/m²/a and most of China’s CST capacity - currently there are 3 GWe under construction - is deployed at sites with less than 2,100 kWh/m²/a.

CST Value
The inherent “value” of CST is the result of the particular services and benefits it offers. Further, the key features the value proposition has in common across end use applications is that CST will generally complement other renewable generation options and justify higher generation cost compared to Variable
Renewable Energy (VRE) options. As such CST is regarded by both the International Energy Agency (IEA) and the International Renewable Energy Agency (IRENA) as one of the technology options to support the required energy transitions towards net zero, with several 100 GW e of CST capacity required. Thermal energy storage is the key and distinctive feature CST offers, forming the basis for several sources of value.

S774Doc-676563417-363 / v0.2 21 CST Value Proposition
In economic terms the financial ‘’value’’ of CST can be expressed as the sum of its lifetime financial benefits suitably quantified and discounted to the time of construction. This can be compared to its total cost being the capital cost plus the lifetime sum of operating cost discounted to the time of construction.

The respective value and overall contribution CST can offer depend on the prevailing boundary conditions and the type of end-use application.

CST Cost
To calculate the CST cost for the present study, a cost model for CST in Australia has been developed. The model is broken down into the three main sub-systems: solar field, thermal energy storage and power block. By breaking down the cost into these three subsystems, the cost for CST plants can be calculated for a wide range of configurations by simply providing the solar field size, the thermal energy storage size, and the power block size.

The underlying specific cost figures have been determined for a green field solar tower reference plant, using a more detailed cost model. The cost factors, cost multipliers and scaling exponents are based on international CST projects Fichtner has been involved in, supplier budget quotes for main equipment, and recent stakeholder engagements.

Considering a plant configuration for mainly night-time dispatch (14 hours of storage) and making use of the economies of scale, in particular in the power cycle, a reference plant with the following key parameters has been defined.

Item Unit Reference Specific Scaling Reference
Value Cost Exponent Cost2
[A$/UNIT] [mA$]

Power block (net) MWe 140 2,028,795 0.8 284

Thermal energy MWht 4,667 35,880 0.85 167
Storage

Solar Field MWt 720 644,230 0.88 464

To allow the calculation of CST cost for a wide range of configurations, scaling exponents are provided for each of the three cost portions. Further, the model includes regional cost factors to adapt the reference cost to different regions in Australia.

2
The reference location was nominated as an inland location in NSW, I.e., the “NSW medium” region is the nominated reference region with all other regions then scaled accordingly.

S774Doc-676563417-363 / v0.2 22 CST Value Proposition
Grid Connected Power Generation

For large-scale grid systems typically CST systems of 100 MWe and above are deployed. Large scale grid connected CST options have been investigated for both the National Electricity Market (NEM) and the
Western Australian Southwest Interconnected System (SWIS). The assessment has been conducted by means of capacity expansion modelling, considering multiple expansion scenarios.

Key findings and conclusions:
▪ Modelling capacity expansion in the NEM with OpenCEM using the Fichtner CST cost model and
AEMO's draft cost models for other technologies, shows CST uptake of 5.6 GW by 2050, dispatching
around 10% of total electricity.
▪ Approximate modelling of capacity expansion in the SWIS suggests rapid CST uptake from 2030
onwards growing to 840 MW by 2050 and dispatching as much as 20% of total electricity. This is a
greater fraction compared to the NEM, linked to an absence of existing hydro and smaller levels of
coal generation.
▪ Modelling whole electricity system OpenCEM results for CST capacity uptake in the SWIS and NEM

capacity expansion with and without 7,000

the inclusion of CST allows the financial 6,000

value of a CST system to be deduced 5,000
Installed Capacity [MWe]

via the overall reduction in total system 4,000

annualised cost. 3,000

▪ Direct comparison of CST system 2,000

lifetime financial value with lifetime CST
1,000

system cost shows that the 2023 deficit
0
between cost and value is lowest in the 2023 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050

Year
SWIS and shows the earliest break even SWIS NEM

between cost and value in 2025. In the
longer term Southwest NSW offers the greatest net value.
▪ The inherent value of CST systems in grid connected applications is their ability to fill in the gaps with
long intraday storage when variable wind and solar photovoltaics (PV) generation cannot meet
demand. This becomes most apparent as large legacy fossil fuel generation systems are progressively
retired.
▪ AEMO’s past ISP modelling used fixed dispatch traces for CST as opposed to optimal dispatch. This
results in predictions of no CST uptake. However, uptake is predicted to occur when using optimal
dispatch.
▪ Testing varied CST system configurations indicates that longer durations of storage and higher solar
multiples deliver the best overall system value, even though this results in some summer dispatch
during the day.
▪ The extra societal benefits from regional employment and diversity in technology options should more
than justify early actions in Australia to ensure the establishment of a viable pipeline of CST projects.
▪ To capture the whole of system benefits that CST would deliver, a realistic growth rate of installation is
needed to establish capability and supply chains. A realistic uptake trajectory beginning around 2025,
while likely to require policy intervention, would see steady CST system growth that would leave the
system around A$10b better off by 2050.

S774Doc-676563417-363 / v0.2 23 CST Value Proposition
Remote Area Power Generation

For the remote area power supply end-use application, hybrid systems, with and without a CST element, have been assessed. For this purpose, two representative mining areas (Pilbara and Mt. Isa region) have been selected and three options have been assessed for a 100 MWe baseload reference case. The three options are: (i) CST incl. TES in combination with solar PV, (ii) CST incl. TES in combination with solar PV and wind and (iii) PV and wind in combination with a battery energy storage system (BESS).

Key findings and conclusions:
▪ Australia with its large mining sector, mainly located in remote locations, and ESG driven emission
reduction targets, has a huge demand for remote area renewable power generation - ultimately
requiring 100% renewable and hence emissions free mining operations.
▪ Adding CST to the power mix becomes increasingly beneficial when higher renewable shares are being
targeted and, hence, medium to large storage
LCOE and annual average generation profile for 90% RE share
capacities are required. (Newman case study – 2023 cost basis)
▪ For high renewable shares, combining CST with CST+PV
144 $/MWh
solar PV and (in some cases) wind, results in the
lowest LCoE for remote area power generation.
▪ High back-up generation cost - to provide the
last and most expensive 10 - 20% percent of
generation - strengthens the CST business case. CST+PV+Wind
149 $/MWh
The higher the back-up generation cost, the
more beneficial are higher RE shares and, thus,
larger storage capacities. Avoidance of high
back-up generation costs also justifies higher
regional / remote area cost multipliers
associated with higher renewable shares PV+Wind
160 $/MWh
(considered as part of the assessment).
▪ The higher the complementarity between PV
and wind resources, the lower the benefit of
adding CST into the energy mix.
▪ Additional values from capacity value, grid
services or combined heat and power generation (in case of additional process heat demand), when
factored in will further support the value proposition of CST for 100% RE (island) grids. Consideration
should be given to these additional values.
▪ When compared to solar PV and wind, the system size (via economies of scale) has a bigger impact in
case of CST. Thus, for smaller remote area systems, the advantage of including CST in the power mix
will reduce, while for larger systems it will increase.
▪ CST should be considered (and adequately assessed) as a dispatchable option for high RE shares,
especially for larger systems, when PV and wind do not complement well and there is high back-up
energy cost.

S774Doc-676563417-363 / v0.2 24 CST Value Proposition
Industrial Process Heat

The process heat market comprises a wide range of industries and end-use applications, requiring process heat at different temperatures and types, such as hot water, steam, or hot air. Required capacities range from single digit MWt systems to GWt systems. Thus, CST process heat systems have been investigated for different temperature ranges, capacities, and sites (solar resource levels). Further, a comparative assessment has been conducted considering both solar PV in combination with an electrically heated thermal energy storage (eTES) and gas fired boilers.

Key findings and conclusions:
▪ The process heat market comprises a wide range of industries and end-use applications, requiring
process heat at different temperatures, in different forms (hot water, steam or hot air) and at different
times.
▪ Considering today’s cost level and a good DNI (>2,500 kWh/m²/a), CST becomes competitive for fuel
prices above A$ 60 /MWh (A$ 16.7 /GJ) for solar shares of up to 70 - 75%.
▪ For CST cost projected for 2030, the combined LCoH (solar and back-up fuel) will reduce by 10% - 20%
depending on the solar share. The breakeven Combined LCOH for different solar shares and fuel cost
(site with good solar resource >2,500 kWh/m²/a)
fuel price consequently reduces to around A$ 45
2023
/MWh (A$ 12.5 /GJ) by 2030.
▪ The location, i.e., solar resource (DNI level) but
also the latitude, which impacts the solar field
efficiency, has a strong impact on the LCoH.
Within the considered DNI range (2,000 - 2,900
kWh/m²/a), the LCoH is nearly proportional to
the DNI.
▪ CST offers the advantage of providing thermal
2030
energy directly and efficiently. As such CST
requires significant less land area when
compared to PV for power-to-heat (PtH) in
combination with eTES.
▪ For the investigated sites with a good solar
resource, the CST option results in a 20 - 30%.
lower LCoH when compared to the PV-eTES
option. For the site with a comparatively low
solar resource (2,000 kWh/m²/a), the PV-eTES option results in similar LCoH in case of a solar share of
50%. For decreasing DNI levels PV-eTES delivers a progressively lower LCoH.
▪ The optimal process temperature range for CST is between 150°C to 500°C. However, different types of
CST technologies can cover also higher process temperatures of up to >1,000°C.
▪ CST should be considered (and adequately assessed) for industrial sites with sufficient adjacent land.

S774Doc-676563417-363 / v0.2 25 CST Value Proposition
Green Fuel Production

Green fuels (also referred to as e-fuels or sustainable fuels) are carbon-neutral or carbon-free alternatives to fossil fuels and can be produced from hydrogen made with renewable electricity, via biosynthesis or via thermochemical processing of water or other feedstocks. This study focuses on the green fuels of ammonia and methanol, as well as hydrogen produced via water electrolysis using CST in combination with other renewable sources.

Key findings and conclusions:
▪ Green fuels production benefits from CST as a source of renewable power and heat. Overall, a
reduction of green fuel production cost of approximately 8% to 40% was achieved by adding CST to
the energy supply mix and combining CST with other renewable sources.
▪ The ability to deliver a combined heat and power (CHP) solution within the one technology has major
benefits and is one of the key value propositions of CST. Specifically, CST can provide an optimal
balance of renewable heat and power, as required for the production of different renewable fuels.
▪ Hydrogen production systems based on solid oxide electrolyser cells (SOEC) technology explicitly
profit from CST as a result of the Relative change of levelised cost of the final product (for site with high
solar resource 2,900 kWh/m²/a)
low-cost heat supply and Relative change of levelised cost of final product

increased hydrogen production
120%

100%

yields, leading to a cost reduction
Relative Cost [%]

80%

of approximately 10% (compared 60%

40%
to the PEM case without CST). As
20%

hydrogen is the key feedstock for 0%
Case H₂ Case H₂ Case H₂ Case CH₃OH Case CH₃OH Case CH₃OH Case NH₃ Case NH₃ Case NH₃
methanol and ammonia +CST +CST +SOEC +CST +CST +SOEC +CST +CST +SOEC

production, these fuels also profit from cheaper hydrogen.
▪ Firm capacity of CST is enabled by the TES, which provides flexibility and enables the dispatchability of
the plant. The dispatchability allows for the power and heat demand to be met during times of low
sunlight and or wind resource. This results in a lower requirement of installed capacity (reduction of
approximately 15% to 30%) of solar PV and wind and less curtailment of energy overall.
▪ The flexibility introduced with the TES of the CST plant is especially beneficial for less flexible
downstream processes to produce hydrogen derivatives, such as methanol or ammonia, by reducing
the need for other storage systems, such as BESS, that are often significantly more expensive.
▪ Overall, the production cost reduction of green fuels heavily depends on the green fuel itself and the
chosen boundary conditions. Generally, systems with a high heat demand and low operational
flexibility profit most from CST.
▪ For systems with a high heat demand and low operational flexibility CST should be considered as part
of the energy mix as they show the largest production cost reduction compared to systems without
CST.

S774Doc-676563417-363 / v0.2 26 CST Value Proposition
Conclusion

Australia offers prime conditions to deploy CST competitively, while benefiting from its multiple sources of value. CST can play an important role in Australia’s transition towards net zero, providing both dispatchable power and heat, and, thus, being applicable for a broad range of end use sectors requiring rapid decarbonisation.

CST is a complementary technology and delivers the best energy system outcome, when it is integrated with the VRE technologies, solar PV and wind. CST can become a key enabler to fully replace dispatchable coal and gas-fired generation. In assessing costs and benefits it needs to be compared with fossil fired and other dispatchable renewable generators rather than variable solar PV or wind.

In order to unlock Australia’s CST potential in the different sectors, CST must (i) be recognised as a mature technology which has been deployed internationally for many decades, (ii) be considered as one of the potential solutions, and (iii) be adequately and fairly assessed in capacity expansion and other modelling - in particular in regard to its dispatchability features and other sources of value.

Establishing policies that encourage an early start and smooth growth of CST uptake in all sectors can save the country many billions of dollars in reaching net zero as well as maximising local economic benefit and adding to diversity in energy supply. New policies, energy market measures and decarbonisation strategies should be technology neutral but should be designed to encourage the long duration energy storage and dispatchable behaviour needed in the long term.

S774Doc-676563417-363 / v0.2 27 CST Value Proposition
1 Introduction
Australia has committed to achieve net zero emissions by 2050 and has adopted a national 43% emission reduction target (below 2005 levels) by 2030. To combat climate change and meet emission reduction targets, Australia must accelerate its energy transition and reduce its reliance on fossil fuels in all sectors.

Given Australia’s access to abundant, coal and gas, there is still a high dependency on fossil fuels. Whilst these fossil fuels have historically been cheap, international market demands are resulting in increased domestic costs. On the other hand, Australia has prime and abundant renewable energy resources, in particular solar and wind, and large areas of land, Thus, decarbonisation should be easier and more cost effective for Australia, in comparison with many other countries. A good example is rooftop solar PV for which Australia has the highest penetration in the world.

Recent investment in utility scale solar PV and wind has also been the highest Australia has seen in many years and there are several large GW scale green fuel (green hydrogen) projects under development, both for the local but in particular also for the export market, aiming to make use of Australia’s competitive advantage in terms of its prime renewable energy resources.

Despite its prime solar resource and more favourable boundary conditions, Australia is lagging behind regarding one distinctive type of renewable energy technology - concentrated solar thermal.

This report investigates the value proposition for CST in Australia at present and into the future and for the four key use cases of; grid connected electricity generation, remote area power generation, industrial process heat and green fuels.

CST technology uses different types of reflective surfaces to concentrate the sun’s rays to a small area (the receiver) to generate medium to high temperature heat (from 150°C up to 1,000°C or beyond). The generated heat can be used immediately or stored in a TES system- for multiple hours or even days. The heat can be used directly as (industrial process) heat or further utilised to generate electric power or to drive chemical processes. Electric power generation is up until today the most common use case, with more than 100 commercial plants in operation (> 6.5 GWe). For power generation, the technology is also often referred to as concentrated solar power (CSP). An overview of the different technology options, their developments, and deployments are provided in Section 2 of this report.

Figure 1-1: Power generation principle using CST with thermal energy storage

The inherent “value” of CST is the result of the particular services and benefits it offers, further described within Section 3.

The key features the value proposition has in common across the different end use applications is that
CST will generally complement other renewable energy options and justify higher generation cost

S774Doc-676563417-363 / v0.2 28 CST Value Proposition
compared to variable renewable energy (VRE) options. Related CST costs and their projected developments are addressed in Section 4.

CST is an inherently flexible technology. It has a wide range of operating temperature, with its thermal energy (heat) able to be used directly or stored as a cheaper form of energy for use when required. As a result of this flexibility, CST has multiple end-use applications.

These are:

▪ District heating and cooling,
▪ Solar (thermal) desalination,
▪ Industrial process heating,
▪ Grid connected power generation (utility scale),
▪ Remote area power generation (off-grid),
▪ Green / solar fuel production (electrochemical via electrolysis - short to medium term), and
▪ Green / solar fuel production (thermochemical - medium to long term).

The four most promising of these end-use applications for Australia have been investigated in more detail as part of the presented study.

Grid connected power generation (Section 5): For large-scale grid systems typically CST systems of
100 MWe and above are deployed. Large scale grid connected CST options have been investigated for both the NEM and the SWIS. The assessment has been conducted by means of capacity expansion modelling, considering multiple expansion scenarios.

Remote Area Power Generation (Section 6): For the remote area power supply end-use application, hybrid systems, with and without a CST element, have been assessed. For this purpose, two representative mining areas (Pilbara and Mt. Isa region) have been selected and three options have been assessed for a
100 MWe baseload reference case. The three options are: (i) CST incl. TES in combination with solar PV, (ii)
CST incl. TES in combination with solar PV and wind and (iii) PV and wind in combination with a BESS.

Industrial Process Heat (Section 7): The process heat market comprises a wide range of industries and end-use application, requiring process heat at different temperatures and types such as hot water, steam, or hot air. Required capacities range from single digit MWt systems to GWt systems. Thus, CST process heat systems have been investigated for different temperature ranges, capacities, and sites (solar resource levels). Further, a comparative assessment has been conducted considering both solar PV in combination with an electrically heated thermal energy storage (eTES) and gas fired boilers.

Green / solar fuel production (Section 8): Also in case of the green fuel production, there is a wide range of processes and green fuels. They can be produced from hydrogen made with renewable electricity, or via biosynthesis or via thermochemical processing of water or other feedstocks. The study focuses on ammonia and methanol production as well as standalone hydrogen production via electrolysis using CST with other RE sources.

S774Doc-676563417-363 / v0.2 29 CST Value Proposition
CST
Linear Fresnel Parabolic Trough Solar Tower Parabolic Dish

Thermal Energy
Storage

Grid Connected Remote Area Process Heat Green Fuel
Power Power Production

Figure 1-2: Use-cases considered for CST in Australia

The Australian Solar Thermal Research Institute (ASTRI) has commissioned Fichtner Australia Pty Ltd - part of the Fichtner Group - and ITP Thermal Pty Ltd - part of the ITP Energised Group, to assess the role and value of CST technologies and systems within Australia’s future energy landscape and to provide this study.

ASTRI is a consortium of leading Australian research institutions, which was established to provide a coordinated, national approach to the development and demonstration of solar thermal technologies. The focus of ASTRI’s activities is on high temperature CST systems. ASTRI was established in 2012 through a contractual arrangement between the Australian Renewable Energy Agency (ARENA) and the
Commonwealth Scientific and Industrial Research Organisation (CSIRO). Under this arrangement, ASTRI’s primary objective is to facilitate the commercial uptake of more efficient, higher temperature solar thermal technologies and CST systems.

S774Doc-676563417-363 / v0.2 30 CST Value Proposition
2 CST Technologies: Technology Options,
Developments and Deployments

2.1 CST Technology Overview
Concentrated solar thermal technologies are used to generate medium or high temperature heat, which can either be used directly as process heat or further utilised to generate electric power or drive chemical processes. In the case of electric power generation, the technology is also often referred to as concentrated solar power (CSP).

Due to the optical concentration process, CST technologies can only make use of the direct normal irradiation (DNI). In comparison, PV makes use of both the direct and the diffuse irradiation, the sum of which is measured by global horizontal irradiation (GHI).

The four main types of CST technologies developed and tested so far are:

▪ Parabolic trough,
▪ Linear Fresnel,
▪ Solar tower, and
▪ Parabolic dish.

Parabolic trough and linear Fresnel systems use single-axis tracking collectors (linear-focusing), whereas solar tower (either single tower or multiple towers) and parabolic dish systems use dual-axis tracking
(point-focusing). The three CST technologies that have been commercially deployed so far are parabolic trough, linear Fresnel, and solar tower. Parabolic trough technology has been the dominating technology up till now and accounts for more than 75% of the operational capacity (by end of 2022 the total is around 6.5 GWe for power generation and around 400 MWt for process heat). However, solar tower technology, using molten salt as the heat transfer fluid, has emerged as the favoured approach in recent years and there are now several commercial projects in operation and more than 10 projects under construction. An overview of the different solar power plant technologies is depicted below.

Depending on the CST technology and required application, different heat transfer fluids and conversion principles are applied. The most common heat transfer fluids are synthetic oil, molten nitrate salts and water/steam, i.e., direct steam generation (DSG). The plant concepts may include heat exchange from a primary heat transfer fluid to the process medium used for power generation in conventional power cycles (e.g., heat exchange from synthetic oil to water/steam) or steam may be produced directly in the solar field or the central receiver for industrial process heat applications.

For steam cycle power plants, cooling systems are required to condense the steam at the turbine exhaust.
Cycle efficiency can be improved by decreasing the turbine back-pressure. Depending on the availability and costs of water as well as other boundary conditions, different condenser cooling systems are applied.
They can be classified in three main categories: wet cooling, dry cooling, and hybrid cooling. Dry cooling is often the only viable option at CST sites, typically being placed in remote and dry locations.

S774Doc-676563417-363 / v0.2 31 CST Value Proposition
Solar Power
Plants
Solar Thermal Photovoltaic
(CSP) (PV)

Non – Linear-focusing Point-focusing
Concentrating (single axis) (dual axis)

Solar- Linear Parabolic Central Concentrating Non-
Dish
Chimney Fresnel Through Receiver (CPV) Concentrating

Concentration Ratio

Thermal Energy Storage (TES) Battery Energy Storage System (BESS)

Rankine Cycle (ST)
Wind
Turbine Integrated Solar Combined Cycle DC-AC Inverter

Brayton Cycle

Stirling
Engine

Electric Power

Figure 2-1: Overview of solar power plant options

The industry standard parabolic trough power plants use thermal oil as the heat transfer fluid (HTF), a steam Rankine power cycle, and two-tank indirect molten salt TES. Given the upper temperature limit of the HTF oil of around (390°C), both the specific storage capacity and the overall plant efficiency is limited.
Thus, there are new concepts under development, using either a silicone based HTF to allow for a wider operating temperature range (e.g., Helisol with -40 to 425°C range) or directly using molten salt as the
HTF, to allow fordirect molten salt storage combined with higher operating temperatures. For industrial heat applications typically linear-focusing systems are used, often providing direct steam or pressurised water.

The most common deployed solar tower design is based on a central externally irradiated receiver, using molten salt as the HTF, a two-tank direct molten salt TES and a steam Rankine power cycle. The state-of- the-art molten salt is a sodium nitrate-potassium nitrate mixture referred to as “solar salt”. Solar salt allows operating temperatures to approximately 565°C, thus, enhancing the specific storage capacity compared to a trough system, hence, reducing specific TES cost, while increasing the overall plant efficiency. Due to the higher HTF and storage temperatures, the first large-scale commercial tower plants, such as Crescent Dunes, experienced some failures in the molten salt system and the hot-salt tanks (NREL,
2020). The issues are now known and can be addressed in new CST projects using molten salt as HTF and storage media.

Given the intermediate conversion of solar energy into thermal energy, CST technologies offer benefits when compared to cheaper but variable renewable energy (VRE) technologies, such as PV and wind. CST plants can be equipped with TES, making them dispatchable and providing firm capacities. Solar heat collected during the daytime can be stored in TES systems, based on various materials (mainly molten salt, but also other emerging storage materials such as particles or ceramics - depending on the CST technology). During times with no or insufficient solar irradiation or during peak demand periods, thermal

S774Doc-676563417-363 / v0.2 32 CST Value Proposition
energy can be extracted from the storage to run a power cycle, industrial process, or green fuel production continuously.

Given the TES is the key feature of CST all CST plants currently under construction or under development are equipped with a TES system. Back-up heaters (fossil and/or using green fuels) can also be incorporated in a CST plant, which further increases the level of power supply stability and the degree of dispatchability. Additional information on the general benefits CST plants offer is provided in Section 3 and further detailed for the respective end-use cases in the subsequent parts of this report.

Another option are CST-PV hybrid plants, combining low-cost (on-sun) electricity generation by PV and the cheap thermal energy storage available to CST plants, which offer additional benefits to the grid (see
Section 3). The combination of both technologies in one plant will allow for dispatchable (solar) power and (if required) heat generation with high-capacity factors of up to 90% at lower (combined) levelised cost compared to a pure CST plant. Depending on the required dispatch profile the CST storage system is optimised accordingly, and a small BESS might be added, to allow for more flexibility between the operating modes. Further, part of the PV generation - which might otherwise be curtailed depending on the system design - can also be stored in the TES using an electrical heater. Depending on the system and
TES design, PV power might be even used to increase the storage temperature, both decreasing storage cost and increasing the overall plant efficiency3, Another positive effect of CST-PV hybrid plant are the additional cost savings both for the initial investment (e.g., development cost, grid connection, shared facilities, etc.) as well as during the operation period.

Research and development activities in CST currently focus on solar tower technologies. The development of next-generation CST plants, with target operating temperatures of more than 700°C, is being led by the
U.S. DOE Gen3 CST program, in which ASTRI and CSIRO are cooperating parties, and the EU through its
Horizon Europe research and innovation funding program.

The Gen3 effort is looking at three pathways, using liquid, particles, or gas in the receiver, to lower costs and increase efficiency. Each pathway is also working to reduce heliostat field costs, which make up the largest part of installed capital costs. The Gen 3 effort is also employing sCO2 power cycles that can operate at higher efficiencies than the steam Rankine cycles currently used in commercial plants
(NREL, 2021). An overview of the three pathways, component cost and performance targets, and risks for
CST Gen 3 technology identified in the Gen 3 road mapping study, is summarised in Table 2-1. When demonstrated and deployed successfully, the new generation of CST plants will enable a step change reduction in the cost of CST. With regards to CST for industrial process heat, there are additional developments, focussing on low-cost line focusing systems.

An Australian example for a Gen 3 CST technology is Vast Solar’s high-temperature sodium-based heat transfer technology and modular tower design, as further outlined in Section 0. A good overview of ongoing Gen 3 developments with Australian participation is provided in ASTRI’s latest Public
Dissemination Report (ASTRI, 2022).

3
For example, in case of parabolic trough power plants using the in a state-of-the-art HTF, the solar field and hence storage
temperature is limited to around 390°C. Using an electrical heater, the TES temperature can be lifted to around 560°C in case
of a state-of-the-art two tank molten salt storage system.

S774Doc-676563417-363 / v0.2 33 CST Value Proposition
Table 2-1: Gen 3 road map pathways, targets, and challenges (NREL, 2021)

2.2 Australia’s Solar Resource
For CST technologies, which use concentrated solar radiation, the direct normal irradiation (DNI) is the parameter that is relevant4. For large-scale CST plants, sites with an annual DNI sum of more than
1,800 - 2,000 kWh/m²/a are generally regarded as suitable5. In case of solar process heat applications, sites with lower DNI potential are also considered. For example, there are CST process heat and district heating projects in central Europe, with an annual DNI as low as 1,000 kWh/m²/a.

As can be seen below, Australia is globally among the areas with one of the best DNI potential. A large portion of Australia offers an excellent DNI for large-scale CST plants for power and green fuel production and the entire mainland is generally suitable for CST process heat applications from a solar resource perspective.

While Australia offers regions with up to 2,900 kWh/m²/a, most of the globally installed CST capacity is at sites with lower DNI. For example, the Spanish6 CST plants are operating with an annual DNI in the range of only 2,000 - 2,200 kWh/m²/a. Most of China’s CST capacity - currently there are 3 GWe under construction - is deployed at sites with less than 2,100 kWh/m²/a.

4
The global irradiation is the sum of direct and diffuse irradiation. Direct radiation is typically measured on a plane normal to
the beam as Direct Normal Irradiation (DNI).
5
Also depending on the latitude of the project site, determining the optical efficiency of the solar field
6
Spain is the country with currently the largest installed CSP capacity (2.4 GW) - see Section 2.3

S774Doc-676563417-363 / v0.2 34 CST Value Proposition
Figure 2-2: Global and Australian direct normal irradiation (SolarGIS, 2023)

Figure 2-3 shows the cumulative distribution function for the direct normal irradiation, based on the
Global Solar Atlas using SolarGIS data (SolarGIS, 2023). As can be seen, more than 50% of the Australian land area (approx. 3.7 million km²) has a DNI level above 2,500 kWh/m²/a and more than 85% has a DNI level above 2,000 kWh/m²/a.

Figure 2-3: Australian direct normal irradiation distribution by land area (SolarGIS, 2023)

A research study conducted by Dupont et al. (Dupont et al., 2020) titled “Global available solar energy under physical and energy return on investment constraints” concluded that Australia (Oceania) holds 19%

S774Doc-676563417-363 / v0.2 35 CST Value Proposition
of the global CST potential, while accounting for only 6% of the global land mass. In another study Trieb et al. (Trieb et al., 2009) estimated the global CST resource potential. They assumed a cut-off of 2,000 kWh/m2/a as being sufficient for CST (electricity) generation and further excluded land with slope, water, forests, shifting sands, protected areas, etc. The global CST resource potential identified amounts to around 3 million TWhe/a, out of which Australia contributes around 700,000 TWh/a or 24%. While the global potential identified covers a bit more than 100 times the global demand, the Australian technical
CST potential could theoretically cover the current demand of around 190 TWh/a (2022) more than 3,500 times.

2.3 International Deployment
By Q1 2023 there were more than 100 concentrated solar power plants in commercial operation with a total installed capacity of around 6.5 GWe, in addition to around 4 GWe of additional capacity under construction, summing up to more than 10 GWe. An overview of the global CST deployment is depicted below, and milestone CST projects listed in Table 2-2. Further, there are around 400 MWt of CST process heat capacity in operation and several medium (5 - 40 MWt) to large-scale > 1GWt projects under construction and under development.

The technology goes back to 1907 when the first patent of a parabolic trough collector was filed in
Stuttgart, Germany. Six years later a 55 kW solar powered pumping station, using parabolic trough technology, was brought online in Egypt. Since its commercialisation in 1980s, the CST industry has gone through three deployment cycles. The first wave of deployment started in 1984 in the U.S. (California) with the construction of the solar electricity generating system (SEGS) plants. In total nine parabolic trough plants with a total capacity of 354 MWe have been put in operation, most of which operated for more than 30 years. This initial CST deployment period ended in 1991 after falling natural gas rates priced CST out of the market (Augustine et al. 2021).

Following a 15-year period with no additional CST plants, the second deployment wave began in 2007 primarily in the U.S. and Spain. The CST deployment in the U.S. was primarily supported through a loan guarantee program, as well as some further state incentives. During this period seven new plants were built and around 1.4 GWe of installed capacity was added. In Spain CST deployment was supported by an attractive feed-in tariff (FIT), which supported the construction of nearly 50 CST plants with a total capacity 2.3 GWe 7. Besides the U.S. and Spain several other countries, including India, the UAE, South
Africa, and Morocco, started during this period to deploy their first CST plants. Given large-scale CST was at that time still competitive with “daytime” PV, only less than half of the plants built during the 2nd wave included thermal energy storage. Storage sizes ranged between 7.5 full-load hours (most parabolic trough plants) and 15 full-load hours for the first Spanish molten salt solar tower (Lilliestam et al., 2023).

7
While there was no plant capacity limitation in the U.S., the Spanish FIT limited the single plant size to 49.9 MWe

S774Doc-676563417-363 / v0.2 36 CST Value Proposition
Figure 2-4: Overview of global CST deployment as of today (Q1 2023)

The end of the Spanish FIT and U.S. loan guarantee program curtailed the development of new CST projects in these two key markets. Further, the rapidly declining PV prices and the fact that most countries have only started in recent years with the implementation of large-scale renewables (i.e., looking first at the price per kilowatt hour and not at the least cost option for the entire future energy system), has slowed down CST deployment in recent years. All projects, which started operation in recent years, or which are under construction, are equipped with a large-scale thermal energy storage system.

Table 2-2: Summary of major CST milestones worldwide

Name/ Country COD Technology Capacity Storage Remark
Location gross Capacity
[MWe] [MHW
/hours]

SEGS I - IX U.S. 1984- Parabolic 354 No Several SEGS plant reached
1990 trough over 30 years of operation,
the last SEGS IX plant still
being in operation

Nevada U.S. 2007 Parabolic 72 No First CST plant of the 2nd
Solar One / trough deployment wave
Boulder
City,
Nevada

Andasol / Spain 2008 Parabolic 50 1,010 First commercial CST plant
Guadix trough with large-scale TES
7

Gemasolar Spain 2011 Molten salt 20 670 / 15 First commercial solar
/ Sevilla power tower power plant

S774Doc-676563417-363 / v0.2 37 CST Value Proposition
Name/ Country COD Technology Capacity Storage Remark
Location gross Capacity
[MWe] [MHW
/hours]

PE 2 / Spain 2012 Linear 30 100 / 0.5 First commercial linear
Puerto Fresnel Fresnel plant
Errado

Shams One UAE 2013 Parabolic 125 No First commercial CST plant
/ Abu trough built in the Middle East
Dhabi

Solana / U.S. 2014 Parabolic 280 4,240 / 6 Largest CST plant in US
Arizona trough with TES

Ivanpah U.S. 2014 Direct steam 392 (1 x No Largest DSG tower
(California) power tower 126, 2 x
/ Nevada 133)

Noor I / II Morocco 2016/20 Parabolic 160/200 1,300 / 3 Part of largest CST
Ouarzazate 17 trough complex in Africa
3,600 / 7

Noor III / Morocco 2017 Molten salt 150 2,800 / 7 Largest single tower unit
Ouarzazate solar tower capacity, Part of largest
CST complex in Africa

Atacama-1 Chile 2018 Molten salt 110 4,600 / First large-scale CST plant
/ Calama solar tower 17.5 in South America,
Developed as CST-PV
hybrid

Supcon China 2018 Molten salt 50 10 hours One of the first CST
tower / solar tower projects supported under
Delingha the initial FIT round

Noor Dubai 2022/20 3 x 200 MW 950 Avg. ~ 10 Largest single CST IPP incl.
Energy 1 / 23 PT + 1 x 100 hours four CST units. At the time
DEWA MW ST + lowest CST tariff at US$
Phase 4 250 PV 0.073 /kWh

A new deployment wave is now underway in China. In 2016 China announced first a FIT to support the construction of 20 CST projects to test different technologies and concepts. As a result, by Q1 2023 around 600 MW of CST capacity has been put in operation, out of which nearly two thirds are based on

S774Doc-676563417-363 / v0.2 38 CST Value Proposition
solar tower technology. Following a new regulation by China’s National Energy Administration, new large- scale renewable installations require a dispatchable portion and CST with TES is considered by the
Chinese administration as one of the options - where applicable from a solar resource perspective. In consequence, as of Q1 2023 there are more than 30 new projects under construction, resulting in more than 3 GW (CSTA, 2023). This means China is taking the lead also in CST deployment and will become soon (within the ongoing 14th Five-Year Plan period) the country with the largest installed CST capacity - as can be seen also in Figure 2-5.

Figure 2-5: Global installed CST capacity by country

Figure 2-6: Global installed CST capacity by technology

Besides China, international deployment is currently lagging behind the deployment levels projected in the two major world energy outlooks. Both the International Energy Agency (IEA) World Energy Outlook
(WEO) (IEA, 2022) and the International Renewable Energy Agency (IRENA) World Energy Transitions

S774Doc-676563417-363 / v0.2 39 CST Value Proposition
Outlook (WETO) (IRENA, 2022a) project substantial CST deployment in their respective net-zero by 2050 scenarios.

IEA In their global energy models, CST is one of the chosen options - for regions with sufficient solar potential - to close the cap in providing long duration energy storage capacities and dispatchable RE generation, thus, increasing the offtake of variable renewables energy technologies.

Case Study - Noor Energy 1 IPP (DEWA Phase IV - 950 MWe CST and PV Hybrid)

Location MBR Solar Park, Dubai
DNI: 2,100 kWh/m²/a

Capacity 950 MWe

Technology CST 750 MWe
Breakdown - Parabolic trough 3 x 200 MW -
- Solar Tower 100 MW
PV 250 MWe

Storage Capacity Parabolic trough 3 x 6.5 GWh
(approx. 12 hours each)
Solar Tower 1 x 3.5 GWh
(approx. 15 hours)

Type / Ownership Build Own Operate / DEWA
(51%); ACWA Power (25%); Silk
Road Fund (24%)

PPA CST: 0.073 US$/kWh
PV: 0.024 US$/kWh

Total investment US$ 4.3 billion

The Noor 1 project, nearing completion (as of Q1 2023), is the largest single-site CST plant in the
world. The project holds the record for the lowest-priced CST plant at US$0.073 per kWh under a 35-
year power purchase agreement (PPA), showing how far CST costs have come down.
Noor 1 is designed to cover best the United Arab Emirates’ load profile, which has a pronounced
evening peak. The 250 MW PV plant covers part of the demand during the daylight hours whereas the
larger CST part serves demand during the evening and night hours due to its long duration thermal
energy storage.

A summary of the CST capacities required in the respective scenarios is provided in Table 2-3. As can be seen the IEA considers up to 437 GW of CST deployment necessary to reach net zero by 2050 (NZE)
IRENA increases the deployment of CST to 842 GW in its Deeper Decarbonisation Perspective (DPP) scenario, which includes further measures beyond the Transforming Energy Scenario to reduce energy and process-related CO2 emissions to zero by 2050 - 2060.

Considering the CST capacities required until 2050, the compound average annual growth rate (CAAGR) between today and 2050 ranges between 14% and 18%. Given the comparatively small CST capacity installed today (6.5 GW) the required CAAGR until 2050 is significantly higher when compared to solar PV or wind - with PV already above the one TW mark and wind about to pass this mark. For example, the IEA

S774Doc-676563417-363 / v0.2 40 CST Value Proposition
Net Zero by 2050 scenario results in CAGR of 10% for solar PV and 8% for wind, compared to 15% for
CST.

Table 2-3: Global CST capacity (GW) deployed in relevant IEA and IRENA scenarios towards net zero

Scenario 2030 2040 2050 CAAGR

IEA WEO 2022 - Announced Pledges 35 177 318 14%
Scenario (APS) 8

IEA WEO 2022 - Net Zero Emissions by 64 283 437 15%
2050 Scenario (NZE)9

IRENA WETO 2022 - Transforming Energy 76 193 309 14%
Scenario (TES) 10

IRENA WETO 2022 - Deeper 196 519 842 18%
Decarbonisation Perspective (DPP) 11

Considering that by 2025 around 10 GW e of CST generation capacity will be operational, the required capacities for 2030 - projected in these scenarios - appear very ambitious. To accelerate deployment of
CST and to meet at least the lower bound by 2030, a better remuneration of the value of CST, with storage, will be necessary. CST needs to be considered as a replacement option for conventional dispatchable capacity (in particular natural gas) rather than an alternative to solar PV. Countries with a good solar resource will need to build their first large-scale CST plants to get familiar with the technology or continue with their initial programs and meet their stipulated CST targets. For example, some of the
National Energy and Climate Plans (NECPs) for EU Member States consider CST, led by Spain planning to add 5 GW and Italy 880 MW of new CST capacity by 2030 (EC, 2023).

8
The IEA WEO Announced Pledges Scenario (APS) aims to show to what extent the announced ambitions and targets, including
the most recent ones, are on the path to deliver emissions reductions required to achieve net zero emissions by 2050. It
includes all recent major national announcements as of September 2022 for 2030 targets and longer term net zero and other
pledges, regardless of whether these have been anchored in implementing legislation or in updated NDCs. In the APS,
countries fully implement their national targets to 2030 and 2050, and the outlook for exporters of fossil fuels and low
emissions fuels like hydrogen is shaped by what full implementation means for global demand.
9
The IEA WEO Net Zero Emissions by 2050 Scenario (NZE) is a normative IEA scenario that shows a pathway for the global
energy sector to achieve net zero CO2 emissions by 2050, with advanced economies reaching net zero emissions in advance of
others. This scenario also meets key energy-related United Nations Sustainable Development Goals (SDGs), in particular by
achieving universal energy access by 2030 and major improvements in air quality. It is consistent with limiting the global
temperature rise to 1.5 °C with no or limited temperature overshoot (with a 50% probability), in line with reductions assessed
in the IPCC in its Sixth Assessment Report.
10
The IRENA WETO Transforming Energy Scenario (TES) describes an ambitious, yet realistic, energy transformation pathway
based largely on renewable energy sources and steadily improved energy efficiency. This would set the energy system on the
path needed to keep the rise in global temperature to well below 2°C and towards 1.5°C during the century.
11
The IRENA WETO “Deeper Decarbonisation Perspective (DPP)” scenario provides views on additional options to further reduce
energy-related and industrial process CO2 emissions beyond the Transforming Energy Scenario. It suggests possibilities for
accelerated action in specific areas to reduce energy and process-related CO2 emissions to zero in 2050-2060.

S774Doc-676563417-363 / v0.2 41 CST Value Proposition
One can expect that China will continue to lead CST deployment, stating in its ‘Action Plan for Carbon
Dioxide Peaking Before 2030’, that China will actively further develop CST and promote its deployment where CST, solar PV and wind complement each other (CSTA, 2023).

Compared to CST for power generation, the deployment for CST process heat is still comparatively small and in case of green fuels just at the beginning with the first project underdevelopment.

The largest commercial CST process heat project in operation is the Miraah Project in Oman with a peak output of 330 MWt . The project, which started operation 2017, uses an enclosed trough design developed by GlassPoint to generate steam for enhanced oil recovery. In 2022, GlassPoint announced the world’s largest CST process heat project, Ma’aden Solar 1. The project is planned to have a thermal capacity of
1.5 GWt and should generate steam for refining bauxite ore into alumina.

Besides a few large scale projects, the majority of the CST process heat projects are projects in the range of 5 - 50 MWt,, providing solar heat to the food, beverage and textile industries. The market is predicted to grow substantially in the next years with a solid project pipeline, which is dominated by Europe and the
Middle East, not considering China for which no data is available (Solarthermalworld, 2023).

Case Study - CST Process Heat for a Brewery

Location Heineken Brewery Seville, Spain
DNI: ~ 2,100 kWh/m²/a

Capacity 30 MWt

Technology Solarlite parabolic troughs
Breakdown SL5770 (5.77 m)
43.414 m² total aperture area
Pressurised water (35 bar /
250°C)

Storage Capacity 69 MWht

Type / Ownership Build Own Operate Transfer
(BOOT) / Engie

HPA Confidential

Total investment Confidential

The Heineken solar process heat project, nearing completion (as of Q1 2023), is one of the largest CST
process heat projects currently under construction and the largest project providing solar heat to
decarbonise a brewery. The new facility will enable Heineken Spain to reduce its fossil gas
consumption by 60% and its carbon footprint by 7,000 tonnes CO 2-e a year. ENGIE is responsible for
the design, installation, management, maintenance, and financing of the project, materialised in a
long-term BOOT (Build, Own, Operate, Transfer) contract to supply 100% renewable energy to
Heineken Spain for the next 20 years.

S774Doc-676563417-363 / v0.2 42 CST Value Proposition
2.4 Developments in Australia
Despite its world leading solar resources, Australia is yet to be a major player in the CST industry and does not have any ‘utility scale’ CST systems in place. Traditionally Australia’s grid connected power generation has been dominated by abundant, and relatively low cost, coal, and gas. The national Renewable Energy
Target (RET) has seen large growth in wind and photovoltaic deployment due to the strong investment incentive it creates for renewable generation, irrespective of the time of generation or correlation with demand. With the retirement of coal fired power plants fast approaching, network stability issues and the need for firm generation at night are creating increased government and energy sector interest in multi- hour, intra-day dispatchable renewables.

CST is also able to provide a multi-hour renewable heat solution for Australian industries. When linked with conventional storage technologies (i.e., steam tanks), CST can deliver a 24/7 renewable heat solution for a diverse range of Australian manufacturers. There are now a number of CST process heat solutions under active development in Australia.

However, despite the lack of commercial deployment, Australia has been a consistent player in its contributions to CST RD&D.

In terms of existing deployments, the largest successful CST system in Australia is the Sundrop farms
36 MWt system near Pt Augusta in SA (Figure 2-7). While this system only generates 1 MWe of electricity, the bulk of its energy is used for desalination and heating to support the production of glasshouse produce (i.e., tomatoes). The EPC contractor for the whole project was John Holland, with the Danish company Aalborg CSP responsible for the solar installation, for which they in turn sourced heliostats from the US company eSolar.

Figure 2-7: Sundrop Farms greenhouses and concentrating solar power plant (courtesy of Aalborg CSP)

S774Doc-676563417-363 / v0.2 43 CST Value Proposition
Beginning in the 1980’s small pilot systems were built, trough units at Meekathara in WA and dishes at
White Cliffs in NSW. Liddell power station in the hunter valley was home to a 9.3 MWt linear Fresnel array that was completed in 2012. The Victorian company Solar Systems installed multiple dish based concentrating PV systems in remote towns from 2000 onwards and also a large array in Mildura (40 dishes
1.5 MW), before going out of operation.

What was to be the largest CST array in Australia was largely fabricated for the Kogan Creek power station solar boost project in Queensland from 2010 to 2014. This 135 MWt linear Fresnel array was intended to feed steam to the coal fired power station and boost its output by 44 MWe. It was however never completed following the exit of AREVA from the solar business and a change of strategic direction by CS
Energy, the client.

The US company SolarReserve attracted a lot of interest with a proposal for a 110 MWe tower system intended for a site north of Pt Augusta in South Australia. This was to be very similar to the Crescent
Dunes project in Nevada. This project never reached financial closure.

Regarding current Australian based CST activity, the Sydney based company Vast Solar is showing strong progress. It has been a world leader in progressing the use of Sodium as a heat transfer fluid for towers. It has a 6 MWt, / 1 MWe pilot multi-tower system at Jemalong in NSW and is working to develop the VS1,
30 MWe / 288 MWh, project in Port Augusta, South Australia, at the site of the previously proposed Solar
Reserve project. The Australian government has announced it will support the project with concessional financing, as well as a grant from the Australian Renewable Energy Agency (ARENA). Vast Solar have indicated that they are aiming to reach financial close and commence construction in 2024.

Figure 2-8: Vast Solar’s CST demonstration power station at Jemalong, New South Wales

The Melbourne based company RayGen is also working with modular tower systems. They are using high performance concentrating PV cell receivers but also combining this with capture of heat from receiver cooling. Pit storage of both hot water and chilled water is used for power generation using an ORC cycle in non-solar periods. They have established the RayGen Power Plant Carwarp (RPPC) project which will add 4 MW of direct PV generation and 3 MW / 50 MWh (17 hours) storage to the West Murray grid at
Carwarp in northern Victoria.

S774Doc-676563417-363 / v0.2 44 CST Value Proposition
In other commercial activity, Woodside has invested in the US CST company Heliogen and it is known that a number of companies have been considering CST systems for process heat.

In addition to these commercial activities, the Australian government via, first the Australian Solar Institute and later the ARENA (directly and via the Australian Solar Thermal Research Institute), has invested close to A$ 200 million on CSP RD&D in Australia. Among the research institutions, CSIRO, ANU, University of
Adelaide, and University of South Australia have the largest CST research programs. There are also CST research groups at the University of Queensland, RMIT, UNSW and other organisations, with growing industry linkages. CSIRO in Newcastle is the biggest R&D group and operates two experimental tower systems at its site. The Australian National University is home to the world’s largest dish concentrator as a result of an active R&D group dating to the 1980s.

There have been many studies over the past decades looking at aspects of how large-scale commercial
CST might be realised in Australia. As part of a 2018 CST roadmap study, ITP has provided a detailed review of past studies dating back to 2008 (ITP, 2018).

In 2017 ARENA released a formal request for Information to the global CST industry (ITP, 2017). This resulted in 31 responses building on experience from every significant CST system globally. The responses expressed universally positive views on the future of CST and expressed interest in involvement in CST deployment in Australia. This RFI process provided a clear indication that a large-scale competitive CST process in Australia, especially given the high solar radiation, would be well subscribed.

S774Doc-676563417-363 / v0.2 45 CST Value Proposition
3 Value of CST
The inherent “value” of CST is the result of the particular services and benefits it offers, further described within this section. The respective value and overall contribution of these depends on the prevailing boundary conditions and the type of end-use application. For example, the value proposition changes with geographical area (i.e., available solar resource, and the particular needs of the network or end-user at a particular location), market rules and the other generating assets in place. The key features the value proposition has in common across end use applications is that CST will generally complement other renewable generation options and justify higher generation cost compared to VRE options - although partly difficult to quantify. As such CST is regarded by both the International Energy Agency (IEA) and the
International Renewable Energy Agency (IRENA) as one of the technology options to support the required energy transitions towards net zero - see also Section 2.3.

In economic terms the financial ‘’value’’ of CST can be expressed as the sum of its lifetime financial benefits suitably quantified and discounted to the time of construction. This can be compared to its total cost being the capital cost plus the lifetime sum of operating cost discounted to the time of construction.
The challenge is that, first, the full range of benefits that CST offers is in parts complex and difficult to quantify. Furthermore, these benefits are not all being recognised and rewarded by current market and policy settings.

Diversification Firm Capacity

Socio- CST Medium & Long
economic Duration Storage

Power and Grid Services
heat

+

Figure 3-1: Key services and benefits of CST

S774Doc-676563417-363 / v0.2 46 CST Value Proposition
3.1 Competitive Medium-Duration to Long Intraday Energy Storage
Central to the value proposition for CST is its incorporation of cost effective thermal energy storage.
Affordable storage of renewable energy (both heat and electric power) will be one of the key pillars in the transition towards net zero. According to the Renewable Energy Storage Roadmap, published by CSIRO in early 2023, the national electricity market alone could require 44 – 96 GWe / 550 – 950 GWh of dispatchable electricity storage capacity by 2050, and Western Australia (WA) could require another
12 – 17 GWe / 74 – 96 GWh (CSIRO, 2023a)12 In addition, the energy transition will require renewable energy storage for other end-use sectors such as the process industry, remote area mining and the green fuel production.

Due to the intermediate conversion of solar energy into thermal energy, CST offers the advantage that heat can be stored directly, i.e., in a low-cost form of energy. Heat can be stored in a thermal energy storage system for hours or even days with very little thermal loss (<1%/d) before it is either directly used as process heat or utilised to generate electric power or to drive chemical processes. Given thermal energy storage is the key feature CST offers when compared to other renewable energy options, all CST plants constructed in the recent past, which are under construction or under development include a large- scale thermal energy storage system, some of which have a storage capacity of several GWht - see project examples in Section 2.3.

CST with thermal energy storage (CST-TES) benefits from economies of scale more than some other renewable energy storage options, making it attractive for both medium duration (4-12h) and long intraday (12 - 24h) storage (CSIRO, 2023a):

▪ Medium-duration storage systems can play an important role providing major grids with the flexibility
to manage any imbalances between supply and demand, as well as supporting grid capacity. Medium-
duration storage systems can be used for applications such as network support, time shifting energy
and helping avoid or defer T&D investment.
▪ Long intraday (from >12 to 24 hours) storage is used for network support to help stabilise day-to-day
variation in electricity supply and the time shifting of energy to manage differences between peak VRE
generation and peak energy use times each day.

Using the Levelised Cost of Storage (LCoS) definition proposed in the Renewable Energy Storage
Roadmap13 and the cost data and economies of scale factors proposed in this report (see Section 4) LCoS have been calculated for a 100 MWe reference case with 230 cycles per year and different storage capacities. As can be seen, the LCoS can be reduced by nearly 30% when doubling the storage capacities.
Besides the individual economies of scale impact for each of the subsystem, main driver is the fixed power capital cost (such as power block, balance of plant systems and grid connection), for which their cost per
MWh output reduces for larger storage capacities. The LCoS reduction with increasing storage duration is a clear advantage when compared to most other storage options. Further, any CST-TES option can be equipped with an additional e-heater in order to make use of any low cost solar PV or wind generation, otherwise curtailed.

12
This roadmap uses a scenario-based approach, building on pathways developed in the Australian Energy Market Operator’s
(AEMO) 2022 Integrated System Plan (ISP) that could materially impact Australia’s energy sector. Data referenced are based
on the Step Change and Hydrogen Superpower scenarios
13
See Appendix C of the Renewable Energy Storage Roadmap

S774Doc-676563417-363 / v0.2 47 CST Value Proposition
Figure 3-2: LCoS for different storage capacities (full-load-hours - FLH)

The Renewable Energy Storage Roadmap considers a wide range of storage technology options and end- use sectors. For the specific 8-hour (230 and 285 annual cycle) storage duration cases, PHES was estimated to have the lowest cost in the near term, closely followed by the CST storage option. In the long term, CST storage was estimated to have the lowest cost. Also, for the specific 24-hour (117 annual cycle) storage duration case, PHES and CST storage were estimated to have the lowest costs in the near term. In the long term, again CST storage was estimated to have the lowest cost. Further, also for the remote area mining case study the roadmap concluded that CST Storage was estimated to be the least cost system in the near and long term. In case of industrial process heat, the storage options are limited, making CST with TES an attractive option (where applicable from solar resource and land area perspectives).

Figure 3-3: LCoS results, medium-duration storage (8-hour storage duration, 230 annual cycles) (CSIRO, 2023a)

S774Doc-676563417-363 / v0.2 48 CST Value Proposition
Figure 3-4: LCoS results, long intraday storage (24-hour storage duration, 117 annual cycles) (CSIRO, 2023a)

3.2 Firm Capacity
What end use applications require is not energy storage as such, but rather the supply of energy in a reliable manner when it is most needed. Starting with a solar radiation input, it is energy storage that allows this to be achieved. The result of being able to deliver the rated capacity (MW e or MWt as appropriate) in a reliable or firm manner is a valuable capability that can be quantified.

In order to cover the demand within an electricity network and to allow for stable network operation, there is the need for sufficient firm capacity. I.e., firm capacity is the generation capacity available for production at any time it is needed and which can be guaranteed. Traditionally, much of this role has been filled by coal or gas fired power plants.

CST plants operate like conventional fossil fuelled thermal power plants using similar (but more flexible) steam turbines. Central to a dispatchable CST plant providing firm capacity, is its ability to store high temperature heat in a TES system. Solar heat collected during daytime and stored in the TES can be extracted from the storage to produce steam to run the power cycle on demand, i.e., during night-time, during times with no or insufficient solar irradiation or during peak demand periods - depending on the design of the plant. In addition, back-up heaters, using any type of renewable or fossil fuel, can also be incorporated in a CST plant to ensure even greater security of supply at any time - i.e., when the storage is empty - and to avoid unnecessary oversizing of the TES system.

The option to incorporate an electrical heater allows CST plants to store otherwise curtailed solar PV and/or wind power, further increasing their utility and complementarity with those variable renewable technologies.

Whilst the qualitative value of a zero emissions technology that is reliably available when needed, is apparent, quantifying this and rewarding firm capacity is complex. The range of terms that are used can also be confusing. Capacity value, capacity credit, equivalent firm capacity, and other terms are used by various authors to quantify the equivalent statistical fraction of availability to generate, offered in critical times of need, compared to an ideal always available generator. The use of stored thermal energy of sufficient duration allows CST to be largely available when required to meet system needs.

S774Doc-676563417-363 / v0.2 49 CST Value Proposition
Firm Capacity value should not be confused with capacity factor which is simply the annual average fraction at which a generator produces at its nameplate capacity. For example, a wind generator in a good site might achieve a capacity factor of 50%, but if there is a lot of wind generation in the system then the critical times when energy is required will be precisely when wind generation is low, thus the firm capacity value of wind will be very low. A standby diesel generator with a large fuel reserve may be rarely used so its capacity factor will be low. However, it can be relied on at critical times so its firm capacity value will be close to 100%.

Some large electricity systems operate a ''capacity market'' in parallel to an ''energy market''. In a capacity market, generators receive an annual payment per MW in proportion to their firm capacity value compared to an ideal dispatchable generator. In addition, they are paid for the actual electricity they generate via the energy market. The NEM has no capacity market, however the national Capacity
Investment Scheme under development is a potential initiative to establish one. The SWIS through its
Reserve Capacity Mechanism (RCM) does reward capacity value.

The challenge with these capacity market payments however is that if they don’t also recognise the need for fully zero emissions solutions, they will tend to be filled by low CAPEX high fuel cost fossil fired generators like diesel generators or gas turbines. If there is a goal to reduce emissions to zero, then a capacity market mechanism is required that ensures a growing share is met by firm capacity dispatchable renewable generation.

A comparative example of a large-scale CST-PV hybrid plant, developed in the MENA region, and a combined cycle gas turbine (CCGT) power plant is depicted in Figure 3-5. Making use of a CST-PV hybrid solution, cheap on-sun PV is dispatched during the day and the CST system, utilising a thermal energy storage, is dispatched during the (peak) night-time hours. A techno-economic analysis was undertaken based on dispatch optimisations for multiple plant configurations and required minimum peak shares.
Depending on the required peak share and assuming on-sun PV generation is maximised to the set limit; the total annual net generation increases with increasing peak share. Increasing annual net generation means increasing capacity factor as depicted below.

As can be seen, the CST-PV hybrid option becomes competitive even for the low gas case
(US$ 5.5 /MMBTU or US$ 5.2/GJ). Depending on the peak share requirement, as the CST portion increases so does the weighted LCoE. TES sizes for the indicated peak share range varies between 7 and 11 hours.
PV-BESS options were also investigated, but resulted in higher LCoE, when compared with the CST-PV options.

Given Australia’s high solar resource (in many parts above the level of the shown example) and higher wholesale gas prices (as of Q1 2023 capped at A$ 12 /GJ or approx. US$ 8.6 /GJ or US$ 9.1 /MMBTU),
CST-PV hybrids should already be competitive in Australia today with an LCoE at less than A$ 100 / MWh.

S774Doc-676563417-363 / v0.2 50 CST Value Proposition
Figure 3-5: Comparative example of CST-PV hybrid and combined cycle (CCGT) providing firm capacity

Given the CST part is dispatchable, the comparison of the CST-PV plant with the CCGT has been done both with and without consideration of the additional capacity value, expressed in the example as a capacity credit (CC). The capacity credit is defined as the capacity share of a power plant contributing to the reliability of the system for load covering 14.

The higher the peak share, (i.e., the capacity factor), the higher the dispatchable portion and hence the higher the capacity credit. In addition, the comparison includes the results for a 100% capacity credit, which can be reached if the CST-PV plant is equipped with a back-up heater. As can be seen, LCoE improves by around 10% in case of a 100% capacity credit allowance.

Table 3-1 provides an example of the capacity credits considered in the latest Western Australian
Wholesale Electricity Market (WEM) Electricity Statement of Opportunities (ESOO) (AEMO, 2022a). The
WEM ESOO is prepared annually and provides forecasts and analysis of peak demand and energy use in the SWIS in Western Australia for the next 10 years. As can be seen, CST is not yet included. If it were included, then CST system with long duration (15hr +) storage should be ranked at over 90%, similar to dispatchable fossil fired plants.

14
Capacity credit (CC) is a term particularly used in power expansion planning. For each power plant that is added to a power
system, it would be helpful to know how this power plant addition will contribute to the reliability of its connected power
system. The reliability of a power system increases when the probability is raised that there will always be sufficient total
power supply available for covering the current system load. The contribution of power plants to system reliability is
expressed in terms of “capacity credit”, that is defined as: The capacity portion of a power plant which is contributing to the
reliability of the system for load covering. In order to see the implication of the capacity credit in the overall techno-
economic assessment, i.e., in the tariff calculation for each of the potential configurations, the avoided “conventional”
capacity additions - keeping system load with the same reliability - are determined, based on the calculated capacity credit.
For the capacity credit allowance calculation, a CCGT plant has been taken as reference plant.

S774Doc-676563417-363 / v0.2 51 CST Value Proposition
Table 3-1: Capacity credits by energy source for 2023-24 as per the WEM ESOO (AEMO, 2022a)

Fuel type Maximum capacity [MW] Capacity Credits Capacity Credit factor
assigned

Gas 1647.1 1597.8 97.0%

Dual (Gas / Distillate) 1326.0 1277.9 96.4%

Coal 1371.1 1362.4 99.4%

Distillate 132.2 121.6 92.0%

Wind15 1010.8 150.0 14.8%

Solar16 150.8 14.9 9.9%

Waste-to-energy 65.0 59.0 90.8%

Landfill gas 21.6 12.9 59.8%

DSP17 85.0 83.8 98.6%

Electric Storage 100.0 46.3 46.3%
Resource (ESR)

A team led by the University of Technology Sydney (UTS) investigated the firm capacity that CST plants could offer as part of a network benefits study (Rutovitz et al., 2013). The study was done across the
Australian NEM, for a number of storage levels and times of the year. Indicative firm capacity was evaluated by considering the average capacity fraction during the top 21 load events of the season using historic NEM data and a simplified CST plant model with standard dispatch with hourly solar data based on BOM satellite based DNI data. As one of the key results the study showed that with just five hours of storage capacity, very high indicative firm capacity values (capacity fractions) of above 70% would be obtained in many areas. This increases further with larger amounts of storage. Better results would likely be obtained in WA and NT given the relatively higher DNI in key areas of these states / territories. It should be noted that this simple modelling was carried out assuming ‘’dumb dispatch’’ i.e., without factoring in the additional capacity value that would accrue from a real system actively strategically dispatched to target the most critical times.

15
Wind includes the wind and solar hybrid facilities
16
Solar includes the solar and electric energy storage hybrid facilities
17
Demand Side Program (DSP) Due to nature of DSPs, the maximum capacity of DSP here has been defined as the contracted
quantity.

S774Doc-676563417-363 / v0.2 52 CST Value Proposition
Figure 3-6: Indicative firm capacity of CST plants across the eastern states (Rutovitz et al., 2013)

3.3 Grid Benefits
CST plants use rotating power generation equipment (turbines) and, thus, offer similar grid benefits compared to conventional fossil fired dispatchable thermal power plants. Grid benefits can be divided into two categories: Ancillary services and avoided grid cost (for the whole network).

3.3.1 Ancillary services
Ancillary services are those services, other than electricity supply, that are needed to run an electrical network in a stable manner. Ancillary services include, amongst others:

▪ Frequency control,
▪ Voltage control,
▪ Spinning reserve,
▪ Emergency control action, and
▪ Black start capability.

In the NEM, a range of ancillary services are already traded on a competitive basis in parallel with energy sales. The services are typically supplied by thermal (fossil) power stations characterised by synchronous generators, large amounts of angular momentum and the ability to ramp up and down as required.

As is the case in many electrical networks, ancillary services have historically commanded low prices in the
NEM, because they were in oversupply due to the dominance of fossil fuelled generation. As the levels of intermittent renewable generation such as solar PV and wind has increased, the market price of ancillary services has increased. However, it should be noted that the value of ancillary services traded is only a small fraction of the value of energy sales.

The inertia provided by the angular momentum of turning machinery is an important contributor to maintaining frequency control under unexpected events. One approach to maintaining frequency control,

S774Doc-676563417-363 / v0.2 53 CST Value Proposition
which has been adopted in areas of high wind and PV penetration, is a synchronous condenser (syncon).
Essentially a large flywheel coupled to a synchronous generator. These are expensive and energy consuming devices.

When a CST system is operating, it is a synchronous form of generation that provides frequency control as an additional operational benefit (just like coal fired power plants). Adopting CST steam turbines with a clutch between the turbine and generator can also allow for CST systems to act like a syncon and so provide grid services when the CST plant is not generating electricity, thus, enhancing revenues while avoiding the installation of additional synchronous condensers. The concept is, for example being proposed by Vast Solar for its VS1 project in South Australia, where several synchronous condensers have already been installed in order to increase the VRE share in the generation mix.

3.3.2 Avoided grid cost
High firm capacity values offer the possibility that an appropriately located CST plant could relieve the strain on network assets that are at the fringe of grid for example. This was the focus of the network benefits study carried out by the UTS team. The UTS study found that in the East Coast NEM, CST could avoid the need for augmentation in 72% of constrained sites examined, while 25% of the constrained sites were assessed as being cost effective. In the most extreme cases, the network benefit was sufficient to justify
100% of the CST plant cost (Rutovitz et al., 2013). A hypothetical network benefit payment was determined from avoided cost of upgrade (noting that this monetisation of value is not automatically accessible under current market rules). They found that the hypothetical value of a network support payment justified on avoided network investments at identified constrained locations in the NEM would be A$ 15 /MWh on average but in specific locations was found to be up to A$ 134 /MWhe. A total of 533 MWe of systems were predicted to be cost effective out to 2023, with the plant capacity ranging from 8 MWe to 120 MWe and storage between 5 and 15 hours.

3.4 Combined Heat and Power
Combined heat and power (CHP), also known as cogeneration, is the simultaneous production in the same plant of heat (thermal energy) and power (electrical energy), hence, increasing the overall conversion efficiency. Gas and coal fired CHP plants are widely used globally, both for process heat and district heating.

Given the intermediate conversion of solar energy into thermal energy, CST offers multiple options for
CHP. For example, a CST or CST-hybrid plant can be equipped with a back-pressure turbine or steam can be extracted at multiple pressure levels. High temperature solar heat can also be used directly, and the remaining stored heat utilised for dispatchable power generation, together with other RE sources such as on-sun PV or fluctuating wind.

Also in the field of green fuel production CST is particularly interesting for applications which require both heat and power, such as renewable methanol production or high temperature electrolysis, using solid oxide electrolyser cells. This advantage of combined heat and power is analysed in detail in Section 8.5.2.

S774Doc-676563417-363 / v0.2 54 CST Value Proposition
3.5 Socio-Economic Benefits
The high labour content and outright costs of construction for CST plants have obvious positive effects in terms of employment generation and economic stimulation. For example, one of the previous major CST developers Abengoa reported that in the U.S., with two projects (Mojave and Solana), the supply chains have created the following socio-economic benefits across the U.S. (ITP, 2018):

▪ More than 70% American goods and services,
▪ A national supply chain that includes 300 companies in 31 states and more are anticipated,
▪ Over 3,000 supply chain jobs created across America, and
▪ US$ 1.8 billion total investment into the local economy of those 31 states.

Some of this supply chain represents demand for non CST specific components and services. Some will represent businesses adapting to the provision of CST specific products and services, this capability building will potentially contribute to cost reductions for subsequent projects. This apparent benefit however may be lost if a first project is not followed by a pipeline of projects.

During the CST boom period in Spain, a definitive study of this aspect was carried out by Deloitte for the
Spanish industry association Protermosolar (Deloitte, 2011). This study found that the economic benefit to the Spanish economy was well in excess of the net extra cost of the tariff support measures that were used to stimulate the CST industry.

Since the commencement of the FIT in 2007, investment grew and contributed 1.65 billion Euro to Spain’s
GDP in 2010 – a period when Spain was significantly affected by the global financial crisis 18. Of this 1.65 billion Euros, it was reported that 89.3% was in construction and most of the remainder was for ongoing expenditure for operation and maintenance of completed systems, with 2.67% being for R&D.

This activity was spread over a variety of sectors with about 70% of the investment remaining in Spain.
The report predicted that if the targets proposed for the period 2011 to 2020 would have been met, the contribution to GDP in 2020 could have been in the order of 3.5 billion Euros.

A total of 23,844 people were employed in CST related activity in Spain in 2010 according to the Deloitte study. Much of the employment in construction directly helped a sector most affected by the overall economic contraction at the time. It is estimated that 176 million Euros in employment subsidies were offset in 2010 as a result.

The job creation potential of the CST industry in Australia will also be relevant especially during the construction phases where the labour needs are largest. Significant labour will also be required during operation and maintenance. Furthermore, indirect jobs will be generated in the rest of the economy as a consequence of these activities.

Estela’s Solar Thermal Electricity Global Outlook estimated that under a moderate scenario between 9 and
11 jobs per MW would be created for installation activities and 1 job per MW for O&M activities. These job creation numbers are based on the experiences in Spain and the United States (ESTELA, 2016).

18
Note that the 2010 analysis is a snapshot of an industry in a growth phase, and so the cost of the FiT is small relative to the
investment being made in construction. Over time the FiT cost will continue to accumulate, albeit discounted over time.

S774Doc-676563417-363 / v0.2 55 CST Value Proposition
Skills related to the construction and operation of CST plants are similar to skills related to operation of thermal power plants fuelled by coal and natural gas. Thus, skills developed in fossil fuel industries will be applicable and transferable to most CST related jobs.

3.6 Diversification of Supply Options
Countries that develop CST capability, even though this has significant lead time and begins with some early and more costly projects, gain the option of being able to access the technology more cost- effectively in later years when the need for firm capacity and ancillary services, e.g., replacement of an existing conventional power station fleet, is likely to be higher. This is often called option value. Option value can best be thought of as an insurance policy value. The value can be linked to the extra cost that would be encountered via the alternative of importing components and capabilities from other countries at a future time.

At present, the global availability of resources and of the production value chains, for example, of PV panels, are all very much a part of international geo-political tensions. As such, the option value of domestic deployment of CST is further enhanced by a direct national security element.

Concrete, aggregate, carbon steel, sodium nitrate and solar glass are the most consumed materials in the manufacturing and construction of CST plants. According to the U.S. Department of Energy, supply scarcity is unlikely to impact the raw material requirements for CST installation (DOE, 2021).

S774Doc-676563417-363 / v0.2 56 CST Value Proposition
4 CST Cost

4.1 Today’s CST Cost
Total installed CST project cost depends on several variables and, thus, varies significantly between projects. Comparing cost figures from IRENA’s renewable cost database and Fichtner’s cost database, total installed cost figures per kW of capacity, can vary by more than +/- 50%.

The main variables impacting the total installed cost are:

▪ the CST technology,
▪ the project size (economies of scales),
▪ the plant configuration, i.e., main equipment / system sizes (solar multiple19 and storage size),
▪ the operating conditions,
▪ the project country, and
▪ the site location (i.e., the prevailing boundary conditions, site preparation cost, grid connection).

Hence, specific CST cost figures should always be provided together with at least the main determining variables. A CST plant can be broken down ultimately into three main subsystems: the solar field, the thermal energy storage and power block (incl. the balance of plant systems). The size (capacity) of these three subsystems drives the overall cost of a CST plant and the subsystem size (capacity) can be chosen independently when projects are optimised to market conditions and customer requirements.

According to the latest Renewable Power Generation Cost report by IRENA, total installed costs for CST plants fell between 2010 and 2020 by around 50% (IRENA, 2022b). This decrease in total installed cost is considered significant, in particular given that only around 6.5 GW of CST capacity has been deployed so far and over the years the average storage duration and solar multiples have increased. Based on IRENA’s
Renewable Cost Database, the current weighted average total installed cost for CST plants, for power generation, amounts to US$ 4,746/kW (IRENA, 2022b).

To calculate the CST cost for the present study, a cost model for CST in Australia has been developed 20.
The model is broken down into the three main systems solar field, thermal energy storage and power block. By breaking down the cost into these three systems, the cost for CST plants can be calculated for a wide range of configurations by simply specifying the solar field size, the thermal energy size, and the power block size.

The underlying specific cost figures have been determined for a green field solar tower reference plant, using a more detailed cost model. The cost model developed is based on international CST projects
Fichtner has been involved in, supplier budget quotes, and recent stakeholder engagements. Industry and stakeholder engagement conducted for this study include amongst others Vast Solar, Cosin, Brightsource
Energy, SBP, Sener, JC, BASF, NREL and IRENA.

19
Solar multiple is a measure of the solar field size (aperture area) as a function of the power block's nameplate capacity.
20
The CST cost model spreadsheet can be made available upon request contacting the authors of this study.

S774Doc-676563417-363 / v0.2 57 CST Value Proposition
International cost data has been transferred to Australia, applying exchange and country factors, published by Compass International (Compass International, 2022). Further, historical cost data has been adjusted using price escalation rates, based on the Chemical Engineering Plant Cost Index (CEPCI).

Considering a plant configuration for mainly night-time dispatch (14 hours of storage) and making use of the economies of scale, in particular in the power cycle, a reference plant with the key parameters in Table
4-1have been defined.

Table 4-1: CST reference configuration

Item Unit Reference Value

Power block (gross) MWe 150

Power block (net) MWe 140

Thermal energy storage MWht 4,667

Solar field MWt 720

Based on the assumption of an EPC procurement model, in total 20% of indirect cost are applied on top of the direct cost. Further, on top of the EPC 5% of owner’s cost are added, incl. land cost, development cost, utility connections and additional owner’s costs during construction and commissioning. Given current global supply chain issues and related cost escalations, an additional escalation factor of 13% has also been included. This escalation factor is consistent with the Draft 2023 Inputs, Assumptions and
Scenarios Report (draft 2023 IASR), proposed to be used in AEMO’s 2023/2024 Forecasting and Planning publications for the NEM, including the 2024 Integrated System Plan (ISP) (AEMO, 2022b).

To allow the calculation of CST cost for a wide range of configurations, scaling exponents are provided for each of the three cost portions. For any new plant configuration considered in this study, (i.e., capacities for each of the three main systems) the new cost figures are calculated, applying the scaling exponents on the specific cost figures of the reference plant. The principle is:

Subsystem Capital Cost = Base Subsystem Capital Cost x (Subsystem size/Base Subsystem size)^n

I.e., the lower the scaling exponent, the lower the specific cost will be for a larger size of the subsystem.
Scaling exponents have been determined using the detailed cost model and the available cost data base including different project and system sizes. For the three main systems exponents are:

▪ Power block: 0.8
▪ Thermal energy storage: 0.85
▪ Solar field: 0.88

In order to account for the (in parts) large differences in total installed cost for different regions in
Australia, the model allows for the selection of different regions. Regional factors have been derived from
AEMO’s draft 2023 IASR (AEMO, 2022b). For WA and NT (not included in AEMO’s draft 2023 IASR), data from a previous AEMO report has been used (GHD, 2018).

S774Doc-676563417-363 / v0.2 58 CST Value Proposition
An Australian reference location was established, in order to convert international cost data to Australia.
The reference location was nominated as an inland location in NSW, I.e., the “NSW medium” region is the nominated reference region with all other regions then scaled accordingly. This is important to note, as the reference region used in the AEMO draft 2023 IASR is VIC low. I.e., when determining the reference cost data equivalent to AEMO’s draft 2023 IASR VIC low must be selected. NSW medium has been selected as best aligning with the locations where CST plants in other countries have been built. I.e., CST plants are generally located in remote locations, however, still in reach of required infrastructure and labour force. Considering the above methodology and underlying data, the specific and total cost for the reference configuration built in a “NSW medium” reference location with FID at the end of 2022 (FY 2022-
23) is summarised in the table below. The total cost of a 140 MWe CST System with 14 hours of storage, targeting firm, night-time power generation is estimated at A$ 915 million.

Table 4-2: Subsystem specific cost and total cost for the CST reference configuration (NSW medium)

Item Unit Reference Specific Total Cost
Value Cost [A$/UNIT] [mA$]

Power block MWe 140 2,028,795 284

Thermal energy storage MWht 4,667 35,880 167

Solar field MWt 720 644,230 464

The two tables below provide further CST cost examples for different CST plant sizes and configurations.
Table 4-3 shows three examples with different power block sizes for nighttime dispatch, i.e. solar heat is collected during the day, stored and used during the night to generate power.

Table 4-3: Example CST cost for nighttime dispatch configurations (2023 / VIC Low)

Item Unit Night 1 Night 2 Night 3

Power block MWe 100 140 180

Thermal Energy storage Hours / MWht 16 / 3,810 15 / 5,000 14 / 6,000

Solar field SM / MWt 2.6 / 619 2.4 / 800 2.2 / 943

Total Cost mA$ 726 913 1,067

Power Block mA$ 211 267 319

Thermal energy storage mA$ 133 167 195

Solar field mA$ 382 479 554

Total specific cost A$/kW 7,261 6,525 5,930

S774Doc-676563417-363 / v0.2 59 CST Value Proposition
Table 4-4 shows three examples for day and nighttime operation. The power block size is fixed and the solar multiple and thermal storage size differ. For better comparison with GenCost and AEMO cost data, the cost data for the different CST configurations is provided using the VIC low regional factor. The difference between VIC low and NSW medium amounts to 6.2%.

The configuration “Night 2”, with a net plant capacity of 140 MWe, a storage capacity of 15 full load hours and a solar multiple of 2.4, is the configuration considered by the GenCost team as the CST reference plant in its annual electricity cost estimates for Australia (CSIRO, 2023b). Each year, Australia's national science agency CSIRO, and AEMO, work with industry to give an updated cost estimate for large-scale electricity generation in Australia. Following on the publication of draft numbers in December 2022, ASTRI submitted the cost model for CST developed in this project to the Aurecon, GenCost and AEMO teams. It was subsequently accepted by them but published only for a single configuration of 140 MWe, 15 hr storage SM2.4. Additional CST configurations have been proposed to be included in the future GenCost publications using the provided cost model.

Table 4-4: Example CST cost for day and nighttime configurations (2023 / VIC Low)

Item Unit Day & Night 1 Day & Night 2 Day & Night 3

Power block MWe 100 100 100

Thermal Energy storage Hours / MWht 10 / 2,381 15 / 3,571 20 / 4,762

Solar field SM / MWt 2.8 / 667 3.4 / 810 3.6 / 857

Total Cost mA$ 708 821 881

Power Block mA$ 211 211 211

Thermal energy storage mA$ 89 126 160

Solar field mA$ 408 484 509

Total specific cost A$/kW 7,082 8,208 8,805

4.2 Future CST Cost Developments
The determined CST cost reduction curve was based on the latest IEA World Energy Outlook and the provided CST deployment projection of the Announced Pledges Scenario (APS) Scenario (IEA, 2022) and a defined learning rate for CST.

As shown in Section 2.3, the compound average annual growth rate for CST in the APS amounts to 14%, resulting in 318 GWe of installed capacity in 2050. Compared to some other scenarios, the APS scenario can be regarded as a moderate scenario, concerning CST deployment. It should be noted that the scenario only includes the CST capacity projected for power generation, thus, any future CST deployments in other sectors are not considered for the CST cost development in this study.

S774Doc-676563417-363 / v0.2 60 CST Value Proposition
For the learning rate until 2035 15% cost reduction per doubling of installed capacity is assumed and thereafter 10% was considered. Comparing this learning rate with the learning rate determined by IRENA for the total installed cost of CST of more than 20% (IRENA, 2022b), the approach taken is regarded as conservative.

Figure 4-1 provides the resulting cost reduction curve considered in this study “Fichtner Scenario” as well as a comparison with other cost reduction scenarios. As can be seen, the considered cost reduction scenario, lies in the middle when compared to the other cost reduction scenarios. Both the IEA and IRENA
Scenarios consider a learning rate of 20%. The Moderate NREL scenario is based on cost reductions provided by NREL in its Annual Technology Baseline (NREL, 2022). AEMO NZW post 2050 relates to
AEMO’s cost reductions provided as part of the Draft 2023 Inputs, Assumptions and Scenarios Report
(AEMO, 2022b).

Figure 4-1: Considered CST cost reduction scenario in comparison to other reduction scenarios

4.3 GenCost Comparison
The 2023 GenCost study (CSIRO, 2023b) reports cost estimates for a range of energy generation and storage technologies (see Section 4.1). Following on the publication of the draft version in December
2022, ASTRI submitted the cost model for CST developed in this project to the GenCost and AEMO teams to adequately reflect CST in the GenCost study and consequently in the AEMO ISP assumptions (AEMO,
2023). The model was accepted for a single configuration of 15 hr, SM 2.4, with specific cost A$ 6,525 /kW
(VIC low).

In the following, the work done as part of this study is used to complement the GenCost LCoE and capital cost of storage comparison. As part of the assessment of the grid connected power generation end-use case (Section 5), different CST plant configurations at different locations within the NEM have been modeled. Using the CST cost model developed as part of this study and the same financial assumption considered in the GenCost study (CSIRO, 2023b), the LCoE of a wide range of CST plants have been calculated. Besides different storage and solar field sizes, also the power block size was varied to show the full possible range for CST.

S774Doc-676563417-363 / v0.2 61 CST Value Proposition
Figure 4-2 shows the 2023 LCoE for various technologies from the GenCost 2023 report compared with the newly added CST range. The lower bound is a 180 MWe plant (net) in a location like Longreach with
DNI 2,600 kWh/m2/a and the upper bound is a 54 MWe plant (net) in a location like Wagga Wagga with a
DNI of 2,200 kWh/m2/a. It should be noted GenCost didn’t report an estimate for the LCoE for CST and the LCoE was deduced using the same assumptions provided by the report:

▪ WACC 5.99%
▪ IDC Included into CAPEX per year
▪ LCoE method Simple LCoE.

Figure 4-2: 2023 GenCost LCoE for various technologies compared with CST

Figure 4-3: 2030 GenCost LCoE for various technologies compared with CST

S774Doc-676563417-363 / v0.2 62 CST Value Proposition
The CST of the lower bound is a configuration with a comparatively large solar multiple and thermal energy storage of 16 full load hours, allowing both for nighttime and daytime dispatch, resulting in the lowest LCoE of a stand-alone CST plant.

It can be seen that both in 2023 and in 2030, CST offers a competitive LCoE range compared to other flexible, high-capacity factor, low emissions options. It is also a lower cost than open cycle gas peaking plants even without considering GHG emissions costs.

The GenCost report deals also with current storage technology capital costs. In comparing the capital costs of storage technologies, GenCost has divided the total cost of a system by the hours of storage to arrive at a cost per kWh metric. This should not be confused with the kWh linked cost factor that is then added to a cost per kW factor in other analysis, such as the CSIRO Energy Storage roadmap. In the case of
CST, the GenCost storage cost is not a correct interpretation as the full system cost including the solar field contribution has been divided by the storage duration, resulting in an incorrectly high value relative to, for example, PHES of the same duration.

In Figure 4-4, the analysis for CST has been corrected by dividing only the sum of the storage and power block cost contributions by duration so that it is presented on the same basis as the other storage technologies. This is the approach taken in the CSIRO Energy storage roadmap. With this analysis, CST storage is seen to be very close to PHES of the same duration.

Storage technologies like batteries and PHES do not include the CAPEX of the PV field needed to charge them in their capital cost.

Figure 4-4: Capital costs of storage technologies using GenCost2023 methods and assumptions with CST corrected.

S774Doc-676563417-363 / v0.2 63 CST Value Proposition
5 Grid Connected Power Generation

5.1 Key Findings
The list below provides the key findings made during the techno economic analysis of the grid connected power generation end use case and provides an overview of the value proposition CST can offer within this sector in Australia.

▪ Modelling capacity expansion in the NEM with OpenCEM using the Fichtner CST cost model and
AEMO's draft cost models for other technologies, shows CST uptake of 5.6 GW by 2050, dispatching
around 10% of total electricity.
▪ Approximate modelling of capacity expansion in the SWIS suggests rapid CST uptake from 2030
onwards growing to 840 MW by 2050 and dispatching as much as 20% of total electricity. This is a
greater fraction compared to the NEM, linked to an absence of existing hydro and smaller levels of
coal generation.
▪ Modelling whole electricity system capacity expansion with and without the inclusion of CST allows the
financial value of a CST system to be deduced via the overall reduction in total system annualised cost.
▪ Direct comparison of CST system lifetime financial value with lifetime CST system cost shows that the
2023 deficit between cost and value is lowest in the SWIS and shows the earliest break even between
cost and value in 2025. In the longer term South West NSW offers the greatest net value.
▪ The inherent value of CST systems in grid connected applications is their ability to fill in the gaps with
long intraday storage when variable wind and PV generation cannot meet demand. This becomes
most apparent as large legacy fossil fuel generation systems are progressively retired.
▪ The uptake of household BESS that is taken as a boundary condition by AEMO and reaches 30 GW in
the NEM by 2050, (17% of total), is based on questionable assumptions and when this is removed,
results in the choice of other technology mixes being preferred including an extra 1.5 GW of CST by
2050.
▪ CST is ideal at filling the important shortfalls in variable renewable generation. The optimal dispatch
profile for CST shows a preference to times when PV is not dispatching, and wind generation is lower
relative to demand.
▪ AEMOs past ISP modelling used fixed dispatch traces for CST as opposed to optimal dispatch. This
results in predictions of no CST uptake. However, uptake is predicted to occur when using optimal
dispatch.
▪ Testing varied CST system configurations indicates that longer durations of storage and higher solar
multiples deliver the best overall system value, even though this results in some summer dispatch
during the day.
▪ The extra societal benefits from regional employment and diversity in technology options should more
than justify early actions in Australia to ensure the establishment of a viable pipeline of CST projects.
▪ To capture the whole of system benefits that CST would deliver, a realistic growth rate of installation is
needed to establish capability and supply chains. A realistic uptake trajectory beginning around 2025,
while likely to require policy intervention, would see steady CST system growth that would leave the
system around A$10 bn better off by 2050.

S774Doc-676563417-363 / v0.2 64 CST Value Proposition
5.2 Introduction
In Australia, discussion of grid connected CST systems focusses largely in the NEM regions which actually covers only the Eastern states (indicated by the spread of transmission lines in Figure 5-1). In Western
Australia, the SWIS, that is centred around Perth is also an important opportunity for grid connected systems. The Northwest Interconnected System (NWIS) and Darwin Katherine Interconnected System
(DKIS) are smaller grids with aspects in common with remote area application in general.

Figure 5-1: Transmission system coverage compared to direct normal irradiation

CST systems work best in high DNI locations. The DNI contours in Figure 5-1 therefore suggest that in the
NEM the likely location of future CST systems would be towards the inland fringes. While these locations benefit from higher DNI, it comes with a trade-off in terms of less optimal transmission connections. This noted, AEMO has designated a range of future Renewable Energy Zones that are to be the target for future transmission initiatives. A number of those are in suitable inland locations, with high DNI and in areas well suited to CST systems.

AEMO publishes an ‘’Integrated System Plan’’ (ISP) for the NEM every 2 years. The ISP uses capital cost and performance information developed by CSIRO’s GenCost program in the 12 months prior to the ISP.
This input is then used by AEMO in an internal least cost capacity expansion model run using the commercial package PLEXOS21. The most recently published ISP was released in June 2022
(AEMO, 2022d). It predicted no future uptake of CST in the NEM. In December 2022 AEMO published their

21
https://www.energyexemplar.com/plexos

S774Doc-676563417-363 / v0.2 65 CST Value Proposition
draft 2023 IASR (AEMO, 2022b) (AEMO, 2022c) in preparation for the 2024 ISP. Aurecon also published the work they had done for the GenCost in support of those draft assumptions at around the same time
(Aurecon 2022). Subsequent to the completion of the modelling presented in this report, final assumptions were published (AEMO, 2023) (Aurecon, 2023) (CSIRO, 2023b).

For this study ITP's open source capacity expansion model ''OpenCEM'' has been used to examine the impact of assumptions and parameter values on the uptake and potential role of CST in the NEM. It has allowed a quantitative assessment of CST cost vs value to be established year by year. OpenCEM carries out a similar simultaneous optimisation of dispatch and determination of least cost uptake of new generation as AEMO carries out using Plexos. A major difference in this work is that in past years, AEMO’s modelling of CST systems in Plexos has used a single pre-determined daily dispatch profile rather than strategically optimised dispatch. On the other hand, the modelling carried out with OpenCEM has modelled CST dispatch that is strategically optimised. As explored further in Section C4 this can make a large difference to predicted uptake.

A simplified proxy model has also been used to approximate the situation for the SWIS. A detailed description of the modelling approach in OpenCEM can be found in Appendix B.

The starting point to the investigation was a benchmark comparison of the 2022 ISP with OpenCEM results using the same input assumptions. The 2022 ISP results have been reconciled with OpenCEM as described in Appendix B Section B7. The 2022 ISP predicted no future uptake of CST in the NEM. Re- running OpenCEM with a range of plausibly lower CST CAPEX models and testing other configurations, did not change this outcome. Investigations were conducted to understand why CST was not being selected. This revealed that what appeared to be an unrealistically low CAPEX model for battery systems appeared to be locking out CST and also to a large degree pumped hydro systems from uptake. The battery model appeared to be both fundamentally low, based on the 2 hour battery cost, but also have an unrealistically low incremental increase for BESS of longer duration.

The publication of the draft 2023 IASR in December 2022, provided an opportunity to re-visit the analysis with more up to date numbers for other technologies. It transpires that there are considerable changes between the 2022 and draft 2023 IASR data sets. One of the most striking and pertinent changes is that the underlying cost model for BESS shows a significant increase and, most importantly, the dependence on cost increase on storage duration seems much more realistic.

5.3 Predicted CST Uptake
Optimal capacity expansion in the NEM was modelled using OpenCEM based on published draft inputs and assumptions for the draft 2023 IASR. The Analysis resulted in predicted uptake of 2 GW of CST (at 15h of storage) by 2050.

However, the cost model and configuration for CST in the Draft 2023 Inputs, Assumptions and Scenarios
Report remains very much in question in the view of the authors of this study.

As presented in Section 4, the present study has established a revised capital cost model for CST that takes into account the best available information globally and presents it on the same basis for plants hypothetically contracted in FY2023. The new cost model has separate cost terms for the solar field, the thermal storage and the power block and balance of plant. This cost model can be used to determine the

S774Doc-676563417-363 / v0.2 66 CST Value Proposition
capital cost of CST plant configurations of varying storage duration and solar field size relative to the power block. Thus, the revised model has now been used within the OpenCEM modelling and the configuration has been considered as part of the sensitivity analysis.

Figure 5-2: Comparison of dispatchable renewable LCoEs using AEMO 2023 draft cost data for ISP2024 with
addition of new CST cost model. (12 hour duration using ISP Draft 2024 Data 6.5%WACC)

The new cost model for CST together with the costs for other technologies presented in AEMO’s draft have been used to compare technologies using the dispatchable LCoE analysis adapted from (ITP, 2018).
The comparison, in Figure 5-2 shows that at 12 hours duration, the CST option is considerably cheaper than battery solutions and on par with pumped hydro.

It is important to note that even with the changes in AEMO’s draft 2023 IASR, questions remain as to if the new cost model for BESS is not still a bit on the low side.

Figure 5-3 shows results for predicted optimal capacity and dispatch evolution in the NEM with the baseline CST configuration adopted, of 20 h storage duration and solar multiple (SM) 3.88, CST costings with the new model, as well as other costs and assumptions from the draft 2023 IASR costs in the Step
Change scenario. The results show an uptake of 5.57 GWe of CST in 2050. Since construction of this amount of capacity where none existed before is considered physically impossible in a single year, the approach taken is to apply a fixed compound annual growth rate (CAGR) of 35% starting from 2030 that then results in the predicted 2050 capacity being built.

It should be noted that computational effort precluded investigating options where CST plants also included electric, or gas fired supplementary heating. Any such additions could only add to the predicted uptake. Whilst CST with 20h of storage and a solar field of SM 3.88, appeared to be the best baseline configuration, it is likely that systems actually constructed will have a range of configurations according to the circumstances of each project.

The outcome of the analysis is 5.57 GWe of CST (20h) out of a total system capacity of 180 GWe. While this is only a small proportion of total capacity (3.2%), this is still a significant amount for the Australian CST

S774Doc-676563417-363 / v0.2 67 CST Value Proposition
industry. Furthermore, the installed capacity needs to be considered in terms of its share of dispatchable / firm capacity and share of total energy dispatched.

Figure 5-3: OpenCEM results for capacity (top) and dispatch (bottom) in the NEM, for 20h, SM 3.88, Fichtner cost for
CST 35%CAGR , other costs and assumptions from the draft ISP 2023 costs in the Step Change scenario

It can be seen that as coal and gas fired generation progressively leaves the system and the amount of variable PV and wind increases, the fraction of the total capacity that can be defined as ‘’dispatchable’’ settles at around 30% of the total. In 2050 this is approximately 53 GWe. The CST component then is
10.4% of the total dispatchable (firm) capacity.

When dispatched energy is examined it is seen that the CST contribution is a larger fraction than its capacity. It is around 10% of total electricity dispatched and 49% of the dispatchable electricity

S774Doc-676563417-363 / v0.2 68 CST Value Proposition
dispatched. It is apparent that CST is predicted to provide a critical high value role that facilitates a full transition to zero emissions.

5.3.1 Approximate analysis for SWIS
The analysis of the NEM appears to indicate that under almost any scenario, CST should be taken up and fill a critical role but not until the majority of coal plants retire and demand grows further. However, this is not the case with the SWIS, which shows CST uptake earlier.

The SWIS is an interesting system in terms of timing and available options for meeting dispatchable renewable energy requirements. The SWIS has not been subject to the same level of modelling input by
AEMO as has been done for the NEM. As a consequence, developing a detailed OpenCEM model is an onerous task outside the scope of the present project. However, to gain some insights, a simplified proxy model was established as detailed in Appendix B, Section B6.

Figure 5-4 shows capacity and dispatch outlooks for the SWIS produced by OpenCEM with the new CST cost model and other technology costs and assumptions from the AEMO draft 2023 IASR. CST uptake is predicted beginning from about 2030 and reaching a total capacity of approximately 840 MWe by 2050.
The corresponding dispatch mix forecasted indicates that CST plays a critical role in balancing variable generation as the system decarbonises, contributing over 21% of generated electricity by 2050.

It is apparent that the SWIS, with the absence of existing hydro capacity with no Snowy 2.0 and with a smaller contribution from coal, sees an earlier uptake of CST predicted than in the NEM case. It also shows that CST uptake results in growth that delivers a larger fraction of generated energy.

S774Doc-676563417-363 / v0.2 69 CST Value Proposition
Figure 5-4: OpenCEM results for capacity (top) and dispatch (bottom) from SWIS basic model for 20h, SM 3.88,
Fichtner cost for CST

5.3.2 Combined uptake
The previous sections provide a separate analysis for CST uptake in the NEM and in the SWIS. The combined national uptake of CST, including the separate trajectories is shown in Figure 5-5.

Figure 5-5: OpenCEM results for CST capacity uptake in the SWIS and NEM

S774Doc-676563417-363 / v0.2 70 CST Value Proposition
As can be seen in Figure 5-5, it is the SWIS where initial uptake is suggested to be favoured, but this levels off and as the value proposition in the NEM catches up, subsequent deployment in the NEM dominates in later years.

5.4 Quantifying CST Value
Section 3 discusses the various sources of value that a CST system offers. Under present market and policy settings in the NEM and SWIS, not all these sources of value are fully rewarded. The financial income that can be estimated is uncertain and not yet sufficient to cover the financing of a CST system.

The OpenCEM model works by minimising the overall annualised cost of meeting demand year by year.
Given that it is predicting uptake of CST in both the NEM and SWIS would be part of an optimised system in coming years, it implies that the total system cost of ownership to society will be lower if CST is included.

To arrive at a quantitative assessment of value, a series of runs were conducted where CST was removed from the technology options for optimisation, but a single 100 MWe CST system was forced in for each year in turn, at zero cost, at various location. This then resulted for each case, in a reduction in annualised system costs in each year thereafter. This reduction in costs is a consequence of the avoided construction of other technologies and the avoided dispatch of other generators. It is made up of reductions in capital recovery, fuel costs and O&M from these reductions in capacity and dispatch.

This reduction in annualised costs effectively represents the annualised benefit or value of a hypothetical
CST system. To quantify a total value of a system at its time of construction, the lifetime NPV of the annual benefit was determined. For systems built with an economic life that extended beyond the 2050 modelling window, it was assumed that the annual benefit continued at the level determined by the model for 2050.

This quantification of value is capturing the value of CST that is linked to its ability to offer firm capacity, long duration storage and implicitly, ancillary services, in the grid. This value is being determined by comparison with the cost of other technology mixes that achieve all these things. What is not counted in this assessment is the values that could be attributed to; societal benefits, diversity of technology, and reduction of emissions below current targets.

This lifetime financial value of a CST system can be compared to a corresponding cost to allow an assessment of net benefit. The appropriate comparison is the sum of the capital cost in the year of construction of the CST system plus the NPV in that year of the lifetime O&M costs. The net value that is the difference of these two is an estimate of the societal net financial benefit of a CST system. The net value benefit would manifest in lower prices to energy of consumers and / or an increased return on investment for project owners, depending on the market at the time.

This quantification of predicted value is also independent of the market value that might be achieved in the wholesale electricity market along with any future capacity market. Rather it is to be hoped that future adjustments to the NEM and SWIS market designs are such that they will more closely reflect the inherent values predicted here.

S774Doc-676563417-363 / v0.2 71 CST Value Proposition
Figure 5-6 shows the predicted year by year change in total CST system lifetime value per kW of capacity for a 20 hour storage, SM3.88 CST system in the NEM compared to the lifetime costs. Figure 5-7 shows the same analysis for the SWIS.

It can be seen that the overall trend is an increase in value over time, in early years. This is as expected as existing fossil fired plants gradually remove from the system and the allowable emissions budget reduces in parallel with a continued increase in variable RE and demand. At the same time the CAPEX and hence total lifetime cost of a system reduces with time as a consequence of the learning curve model of cost reduction described in Section 4.2. The point in time where the lifetime value exceeds lifetime cost marks the point where from a societal perspective, CST has become a profitable investment. These results indicate a crossover in 2026 in the SWIS and 2028 in the NEM.

The considerable uncertainty in these estimates is indicated approximately by the error bars. It results from the uncertainty in the capital cost estimates for all technologies plus uncertainty over time in estimating their respective cost reduction trajectories.

It is notable that the SWIS offers considerably higher net value (value minus cost) than the NEM at present and continuing to the 2032. This is in line with the earlier uptake predicted (see Figure 5-4).

However, the value for a first 100 MWe plant in the SWIS peaks in 2030 and then drops and levels off. In the NEM, value continues to climb out to 2045 before dropping. Noting that this determination of value is for a single CST plant forced in for each year in turn, where no other CST plants are allowed prior to it, the explanation hypothesised is that long duration storage is needed earlier in the SWIS and so if the CST option is not in the OpenCEM technology list, it will choose BESS and new gas plants as a matter of necessity. Once assumed as part of the system, these then reduce the value of subsequently adding further CST.

Figure 5-6: CST value compared to CAPEX and lifetime total cost for the NEM.

S774Doc-676563417-363 / v0.2 72 CST Value Proposition
Figure 5-7: CST value compared to CAPEX and lifetime total cost for Western Australia SWIS.

The point in time at which the value and cost cross over does not mean OpenCEM will predict construction at that time as other technology combinations may be fractionally better on net value.
However, the higher the net value the higher the likelihood that CST will be in the optimal mix.

The OpenCEM model, individually assesses technology uptake in each renewable energy zone in the NEM, the results in Figure 5-6 represent the results from the apparently most favourable zone in each year. The trend was that Central Queensland was more favourable in earlier years and then moving to SW NSW.
However, given the strong sensitivity to input assumptions and other factors, drawing zone specific conclusions is highly uncertain. Considering the results as representative of the whole NEM is safest.

As discussed earlier, construction of CST requires a realistically achievable growth trajectory to reach an identified target capacity. To illustrate the impact to society of facilitating the required uptake of CST,
Figure 5-8 shows the change in 2023 NPV of total net value for the total growth trajectory for NEM and
SWIS combined that is shown in Figure 5-5.

The final result is a suggested A$10 bn benefit in 2023 NPV to the system overall. Only the first few small plants would require some support to account for a deficit between value and total cost. For such a scenario to play out however, energy market and policy settings will need to adjust to properly reflect the value that has been identified. Adjustments to the WA Reserve Capacity Mechanism and settings on the proposed national Capacity Investment Scheme could provide a way that this comes to pass.

S774Doc-676563417-363 / v0.2 73 CST Value Proposition
Figure 5-8: Accumulated cost and value in the NEM and SWIS combined

5.5 CST’s Role in the System
To better illustrate the role that CST plays in an optimised system, Figure 5-9 and Figure 5-10 show the hour by hour dispatch of different technologies in representative weeks in winter and summer respectively.

In the winter case in Figure 5-9 it can be seen that the optimisation results in CST dispatch that ramps up as the PV output declines at the end of each day. It then continues largely for as long as stored energy is available. In many cases that means generation all night and then ramping down as PV rises again the next morning. The night of June 4th (early hours of June the 5th) provides an example of a situation when there is a large amount of wind generation during the night, almost sufficient to meet all demand. In this case the CST dispatch is zero. During the day on June the 6th there is some CST dispatch, suggesting that energy remained in storage from the night before and daytime dispatch was preferred over curtailment.

In the summer case shown in Figure 5-10, dispatch during the night is still favoured however the extra energy collected during the day, means that dispatch during the day is predicted on many days, as there is simply too much energy available. Building a system with a lower solar multiple would tend to reduce the daytime dispatch however this modelling with OpenCEM does not favour choice of such reduced size solar multiples. What appears to be the case is that the larger solar multiples allow for greater critical high value generation during winter nights and on days when the overall solar resource is poor. Then frequent daytime dispatch in summer is better than simply wasting energy.

S774Doc-676563417-363 / v0.2 74 CST Value Proposition
Figure 5-9: Winter week 2050 dispatch for 20 hour SM 3.88 CST system

S774Doc-676563417-363 / v0.2 75 CST Value Proposition
Figure 5-10: Summer week 2050 dispatch for 20 hour SM 3.88 CST system

S774Doc-676563417-363 / v0.2 76 CST Value Proposition
This interpretation of CST dispatch is examined more rigorously in the statistical analysis in Figure 5-11, which illustrates the seasonal average dispatch by hour of the day. In all cases dispatch is being favoured outside of daylight hours. Daytime dispatch is relatively small during winter, but larger in other seasons.
However even in summer, night dispatch is favoured over daytime.

There is still between 25 - 50% average dispatch capacity factors during the daylight hours however, which is assumed to result from contributions on those days when the overall PV output is intermittent or low. Plus, from situations where the CST systems have sufficient energy in storage that they would otherwise curtail. Examining the plots for summer, it can be seen that the average level of dispatch during daylight hours increases as would be expected for the higher input solar resource levels.

Figure 5-11: Normalised average hourly CST dispatch for South West NSW (SWNSW) zone

A more in depth understanding of the manner in which CST is dispatched in an optimal way is provided by considering the frequency distribution of CST dispatch vs that of PV or wind as shown in Figure 5-12 and Figure 5-13 respectively and against residual demand in Figure 5-14.

In the case of the comparison with PV, it can be seen that by far the majority of dispatched CST MWh are at times when the PV dispatch is zero. Conversely there is zero dispatch of CST at times when PV is at its maximum. There are however very small amounts of dispatch of CST at the low to moderate levels of PV dispatch that are consistent with the explanations above. That is to say, CST is always best dispatched when the sun is not shining unless its storage is full and there is nowhere for the energy to go.

S774Doc-676563417-363 / v0.2 77 CST Value Proposition
Figure 5-12: Frequency distribution of CST dispatch compared to PV

The correlation of CST dispatch with wind dispatch is more subtle. In this case all CST dispatch occurs when there is some level of wind dispatch, they actually show a strong correlation overall. This largely reflects the fact that there is almost always some wind being dispatched, it is virtually never zero, particularly at night when most CST is dispatched. When overall wind dispatch level is very high, CST dispatch drops to zero, consistent with the interpretation of the June 6th example above where wind was sufficient to meet all load at night.

Figure 5-13: Frequency distribution of CST dispatch compared to wind dispatch level.

S774Doc-676563417-363 / v0.2 78 CST Value Proposition
Figure 5-14 examines the volume of CST dispatched energy vs the residual demand level. Residual demand is the underlying demand plus the demand for electricity charging of other storage less the output from wind and PV combined. It is effectively the demand that must be met from dispatchable generation including energy storage systems.

Noting that the predicted installed capacity of CST for this is 5.56 GWe, CST can at any one time only contribute 5.56 GWe of power. Thus, for example if residual demand is 15 GW, it can at most provide 30% of that, at 8 GW it could contribute up to 62.5% and at 20 GW 25%. Whilst it is hard to see in the figure, it is the case that CST is contributing at close to its maximum capacity for all residual demand levels above around 10 GW and crucially, all the very high residual demand outliers. These are the most critical times.

Figure 5-14: Frequency distribution of CST dispatch in the NEM compared to residual demand level.

5.6 Sensitivity to Configuration, Cost, and other Factors
A range of scenarios and parameter variations have been examined for the NEM case. The results of these are detailed in Appendix C and summarised here.

For a given nameplate electrical capacity, CST plants can be configured with various amounts of thermal storage and solar field sizes (measured with the Solar Multiple). These parameters were examined independently for their impact on 2050 capacity uptake, dispatched energy level and overall system net present cost. It was found that all metrics showed very strong benefit from increasing storage level to at least 10 hours with SM favoured at a size that corresponds closely to one that minimises the LCoE.
20 hours with SM 3.88 was chosen as the baseline for the investigation overall.

Other parameters and scenarios were examined for their impact on predicted uptake and dispatched energy of CST as summarised in Table 5-1.

S774Doc-676563417-363 / v0.2 79 CST Value Proposition
Table 5-1: Summary of results to NEM 2050 CST uptake from various assumption changes.

Change Change in Change in Other impacts
CST CST
Capacity Dispatch
[MW] [MWh]

Draft ISP2024 assumptions vs Reduced uptake of PV and
0 0
ISP20222 assumptions BESS, increased uptake of wind

introducing Fichtner model for CST CST uptake replacing some
with other draft ISP2024 vs all draft +2,081 +8,281,835 Wind PV and BESS
ISP2024 assumptions

Increased CST uptake,
CST 10% CAPEX reduction vs draft +2,973 +1,5434,382
displacing some PV, Wind,
ISP2024 + Fichtner CST CAPEX
BESS, and PHES

CST 10% CAPEX increase vs draft Decreased CST uptake,
-2,053 -1,2174,519
ISP2024 + Fichtner CST CAPEX increasing PV and wind uptake

Earlier and additional CSP
+1,238 +4,597,477 uptake. Additional gas
Removing Snowy 2.0 vs draft
dispatched in 2025 to 30, Some
ISP2024 + Fichtner CST CAPEX
uptake of PHES 24h from 2040.

Removing Coordinated DER vs draft Greater uptake of 4h BESS and
+1,503 +8,688,597
ISP2024 + Fichtner CST CAPEX CSP

Removing Coordinated DER with Less gas built/dispatched
maintained emissions pathway draft +2,543 +12,009,929 leading to more PV, wind and
ISP2024 + Fichtner CST CAPEX 4h BESS built

Accelerated rollout of Wind, PV,
Zero emissions from 2040 vs draft +2,364 -10,259,199 4h BESS and PHES and CSP, less
ISP2024 and Fichtner CST CAPEX Gas

Increased BESS CAPEX vs draft Some additional uptake PV and
+968 +4,228,413
ISP2024 + Fichtner CST CAPEX 48h PHES

No CSP is built with pre-
OpenCEM optimal dispatch for an computed dispatch, capacity is
-344 -1,097,437
8 h SM2.4 system vs pre- AEMO ISP taken up by PV and Wind, PHES
2022 computed CSP dispatch traces and 8h BESS

S774Doc-676563417-363 / v0.2 80 CST Value Proposition
Notably, only removing Snowy 2.0 completely, resulted in a prediction of earlier deployment of CST. As can be seen, many of them do predict greater CST uptake. Examination of sensitivity to cost gives an indication of how the high uncertainty in the costs estimates for all technologies affects the technology mixed suggested as optimal.

The coordinated DER (i.e., distributed small low duration BESS) that is shown as a significant share of dispatchable capacity in the results is not a prediction of the model. Rather it is an assumed result of the prediction of consumer uptake that AEMO is using. Removing it as an assumption whilst fixing the emissions trajectory at the same levels shows that it is not immediately replaced by short duration BESS in the model predictions. Instead, an extra 2,543 MW of CST is chosen along with more PV wind and 4- hour BESS. The DER BESS are thus not a least-cost approach for society. It is hard to see why such high uptake by consumers would actually occur.

The bottom row in Table 5-1 indicates a test of reverting to the predetermined dispatch trace for CSP previously used by AEMO for an 8 hour duration system. It shows that such an approach completely fails to recognise the strategic value of a CST plant and then leads to prediction that none will be built.

S774Doc-676563417-363 / v0.2 81 CST Value Proposition
6 Remote Area Power Generation

6.1 Key Findings
The list below provides the key findings made during the techno economic analysis of the remote area power end use case and provides an overview of the value proposition CST can offer within this sector in
Australia.

▪ Australia with its large mining sector, mainly located in remote locations, and ESG driven emission
reduction targets, has a huge demand for remote area renewable power generation - ultimately
requiring 100% renewable and hence emissions free mining operations.
▪ Adding CST to the power mix becomes increasingly beneficial when higher renewable shares are
being targeted and, hence, medium to large storage capacities are required.
▪ Examination of options using solar and wind data, for example, for the Pilbara (Newman) indicates
that for 90% RE share a combination of CST and PV is lowest cost and can provide a weighted average
LCoE of A$144/MWh.
▪ The study results are in line with the conclusion drawn in the CSIRO Renewable Energy Storage
Roadmap that CST Storage offers the least cost solution in the near and long term.
▪ For high renewable shares, combining CST with solar PV and (in some cases) wind, results in the
lowest LCoE for remote area power generation.
▪ High back-up generation cost - to provide the last and most expensive 10 - 20% percent of
generation - strengthens the CST business case. The higher the back-up generation cost, the more
beneficial are higher RE shares and, thus, larger storage capacities. Avoidance of high back-up
generation costs also justifies higher regional / remote area cost multipliers associated with higher
renewable shares (considered as part of the assessment).
▪ The higher the complementarity between PV and wind resources, the lower the benefit of adding CST
into the energy mix.
▪ Based on the future cost projections considered, the relative advantage to add CST to the generation
mix remains nearly constant over time.
▪ A more detailed and site specific assessment, would examine additional benefits not assessed in the
present study. These benefits, which would further support the value proposition for high renewable
systems with CST, include:
- avoidance of costs for grid stabilising measures; and
- reduction or avoidance in required back-up generator capacity if cheaper back-up heaters are
added to the heat transfer system of the CST plant.
- combined heat and power generation in case of an additional process heat demand.
▪ Using curtailed PV and/or wind power to further heat up the thermal energy storage, can result in a
reduced solar field size and, depending on the CST technology, in increased operating temperatures
and, thus, higher solar-to-electricity efficiencies. As part of the present high-level assessment, such
options have not been analysed in detail, but could increase the value proposition.
▪ When compared to solar PV and wind, the system size (via economies of scale) has a bigger impact in
case of CST. Thus, for smaller remote area systems, the advantage of including CST in the power mix
will reduce, while for larger system it will increase.

S774Doc-676563417-363 / v0.2 82 CST Value Proposition
6.2 Introduction
In order to assess and quantify the value of CST for a remote area power system, a reference case with a
100 MWe baseload electricity demand has been used. Specifically, the reference case reflects a remote area power system, where the constant electricity demand needs to be covered at all times - through the installed renewable energy generators or by means of a back-up generator.

The following three options have been modelled for two reference locations:

I. CST including TES in combination with PV;

II. CST including TES in combination with PV and wind; and

III. PV and wind in combination with a BESS.

Two representative reference sites were selected further described below: Newman representing the
North-West (Pilbara) mining region and Mount Isa representing a major remote mining area in
Queensland.

For each of the technologies and considered sites, at first a resource assessment has been conducted and the most representative year been identified, considering both solar (DNI and GHI) and wind data. For
CST (DNI) and PV (GHI) Solargis data was used. For wind ERA5 data was used for the yield calculations.

Using the correlated solar and wind data sets, hourly heat (CST) and electricity (PV and wind) generation profiles were produced, using state-of-the-art simulation software tools. In case of CST, the solar field was modelled using the System Advisor Model (SAM), which has been benchmarked by the authors in numerous other CST projects. For the PV yield simulations, PVsyst was used, and wind profiles were generated using WindPRO. CST heat profiles were generated using a single-tower reference plant configuration. An overview of the three considered reference configurations to generate the reference heat and electricity profiles is provided in Table 6-1.

Table 6-1: Reference configurations for each technology

Item CST PV Wind

Technology Single tower (central Bifacial single axis V162-6.0 (hub height
receiver) tracked 125 m)

Single unit / module size 580 MWt 540 Wp 6 MW

Reference capacity 580 MWt 100 MWe 100 MWe

The other CST components (incl. thermal energy storage) have been sized based on the reference plant design as provided in Appendix A2. The reference BESS system is based on the data provided in the latest AEMO Costs and Technical Parameter Review (Aurecon, 2022).

Based on the reference configurations, related generated profiles, reference cost (CST cost model and
AEMO’s draft 2023 IASR (AEMO, 2022b) for all other technologies) and scaling factors, a two-step process was conducted. In the first step, dispatch optimisation runs were conducted for each of the three options, using the optimisation software tool BOFIT. This included parameterising the CST, PV, and wind capacities

S774Doc-676563417-363 / v0.2 83 CST Value Proposition
as well as the TES and BESS sizes (power and energy). In order to meet the constant power demand, back-up generators were considered for the remaining load.

In the second step, for each of the dispatch optimisation runs, the required tariff was calculated in a techno-financial model, using a WACC of 7.5%, a 25 year PPA term and different back-up generation costs. Results were generated for the two reference sites, considering both current (2023) and future
(2030) costs.

For the back-up generation no specific generation technology or back-up fuel was specified. Rather, the total cost of the back-up system was included as a parameter. Given that power generation in remote locations is expensive, regardless of whether Diesel fuel, bio fuel or in the future green fuels are used, back-up generation costs in the range of A$ 200 - 600 /MWhe have been considered.

6.3 Reference Sites
The two representative reference sites selected are Newman representing the North-West (Pilbara) mining region and Mount Isa representing a major remote mining area in Queensland. As can be seen in
Table 6-2, Newman has excellent DNI, above 2,900 kWh/m2/a, while the DNI resource at Mount Isa is somewhat lower, but still very good with 2,649 kWh/m2/a.

Table 6-2: Remote area reference sites

Item Newman Mount Isa

Coordinates22 Lat -23.41° / Long 119.8° Lat: -20.74° / Long 139.53°

Elevation [m.a.s.l] 526 384

DNI [kWh/m²/a] 2,919 2,649

GHI [kWh/m²/a] 2,291 2,268

Wind speed [m/s] 6.12 7.1

Wind density [W/m²] 224 358

Combined PV and wind capacity 64% 67%
factor

Both sites have a high combined PV and wind capacity factor of 64% and 67% respectively. Further, the wind profile complements comparatively well the daytime solar generation, as can be seen in Figure 6-1.

The average wind capacity factor for both Newman and Mount Isa is indicated below, showing a slightly better match with a daytime PV generation in case of Mount Isa. As such, the combined capacity factor is slightly higher in Mount Isa.

22
Coordinates relates to the solar resource. For the wind profile a suitable location within proximity has been selected

S774Doc-676563417-363 / v0.2 84 CST Value Proposition
Figure 6-1: Average wind capacity factor for Newman (left) and Mount Isa (right)

6.4 Newman Reference Location
Based on the parameter runs conducted for the three options, renewable energy shares close to 100% can be achieved. The resulting Levelised Cost of Electricity (LCoE) as a function of the renewable energy share (RE share) is depicted below. While Option 1 (CST+PV+TES) and Option 3 (PV + wind + BESS) initially result in a similar LCoE, with increasing RE share first Option 1 and then Option 2
(CST + PV + wind + TES) provide the lowest LCoE. The main reasons for this outcome are the increasing cost advantage of the thermal energy storage as you move to a higher RE share (Option 1) together with a high combined PV and wind capacity factor (Option 2).

Figure 6-2: LCoE based on renewable energy share without consideration of back-up cost (Newman - 2023)

Figure 6-3: below depicts the lowest combined LCoE for the three options for different back-up generation costs (back-up generators running on fossil fuels or alternative fuels). As can be seen, LCoEs are close in the case of a back-up generation cost of A$ 200 /MWhe, with Option 1 resulting in the lowest combined LCoE of A$ 150 /MWh. In the case of higher back-up generation costs, Option 1, and Option 2
(both of which include CST) result in lower combined LCoE when compared to Option 1. In case of
A$ 400 /MWh back-up generation cost the benefit is in the range of 10% and increases in case of
A$ 600 /MWh to 15%.

S774Doc-676563417-363 / v0.2 85 CST Value Proposition
The optimum RE share resulting in the lowest combined LCoE varies depending on the back-up generation cost and the option. In case of low back up generation cost (A$ 200 /MWhe) the “lowest
LCoE” RE shares vary between 80% (Option 3) and 90% for (Option 1 and Option 2 - both including CST).
In case of high back-up generation cost (A$ 600 /MWhe) the “lowest LCoE” RE shares increase up to 95% in case of the CST options. The LCoEs (not considering any back-up generation cost) in the case of a 90%
RE share amount to A$ 144 /MWh for Option A$ 1,149 /MWh for Option 2 and A$ 160 /MWh for Option
3.

Figure 6-3: LCoE based on different back-up generation cost (Newman - 2023)

Considering the future cost reduction provided for solar PV, wind, and BESS in the AEMO draft 2023 IASR
(AEMO, 2022b) and the CST cost reductions (see Section 4.2), the advantage to add CST to the generation mix remains nearly constant. Depending on the back-up generation cost, the resulting LCoE are up to14% (A$ 600 /MWh back-up generation cost) lower when adding CST to the power mix, as depicted in Figure 6-4.

Figure 6-4: LCoE based on different back-up generation cost (Newman - 2030)

S774Doc-676563417-363 / v0.2 86 CST Value Proposition
Figure 6-5 depicts the generation split for a 90% RE scenario. Specifically, the lowest LCoE case for each option for which at least 90% of the annual load is covered by renewable sources - either directly or through a storage system. As can be seen, in the case of Option 1 more than half and in the case of
Option 2 still 23% of the annual generation is covered by CST.

Figure 6-5: Annual generation split for 90% RE share (Newman - 2023)

The annual average hourly distribution for the three options is provided in Figure 6-6. As one can expect, the CST portion is primarily dispatched during the night time hours, while day time generation is primarily covered by solar PV. In the case of Option 2, there is also substantial wind generation during the day, which means that more PV generation gets curtailed. As shown in Section 6.3, the wind resource is better during night-time, which can be seen by the average wind generation of Option 3 which is significantly higher during the night.

In summary, there is a CST value proposition in remote area power generation considering the prevailing boundary conditions of the Pilbara. For RE shares to around 93% CST in combination with PV (Option 1) offers the lowest LCoE. In case of high back-up generation cost and higher RE share it is beneficial to add wind to the mix (Option 2).

A more detailed and site-specific assessment, would examine additional benefits not assessed in the present study. These benefits, which would further support the value proposition for high renewable systems with CST, include the avoidance of costs for grid stabilising measures and the reduction or avoidance in required back-up generator capacity - if cheaper back-up heaters are added to the heat transfer system of the CST plant. Further, curtailed PV and/or wind power can be used to heat up the thermal energy storage, which can result in a reduced solar field size and, depending on the CST technology, in an increased operating temperature and, thus, higher solar-to-electricity efficiencies. Last, in case there is also a process heat demand at site, cogeneration of both power and heat will further enhance the value proposition. As part of the present high-level assessment, such options have not been analysed in detail, but can increase the value proposition.

S774Doc-676563417-363 / v0.2 87 CST Value Proposition
Figure 6-6: Annual average generation profile for 90% RE share (Newman - 2023)

6.5 Mount Isa Reference Location
The second reference site - Mount Isa – provides, in comparison to Newman, a higher combined PV and wind capacity factor of 67%. As a result, there is a lower value of adding CST to the generation mix.

Figure 6-7 depicts the LCoE as a function of the RE share for Mt Isa. As can be seen, there is always a benefit to have wind as part of the generation mix. Option 3 results in lower LCoE for RE shares up to
93%. Only in case of higher RE shares it is beneficial to add CST to the mix (Option 2).

S774Doc-676563417-363 / v0.2 88 CST Value Proposition
Figure 6-7: LCoE based on renewable energy share (Mt. Isa - 2023)

As such, there is also only a benefit to add CST to the mix in case of high back-up generation cost - as depicted for the selected 90% RE share threshold. Again, a more detailed and site-specific assessment, would deliver additional benefits not assessed in the present study, such as avoided stabilising measures, avoided back-up generator capacity or the utilisation of curtailed PV and wind generation.

Figure 6-8: LCoE based on different back-up generation cost (Mt. Isa - 2023)

6.6 Market Size
The mining sector is an enduring cornerstone of the Australian economy. Today, there are about 350 operating mines, and mining contributes more than 10 per cent to national gross domestic product.
Resource and energy export earnings in 2021-22 are forecast to reach a record A$425 billion (CEFC,
2022).

S774Doc-676563417-363 / v0.2 89 CST Value Proposition
Figure 6-9 (adapted from ITP, 2019) shows the relative magnitude of annual fuel use (excluding vehicles) at major mining industry sites along with those sites classified as power generation which are not large generators in the grid. In the mining sector the stationary fuel use is mostly for onsite power generation.
Gold and iron ore mines are significant energy users. Gold ore mines are not expected to use heat for ore beneficiation and refining. There are large bauxite mines at Gove and Weipa. There are large Zinc lead and silver mining sites at Mt Colin, Broken hill. McArthur has copper operations. Parkhurst Magnesia
Plant is under other non-metallic minerals mining. Citic pacific mining (Sino Iron ore site) at Mardie is a significant consumer of gas for power generation and also a major emitter of greenhouse gases. There is a power plant at Telfer (New Crest mining) for gold operations. Other significant gold mining operations include Newmont Tanami, Tropicana gold mines, Carouse dam, Sunrise dam, Darlot and Gruyere (ITP,
2019).

The figure includes power generators in the Northwest interconnected system, Darwin Katherine interconnected system and Mt Isa. The Diamantina power plant at Mt Isa is the largest consumer of gas at Mt Isa supplying Glencore. Some mines have dual fuel engines that are capable of using gas and diesel. There are remote sites that run on diesel and LNG.

Figure 6-9: Overview of distribution of mining energy demand and correlation with DNI potential (adapted from
ITP, 2019)

Already today there are about 350 operating mines and the average annual growth rate in energy consumption in the mining sector was 8.6 per cent over the past 10 years, far outstripping growth in other sectors. Further, the energy transition is driving new demand in a range of sectors and minerals.

S774Doc-676563417-363 / v0.2 90 CST Value Proposition
Commodities such as lithium, graphite, cobalt and nickel are expected to benefit from significant tailwinds driven by increasing battery demand, more than offsetting the headwinds posed by declining demand for high emitting fossil fuels (CEFC, 2022). The increase in demand combined with ESG driven emission reduction targets, results in a significant demand for remote area renewable power generation in Australia.

The identified existing remote area power generation sites have been approximately assessed for potential competitiveness of CST considering the DNI level, the indicated size of the CST system, the existing fuel cost and the impact of cyclone region. A ‘’profitability score’’ metric equal to the estimated cost of current fuel contribution to electricity cost divided by a high-level estimate LCoE of CST was used, to see if installing a CST system would deliver a cost saving relative to an existing fossil fuelled system.

Considering CST cost data for 2023 and a reference (wholesale) fuel price of A$12/GJ23 the breakeven replacement potential for CST for remote power amounts to around 105 PJ/a of primary fuel use equivalent to an approximate average load of 1.1 GWe - split to around 20 remote sites. Noting that the most cost effective approach is a hybrid system with around 50% of the energy provided by CST, the
1.1GWe of high capacity factor conventional generation capacity would be replaced by CST plus similar capacities of PV and /or wind, each operating at lower capacity factors.

Moving to 2030, the anticipated decrease in CST CAPEX along with a general increase in competing fossil fuel prices from an indicative cost of CO2e of A$40/t, increases the number of sites with a profitability score greater than 1. In 2030 market potential would be 212 PJ/a equivalent to an approximate average load of around 2.2 GWe.

23
Adapted to the annual demand figures - see Section 7.5

S774Doc-676563417-363 / v0.2 91 CST Value Proposition
7 Industrial Process Heat

7.1 Key Findings
The list below provides the key findings made during the techno economic analysis of the industrial process heat end use case and provides an overview of the value proposition CST can offer within this sector in Australia.

▪ The process heat market comprises a wide range of industries and end-use applications, requiring
process heat at different temperatures, in different forms (hot water, steam or hot air) and at different
times.
▪ Considering today’s cost level and a good DNI (>2,500 kWh/m²/a), CST becomes competitive for fuel
prices above A$ 60 /MWh (A$ 16.7 /GJ) for solar shares of up to 70 - 75%.
▪ For CST cost projected for 2030, the combined LCoH (solar and back-up fuel) will reduce by 10% -
20% depending on the solar share. The breakeven fuel price consequently reduces to around A$ 45
/MWh (A$ 12.5 /GJ) by 2030.
▪ The location, i.e., solar resource (DNI level) but also the latitude, which impacts the solar field
efficiency, has a strong impact on the LCoH. Within the considered DNI range (2,000 - 2,900
kWh/m²/a), the LCoH is nearly proportional to the DNI. This noted, locations with higher DNI are
typically more remote and, thus, the regional cost factors are higher which increases the LCoH to a
small degree.
▪ CST offers the advantage of providing thermal energy directly and efficiently. As such CST requires
significant less land area when compared to PV for power-to-heat (PtH) in combination with eTES.
▪ For the investigated sites with a good solar resource, the CST option results in a 20 - 30%. lower LCoH
when compared to the PV-eTES option. For the site with a comparatively low solar resource (2,000
kWh/m²/a), the PV-eTES option results in similar LCoH in case of a solar share of 50%. For decreasing
DNI levels PV-eTES delivers a progressively lower LCoH.
▪ The optimal process temperature range for CST is between 150°C to 500°C. However, different types
of CST technologies can also cover higher process temperatures of up to >1,000°C.

S774Doc-676563417-363 / v0.2 92 CST Value Proposition
7.2 Introduction
The process heat market (see Section 7.6) comprises a wide range of industries and end-use applications, requiring process heat at different temperatures and types such as hot water, steam, or hot air. Required capacities range from single digit MW systems to GW scale systems. Thus, to assess and quantify the value CST may offer for process heat systems, different analyses have been conducted for different temperature ranges, capacities, and sites within Australia, with different cost and different solar resource levels. The assessment has been conducted using two generic CST systems: a parabolic through and a solar tower based system to provide process heat.

The parabolic trough process heat system has been modelled for a mid-temperature medium size end- use case of around 240°C and a constant heat demand of 10 MWt, which corresponds to an annual demand of 87.6 GWht or 315 TJ. To show the impact on the location within Australia three different regions have been investigated as summarised in Table 7-1.

Table 7-1: Reference sites for process heat assessment

Item SA low QLD medium WA high

Region South Australia with Queensland with WA with “very good”
“low” / moderate “good” DNI DNI
DNI e.g., Adelaide
region

Reference location Adelaide Hughenden Newman

DNI [kWh/m²/a] 2,000 2,600 2,900

GHI [kWh/m²/a] 1,850 2,250 2,300

For each of the sites different solar field sizes and different thermal energy storage sizes were considered to enable assessment of different solar shares and capacity factors. Further, a comparative assessment has been conducted considering two alternative technology options. The first is a renewable heat solution using solar PV in combination with an electrically heated thermal energy storage (eTES). The second option is a conventional gas fired boiler. Technology costing was based on the simple cost model developed, considering both the economies of scale for each of the sub-systems (energy collection and storage) and regional cost adjustments.

The solar tower based CST system has been used to show the impact of different process heat supply temperatures levels and the economies of scale of CST systems required for large process heat demands.

7.3 Mid Temperature Medium Size End-Use Case
Figure 7-1 shows the LCoH results for three reference sites and different solar field sizes, for the parabolic trough. For each solar field size (i.e., SM2, SM3, SM4) the configurations with the lowest LCoH have been determined, considering different storage sizes and hence capacity factors.

As can be seen, the capacity factors/solar shares of the systems (i.e. the proportion of renewable heat available in average over a year considering a constant demand profile), increase substantially with increasing solar field sizes, but with only a relatively small increase in LCoH. In essence, this indicates that

S774Doc-676563417-363 / v0.2 93 CST Value Proposition
a larger solar field has minimal impact on the cost of heat, but substantially increases the amount of heat available to a facility. The impact of solar resource (i.e. DNI) is also quite evident. The higher the DNI the lower the LCoH and the higher the capacity factor. In essence, CST delivers more daily heat at a lower
LCoH the higher the DNI. This noted, locations with higher DNI are typically more remote and, thus, the higher regional cost factors will have a negative impact on LCoH. Any such impact on LCoH is expected to be relatively minor. Further, also the latitude has an impact on the LCoH (in particular for line- focussing systems), as the lower the latitude the higher the solar field efficiency.

Figure 7-1: LCoH and solar shares for three investigated sites and different solar field sizes

The value CST can offer to an industrial process heat user will depend on the cost of fuel CST can substitute. Figure 7-2 below provides the results for the investigated sites, considering a wide range of
CST configurations (solar field size and the TES size) as well as different fuel costs to cover the remaining heat supply to allow for continuous process heat operations. As can be seen, the cost of the fuel being replaced by CST has an impact on the combined LCoH. Where the fuel cost is lower than the specific
LCoH for CST at the respective site, the LCoH increases as the proportion of CST increases. Where the fuel cost is higher than the specific LCoH for CST, the combined LCoH decreases as the proportion of CST increases.

The figure also indicates that where the cost of the fuel being replaced is high, CST can deliver an economically viable reduction in LCoH up to a capacity factor of approximately 75 - 80% for QLD Med and WA High or 60% for SA Low. To achieve a higher capacity factor, an end user would need to pay relatively more for a larger solar field for increasingly lower solar heat gains. This is clearly demonstrated by the sharp LCoH cost increase at capacity factors above 75% for QLD Med and WA High.

S774Doc-676563417-363 / v0.2 94 CST Value Proposition
Figure 7-2: Combined LCoH for all sites for different solar shares and fuel cost (2023 cost)

The future cost reduction potential of CST as presented in Section 4.2 significantly increases the value that CST can provide to process heat users over time. As can be seen comparing Figure 7-3 to Figure 7-2,, the combined LCoH is expected to reduce in the order of 20% over the next few years in case of high solar shares. This reduction will lower the fuel cost where CST becomes economically viable in

S774Doc-676563417-363 / v0.2 95 CST Value Proposition
comparison with current fuel costs. For example, the break-even fuel price reduces from around
A$ 60 /MWh (A$ 16.7 /GJ) in 2023 to around A$ 45 /MWh (A$ 12.5 /GJ) by 2030 for QLD Med. An increasing price on CO2 emissions will further improve the viability of CST in comparison to conventional fuels. Each A$ 20 /tCO2 will add approximately A$ 1 /GJ to the cost of gas.

Figure 7-3: Combined LCoH for all sites for different solar shares and fuel cost (2030 cost)

S774Doc-676563417-363 / v0.2 96 CST Value Proposition
When comparing CST and fossil fuel price scenarios, process heat users should also consider the economies of scale of CST and the fact that the cost of conventional fuels (e.g. gas) will differ based on the amount used. If CST reduces the amount of gas used, the cost of any smaller amounts of consumed gas might be higher per GJ. The related impact is explained further below and related cost and NG price functions considered in the assessment.

7.4 Comparative Assessment with PV-eTES
A comparative assessment of CST-TES and PV-eTES, (i.e. using solar PV to heat up a thermal energy storage system to generate process heat on demand), has been conducted for all three sites using a reference gas price of A$ 12.5 /GJ (A$ 45 /MWh). Figure 7-4 below summarises the results.

The comparison shows that for locations with higher DNI (QLD Med and WA High), CST offers a lower
LCoH compared to PV-eTES. In case of QLD Med the LCoH for PV-eTES are around 20% higher within the considered range (50 - 75% solar share). In case of WA High the delta increases to around 30%. The main reason is the comparatively higher increase of DNI (relevant solar irradiation for CST) when compared to the small GHI increase (relevant irradiation for PV).

The differences in LCoH for CST between QLD Med and WA High are only marginal, despite the higher
DNI resource in WA High. This is due to the regional cost factors that increase the investment costs in
WA compared to QLD. In the calculated cases this increase in investment is equalised by the increased solar heat from the higher DNI. In contrast, the PV-eTES LCoH is increasing for WA High compared to
QLD Med, due to the smaller increase in GHI not compensating the higher cost.

For SA Low with a comparatively low solar resource (2,000 kWh/m²/a), the PV-eTES option results in similar LCoH in case of a solar share of 50%. For increasing solar shares PV-eTES delivers a progressively lower LCoH in case of the SA low site. The main reason is the lower seasonal variation of the PV generation when compared to the CST generation profile. Given the lower solar resource, the range of viable solar shares / capacity factors is generally reduced.

It should also be noted that no land costs have been considered for the assessment and typical collector and PV spacings have been applied. This is relevant as CST requires significant less land area when compared to PV for power-to-heat (PtH) in combination with an electrical thermal energy storage (eTES).
Depending on the boundary conditions the difference can be well above 50%.

S774Doc-676563417-363 / v0.2 97 CST Value Proposition
Figure 7-4: LCoH for all sites for CST-TES and PV-ETES (solar only LCoH)

S774Doc-676563417-363 / v0.2 98 CST Value Proposition
7.5 Impact of Project Size and Operating Temperature
Using a generic solar tower based CST system (able to cover a wide range of process temperatures), the impact of different required heat supply temperatures (hence CST operating temperatures), different DNI levels and project sizes has been assessed. The assessment considered a wide range of solar field and storage sizes to meet the respective heat demands always with the lowest LCoHs. Cost data is based on the CST cost model presented in Section 4.1.

Figure 7-5 below shows the results for a process heat supply at 200°C for different project sizes
(consumption levels) and three different DNI levels. It should be noted that the graph has a logarithmic scale in the x axis. The model for cost of gas versus consumption level that has been adopted shows a similar dependence on consumption level to the LCoH of the CST system - based on the applied economies of scale factors in the CST cost model. Still, CST becomes increasingly competitive with increasing consumption level.

Considering a medium size project (of around 100 GWht/a), the depicted gas price range varies between
A$ 32 /MWh and A$ 58 /MWh24 (A$ 9 /GJ and A$ 16 /GJ). Thus, for good solar resource sites and gas prices at the upper range, already today the breakeven point can be reached25. In case of projects with an annual consumption above 1,000 GWht/a, CST is, even for sites with only a moderate solar resource, worthwhile to be considered.

In case of more expensive fuel oil, CST offers an advantage already today for small projects and sites with only a moderate DNI level.

Figure 7-5: LCoH of CST for various DNI levels and process heat consumption (200°C level)

24
Boiler efficiency 80%
25
Compared to the previous comparative assessment, based on parabolic trough technology, the LCoHs of the tower based
system differ slightly, given the slightly boundary conditions (DNI level, operating temperature) and the difference in CST
technologies.

S774Doc-676563417-363 / v0.2 99 CST Value Proposition
Figure 7-6 shows the impact of the operating temperature for a site with a moderate DNI level of
2,000 kWh/m²/a (comparable with the previous SA Low reference). Still in case of fuel oil as the current heat source, CST offers an advantage today even at sites with only a moderate DNI level. In case of natural gas, breakeven can be reached today for larger systems with an annual consumption above
1,000 GWht/a and medium to high operating temperatures. At very high operating temperature, thermal losses are too high and CST is not competitive.

Figure 7-6: LCoH of CST for various temperature and process heat consumption (DNI 2,000 kWh/m2/a)

7.6 Market Size
The National Pollution Inventory (NPI) has exact locations of industry facilities, along with their industry sector. Whilst actual levels of energy use are not provided, the levels of combustion-related pollutants, together with the overall fuel energy use that can be identified for a sector, have been used to deduce an indicative distribution as shown in Figure 7-7 (adapted from ITP, 2019). Also depicted is the DNI level, indicating a good solar resource level for many sites.

Clearly alumina and other nonferrous, iron and steel, petroleum refining, LNG plants, ammonia plants, sugar mills and cement and lime are significant energy consumers. Alumina, ammonia and LNG plants use gas. Iron and steel use coal as the main fuel. Sugar mills use bagasse waste as fuel. The other significant beige dots are pulp and paper industry sites which use mainly wood waste-based fuels for a lot of their energy. Energy use for mining is mainly for power generation as discussed in Section 6.

S774Doc-676563417-363 / v0.2 100 CST Value Proposition
Figure 7-7: Locations and intensity of energy use and DNI level in Australia 2019-20

Figure 7-8 below provides an overview of the total annual heat use in PJ for the different industry sectors, split into three different temperature ranges.

The total annual heat use split into the three temperature range amounts to 69 PJ/a for <150°C, 365 PJ/a for 150 - 500°C and 292 PJ/a for >500°C. The total market size amounts to around A$ 10 billion considering a gas price of A$ 12 /GJ.

It is apparent that some sectors are very large users of process heat. The top three sectors in the 150°C to
500°C range are ‘Alumina and other non-ferrous minerals’, ‘Food and beverage’, and ‘Oil and gas extraction’. However, ‘Food and beverage’ drops to number six if the heat provided by the combustion of fossil fuel only is considered, as there is a large use of sugar cane waste (bagasse) at sugar refineries. The
‘Pulp and paper’ and ‘Wood products’ sectors also use significant amounts of biomass. Australia’s six
Alumina refineries are responsible for a large share of process heat, approximately half of their use falls at below 200°C, for digestion processes.

S774Doc-676563417-363 / v0.2 101 CST Value Proposition
Figure 7-8: Industrial heat use in Australia 2016-17, by sector and temperature of use (ITP, 2019)26

While there are some industry sectors that have only a few large heat consuming sites, there are others, such as the ‘Food and beverage’ sector, with hundreds of heat consumers, each having a comparatively small heat demand of less than 100 TJ per year. An overview for the main sectors and their size distribution is provided in Table 7-2 below.

26
Most sites within the NEM and SWIS are assumed to import electricity from grid. There are sites that use gas turbines to
generate power and heat for operations within the NEM and SWIS which are hard to predict. The energy use provided has
to be considered as approximate.

S774Doc-676563417-363 / v0.2 102 CST Value Proposition
Table 7-2: Size distribution of sites using heat in the range < 500°C

Sites in range [PJ/year] < 0.1 0.1 to 0.5 to 1 1 to 5 5 to 45 Total Total
0.5 number heat use
Avg power at 80% CUF27 (<5) (5 to (21 to (40 to (200 to of sites [PJ/year]
[MWt] 20) 40) 200) 1,800)

Alumina and other non- 17 5 3 4 7 36 85.5
ferrous

Food and beverage 244 52 1 16 7 320 115.2

Oil and gas extraction 24 34 13 24 3 98 86.4

Petroleum refining 3 2 0 2 3 10 55.5

Ammonia and other 68 7 3 5 0 83 2.1
chemicals

Cement, lime products 30 8 0 0 0 38 5.0

Pulp and paper 21 15 2 5 1 44 20.6

Commercial and services 105 3 0 0 0 108 18.0

Wood and wood products 22 12 4 6 0 44 14.0

Other mining 335 34 5 0 0 374 7.3

Other sectors 313 0 0 0 0 269 24.5

Total 1,182 172 31 62 21 1,468 434

The 1,500 sites have been approximately assessed for potential competitiveness of CST taking into account the DNI level, the indicated size of the CST system, the existing fuel cost and the impact of cyclone region. As was done for the remote area power generation sites, a ‘’profitability score’’ metric equal to the estimated cost of current fuel divided by a high-level estimate LCoH of CST was used, to see if installing a CST system would deliver a cost saving relative to an existing fossil fuelled system.

Considering CST cost data for 2023 and a reference (wholesale) fuel price of A$12/GJ 28 the replacement potential for CST for remote power amounts to around 60 PJ/a of primary fuel use equivalent to an approximate average load of 3.2 GWt based on a CUF of 60%.

Moving to 2030, the anticipated decrease in CST CAPEX along with a general increase in competing fossil fuel prices from an indicative cost of CO2e of A$ 40/t, increases the market potential around 220 PJ/a equivalent to an average load of nearly 12 GWt.

27
Capacity utilisation factor (CUF)
28
Adapted to the annual demand figures - see Section 7.5

S774Doc-676563417-363 / v0.2 103 CST Value Proposition
8 Green Fuels

8.1 Key Findings
The list below provides the key findings made during the techno economic analysis of the green fuel end use case and provides an overview of the value proposition CST can offer within this sector in Australia.

▪ Green fuels production benefits from CST as a source of renewable power and heat. Overall, a
reduction of green fuel production cost of approximately 8% to 40% was achieved by adding CST to
the energy supply mix and combining CST with other renewable sources.
▪ The ability to deliver a combined heat and power (CHP) solution within the one technology has major
benefits and is one of the key value propositions of CST. Specifically, CST can provide an optimal
balance of renewable heat and power, as required for the production of different renewable fuels.
▪ Hydrogen production systems based on SOEC technology explicitly profit from CST as a result of the
low-cost heat supply and increased hydrogen production yields leading to a cost reduction of
approximately 10% (compared to the PEM case without CST). As hydrogen is key feedstock for
methanol and ammonia production, these fuels also profited from cheaper hydrogen.
▪ Firm capacity of CST is enabled by the TES, which provides flexibility and enables the dispatchability of
the plant. The dispatchability allows for the power and heat demand to be met during times of low
sunlight and or wind resource. This results in a lower requirement of installed capacity (reduction of
approximately 15% to 30%) of solar PV and wind and less curtailment of energy overall.
▪ The flexibility introduced with the TES of the CST plant is especially beneficial for less flexible
downstream processes to produce hydrogen derivatives, such as methanol or ammonia, by reducing
the need for other storage systems, such as BESS, that are often significantly more expensive.
▪ Overall, the production cost reduction of green fuels heavily depends on the green fuel itself and the
chosen boundary conditions. Generally, systems with a high heat demand and low operational
flexibility profit most from CST.
▪ For systems with a high heat demand and low operational flexibility CST should be considered as part
of the energy mix as they show the largest production cost reduction compared to systems without
CST.

S774Doc-676563417-363 / v0.2 104 CST Value Proposition
8.3 Introduction
This section analyses and quantifies the benefits of CST, as a technology that can provide both power and heat, for the production of green fuels, particularly in combination with other renewable energy sources, such as solar PV and wind power.

Green fuels (also called electrofuels, e-fuels or sustainable fuels) are carbon-neutral or carbon-free alternatives to fossil fuels and can be produced from hydrogen made with renewable electricity, via biosynthesis or via thermochemical processing of water or other feedstocks. Hydrogen itself can be considered as a green fuel and as a core input fuel for other green fuels, such as ammonia and methanol.
Green fuels will play an important role in decarbonising hard-to-abate sectors, such as aviation, shipping, and long-distance transport in the future. The chemical synthesis of green fuels typically requires large amounts of energy and or heat, via high temperature chemical reactions 29.

This study focuses on the green fuels; hydrogen, ammonia, and methanol, which are analysed in the following sections.

8.4 Energy Requirements and End Use Cases

8.4.1 Overview
Green fuels are used in a variety of ways and sectors, with some able to be used as direct replacements to fossil fuels in internal combustion engines (ICE) in cars, trucks, ships, and planes (e.g., methanol, and sustainable aviation fuels), while others require engine modifications. Hydrogen and in some cases ammonia and methanol can also be used in fuel cells (for electricity production), which have the additional benefit of higher conversion efficiencies compared to ICEs.

There are many chemical compounds and mixtures that can be used as green fuels and can be produced in large quantities from many different chemical processes beginning with hydrogen and other feedstocks. Table 8-1 outlines a summary of key properties of the green fuels investigated in this study.

Table 8-1: Key properties for the green fuels considered in the study at ambient conditions

Parameter Unit Hydrogen Ammonia Methanol

Chemical formula - H2 NH3 CH3OH

Physical state (STP) - Gas Gas Liquid

Density (STP) kg/m3 0.0899 0.760 786

Density (liquid) kg/m3 70.8 698 786

Higher heating value kWh/kg 39.4 6.30 6.40

Lower heating value kWh/kg 33.3 5.17 5.53

29
Some chemical processes such as the Haber-Bosch-Process for ammonia production are exothermic and do not require a
constant external heat supply.

S774Doc-676563417-363 / v0.2 105 CST Value Proposition
Parameter Unit Hydrogen Ammonia Methanol

Volumetric energy density30 kWh/m3 2,360 3,530 4,300

Boiling point °C -253 -33.4 64.8

8.4.2 Hydrogen
Today, hydrogen is largely produced from natural gas through the steam reforming process. Hydrogen made from natural gas and without any form of carbon capture is considered “grey hydrogen”. This study, however, is focused on the generation of “green hydrogen” through water electrolysis powered from renewable energies.

End-Uses of Hydrogen
Hydrogen has a wide range of potential end-uses across several key sectors, which can be summarised as follows:

▪ Industrial processes: hydrogen is used in a variety of industrial processes, such as oil refining and
chemical manufacturing as a feedstock to produce base chemicals, such as ammonia or methanol.
More than 350 kilo-tonnes of hydrogen (equivalent to approximately three quarters of Australia’s total
hydrogen production) is used each year in Australia for ammonia production alone (COAG Energy
Council, 2019).
▪ Mobility: hydrogen can be directly used as a fuel for fuel cell electric vehicles (FCEVs). Furthermore,
hydrogen is used as a feedstock for the production of green fuels, such as methanol, ammonia, or
sustainable aviation fuel (SAF).
▪ Power generation: hydrogen can be used to generate electricity in fuel cells or gas turbines. This can
help decarbonise the power sector and provide a reliable source of backup power.
▪ Heat generation: hydrogen can be used to provide high temperature process heat or to heat
buildings.
▪ Storage of renewable energy: hydrogen can be used to store renewable energy, which will aid in the
integration of intermittent sources of energy with the existing electrical grid.
▪ Other applications: hydrogen has other potential applications, such as in rocket fuel, steel
manufacturing, metal cutting, and food processing.

Focus Sustainable Aviation Fuel: According to The International Air Transport Association (IATA), SAF could contribute approximately 65% of the reduction in emissions required by the aviation industry to reach net-zero by 2050. Consequently, there is a strong push towards the utilisation of SAF in the sector.
The European Union for example has set a target of blending 2% SAF in all jet fuel by 2025 and 5% by
2030. Similarly, the United States has set a target of blending 10% SAF in all jet fuel by 2050. Countries such as Japan, China and South Korea are investing in SAF production and blending. Today, SAF is primarily produced from hydro-processed esters and fatty acids (HEFA) using vegetable oils, animal fats, or greases, however, theses feedstocks are limited and not expected to be sufficient for the anticipated uptake of the fuel in the future. As an alternative pathway, Fischer-Tropsch synthesis can be used to

30
liquid and LHV basis

S774Doc-676563417-363 / v0.2 106 CST Value Proposition
produce SAF using synthesis gas (a mixture of carbon monoxide and hydrogen), which could potentially generate high demand for hydrogen for the aviation sector.

Role of Hydrogen
In recent years, hydrogen has been the focus of several industries for the decarbonisation of the energy supply chain. Hydrogen as an energy carrier exhibits important advantages and disadvantages that must be considered as the technology rapidly develops across the world.

Hydrogen is a versatile substance that can be used in a variety of applications, including transportation, power generation, and industrial processes. Hydrogen as a fuel source is cleaner than any other fuel type
(if combusted or used in a fuel cell), only producing water vapour as a byproduct and with no carbon emissions. Additionally, hydrogen has a very high gravimetric energy density, meaning it can store massive amounts of energy per unit mass, which is important for applications where weight is limited, such as in transportation.

Key challenges associated with the use of green hydrogen, however, include the high specific levelised cost of product today, as well as the difficulty and expense of its storage and transportation. The transport of gaseous hydrogen typically involves the compression to 200 - 700 bar, due to its extremely low volumetric energy density at atmospheric pressure. This compression is energy intensive and requires expensive equipment. The option for liquification also comes with several challenges, such as the extremely low boiling point of hydrogen of -253ºC and consequently the need for energy intensive and expensive cryogenic equipment. Additionally, there is a lack of infrastructure in place for the use of hydrogen as an energy carrier today. Furthermore, hydrogen presents other safety and operational issues, due to its high flammability, potential to result in embrittlement of some materials, tendency to leak (due the small size of the molecule) and ability to boil off easily.

Despite these issues, however, hydrogen is expected to play a significant role in decarbonising the global economy, especially in hard-to-abate sectors with no or little potential for electrification.

Production of Hydrogen
Hydrogen can be produced in many different ways with common methods including:

▪ steam methane reforming (grey hydrogen), which is inexpensive but environmentally costly and is the
typical method of hydrogen production today;
▪ water electrolysis from grid power (yellow hydrogen),
▪ CST driven multi-step solar thermochemical water splitting, and;
▪ water electrolysis from renewable energy (green hydrogen).

Electrolysis involves the decomposition of water using a large amount of electricity to form its constituent gases (hydrogen and oxygen), which can both then be utilised for downstream value chains. Electrolysis is currently achievable via two mainstream technologies, proton exchange membrane (PEM) electrolysis and alkaline water electrolysis (ALK). Additionally, high temperature electrolysis using solid oxide electrolyser cells (SOEC) is a third technology that has the potential of becoming mainstream in the near future. SOEC has recently moved from the laboratory to the commercialisation stage, with companies such as Bloom Energy and Sunfire offering SOEC systems today. Each of these hydrogen generation technologies are discussed further below, with KPIs described in detail in Appendix D1.2.

S774Doc-676563417-363 / v0.2 107 CST Value Proposition
▪ Alkaline water electrolysis: ALK electrolysers are comprised of metal electrodes surrounded by an
alkaline liquid medium and separated by a permeable membrane. ALK has been used for more than
120 years for industrial hydrogen production, with a technology readiness level (TRL) of 9 (IEA, 2023)
and various suppliers that can provide modular systems in the megawatt scale.
Although ALK electrolysers have a lower specific CAPEX than PEM and SOEC technologies, it is not the
preferred technology for the boundaries defined in this study (off-grid green fuel production from
renewable energies). This is due to the low operational flexibility offered by the technology, which is
an important argument for the underlying use case.
▪ Proton exchange membrane electrolysis: in contrast to ALK, PEM operates in an acidic
environment, making it necessary to use precious metals for the electrodes to prevent corrosion.
Development of PEM electrolysis started in the 1960s, where after 20 years of development for
military and aerospace industries, R&D activities led to first commercial application in the beginning
of the 21st Century. PEM is considered to have a TRL of 9 (IEA, 2023) with various suppliers capable of
providing modular systems in the megawatt scale.
A PEM electrolyser was selected as a base configuration in this study for the following key reasons.
The technology has particularly fast response times and high ramp rates (load gradients), allowing for
high dynamic operation. This is especially relevant with fluctuating, intermittent electricity supply from
renewable sources. PEM further demonstrate a rapid start-up time and low turn down ratio (minimum
load), making the technology well-suited for applications where the demand for hydrogen is variable.
PEM electrolysers are commercially available for large-scale applications and are a proven technology
with a high durability, low degradation, and low maintenance requirement relative to ALK.
▪ Solid oxide electrolysis: SOEC operates at high temperatures of approximately 700ºC and is unlike
the previous technologies in that it utilises both electrical and thermal energy to achieve the
decomposition of water. First developments of SOEC took place in the late 1960s, with new interest
and increased R&D activities in the last decade leading to the first commercially available systems
from suppliers today. SOEC is considered to have a TRL of 8 (IEA, 2023) with only a limited number of
suppliers capable of providing systems in the low megawatt scale. Today, a 4 MW unit from Bloom
Energy is the world’s largest solid oxide electrolyser installation (IEA, 2023). It produces 20% to 25%
more hydrogen per unit of energy input compared to PEM or ALK.
SOEC electrolysers show important advantages over the other technologies, especially when
combined with heat generation technologies, such as CST. A key argument for the use of SOEC
electrolysers is the comparably low specific energy consumption. . While PEM and ALK electrolysers
operate on electricity alone, SOEC partially substitutes the input electricity requirement with high
temperature heat, which can be provided by a CST plant, thereby producing hydrogen in a very
energy efficient way. For this reason, hydrogen production with an SOEC electrolyser in combination
with CST was included in this study.

As a result of the potential that PEM and SOEC offer towards the goal of this project, these were the selected electrolyser technologies considered further in the study.

S774Doc-676563417-363 / v0.2 108 CST Value Proposition
Figure 8-1 shows a high-level overview of the mass and energy requirements to produce green hydrogen using PEM electrolysis.

PEM Electrolysis

Electrical Energy Hydrogen Gas
54,500 kWh 1t

Fresh Water Oxygen Gas
12 t 8t

Figure 8-1: PEM-electrolyser - a high-level overview of the mass and energy requirements for green hydrogen
production

The diagram shows the energy flows of a typical PEM electrolyser with an efficiency of 61.2% and is normalised to one tonne of hydrogen output. Although the splitting of water is an endothermic process, a significant amount of waste heat is generated, which must be removed to maintain the unit’s operating temperature. An overview of the operating conditions of PEM electrolysis is provided in Table 8-2.

Figure 8-2 shows a high-level overview of the mass and energy requirements to produce green hydrogen using SOEC electrolysis.

SOEC Electrolysis
Electrical Energy
36,800 kWh Hydrogen Gas
1t

Thermal Energy
5,900 kWh Oxygen Gas
8t

Fresh Water
12 t

Figure 8-2: SOEC-electrolyser - a high-level overview of the mass and energy requirements for green hydrogen
production

S774Doc-676563417-363 / v0.2 109 CST Value Proposition
Table 8-2: An overview of the operating conditions for the selected electrolyser technologies

Parameter Unit PEM SOEC

Specific energy consumption kWhel/kgH2 54.5 36.8
kWhth/kgH2 - 5.9

Energy efficiency31 % 61.2 78.0

Operating temperature °C 80 500 - 900

8.4.3 Methanol
Methanol (CH3OH) is a liquid alcohol that is produced at large-scale and is used in many products in the chemical industry around the world. Given that methanol is one of the chemicals traded in the largest volumes globally, producing methanol with net-zero emissions (green methanol) at a low cost would offer significant business opportunities and global greenhouse gas emissions reduction. It has therefore become of significant interest to several industries today.

End-Uses of Methanol
There are multiple end-use applications for methanol, which include the following:

▪ Methanol is used to produce a variety of petrochemicals, such as acetic acid, formaldehyde, and
methyl tert-butyl ether (MTBE).
▪ Methanol is an important feedstock for a variety of chemical processes, such as the production of
plastics and polymers and in the refining of oil to remove impurities.
▪ Methanol is also used to produce pharmaceuticals and explosives.
▪ Methanol can be used in the transport sector as a combustible fuel. Unlike hydrogen and ammonia,
methanol is similar to traditional fossil fuels, in that the substance can be mixed with petrol for direct
use in current ICEs. Further, it can be used in its pure form in modified engines or be further
processed to SAF.

Focus Sustainable Maritime Fuel and SAF: In the maritime shipping industry, 61% of all newbuild ordered ships by tonnage were alternative fuel capable in 2022 (Gordon, 2023). Of this, approximately
11% were ammonia-ready vessels and 7 % were methanol vessels. A.P.Moller-Maersk, an international integrated logistics company, recently announced 19 methanol dual-fuel containerships on order as of
October 2022, with delivery of the ships planned between 2023 and 2025. This announcement was followed by other large shipping companies ordering methanol ships, such as CMA
CGM, Cosco and Cargill. It can be concluded that the maritime mobility sector is an important market for green fuels like methanol with significant growth potential. Besides the use in maritime transport, methanol can also be blended with conventional jet fuel to create a more sustainable fuel blend. Further, several companies including BASF, Thyssenkrupp and Exxon are developing methanol to SAF processes.
The enormous growth potential of SAF is elaborated on in Section 8.4.2.

31
For SOEC: electrical and thermal energy.

S774Doc-676563417-363 / v0.2 110 CST Value Proposition
Role of Methanol
Among the selected green fuels in this study, methanol has the highest volumetric energy density and is a liquid at ambient conditions (see Table 8-1), which is an attractive trait for a transport fuel. This allows methanol to be used as a (long term) energy storage and a medium to transport/export energy as another end-use application. Methanol will help overcome typical drawbacks of transporting or storing hydrogen in the future, which will face issues such as embrittlement of metal and boil off if otherwise transported in its natural form.

Production of Methanol
Steam reforming of natural gas is the most common process for producing methanol today. An alternative sustainable method for the production of methanol is via catalytic synthesis with the feed hydrogen provided using electrolysis and carbon dioxide from direct air capture (DAC), both using renewable electricity and heat.

Figure 8-3 outlines the mass and energy requirements for the synthesis of renewable methanol,, including DAC for the supply of carbon dioxide and normalised to one tonne of methanol output. The electrical energy demand of the proposed system excluding electrolysis is 1.26 MWh per tonne of methanol, which when combined with electrolysis totals to 12.2 MWh per tonne.

Methanol Reactor
Electrical Energy
Thermal Energy 236 kWh
1,850 kWh Methanol
Hydrogen Gas
1t
0.2 t

Electrical Energy Direct Air Capture Plant Carbon Dioxide
1,022 kWh 1.46 t

Thermal Energy
3,212 kWh
Dry Air
2,330 t

Figure 8-3: A high-level overview of the mass and energy requirements for the production of renewable methanol

8.4.4 Ammonia
Ammonia (NH3) is a gas at ambient conditions, although it is most commonly stored and transported as a liquid at modest pressure (or reduced temperature) and is along with methanol, amongst the chemicals produced and traded in the largest volumes in the world. Today, the majority of ammonia produced is used for the production of fertilisers, explosives, and other chemical compounds (USGS, 2020). Due to its high hydrogen content, ammonia has the potential to serve as an excellent medium for the transport of hydrogen, similar to methanol. Furthermore, with ammonia production accounting for a large amount of carbon emissions as a result of the grey hydrogen feedstock, there is significant potential for emissions reductions in the sector made possible by the utilisation of green hydrogen.

S774Doc-676563417-363 / v0.2 111 CST Value Proposition
End-Uses of Ammonia
Ammonia today is used all around the world for important applications, which include the following:

▪ Ammonia is produced today primarily for fertiliser production. Ammonia is a critical chemical for crop
production world-wide and is therefore a widely traded commodity.
▪ Ammonia is also widely used today for refrigeration, with key examples including cold storage
facilities or vehicles, supermarkets and in air conditioners.
▪ Ammonia is also used for the manufacture of explosives, which is another critical component in the
mining sector.
▪ Ammonia is used for the chemical manufacturing of other secondary products, for industrial chemical
processing, and for cleaning. Other secondary products include plastics, textiles, pharmaceuticals,
rubbers, and leathers.
▪ Ammonia can be used in the transport sector as a directly combustible fuel (uncommon), or through
fuel cells to produce electricity.

Ammonia as a Hydrogen Carrier
Green hydrogen that is used as a feedstock for green ammonia can be decomposed at the other end of the supply chain, to reclaim the input hydrogen, introducing an alternative transport opportunity for green hydrogen. Using this method, green hydrogen can be safely stored on ships for transport in large quantities, whilst avoiding some of the critical disadvantages of hydrogen discussed in Section 8.4.2. In particular, with the existing ammonia trade, being so large today, utilising the substance as a hydrogen carrier will allow hydrogen to take advantage of existing infrastructure, which is a significant advantage compared to transporting hydrogen in its pure form. Such an opportunity offers the potential to minimise the levelised cost of hydrogen delivered, which is significantly influenced by transportation costs today. As a result of this economic potential, the interest in conversion or upgrading of hydrogen to ammonia has gained traction in the previous few years.

Role of Ammonia
The use of ammonia as a green fuel is an attractive option, due to the substance’s favourable physical properties, such as its relatively high boiling point and volumetric energy density (compared to hydrogen), which are positive traits for transport fuels. Furthermore, ammonia can be created as a direct product for several key industries around the world or used directly as a hydrogen carrier for energy trading. Conversely, there are challenges associated with the handling of ammonia, as the substance is a highly toxic gas, and exposure to even small amounts can cause respiratory problems, burns, and death.
Ammonia is flammable, and its vapours can form explosive mixtures with air. Furthermore, ammonia can corrode metals and plastics, and therefore special materials must be used to store and transport the substance.

Production of Ammonia
Green ammonia can be formed from the combination of nitrogen and green hydrogen via the Haber-
Bosch process using a renewable power supply. The technologies required for the production of green ammonia include the electrolysis of water to produce green hydrogen, and an air separation unit (ASU) to provide the feedstock nitrogen. A high-level mass and energy flow diagram is shown in Figure 8-4 to demonstrate the requirements of ammonia production from the proposed process.

S774Doc-676563417-363 / v0.2 112 CST Value Proposition
Haber Bosch Plant

Electrical Energy
510 kWh
Hydrogen Gas Ammonia
0.179 t 1t

Air Separation Unit
Nitrogen Gas
Electrical Energy 0.827 t
90 kWh

Air
1.095 t

Figure 8-4: A high-level overview of the mass and energy requirements for the production of green ammonia

The values in the diagram are normalised to one tonne of ammonia and include the ASU to supply the nitrogen gas. It is important to note that the waste heat output from the ammonia plant was assumed to be 50% of the total heat produced from the chemical reaction of the ammonia synthesis, with the other
50% directly utilised in the reactor to maintain the high operating temperature of 440°C. Furthermore, the electrical energy required to supply the Haber Bosch plant as well as the ASU is 0.6 MWh per tonne of ammonia. Combining this with the electrical energy demand to produce the hydrogen feedstock that can be derived from Figure 8-4 gives a total energy consumption of 10.36 MWh per tonne of ammonia.32

8.5 Techno-economic Optimisation of Green Fuel Plants
Using the information and data provided in Sections 8.1 and 8.2, the subsequent step in the evaluation of the CST value proposition was the techno-economic optimisation of different hydrogen systems coupled with and without CST. This was achieved using Fichtner’s in-house optimisation tool, the H2-
Optimizer, which can conduct techno-economic evaluations of entire energy system value chains.
Figure 8-5 gives an overview of the capabilities of the tool and highlights the components that were utilised in this study.

32
Based on the hydrogen production with a PEM electrolyser which was selected as base case.

S774Doc-676563417-363 / v0.2 113 CST Value Proposition
Figure 8-5: Energy system model of the Fichtner H2-Optimizer

Using the mass and energy flow diagrams and the technical data tables presented in Appendix D1.1 as inputs to the model, the energy system was optimised to identify and quantify the system with the lowest production cost. The optimisation and dispatch included the renewable energy generation technologies, the hydrogen production system (electrolyser and hydrogen storage 33), as well as the downstream green fuels production facilities and all of their associated components.

8.5.1 Methodology
The optimisation process involved the analysis of three unique cases defined by the output product, hydrogen, methanol, or ammonia. While the case of hydrogen production was independent of the other two cases, both methanol and ammonia were implemented with generated hydrogen as a feedstock. The methodology for the optimisation process is summarised in Figure 8-6 below.

2. Run Optimisation
1. Gather Input Data Optimisation of cases with 3. Analyse Results
Data obtained for the and without CST: Quantified the benefits of
reference year of 2030 from 1. Hydrogen adding CST to a green fuel
AEMO and other sources. 2. Methanol production plant
3. Ammonia

Figure 8-6: Overview of the methodology for the techno-economic optimisation performed

The base case for the techno-economic analysis is represented by an off-grid hydrogen plant in the year
2030, powered by renewable energy sources, including solar PV and wind. The goal of the optimisation was to design the least-cost solution and parametrisation of the components of interest, including energy generation and conversion technologies, as well as storages. The three cases were first optimised with power available from solar PV and wind alone, and subsequently with the addition of CST (single tower and central receiver) for comparison and to quantify the benefits of the technology.

33
Storage technologies were implemented in the modelling as flexibility option to the extend needed to balance demand and
supply. They were considered in the economic analysis.

S774Doc-676563417-363 / v0.2 114 CST Value Proposition
The optimisation shows the costs for electricity generation plants as being the largest cost driver, contributing over 70% of the total CAPEX. Conversely, the share of CAPEX of the electrolyser plant was only approximately 15% across the three cases34. Therefore, the focus of the analysis was set on optimising the electricity supply. The two main performance indicators and cost drivers were the CAPEX for the installed capacity of solar PV and wind, as well as the curtailment of electricity from these sources.
Despite the share of cost of the electrolyser system being small, it was nevertheless analysed and considered in the economic evaluation and optimised for the H2-case.

The optimal utilisation factor of the electrolyser was found to be approximately 60%. By increasing the utilisation and reducing the installed electrolyser capacity, a reduction of the initial investment cost for the electrolyser system was achieved. However, a smaller electrolyser system (running with a higher utilisation) was found to be unable to utilise the peaks of the feed-in from renewables as well as an electrolyser system with a higher capacity (and lower utilisation). This led to a higher requirement of installed capacity of solar PV and wind, which resulted in significantly greater installation costs.
Considering the CAPEX of the electrolyser was only approximately 15% throughout the three cases, it can be concluded that a higher utilisation does not lead to a lower overall cost of hydrogen.

Further, the impact of a change in electrolyser capacity (+/- 20%) and the respective change in utilisation was analysed in a sensitivity study. In the methanol case with CST, the levelised cost of methanol increased by +8% and +7% respectively when changing the electrolyser capacity by +/- 20%. This sensitivity analysis confirmed the chosen approach of using an optimised but constant utilisation for the electrolyser.

When looking at specific projects (use cases), all relevant boundary conditions must be considered for a holistic system optimisation. Such conditions (e.g., land availability, renewable resource, operational requirements, and local demand for power and/or heat) will have an impact on green fuel production. As such, there might be end use cases (e.g., combined power and hydrogen generation), for which a higher or lower capacity factor - compared to what has been calculated in the presented use cases - is beneficial
(i.e., the electrolyser capacity for a given end use case is highly project dependent and needs to be looked at individually - as part of the optimisation process). Such detailed (project specific) optimisation processes that look at multiple further variables, were not within the scope of this study.

Base cases analysed in this study:
The hydrogen system designed was based on an ammonia demand of 200 ktpa, requiring approximately
36 ktpa of hydrogen. This value was then used as the hydrogen demand for all cases, resulting in a methanol production of 179 ktpa.

1. Case Hydrogen (H2): 36 ktpa output
a. PEM without CST
b. PEM with CST

34
Cases without the additional CST plant.

S774Doc-676563417-363 / v0.2 115 CST Value Proposition
2. Case Methanol (CH3OH): 179 ktpa output
a. PEM without CST
b. PEM with CST
3. Case Ammonia (NH3): 200 ktpa output
a. PEM without CST
b. PEM with CST

These cases were also analysed with SOEC in place of PEM, noting however that it is an early stage technology and introduces much greater uncertainty to the results. As presented in Section 8.5.2 the combination of CST and SOEC is particularly beneficial.

8.5.2 Benefits of CST for green fuel production
This section presents the benefits of the implementation of CST into the system energy mix. The key advantages of including CST in the energy system for green fuel production are firm capacity and the availability of green heat.

8.5.2.1 Combined Heat and Power - Availability of Green Heat
One of the key advantages of CST is that the technology can provide combined heat and power. In addition to the increased flexibility (which is discussed in Section 8.5.2.2), the heat demand of the system can be supplied by the CST plant at low costs relative to auxiliary heating, particularly for systems with a high thermal demand, such as for Case CH3OH +CST. In this case, the optimisation resulted in over a third of the total heat generated by CST being distributed directly as heat to satisfy the thermal demand of the methanol synthesis and DAC processes. Figure 8-7 shows this distribution of the generated heat from CST for the Distribution
Case CH3OH +CST.
of Heat Production

36% Heat to Steam Turbine

64% Heat to Methanol

Figure 8-7: Distribution of heat generated by the CST plant in Case CH3OH +CST

Furthermore, Figure 8-8 shows the annual duration curves of the heat supply and electricity generation from the steam turbine for the same case. The turbine is operated on full load at 140 MW for over 2,500 hours of the year and has an overall annual utilisation of 42%, or 3,670 equivalent full load hours per year. For more than 2,000 hours, the steam turbine operates on minimum load, however, still provides enough electricity to supply the methanol synthesis and in order to maintain the methanol demand at all times. The dark-blue graph illustrates the annual duration curve of heat from the CST plant to the

S774Doc-676563417-363 / v0.2 116 CST Value Proposition
methanol plant, which represents 84% of the heat required for methanol production overall, demonstrating the key benefit that CST provides in the energy mix to processes with high thermal demands.
Sorted Duration of Heat Extracted to Off-takers and Electricity Output from Turbine
160
140
Power Output [MW]

120
100
80
60
40
20
0
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000
Time [h]
Heat to Methanol Electricity from Steam Turbine

Figure 8-8: Sorted annual duration curve of heat utilisation from the CST plant in Case CH3OH +CST

Figure 8-9 compares the sources of heat for the methanol synthesis process for the case with and without
CST. The annual demand of heat for the production of methanol of approximately 900 GWh can be supplied by the CST plant at a share of 84%. This heat is provided at a much lower cost than that of an auxiliary heater (using PV or wind power), thereby reducing the levelised cost of thermal energy by 65%.
The remaining 16% of the heat demand is still supplied by the auxiliary heater, which converts electrical energy from the renewable energy sources (including CST in Case CH 3OH +CST) to heat with an efficiency of 98%.
Source of heat for methanol production
1,000
900
800 16%
Heat Supply [GWh]

700
600
500
100%
400 84%
300
200
100
0
Case CH₃OH Case CH₃OH
+CST
CST Heat to Methanol Aux. Heater
Figure 8-9: Sources of thermal energy for methanol production in the case with and without CST

Value proposition of CST: The benefit of CST is greater for systems with a high heat demand, such as methanol production or hydrogen systems based on SOEC technology. This is explained by the

S774Doc-676563417-363 / v0.2 117 CST Value Proposition
substitution of auxiliary heating powered by solar PV and wind with the relatively inexpensive heat generated by CST.

8.5.2.2 Firm capacity
The CST plant adds a dispatchable source of energy to the system that is provided by the steam turbine and enabled by the long duration TES. This dispatchability complements the electricity generated from solar PV and wind and provides support to meet the demand during times of low sunlight and or wind resources. The result is reduced installed capacities of solar PV and wind and less curtailment of energy overall, as demonstrated in the following figures. Firm capacity and dispatchability are important to supply base-load oriented processes, such as methanol synthesis, and to help bridge the gap between periods of variable wind and PV output. The added flexibility reduces dependency for other energy storages, such as BESS, which are typically required in systems consisting of only solar PV and or wind.

Availability of green power with a high capacity factor
The dispatchability of the steam turbine in combination with the TES enables the CST plant to provide green power with a high-capacity factor (in the range of 55% to 65%), compared to solar PV (around
31%) and wind (around 39%) not considering any curtailment. Case NH3 +CST saw the CST plant distribute 100% of its solar thermal energy to the steam turbine, as the downstream ammonia plant did not require input heat. The steam turbine was then able to provide a dispatchable supply of electricity and hence hydrogen, to the ammonia production, which does not have high flexibility itself.

Figure 8-10 compares the annual duration curves of each generation technology for the Case NH3 +CST.
In the figure, the difference in the capacity factor of the steam turbine within the CST plant to the other technologies (solar PV and wind) is substantial, with the turbine operating in full load for over 4,000 hours of the year. Running on a reduced load (20% of nameplate capacity) for approximately 2,000 hours, the steam turbine of the CST plant provides electricity to the base-load oriented ammonia synthesis process based on the optimised dispatch.
Sorted duration curve of capacity factor of all electricity sources
100%
Power Output [% of Pmax]

80%

60%

40%

20%

0%
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000
Time [h]
Solar PV Wind Steam Turbine of CST plant
Figure 8-10: Sorted duration curves of the capacity factor of all electricity generators in Case NH3 +CST

Value proposition of CST: The higher capacity factor of the CST plant is ideally suited to supply inflexible, base-load oriented downstream systems, such as ammonia and methanol production. Without

S774Doc-676563417-363 / v0.2 118 CST Value Proposition
CST, these systems would require a firm energy supply for these processes from solar PV and wind, which
has to be enabled by implementing more BESS capacity. This is significantly more expensive than CST,
inflating the capital costs of the system.

Reduction in capacity of renewable energy sources
The dispatchability of the CST plant, enabled by the steam turbine in conjunction with the long duration
TES, allows for a considerable reduction in the necessary installed capacity of the other renewable energy
sources (solar PV and wind). As the installed capacities of the energy sources were optimised for each
individual case, the impact of implementing CST into each system was independent of the other cases.
Figure 8-11 shows the combined installed capacity of all energy generators in the system for each case,
including CST. It was observed that in all cases, the inclusion of CST in the energy mix reduced the
required overall generation capacity, which can be attributed to the greater capacity factor of the steam
turbine in the CST plant relative to the solar PV and wind generation technologies.
ources Installed
Installed
capacity
capacity
of electricity
of electricity
sources
sources
1.6 1.6
1.4 1.4
Installed Capacity [GW]

Installed Capacity [GW]

1.2 1.2
1.0 1.0
-32% -32%
-18% 0.8 0.8 -18% -18%
-17% -17%
0.6 0.6
0.4 0.4
0.2 0.2
0 0
Case NH₃ Case NH₃ Case H₂
Case H₂ Case H₂
Case H₂Case CH₃OH
Case CH₃OH
Case CH₃OH
Case CH₃OH
Case NH₃
Case NH₃Case NH₃
Case NH₃
+CST +CST +CST +CST +CST +CST +CST

Figure 8-11: Installed capacity of all electricity sources

Value proposition of CST: The reduction in the necessary installed capacity of the other renewable
energy sources (solar PV and wind) leads to a lower CAPEX for these sources. The economic effects on
the overall system are analysed in Section 8.5.3.

Reduction in the Overall Energy Curtailment
In addition to the reduction of installed generation capacities, the dispatchability of the CST plant also
allows for the reduction in the overall energy curtailment in all cases35. Figure 8-12 depicts the curtailed
electricity for each case of the optimised cases, which demonstrates that curtailed electricity is
significantly reduced when CST is integrated into the energy supply. The energy curtailment was reduced
by two thirds in the methanol case and by more than one third in the ammonia case when implementing
CST. The benefits in Case H2 were not as significant, however, still demonstrated a reduction of

35
“Energy clipping” of renewable energy in the analysed off-grid systems is referred to as “curtailment” in this section of the
report.

S774Doc-676563417-363 / v0.2 119 CST Value Proposition
curtailment of 12%, with the reduced impact attributed to the existing flexibility of the electrolyser and
hydrogen storage, which utilise a high share of electricity from solar PV and wind.

The modelling of the specific cases shows that a reduction in energy curtailment could also be achieved
by implementing larger batteries. However, this would result in a significantly higher CAPEX compared to
the dispatchability provided by the CST plant.
Curtailed electricity
Curtailed electricity
1,8001,800
1,6001,600
Curtailed Electricity [GWh]
Curtailed Electricity [GWh]

1,4001,400
1,2001,200
1,0001,000
800 800
-38% -12%-12% -38%
-38%
600 600
400 400 -66%-66%
200 200
0 0
Case NH₃ Case NH₃ Case Case
H₂ H₂ CaseCase
H₂ H₂ CaseCase
CH₃OH CaseCase
CH₃OH CH₃OH
CH₃OHCaseCase
NH₃NH₃ CaseCase
NH₃NH₃
+CST +CST+CST +CST+CST +CST+CST

Figure 8-12: Curtailed electricity from all energy sources for each case in the optimisation

Value proposition of CST: A lower curtailment of electricity as a result of implementing CST leads to a
higher utilisation of the renewable generation plants. This in turn, decreases the levelised cost of
electricity from each source, decreasing the overall levelised cost of product.

Replacement of BESS Systems as an Alternative Source of Flexibility
The flexibility introduced with the thermal energy storage of the CST plant assists downstream hydrogen
derivatives production, and also allows for the reduction of other storage systems, such as the use of
BESS. For example, in Case NH3 +CST, the required capacity of the battery in the system was reduced by
over 90% when CST was integrated into the energy mix. If the hydrogen generation system is based on a
highly flexible PEM system, it does not benefit as much from the additional firming provided by the TES,
as a result of the existing inherent flexibility to cope with varying inputs. This flexibility is manifested in
large amounts of part-load operation of the electrolyser, which was observed in all cases.

Figure 8-13 shows the annual duration curve of the hydrogen production from the electrolyser in
Case NH3 +CST, providing an overview of the operational load over a year. With the electrolyser
optimised in Case H2 as described in Section 8.5.1, a utilisation of 59.9% was achieved, corresponding to
5,250 equivalent full-load hours. From the graph, the electrolyser was observed to operate at full load for
2,260 hours and at part load for 5,500 hours per year, demonstrating its highly flexible operation.36

36
Key electrolyser performance indicators, including the ability to respond to rapid changes of load, are presented in
Appendix D.

S774Doc-676563417-363 / v0.2 120 CST Value Proposition
Sorted Duration Curve of Output H2 Electrolyser Case A +CST
8,000

7,000
Hydrogen output [kg/h]
6,000

5,000

4,000

3,000

2,000

1,000

0
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000
Time [h]

Figure 8-13: Sorted annual duration curve of the hydrogen production in Case NH3 +CST

Figure 8-14 shows the state of charge of the hydrogen storage across the year. The storage is utilised for short term buffering, as well as longer term supply during periods of electrolyser downtime. The analysis demonstrated that the hydrogen storage is used as an important source of flexibility in the green fuel system. The hydrogen storage is used to store hydrogen being produced in times with high solar PV and wind generation to be used in the baseload oriented, less flexible ammonia process, when there is no or low solar PV or wind. As hydrogen storage systems are an expensive source of flexibility, it was considered in the modelling and the evaluation of the economic performance of the system.
Annual Hydrogen Storage State of Charge
200
175
150
State of Charge [t]

125
100
75
50
25
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Time [h]

Figure 8-14: Hydrogen Storage State of Charge over the year in Case NH3 +CST

Value proposition of CST: The flexibility introduced by the CST plant helps to replace other sources of flexibility, such as the relatively expensive BESS. The resultant impact is seen in the CAPEX, as shown in
Figure 8-15.

8.5.3 Economic impacts of CST
The benefits of adding CST to the system in combination with PV and wind were presented and discussed in Section 8.5.2. The following section quantifies these benefits by their economic effects on the

S774Doc-676563417-363 / v0.2 121 CST Value Proposition
production of green fuels. However, as the benefits of CST are interlinked in nature, the individual quantification of each cannot be determined. For example, the flexibility of the CST system is attributed to the TES, which enables the dispatchability and consequently, the high-capacity factor of the plant.

Capex and LCoP37
The technical benefits introduced with the CST plant are ultimately reflected in the levelised cost of product (LCoP37) for each case respectively. As discussed in Section 8.5.2, the dispatchability of the CST plant primarily adds value to downstream subsystems of low flexibility. Thus, the economic viability of implementing CST depends on the flexibility of the parts of the system, in addition to the overall heat demand.

This was observed in the results of the optimisations, where in the methanol and ammonia cases, the higher specific capital costs of the CST plant still resulted in a decrease in the LCoP (i.e., the higher capital cost was offset by the combined provision of low-cost heat and dispatchable power). Conversely, in the hydrogen case based on PEM electrolysis, the additional capital costs of the CST plant were not outweighed by the benefit of the added dispatchability and the availability of combined heat and power.
However, using SOEC electrolysis technology rather than PEM, this was observed to change. In a separate calculation, the effect of introducing CST to a system using the high temperature technology of SOEC was conducted. Results show that the combination of CST and SOEC allow for a reduction of LCoH (see
Figure 8-17).

Figure 8-15 shows the overall investment costs of each case, split into the following four categories:

1. The cost of the energy supply, excluding CST (solar PV and wind combined with the battery)
2. The cost of the CST plant which includes the solar field, TES, and the steam turbine
3. The cost of hydrogen generation, including the electrolyser, the hydrogen storage and all auxiliary
components relating to these38
4. The cost of green derivatives generation, which includes:
- for the methanol plant - the methanol synthesis, methanol storage and DAC unit
- for the ammonia plant - the ammonia synthesis, ammonia storage and ASU

37
“Levelised Cost of Product” (LCoP) refers to the production cost of the respective green fuel analysed in this study (i.e.,
hydrogen, methanol, and ammonia)
38
The optimisation and sizing of the electrolyser system, including its utilisation, is described in Section 8.5.1 (Methodology).

S774Doc-676563417-363 / v0.2 122 CST Value Proposition
Total CAPEX
4.5
4.0
CAPEX [Billion AUD]

3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
Case H₂ Case H₂ Case CH₃OH Case CH₃OH Case NH₃ Case NH₃
+CST +CST +CST
PV, Wind, Battery CST Hydrogen Generation Green Derivatives

Figure 8-15: Overall investment costs of all cases

From the graph, the capital cost of the methanol case is observed to have the greatest reduction as a result of the integration of CST. This can be directly attributed to both the inflexibility of the methanol plant, as well as the high thermal demand of the synthesis process and DAC unit allowing for the large offset of power-to-heat.

Case NH3 +CST also demonstrates a reduction in capital costs, however, not as significant as the methanol case. Unlike for methanol, the provision of low-cost heat is not a key cost driver for ammonia.
Rather, the reduction in capital cost is due to CST's provision of firm power to meet the inflexible base load requirement of the ammonia production. Conversely, for Case H2 +CST (based on PEM electrolysis), the existing flexibility of the electrolyser combined with the lack of a heat demand results in the implementation of the CST plant increasing the capital cost. In a separate calculation, the effects of combining CST and SOEC electrolysis were analysed. As discussed, the additional capital costs of the CST plant were outweighed by the benefit of the added dispatchability and the availability of combined heat and power, resulting in a lower overall LCoH in this case.

Moreover, Figure 8-15 shows that approximately 55% to 80% of the total capital costs (depending on the case) are attributed to the generation technologies (solar PV, wind, and CST). In all cases, wind power was the component contributing the largest share to the overall CAPEX (approximately 23% to 30%). In the
Case CH3OH +CST, the CST plant contributes 21% to the total capital costs, whilst still enabling the benefits previously discussed that ultimately result in a significant CAPEX reduction overall, as well as a reduction in LCoP. Within the other two cases, the CST plant accounts for approximately 20% to 25% of the total CAPEX. The hydrogen production (based on PEM) contributes approximately 20% to the total
CAPEX, with a small variance over all cases. It can be concluded that the added CAPEX by CST is comparably small considering the overall CAPEX and was ultimately outweighed by the associated benefits that the technology introduced (see Section 8.1 and Figure 8-18).

S774Doc-676563417-363 / v0.2 123 CST Value Proposition
CST with Solid Oxide Electrolysis
The emerging electrolysis technology of SOEC presents a technology pathway that can deliver greater yields of renewable hydrogen at lower costs. Specifically, SOEC is a high-temperature electrolysis pathway where the use of thermal energy results in a 30% reduction in the total energy requirements
(54.5 kWh/kg of hydrogen from PEM versus 44 kWh/kg from SOEC39).

Given that SOEC requires high temperature heat and power (with the power requirement partially offset by the heat), it offers strong prospects for the combination with CST. However, as discussed in
Section 8.4.2, the technology is not yet commercially widespread, which introduces a level of uncertainty in the optimisation results.

An analysis was conducted that investigated the impact of adding CST into the energy mix to produce green hydrogen using SOEC. As a result of using this technology over PEM, the installed capacities of the renewable energy sources were significantly reduced. Figure 8-16 shows the total required capacities to satisfy the electricity demand of hydrogen production using SOEC, with and without CST in direct comparison with the previously discussed Case H2 with PEM.
Installed capacity of electricity sources
1.0
0.9
0.8
Installed Capacity [GW]

0.7 -17%
0.6
0.5
0.4 -35%
0.3
0.2
0.1
0
Case H₂ Case H₂ Case SOEC Case SOEC
+CST +CST

Figure 8-16: Change of required installed capacity when using PEM or SOEC in combination with CST

Another advantage of adding CST to the SOEC system was that curtailed electricity was reduced by 55%.
Due to the lower specific electricity demand of the SOEC, the CST plant was able to supply almost half of the required electricity, which explains the reduction of curtailed electricity from solar PV and wind.
Furthermore, the CST plant is capable of providing the entire thermal demand of the SOEC, which would otherwise need to be supplied by auxiliary heating using power-to-heat.

Furthermore, by investigating the LCoH of the two cases, the holistic benefit of utilising CST in combination with SOEC are identified. Figure 8-17 compares the levelised cost of hydrogen from SOEC with and without CST on a relative basis.

39
This number includes heat and power.

S774Doc-676563417-363 / v0.2 124 CST Value Proposition
Relative change of levelised cost of final product
140%

120%

Relative Cost [%] 100%

80%

60%

40%

20%

0%
Case H₂ Case H₂ Case SOEC Case SOEC
+CST +CST

Figure 8-17: Levelised cost of hydrogen when using SOEC in combination with a CST plant

The comparison reveals that the PEM system (without a heat demand) saw a price increase of approximately 10% with the addition of CST. The SOEC system (with a heat demand), however, was able to make use of the inexpensive green heat, leading to a cost reduction of approximately 8% (compared to the PEM case without CST). The reduction of LCoH by combining CST and SOEC was extrapolated using a simplified model for the methanol and ammonia case. In summary, the analysis reiterates the fact that the benefit of implementing CST into an energy mix is maximised in systems with a demand for heat and base-load power.

As previously mentioned, there is a far greater level of uncertainty in the cost model and underlying technical characteristics of SOEC compared to the well-established PEM-based system. This is reflected in the variances in the cases with SOEC. If key parameters such as operational flexibility, technical lifetime and degradation can be confirmed, SOEC based systems in combination with CST will be a promising option for future hydrogen systems, as well as when combined with methanol or ammonia production. A deeper techno-economic analysis of electrolyser technologies is recommended for further studies, which should explicitly involve OEMs of electrolysers, including SOEC.

Levelised Cost of Product for Hydrogen, Methanol and Ammonia
Figure 8-18 summarises the key results by displaying the levelised cost of product for the respective green fuel for each case and represents the most critical results of the techno-economic evaluation. The greatest share of the levelised cost of product for each of the green fuels was contributed by the cost of electricity, which meant that any reduction in the LCoE from the addition of CST led to a reduction in the levelised cost of the final product.

S774Doc-676563417-363 / v0.2 125 CST Value Proposition
Relative change of levelised cost of final product
120%

100%
Relative Cost [%]

80%

60%

40%

20%

0%
Case H₂ Case H₂ Case H₂ Case CH₃OH Case CH₃OH Case CH₃OH Case NH₃ Case NH₃ Case NH₃
+CST +CST +SOEC +CST +CST +SOEC +CST +CST +SOEC

Figure 8-18: Relative change of levelised cost of the final product

Hydrogen: For hydrogen production, PEM and SOEC electrolyser systems must be differentiated from one another. The advantages and disadvantages are discussed in detail in Section 8.4.2. If PEM electrolysis is used, there is less requirement for base-load power as the technology is very flexible in terms of power utilisation, and there is no requirement for heat. As such, the key benefits of CST, including the provision of firm capacity and heat and power, is not a key cost factor in hydrogen production. However, if SOEC electrolysis is used, the use of green heat delivered by CST results in an approximate reduction of 30% of the power requirement. This in turn, results in a strong heat value stream that results in an LCoH that is lower than that of an equivalent PEM system. Given that SOEC requires high temperature heat and power (with the power requirement partially offset by the heat), it offers strong prospects for CST. However, as discussed in Section 8.4.2, the technology is not yet commercially widespread and therefore requires several underlying assumptions for its implementation into the energy system model, which introduces greater uncertainty in the optimisation results.

Methanol: CST provides the greatest advantage in a system that includes methanol production. The reduction in levelised cost of methanol, compared to a system without CST, is the largest of all green fuels that were analysed. This is a result of the methanol production process requiring a large amount of heat and base-load power. CST provides most of the required heat at a low price, and also supports base-load production. This combination of heat and power provided strong value and potential for green methanol production.

Ammonia: For ammonia production, an external heat supply is not a requirement. However, base-load power is essential for the production process, which can be provided by CST at a much lower cost compared to other technologies, such as BESS. CST's ability to provide power for over 85% of the year is a significant advantage. For this reason, the system with CST delivers a lower levelised cost of ammonia than systems with solar PV and wind only.

Value proposition of CST
As a conclusion, the benefits of CST described in Section 8.5.2 collectively influence the production cost of green fuels, which is the ultimate priority to reduce. Comparing production costs of different systems captures the overall benefit of adding CST, which includes the disadvantage of the added capital costs.
The production of all green fuels considered in this study (hydrogen produced from SOEC instead of
PEM, ammonia, and methanol) profit from CST, with methanol production benefiting more so as a result of the heat demand of the synthesis process. Ultimately, systems with a high heat demand and low operational flexibility generally see the greatest benefits from the inclusion of CST in the energy supply.

S774Doc-676563417-363 / v0.2 126 CST Value Proposition
8.6 Market Size
The final step in the analysis of green fuels for the CST value proposition is the modelling and forecast of the green fuels market sizes and sector deployment timelines based on commercially driven uptake. The modelling process for the market research analyses the results of the forecasts with respect to the leading configuration design generated from the techno economic modelling in Section 8.5 and is presented in the following section.

8.6.1 Green fuel market estimates
This section focuses on the potential production capacity of green fuels in Australia, it, includes estimates for both the domestic and international export markets. The estimates are based on public reports, as described below. The aim of the analysis is to enumerate the market size of each of the green fuels in this study in order to better quantify the value proposition of CST in Australia. Table 8-3 summarises the result of the market estimates, with the forecasted market volumes for 2050.

Table 8-3: Estimated green fuel market volumes for 2050

Green fuel Unit Estimated Market Volume in 2050

Hydrogen Mt 22

Methanol Mt 30

Ammonia Mt 34

The figures provided in Table 8-3 for hydrogen explicitly include the hydrogen demand for SAF production, which is elaborated on in Section 8.4.2. The figures for methanol include the demand of methanol as a fuel for transport, and primarily includes the maritime transport sector. More detailed information on the use of methanol is provided in Section 8.4.3.

Hydrogen - demand in 2050
In 2019, Deloitte performed a study to estimate the Australian and global hydrogen demand growth scenarios (Deloitte, 2019). The study produced four scenarios, with varying degrees of green hydrogen adoption. Deloitte’s estimates include both domestic consumption and international export markets and are classified into market segments (for example, transport, steelmaking, industrial feedstock, etc.).
Demand for hydrogen explicitly includes hydrogen as a feedstock for SAF used in the transport sector.
Bloomberg New Energy Finance (BNEF) also published their latest estimates for the global hydrogen market projections in November 2022 (BloombergNEF, 2022). BNEF estimate that the rapid uptake of green hydrogen production in Australia will occur later than Deloitte’s estimate, however, their estimate for 2030 is in line with the Deloitte study (BNEF only published projections up to 2030). Both studies see
Australia’s share in global hydrogen production at approximately 6% to 7%. Deloitte estimate global green hydrogen production will reach up to 295 Mt/a in 2050 (scenario 1 “Hydrogen: Energy of the future"), which can be compared to another study by McKinsey, that is more optimistic than both Deloitte and BNEF, with a global 2050 estimate of 510 Mt/a in 2050 (McKinsey, 2022).

Scenario 1 from the Deloitte study was chosen as the basis for this study, as the study focused on production of hydrogen in Australia (including exports). Additionally, Scenario 1 is more in line with the other, more recent international studies mentioned above. In this scenario, the total annual production of

S774Doc-676563417-363 / v0.2 127 CST Value Proposition
green hydrogen in Australia is expected to reach 18 Mt/a in 2050. As explained in the section below, due to the favourable conditions for the production of green methanol in Australia, it is estimated that
Australia will capture a larger share of the global market for renewable methanol production. This results in an additional demand for green hydrogen of approximately 3.4 Mt/a in 2050. Therefore, the total market for green hydrogen in Australia is forecasted to be 22 Mt/a in 2050. Deloitte expects that a large share of the produced hydrogen (approximately 70% of the 18 Mt/a in the original study) will be exported for use in other countries. The Deloitte study does not discuss the question of which chemical form the hydrogen will be exported as. In this study, it was assumed that half of the hydrogen is exported in the form of compressed or liquefied hydrogen, and the other half as ammonia. The additional ammonia will be produced onshore in Australia and is accounted for in the below ammonia market estimate.

Methanol - demand in 2050
The methanol market potential was estimated based on a study by the International Renewable Energy
Agency (IRENA, 2022, Innovation Outlook: Renewable Methanol). The study provides a market estimate for the global (but not specifically the Australian) market for renewable and conventional methanol in
2050. The study also provides an estimate for renewable methanol produced via electrolysis (referred to as e-methanol by IRENA). The figures provided for methanol include the demand for methanol in the transport sector and primarily involves use for maritime transport. Besides that, methanol can also be blended with conventional jet fuel to create a more sustainable fuel blend as presented in Section 8.4.3.

Due to the favourable conditions to generate renewable heat (which is required in large quantities for methanol production), it is expected that Australia will be able to seize a larger share of the global renewable methanol market, compared to the hydrogen and ammonia markets. It is estimated that the
Australian share in the global renewable methanol production will reach 12% in 2030 (compared to around 7% for hydrogen and 6% for ammonia), which includes domestic consumption and export. This method results in an estimated production capacity of 30 Mt/a of renewable methanol in Australia in
2050.

The hydrogen required to produce methanol is accounted for in the above hydrogen market estimate.

Ammonia - demand in 2050
The ammonia market was estimated in a similar way to the methanol market, based on a second study from IRENA (2022, Innovation Outlook: Renewable Ammonia), which estimated the Australian market as
6% of IRENA’s total global market estimate.

The hydrogen required to produce the ammonia is accounted for in the above hydrogen market estimate. It was also assumed that 50% of the hydrogen produced in Australia is converted to ammonia for export purposes.

This method results in an estimated production capacity of 34 Mt/a of renewable ammonia in Australia in
2050, which includes domestic consumption and export.

S774Doc-676563417-363 / v0.2 128 CST Value Proposition
8.6.2 Potential for CST in green fuels
The potential for CST in 2050 was estimated based on the green fuels market forecast presented (green hydrogen, methanol, and ammonia) and the conducted techno-economic assessment. The estimated future market potential for CST depends on multiple assumptions and is therefore subject to a high level of uncertainty. To account for this, the potential for CST in Australia was calculated as a range rather than a specific value.

Step 1: Number of reference plants required to produce the green fuel demand
For each green fuel, the specific number of reference plants required was calculated. The production output for each reference green fuel plant is presented in Table 8-3. For hydrogen, the electrolyser technology was another consideration important for the market estimates, as the share of SOEC electrolysers was assumed to account for 5% of the total hydrogen production capacity in the Deloitte study (Deloitte, 2019).

Step 2: Specific competitive advantage of CST regarding production cost for green fuels
The effect of CST on the production cost of green fuels (typically given in A$ /kg) differs between fuel type (see Figure 8-18). Green fuel production plants with a green heat demand profit more from CST, as was discussed in Section 8.5.3. The competitive advantage calculated is based on specific site conditions
(feed-in profile, etc.). Not all reference plants that are required to supply the demand for green fuels will share the same conditions that favour CST. It is therefore important to transfer the specific competitive advantage of CST, which differs between green fuels, into a specific share of reference sites with CST.
Renewable methanol demonstrated the greatest advantage from CST, and the assumption is that
30% to 70 % of the methanol reference plants will use CST 40. For the other green fuel plants (green ammonia and green hydrogen based on SOEC technology) the specific share was calculated based on their relative competitive advantage (shown in Figure 8-18) compared to methanol.

Despite the pure economic benefit of CST, competing technologies will still be available and developers may retain individual preferences. Therefore, another factor was implemented to account for the market share of CST against other technologies. This added factor contains high uncertainty, and therefore varied between 30% and 70% so as to provide a range of the precise CST potential.

Results
The overall potential for CST plants to produce green fuels in Australia is presented in Figure 8-19. Given the assumptions listed, the market potential for CST plants exists and will be in the range of 2.6 GWe and
14 GWe in 2050. As demonstrated, green methanol profits most from CST and approximately 70% of the market potential for CST is based on future renewable methanol plants.

40
This range was used to account for uncertainties.

S774Doc-676563417-363 / v0.2 129 CST Value Proposition
16
Total Market Potential
14 70%

12
for CST in 2050 [GW]

Market Share of CST
10 10.0 50%

8
7.1
6 30%

4 4.3

2

0
20% 30% 40% 50% 60% 70% 80%

Share of suitable sites for CST
Figure 8-19: Total Market Potential for CST in 2050 with varying shares of sites suitable for CST and varying market
shares of CST

S774Doc-676563417-363 / v0.2 130 CST Value Proposition
9 Conclusion
Australia offers prime conditions to deploy CST competitively, while benefiting from its multiple sources of value. CST can play an important role in Australia’s transition towards net zero, providing both dispatchable power and heat, and, thus, being applicable for a broad range of end use sectors requiring rapid decarbonisation.

CST’s strength is its dispatchability, namely its ability to generate and store large amounts of heat during daylight hours, for use, on demand, to generate multiple hours of firm, power and to provide reliable heat for industrial processes or green fuel production.

CST is a complementary technology and delivers the best energy system outcome, when it is integrated with the VRE technologies, solar PV and wind. CST can become a key enabler to fully replace dispatchable coal and gas-fired generation. In assessing costs and benefits it needs to be compared with fossil fired and other dispatchable renewable generators rather than variable solar PV or wind.

In order to unlock Australia’s CST potential in the different sectors, CST must (i) be recognised as a mature technology which has been deployed internationally for many decades, (ii) be considered as one of the potential solutions, and (iii) be adequately and fairly assessed in capacity expansion and other modelling - in particular in regard to its dispatchability features and other sources of value.

The analysis presented here shows that in the four sectors investigated; grid connected, off grid, process heat and green fuels; CST can become a key part of a least cost technology mix on the transition to net zero GHG emissions. Establishing policies that encourage an early start and smooth growth of CST uptake in all sectors can save the country many billions of dollars in reaching net zero as well as maximising local economic benefit and adding to diversity in energy supply.

New policies, energy market measures and decarbonisation strategies should be technology neutral but should be designed to encourage the long duration energy storage and dispatchable behaviour needed in the long term. Measures also need to avoid perverse effects from, for example, requiring delivery times that are too short or offering security of financial return that is too short for successful financing of systems with 25+ year economic life.

S774Doc-676563417-363 / v0.2 131 CST Value Proposition
10 References

AEMO. (2022a). 2022 Wholesale Electricity Market Electricity Statement of Opportunities. Australian Energy
Market Operator

AEMO. (2022b). Draft 2023 Inputs, Assumptions and Scenarios Report. Australian Energy Market
Operator https://aemo.com.au/-/media/files/stakeholder_consultation/consultations/nem-
consultations/2022/2023-inputs-assumptions-and-scenarios-consultation/draft-2023-inputs-
assumptions-and-scenarios-report.pdf?la=en

AEMO. (2022c). Draft 2023 Inputs and Assumptions Workbook. Australian Energy Market Operator
https://aemo.com.au/consultations/current-and-closed-consultations/2023-inputs-assumptions-
and-scenarios-consultation

AEMO. (2022d). 2022 Integrated System Plan (ISP)Australian Energy Market Operator.
https://aemo.com.au/energy-systems/major-publications/integrated-system-plan-isp/2022-
integrated-system-plan-isp

AEMO. (2023). 2023 Inputs, Assumptions and Scenarios Report. Australian Energy Market Operator.
https://aemo.com.au/-/media/files/major-publications/isp/2023/2023-inputs-assumptions-and-
scenarios-report.pdf?la=en

ASTRI. (2022). Public Dissemination Report 2021. The Australian Solar Thermal Research Institute.

Augustine, C., Turchi, C., & Mehos, M. (2021). The Role of Concentrating Solar-Thermal Technologies in a
Decarbonized U.S. Grid. National Renewable Energy Laboratory.

Aurecon. (2022). 2022 Costs and Technical Parameter Review - Australian Energy Market Operator.
Aurecon (2022-12-12). https://aemo.com.au/-
/media/files/stakeholder_consultation/consultations/nem-consultations/2022/2023-inputs-
assumptions-and-scenarios-consultation/supporting-materials-for-2023/aurecon-2022-cost-
and-technical-parameter-review.pdf?la=en

Aurecon. (2023). 2022 Costs and Technical Parameter Review - Australian Energy Market Operator.
Aurecon (June 2023)

Bellotti, D., Rivarolo, M., & Magistri, L. (2020). Clean fuels synthesis from green hydrogen: a techno-
economic. Genova. doi:https://doi.org/10.1051/e3sconf/202123803001

BloombergNEF. (2022, November 14). A Breakneck Growth Pivot Nears for Green Hydrogen. Retrieved
May 22, 2023, from https://about.bnef.com/blog/a-breakneck-growth-pivot-nears-for-green-
hydrogen/

CEFC. (2022). The compelling case for decarbonisation, Clean Energy Finance Cooperation and The
Minerals Research Institute of Western Australia (MRIWA),

S774Doc-676563417-363 / v0.2 132 CST Value Proposition
CSIRO. (2023a). Renewable Energy Storage Roadmap. Commonwealth Scientific and Industrial Research
Organisation. https://www.csiro.au/en/work-with-us/services/consultancy-strategic-advice-
services/csiro-futures/energy-and-resources/renewable-energy-storage-roadmap

CSIRO. (2023b) GenCost: annual electricity cost estimates for Australia. Commonwealth Scientific and
Industrial Research Organisation. https://www.csiro.au/en/research/technology-
space/energy/energy-data-modelling/gencost

COAG Energy Council. (2019). Australia's National Hydrogen Strategy. Retrieved October 20, 2022

Compass International. (2022). Worldwide Industrial / Commercial Construction Schedule of Rates
Yearbook (13th Edition). Compass International Inc.

CSTA. (2023). Blue Book of China’s Concentrated Solar Power Industry 2022. China Solar Thermal Alliance.

Deloitte. (2011). Macroeconomic impact of the Solar Thermal Electricity Industry in Spain. Deloitte

Deloitte. (2019). Australian and Global Hydrogen Demand Growth Scenario Analysis.

DOE: (2021). Solar Futures Study. US Department of Energy - Department of Energy Solar Energy
Technologies Office

Dupont. (2020). Global available solar energy under physical and energy return on investment constraints.
ResearchGate - Applied Energy.

EC. (2023). National energy and climate plans. European Commission.
https://commission.europa.eu/energy-climate-change-environment/implementation-eu-
countries/energy-and-climate-governance-and-reporting/national-energy-and-climate-plans_en

ESTELA. (2016). Solar Thermal Electricity Global Outlook 2016, European Solar Thermal Electricity
Association

Eynard, U., Georgitzikis, K., Wittmer, D., & Latunussa, C. (2020). Study on the EU's list of Critical Raw
Materials (2020) Non-Critical Raw Materials Factsheets. doi:10.2873/587825

GHD. (2018). AEMO 2018/2019 Cost and Technical Parameter. AEMO.
https://view.officeapps.live.com/op/view.aspx?src=https%3A%2F%2Fwww.aemo.com.au%2F-
%2Fmedia%2FFiles%2FElectricity%2FNEM%2FPlanning_and_Forecasting%2FInputs-Assumptions-
Methodologies%2F2019%2FGHD-AEMO-revised---2018-19-
Costs_and_Technical_Parameter.xlsb&wdOrigin=BROWSELINK

Gilbert, P., Alexander, S., Thornley, P., & Brammer, J. (2013, September 4). Assessing economically viable
carbon reductions for the production. Journal of Cleaner Production, 64, 581-589.
doi:http://dx.doi.org/10.1016/j.jclepro.2013.09.011

Gordon, S. (2023, January 16). Clarksons Research. Retrieved from Green Technology Tracker: January
2023: https://insights.clarksons.net/green-technology-tracker-january-2023

IEA. (2022). World Energy Outlook 2022. International Energy Agency.

S774Doc-676563417-363 / v0.2 133 CST Value Proposition
IEA. (2023, July 10). Electrolysers. Retrieved from https://www.iea.org/energy-system/low-emission-
fuels/electrolysers

IRENA. (2020). Green Hydrogen Cost Reduction. International Renewable Energy Agency, Abu, Dhabi.
from https://irena.org/-
/media/Files/IRENA/Agency/Publication/2020/Dec/IRENA_Green_hydrogen_cost_2020.pdf

IRENA. (2022a). World Energy Transition Outlook 2022. International Renewable Energy Agency
from https://www.irena.org/Digital-Report/World-Energy-Transitions-Outlook-2022

IRENA. (2022b). Renewable Power Generation Cost. International Renewable Energy Agency.

ITP. (2017). Synthesis of Responses to ARENA’s 2017 Request for Information on CST Projects. Prepared by
ITP for the Australian Renewable Energy Agency (ARENA) August 2018
https://arena.gov.au/assets/2018/03/CST-RFI-Synthesis-Public-Report.pdf

ITP. (2018). Concentrating Solar Thermal Power Technology Status

ITP. (2019). Renewable Energy Options for Industrial Process Heat.

Lilliestam, J., Labordena, M., Patt , A., & Pfenninger, S. (2023). CSP.guru (Version 2023-01-01).

McKinsey. (2022). Global Energy Perspective 2022. Retrieved May 22, 2023, from
https://www.mckinsey.com/industries/oil-and-gas/our-insights/global-energy-perspective-2022

NREL. (2020). Concentrating Solar Power Best Practices Study. National Renewable Energy Laboratory

NREL. (2021). The Role of Concentrating Solar-Thermal Technologies in a Decarbonized U.S. Grid. National
Renewable Energy Laboratory

NREL. (2022). Annual Technology Baseline. National Renewable Energy Laboratory

Rutovitz, J., Peterseim, J., Elliston, B., Harris, S., Mohr, S., Lovegrove, K., Want, A., Langham, E., MacGill, I.,
(2013). Breaking the Solar Gridlock: Potential Benefits of Installing Concentrating Solar Thermal
Power at Constrained Locations in the NEM

SolarGis. (2023). Solar prospecting tool. SolarGis. (accessed 01. 2023

Solarthermalworld. (2023). Solar Industrial Heat Outlook 2023-2026. Solarthermalworld.org
https://solarthermalworld.org/news/promising-solar-industrial-heat-outlook-2023-2026/

Smith, C., Hill, A. K., & Torrente-Murciano, L. (2019, December 28). Current and future role of Haber–
Bosch ammonia. The Royal Society of Chemistry. doi:10.1039/c9ee02873k

Trieb, F., Schllings, C., O'Sullivan, M., Pregger, T., & Hoyer-Klick, C. (2009). Global Potential of Concentrated
Solar Power. SolarPaces.

USGS. (2020). Mineral Commodity Summaries 2020. U.S. Department of the Interior, Reston. Retrieved
October 9, 2020, from https://pubs.usgs.gov/periodicals/mcs2020/mcs2020.pdf

S774Doc-676563417-363 / v0.2 134 CST Value Proposition
Appendix

Appendix A CST Technologies ...............................................................................................................134

Appendix B OpenCEM Overview ...........................................................................................................148

Appendix C Sensitivity Analysis of Uptake in grid connected Systems ..................................163

Appendix D Green Fuels Data .................................................................................................................172

S774Doc-676563417-363 / v0.2 135 CST Value Proposition
Appendix A CST Technologies

A1 Technology Overview

A1.1 Parabolic trough
Parabolic trough power plants consist of many parabolic trough collectors, a heat transfer fluid system, a steam generation system, a Rankine steam turbine cycle, and optional thermal storage and/or backup systems. A schematic of a parabolic trough power plant with indirect thermal energy storage system is depicted below.

Parabolic Solar Steam Steam Turbine
Trough Field Thermal Energy Storage Generator Generator Grid

Superheater
Evaporator

Reheater
Preheater
G

1 3 4 5 6

Steam
2
Extractions

1

2

Condenser
3 Pre-heater

COLD TANK HOT TANK 4 5 6

Figure A-1: Schematic of parabolic trough power plant with storage system

The collector field is made up of a large number of single-axis-tracking parabolic trough solar collectors.
The solar field is modular in nature and comprises many parallel rows of solar collectors, normally aligned on a north-south horizontal axis. Each solar collector has linear parabolic-shaped mirrors that focus the sun’s direct beam radiation on a linear absorber pipe located at the focus of the parabola. The collectors track the sun from east to west during the day to ensure that the sun is continuously focused on the linear absorber.

A heat transfer fluid (HTF) which typically is a synthetic oil mixture is heated up to 393°C as it circulates through the absorber and returns to a steam generator of a conventional steam cycle. The process temperature of state-of-the-art parabolic trough plants is limited due to the temperature decomposition characteristics of current HTF; however, new concepts are under development, using either silicone based
HTF to allow for a wider operating temperatures range (e.g., Helisol with -40 to 425°C) or directly molten salt as HTF to allow for a direct molten salt storage combined with higher operating temperatures.

Parabolic trough technology is the most mature technology amongst the different CST alternatives. The
SEGS plants in California have successfully been operated for more than 30 years and since the new era of CST deployment started in 2007, the vast majority of CST plants are using parabolic trough technology. In Spain alone 45 parabolic trough power plants, each with a capacity of 50 MWe, have been built. The largest parabolic trough power plant built so far is the 280 MWe Solana plant in the U.S., which commenced commercial operation in 2013 and is also equipped with a large TES system (6 FLH).

S774Doc-676563417-363 / v0.2 136 CST Value Proposition
The three main systems of a parabolic trough power plant with indirect molten salt storage are:

▪ Solar field, incl. HTF system
▪ Thermal energy storage system; and
▪ Power block, incl. solar steam generator.

Solar field, incl. HTF system:
The solar field is a modular distributed system of single-axis-tracking solar collector assemblies (SCA) connected in parallel loops via a system of insulated pipes. The SCAs collect heat via a trough of parabolic mirrors, which focus sunlight onto a line of absorbers, welded in line at the focus of the parabola. Each of the loops consists of several SCAs in series, generally aligned on a north-south axis (to harness the solar resource as much as possible), thus tracking the sun during the day continuously from east to west. Considering the SCA size, mainly deployed so far, a loop configuration comprises four SCAs.
Each SCA has its own drive system and consist of several Solar Collector Elements (SCEs) connected in series.

Under consideration of the site conditions and an optimised hydraulic design, several loops are grouped into subfields, with one common header piping for HTF distribution running between the collector loops, placed in the south and north of the header piping. A typical solar field arrangement is depicted below.

1 SCE 1 Loop

Power Block and 1 SCA
Solar Island System

Solar Field Solar Field
Subfield 1 Subfield 2

Solar Field Solar Field
Subfield 3 Subfield 4

Figure A-2: Typical layout of a medium scale solar field

Considering the mostly used types of collectors, one can distinguish between two basic designs, the torque box (e.g., SKAL-ET) and the torque tube (e.g., SENERtrough). It is important to keep the collector geometry and to ensure the optical performance of the solar collectors. The metal support structure must be designed based on the bending moment and torque caused by affecting wind loads, the weight of the various components and the friction of the connecting parts. All forces are transmitted to the ground through the foundation.

During the last decade large efforts have been made to improve parabolic trough collector designs, both from technical and from economical point of view. Collector dimensions have been steadily increased (as

S774Doc-676563417-363 / v0.2 137 CST Value Proposition
depicted below), while reducing the number of required elements (e.g., drives, foundations, etc.), optimising the support structure, increasing the optical efficiency, and optimising manufacturing.

EuroTrough HelioTrough UltimateTrough

Figure A-3: Parabolic trough collector development (source Flabeg)

The absorber tube, the so-called heat collection element (HCE) is a key element of a solar collector. At the
HCE, solar irradiation is absorbed and transferred as thermal energy to the heat transfer fluid. The receiver tube, which has a high-tech selective coating, is located inside an evacuated glass envelope to reduce convective heat losses. Since the glass envelope and the steel tube have different thermal expansion coefficients, both need to be joined via a flexible connection.

All currently operating parabolic trough power plants use highly reflective glass-metal parabolic-shaped mirror panels. Due to its low content of iron oxide, the solar mirror glass is highly transparent. On the back of the thin glass panel, the mirror is chemically coated with a highly reflective silver layer. The silver layer is protected from corrosion by a chemically deposited copper film.

To guarantee the best absorption of the reflected solar radiation, a high-precision drive system is needed. Most of the drive systems (positioned in the centre of each SCA) use hydraulic-cylinder actuation which rotates the SCEs mounted to the drive system. The rotation of the SCA is controlled by a Local
Controller (LOC), which is mounted at the drive pylon in the middle of each SCA and which is connected to the hydraulic unit and the relevant sensors by power and signal cables. The solar field operates as a unit under the control of the Field Supervisory Control (FSC), a computer located in the central control room that communicates with each SCA via LOCs and with the plant's distributed control system.

The state-of-the-art parabolic trough power plant is operating with a synthetic (thermal) oil as HTF, for which there is a commercial track record of around 30 years. For the transportation and treatment of the thermal oil a so-called HTF system is required. Operation at high temperatures results in a decomposition to volatile and heavy chemical components. Therefore, an HTF treatment system is used to separate the decomposition components and to guarantee a sufficient quality of HTF during the whole plant lifetime.

The current HTF is a eutectic mixture of 26.5% diphenyl and 73.5% diphenyl oxide. The solidification temperature of this HTF is 12°C which means that in most locations a freeze protection system is required. Further, the maximum operation temperature is limited to about 393°C. Thus, new concepts are

S774Doc-676563417-363 / v0.2 138 CST Value Proposition
under development, using either silicone based HTF to allow for a wider operating temperatures range
(e.g., Helisol with -40 to 425°C) or directly molten salt as HTF to allow for a direct molten salt storage combined with higher operating temperatures.

Thermal energy storage system
The state-of-the-art thermal energy storage system for parabolic trough power plants is the in-direct two tank molten salt storage, using a eutectic mixture of inorganic nitrates consisting of 60% of sodium nitrate (NaNO3) and 40% of potassium nitrate (KNO3). The salts offer a very favourable combination of availability, density, specific heat, low chemical reactivity, and vapour pressure. Other advantages are its low costs and its harmlessness for the environment. However, the high freezing point of 221°C requires a complex freeze protection system, in particular for line-focusing systems.

In order to store thermal energy, hot HTF from the solar field (393°C) is diverted to the oil-to-salt heat exchangers, where its thermal energy passes to the salt flow coming from the cold tank. It receives the thermal energy of the thermal fluid to heat up and accumulate in the hot tank. In the heat exchanger, the salt is heated up to the hot tank temperature. During night-time or times of reduced radiation, the charging process is reversed and salt from the hot tank is pumped to the heat exchanger, where the salt returns its thermal energy to the cold thermal fluid arriving from the solar steam generator. The thermal fluid heats up to keep producing steam for the turbine, while the cooled salt accumulates again in the cold tank. At both temperature levels, the salt is in its liquid state. The storage technology is fully proven at several power plants which are already in commercial operation since more than 10 years.

Besides the requirement of additional heat exchangers and the fact that during each heat transfer at the
HTF-salt heat exchangers the process temperature decreases, resulting in lower power block efficiency during storage discharge mode, a drawback of the indirect molten salt storage for parabolic trough power plants is its relatively low operating temperature. Due to the temperature limitation of the synthetic oil, i.e., the limited hot molten salt temperature, there is only a temperature difference of 100 K between the hot and cold tank, thus, much more molten salt is required compared to a direct molten salt storage system operated at a higher temperature, which makes the indirect molten salt storage more costly.

Power Block:
In principle, the power cycle of a parabolic trough power plant (as the case for most other CST power plants) is the same as in conventional steam power plants. The conventional part is extended by the so- called solar steam generators (SSG). Basically, the SSG consists, depending on the parameters of the steam, of shell and tube heat exchangers for preheating, evaporation, superheating and generally for re- heating. Depending on the power cycle size, several SSG trains are installed in parallel. The power cycle contains the following major mechanical elements:

▪ Steam turbine, steam turbine generator and steam turbine auxiliaries.
▪ Solar steam generation unit;
▪ Steam turbine condenser, dry or wet cooled;
▪ Feed water/Condensate system, incl. HP and LP pre-heaters.

A1.2 Solar Tower

S774Doc-676563417-363 / v0.2 139 CST Value Proposition
In solar tower (central receiver) power plants, a field of heliostats (large individually two-axis tracking mirrors) is used to concentrate sunlight onto a central receiver mounted at the top of a tower. The field of heliostats, which all move independently of another, can either surround the tower or be spread out on the shadow side of the tower, in the case of smaller systems. Further, there are also multi-tower systems, such as the one developed by Vast Solar, in which multiple tower are connected to a single power block. Due to the high concentration ratios, high temperatures and, hence, higher efficiencies can be reached with central receiver systems, when compared to line focusing systems.

Within the receiver a heat transfer fluid absorbs the highly concentrated radiation reflected by the heliostats and converts it into thermal energy to be used in a conventional power cycle. For solar tower plants typically a Rankine steam turbine cycle is used but there is also the option to connect it to Brayton gas turbine cycle, depending on the applied heat transfer fluid and the receiver concept, respectively.
Initial solar tower developments have focused mainly on four solar tower (central receiver) concepts:

▪ Water/steam (direct steam) solar tower (Rankine cycle);
▪ Molten salt solar tower (Rankine cycle);
▪ Atmospheric air solar tower (Rankine cycle); and
▪ Pressurised air solar tower (Brayton cycle).

In addition, there are advanced HTF options, including particles and liquid metals (mainly sodium), developed with the aim to lift the operating temperature.

Out of the different solar tower concepts, at first the direct steam concept has been commercialised, before the molten salt solar tower concept got the dominating concept. In 2007, the first commercial central receiver power plant started operation in Spain - the PS-10 power plant with a capacity of
11 MWe using water / (saturated) steam as HTF. Following PS-10, a number of projects with superheated steam generation got realised, the largest of which is the 393 MWe Ivanpah projects (3 units), located in
California.

The first commercial solar tower plant using molten salt as HTF started operation in 2011 in Spain. The
Gemasolar plant has a nameplate capacity of 20 MWe and is equipped with a direct two tank molten salt storage system with 650 MWht (around 15 FLH storage capacity). The next molten salt projects realised had already a capacity of 100 MW per unit and have been equipped with large TES systems of several
GWh capacity. Due to the higher HTF and storage temperatures, the first large-scale commercial tower plants, such as Crescent Dunes, experienced some failures in the molten salt system and the hot-salt tanks (NREL, 2020). The issues are now known and are addressed in new CST projects using molten salt as HTF and storage media.

As the direct steam solar tower is lacking a competitive large scale storage solution, the molten salt solar tower, with its favourable direct molten salt storage, is now the dominating solar tower technology. Thus, the following description focuses on the molten salt solar tower concept.

Molten salt solar tower
In a molten salt solar tower power plant the cold salt (290°C) is pumped from the cold tank at ground level to the receiver, which is mounted on top of the tower, where the salt is heated up to around 565°C by the concentrated sunlight. The hot salt then flows back to ground level into a second tank. In order to

S774Doc-676563417-363 / v0.2 140 CST Value Proposition
generate power, the accumulated hot salt is pumped from the hot tank through a solar steam generator to generate superheated steam which powers a conventional Rankine cycle steam turbine. The solar field is oversized to collect more heat than demanded by the steam generator system and the excess thermal energy can be accumulated in the hot storage tank, acting as buffer between heat collection and power generation. A schematic of a molten salt solar tower plant with direct molten salt storage is depicted below.

Central Receiver / Solar Steam Steam Turbine
Heliostat Field Generator Generator Grid

Evaporator
Superheater

Reheater
G

Preheater
1 3 4 5 6

2 Steam
Extractions

1

2

3
HOT TANK
4 5 6

COLD TANK

Figure A-4: Schematic of molten salt solar tower plant with direct storage system

Molten salt combines the benefits of being both an excellent high temperature heat transfer fluid and an affordable high temperature energy storage fluid. The biggest advantages of the direct two-tank molten salt storage system besides the fact that there is no need for expensive heat exchangers between the HTF and the storage fluid, is the high temperature range of the storage system which is nearly 300 K. Thus, for the same storage capacity there is only about one third of salt inventory required in direct systems for solar tower applications compared to the (indirect) storage system, so far applied in parabolic trough power plants, which have only a temperature range of 100 K in the storage system, due to the temperature limitation of the thermal oil. Hence the specific storage costs are significantly lower for the solar tower, making it attractive for CSP projects with long-term storage and high capacity factors.
Further, the operating temperature is increased, enabling higher power cycle efficiencies.

Considering the standard solar salt (60% of sodium nitrate - NaNO3 and 40% of potassium nitrate -
KNO3), theoretical operating temperatures of up to 600°C are feasible, without imposing severe corrosion problems, if used in conjunction with stainless steel MS piping, valves and fittings and as long as impurities, in particular with regard to chlorides (e.g. sodium chloride), are kept low. The main challenge is to avoid freezing of the salt in any of the valves and piping of the receiver-, storage- and steam generation system, given the high freezing temperature (240°C).

The main systems of a solar tower power plant are:

▪ Heliostat field
▪ Central receiver and solar tower
▪ Thermal energy storage system; and
▪ Power cycle, incl. solar steam generator.

S774Doc-676563417-363 / v0.2 141 CST Value Proposition
Heliostat field
The solar field of a solar tower power plant consists of many heliostats, tracking the sun in two axes and reflecting it on the central receiver. The state-of-the-art heliostat design is the glass/metal heliostat design, which consists of one or more flat rectangular float glass mirrors facets, between 1 and 10 m², supported by steel structures. Depending on the supplier’s design, the mirror facets are fastened to metal arms/trusses or mounted in open frames. These in turn are fastened to a torque tube or a space frame.
The torque tube or space frame is finally fastened to a pedestal which is fixed to the ground via reinforced concrete foundation or to a pylon which is directly inserted into the ground. The azimuth/elevation gear drive is affixed to the pedestal and the torque tube to achieve two axis tracking.
Mirror facets are often spherically bent, depending upon their focal distance, distance to the receiver, and are adjusted on the metal arms in such a way that they overlay in a common area on the receiver, also known as canting.

In general, a distinction can be drawn between small to medium scale (few m² to around 50 m²) and large-scale glass/metal heliostat designs with more than 100 m² per heliostat. New heliostat generations are further equipped with remote PV power supply, a small storage (e.g. super capacitor) and the communication is done wireless, thus, cabling in the solar field can be minimised.

Figure A-5: Example of small and large scale glass/metal heliostats (source: SENER, SBP, BSE, Vast Solar, eSolar)

Heliostat fields can be laid out in two different ways depending on the site location (latitude), size of heliostat field and conditions of the site selected. The layout options for the heliostat field are: north field configuration (south field in the southern hemisphere) or a surrounding field configuration. Further, there is the multi-tower concept, where there are several solar towers with smaller solar fields and one common power block.

The solar field efficiency depends on the specifications of the heliostats, site conditions and the heliostat arrangement. There are the following losses:

▪ Intercept factor (deviation from ideal concentration), incl. mirror quality, contour errors, tracking
accuracy and optical aberration.
▪ Cosine effect losses (largest loss factor): In order to achieve reflection the surface of the heliostats face
halfway between the sun and the receiver causing a reduction of the effective reflective surface
depending upon the sun position as a function of the incidence angle.

S774Doc-676563417-363 / v0.2 142 CST Value Proposition
▪ Blocking and shading losses: Losses due to partial blocking of solar irradiation by a heliostat between
the sun and the heliostat as well as due to a heliostat blocking reflected irradiation between another
heliostat and towards the receiver. These losses are minimised by an optimised solar field layout.
▪ Reflection losses: Losses due to reflectivity of the reflective surface selected.
▪ Atmospheric attenuation: Losses of the reflected irradiation on its way to the receiver due to
absorption and dispersion caused by particulates in the atmosphere. Depends upon the distance to
the receiver and turbidity factor, which can vary significantly, e.g. coastal haze at the Gulf coast vs. dry
desert environment in deserts like the Mojave Desert.

The final arrangements of the heliostat field are the result of a techno-financial optimisation of the solar field through sophisticated algorithms which consider, inter alia:

▪ the losses mentioned above;
▪ the tower height;
▪ the receiver capacity / size (based on plant’s capacity and thermal storage desired);
▪ the losses associated to the receiver (i.e. spillage, radiation losses, reflection losses and convective
losses); and
▪ specific cost functions for sub-systems.

Central receiver system and solar tower:
The central receiver is the core element of each solar tower plant, where the highly concentrated solar energy is transferred to the HTF. The central receiver system has a direct impact on the design of the heliostat field and due to the attainable temperatures and pressures of the HTF in the receiver, it is decisive for the conversion efficiency of the power cycle.

Besides direct absorption particle receivers (DAR), which are under development as part of the Gen 3 program, there are only the indirect absorption receivers deployed so far. Indirect absorption receivers absorb the solar energy by means of a material, which transfers the energy to the HTF through convection. In principle, there are two different types of indirect absorption receivers, tube and volumetric receivers, out of which tube receivers are clearly dominating, being applied in molten salt receivers.

A tubular receiver system consists of vertically arranged tubes which can be composed to any number of modules. The modules are mounted on the receivers back wall on the top of the tower. These modules can be flowed through serially or parallel in order to achieve a specific performance concerning pressure drops and temperatures.

Most solar towers built so far are concrete towers (constructed with continuous pour, slip-form concrete), as it is generally the cheaper option, when compared to steel framework towers (depending on the height of the tower). Besides the central receiver, located on top of the tower, ancillary equipment, such as the surge tank, piping, valves, and instrumentation is located in the tower.

Thermal energy storage:
Depending in the central receiver concept, there are different thermal energy storage options applicable.
Besides the direct two tank molten salt storage, described above, there are other storage options, incl:

S774Doc-676563417-363 / v0.2 143 CST Value Proposition
▪ Steam accumulator storage (short term storage applied in DSG concepts)
▪ Solid media storage (mainly considered for concepts using air as HTF)
▪ Particle storage (direct particle storage).

Power block:
The power block for a solar tower power plant does not differ much from the one of parabolic trough power plants, described above. Given that higher operating parameters can be reached in central receiver concepts, live steam conditions and power cycle efficiencies are, however, increased, when compared to parabolic trough power pants.

A1.3 Linear Fresnel
Linear Fresnel reflectors (LFR) use long flat or slightly curved singly axis tracked mirror strips, mounted close to the ground, to concentrate sunlight onto a linear fixed absorber located above the mirror field.
Optionally there is a secondary reflector installed above the absorber.

Due to the linear fixed absorber, i.e. the absence of rotating joints between the collector elements, direct steam generation in the solar field was long in the focus of linear Fresnel technology providers. As there is no competitive large-scale storage solution available for direct steam generating systems, the use of molten salt as heat transfer and direct storage medium is now pursuit for stand-alone linear Fresnel power plants, requiring TES, while for other applications, such as process heat, direct steam integration in the solar field is the preferred choice.

Out of the three main CSP technologies, parabolic trough, solar tower and linear Fresnel, linear Fresnel technology is the CSP technology with the smallest track record. Besides several smaller steam augmentation and process heat projects, there are so far only a few operating linear Fresnel plants. The first commercial linear Fresnel plant is the 30 MWe PE II project (saturated steam), located in Spain, started operation in 2012. The largest plant built so far is the 125 MWe Reliance project in India
(superheated steam), which commenced commercial operation in 2014.

Given LFR technology, based on direct steam generation, is lacking a competitive storage solution, the
LFR technology providers started in 2013 with the development of a DMS (direct molten salt) solution. I.e.
the direct use of molten salt as heat transfer fluid and storage fluid. Following the success demonstration in pilot plants, the DMS LFR technology has been deployed in three smaller projects in Italy as well as a
50 MW project in China. A schematic of a DMS linear Fresnel plant is depicted below. Besides the solar field the sub-systems are the same as for a direct molten salt solar tower.

S774Doc-676563417-363 / v0.2 144 CST Value Proposition
Linear-Fresnel- Field Solar Steam Steam Turbine
Grid
Generator Generator

Superheater

Reheater
Evaporator
Thermal Energy G
Storage System

Preheater
1 3 4 5

2
Steam
Extractions
HOT TANK

1

COLD TANK
2

3

4 5

Air Cooled
Condenser

Figure A-6: Schematic of DMS linear Fresnel plant with direct storage system

S774Doc-676563417-363 / v0.2 145 CST Value Proposition
A2 CST Reference Plant
As a starting point for the conducted optimisation runs, a reference system with an installed net capacity of 140 MWenet, a Solar Multiple of 2.2, and 14 h storage has been defined. The relevant parameters and assumptions for the direct molten salt solar tower reference plant are summarised in below table, many of which are subject to optimisation as conducted as part of the project.

Table A 1: Molten salt tower reference system

Item Unit Value - CR Comments

Solar Field -
Heliostat

Heliostat type/name [-] sbp Stellio:
Pole mounted,
pentagon shape

Net reflective area [m²] 47.5 rectangular shape approximated
per heliostat

Aperture width [m] 6.97

Aperture height [m] 6.97

Number of facets [-] 10+1

Annual mean [%] 94 x 96 x 99 = 89.34 Product of reflectivity, mean
reflectivity (HFLCAL) cleanliness factor, and availability

Total beam error [mrad] 3.16 Sum for heliostat field
optimisation (slope error, tracking
error, sun shape)

Solar Field – System
definition

Field layout [-] Surround

Solar Multiple [-] 2.2 (starting point) To be optimised

Number of heliostats [-] tbd To be optimised

Solar tower

Type [-] Concrete

Number of towers [-] 1

Height [m] 230 To be optimised

Diameter [m] 15 Used for shading

Solar receiver

Receiver type [-] External, cylindrical,
tube receiver

Thermal power @DP [MWt] 720 Based on 150 MW P_e_gross,
gross = 43%,
to be optimized

S774Doc-676563417-363 / v0.2 146 CST Value Proposition
Item Unit Value - CR Comments

Receiver efficiency [-] ~0.89 To be optimised
(vWind=0m/s)

HTF inlet / outlet [°C] 290 / 565
temperature

Mean flux density [kW/m²] 575 Reference is incident flux
@DP

Storage system

Storage type [-] two-tank sensible

Storage medium [-] Solar Salt 60% NaNO3 + 40%KNO3

Storage capacity in [h] 14 To be optimised
full load hours

Total effective [MWht] ~4,667 To be optimised
capacity

Temperature hot / [°C] 565 / 290
cold tank

Heat losses [%ofCap 1.0
/24h]

Power block

Design net electrical [MWe] 140
Power

Design gross [MWe] 150
electrical Power

Design net efficiency [%] ~42 Depending on operating mode

Cooling type [-] ACC

Minimum / [%] 15 / 100
maximum thermal
load

Life steam [°C / 560 / 165
parameters bar]

Reheat steam [°C] 560
temperature

S774Doc-676563417-363 / v0.2 147 CST Value Proposition
Appendix B OpenCEM Overview
OpenCEM is a capacity expansion and dispatch model that simulates the NEM under a set of technical, cost and policy assumptions. Based on those assumptions, OpenCEM computes future capacity expansion (i.e., building large-scale generators and storage systems) and dispatch decisions over a number of years into the future that achieve a system-wide lowest annualised cost of operation.

B1 Modelling Approach
OpenCEM divides the NEM into 16 planning zones to account for differences in renewable energy resources, fuel costs, electricity demand and connection costs. Each zone contains its own list of generator and storage capacity, and aggregates plants by technology in each respective zone. Wind and solar technologies in a given zone have their own hourly power output traces, building and fuel costs.

Figure B-1: NEM zones used by AEMO and replicated in OpenCEM (source AEMO)

A cost minimisation search is performed sequentially for a number of future years (every 5 years starting at 2020 for the pre-run scenarios in opencem.org.au) in which a financial year is simulated using a time- sliced approach to compute capacity decisions and then in full to compute dispatch decisions. New capacity decisions are assumed to be operational during the simulated year. The net of all existing and new capacity computed for one year is carried forward as the starting point to the next. For the first year,

S774Doc-676563417-363 / v0.2 148 CST Value Proposition
initial capacity consists of reported firm capacity by the Australian Energy Market Operator (AEMO) in
2018.

Energy can flow without restriction between all the zones in a region but notional interconnectors of fixed capacity (marked red in the figure) limit the amount of energy transmitted between regions.

By default, OpenCEM uses AEMO Integrated System Plan (ISP) 2018 data for technology and fuel costs, build limits, existing generation, electricity demand traces and renewable energy resource traces (i.e., wind and solar). For CST, OpenCEM by default uses "collector" only traces that estimate thermal output performance from a collector field. With collector only traces, CST plants in OpenCEM can be configured to feature different storage sizes.

B2 Technologies
OpenCEM considers a set of generator technologies as displayed below. There are three classes of technology: Generators, storage, and hybrids. New technologies will be added as required in future iterations of OpenCEM. Users may also add and configure other technologies to simulations.

Table B-1: Technology options included in modelling

Rene
Technology Class Fuel Dispatchable Flexible Constraints
wable

Up to 10.6 TWh of yearly
Biomass Generator Yes Yes Yes Yes generation NEM wide
(CEC 2008)

Combined Cycle Gas
Generator Yes Yes No Mid
Turbine

CCGT with Carbon
Capture and Storage Generator Yes Yes No Mid (No emissions data *])
(CCS)

Moderate penalty on
Black Coal (existing) Generator Yes Yes No Low
operating point change

Moderate penalty on
Black Coal (new operating point change,
Generator Yes Yes No Low
entrant) new entrant costs and
specs

Moderate penalty on
Black Coal with CCS Generator Yes Yes No Low operating point change,
(No emissions data *)

S774Doc-676563417-363 / v0.2 149 CST Value Proposition
Rene
Technology Class Fuel Dispatchable Flexible Constraints
wable

Steep penalty on
Brown Coal Generator Yes Yes No Lowest
operating point change

Steep penalty on
Brown Coal with CCS Generator Yes Yes No Lowest operating point change,
(No emissions data *)

Open Cycle Gas
Generator Yes Yes No Yes
Turbine

Solar PV Dual Axis
Generator No No Yes Yes
tracking

Solar PV Fixed Tilt Generator No No Yes Yes

Build limits per NEM
Solar PV Single Axis Generator No No Yes Yes
planning zone

Build limits per NEM
Wind (low) Generator No No Yes Yes
planning zone

Concentrating Solar Limited zones where
Hybrid No Yes Yes Yes
Thermal 3h storage permitted to build

Concentrating Solar Limited zones where
Hybrid No Yes Yes Yes
Thermal 6h storage permitted to build

Concentrating Solar Limited zones where
Hybrid No Yes Yes Yes
Thermal 12h storage permitted to build

Pumped Hydro Limited zones where
Storage No Yes N/A Yes
Energy Storage 3h permitted to build

Pumped Hydro Limited zones where
Storage No Yes N/A Yes
Energy Storage 6h permitted to build

Pumped Hydro Limited zones where
Storage No Yes N/A Yes
Energy Storage 12h permitted to build

Battery 1 hour Storage No Yes N/A Yes

S774Doc-676563417-363 / v0.2 150 CST Value Proposition
Rene
Technology Class Fuel Dispatchable Flexible Constraints
wable

Battery 2 hour Storage No Yes N/A Yes

Battery 3 hour Storage No Yes N/A Yes

Reciprocating Engine Generator Yes Yes No Yes

Build limits per NEM
Wind (high) Generator No No Yes Yes planning zone, separate
trace to Wind (low)

Restricted GWh per year
Hydro Generator No Yes Yes Yes to emulate 10 year
average behaviour

Moderate Penalty on
Gas thermal Generator Yes Yes No Low change of operating
point
* CCS variants not in use because of incomplete/inaccurate NTNDP data at the time of release

Generators
Generators simulate a range of technologies whose output is affected by either fuel usage or hourly traces. Each generator technology is configured by specifying the following parameters:

▪ Build costs (can be specified per year and NEM zone)
▪ Fixed Operations and Maintenance (FOM) Costs
▪ Variable Operations and Maintenance (VOM) cost
▪ Fuel Costs (can be specified per year and per NEM zone)
▪ Hourly traces (can be specified per NEM zone).

Fuel based generators can dispatch their full nameplate capacity at all times but incur fuel costs when doing so. Trace based generators can dispatch up to the product of their nameplate capacity and hourly trace value.

Storage
Storage simulates large-scale storage devices. OpenCEM manages the charges and discharge of storage capacity in each zone at every hour of dispatch calculations to run the system at the least cost. The following parameters configure storage devices:

▪ Build costs (can be specified per year and NEM zone)
▪ FOM Costs
▪ VOM cost
▪ Hours of charge

S774Doc-676563417-363 / v0.2 151 CST Value Proposition
▪ Round trip efficiency.

B2.1 Hybrid technologies
Hybrid technologies combine a trace based generator with a storage device. The storage device is similar to a storage device, but is charged by a "collector", a trace based generator whose size is proportional to nameplate capacity (usually greater). Energy stored in hybrid technologies is dispatched hourly according to the needs of the system to achieve the lowest system cost. It is possible for hybrid technologies to charge and dispatch simultaneously.

▪ Build costs (can be specified per year and NEM zone)
▪ FOM Costs
▪ VOM cost
▪ Hours of charge
▪ Collection multiple (a ratio of collected power to nameplate power)
▪ Hourly traces (can be specified per NEM zone).

B3 Cost Assumptions
OpenCEM optimises dispatch and capacity expansion decisions by seeking the lowest cost of operating the entire system in a financial year. The optimisation objective is to lower the sum of all annualised costs as described below.

B3.1 Build costs
Capacity expansion decisions incur annualised build costs calculated from the cost per MW of the technology and adjusted by a fixed charge rate. The discount rate is defined by the user and the investment lifetime of technologies is assumed to be 30 years (except for BESS at 15 years and PHES at
50 years).

B3.2 Repayment cost
Annualised capital costs for expansion decisions are carried forward into subsequent years to account for repayment of investments in previous years. They are carried forward as a single lump cost and incremented by the build cost of each simulated year.

B3.3 Operating costs
Simulations account for variable O&M costs in /MWh in proportion to their respective heat rates.
Technologies with limited flexible dispatch also incur a cost each time committed capacity needs to ramp up.

B3.4 Fixed cost
All technologies incur a fixed O&M cost in A$/MW/y for installed capacity.

S774Doc-676563417-363 / v0.2 152 CST Value Proposition
B3.5 Unserved energy cost
OpenCEM assigns a sufficiently high cost to every MWh of demand that is not satisfied by the system to force the optimisation to reduce the cost of operating the system as much as possible. Typically, this cost is set to satisfy or exceed reliability standards.

B3.6 Emission cost
If a cost for emissions in A$/kg is defined in simulations, they will be accounted for in seeking the lowest cost of operating the system and will influence both capacity and dispatch decisions.

B3.7 Shadow costs
Shadow costs are costs used only to adjust the behaviour of simulations, and do not correspond to dollar amounts for a scenario. In other words, shadow costs are a provision to avoid numerical aberrations in simulations. These costs are seldom used and are tuned to have minimal influence on results.

▪ Negligible operational costs for transmission
▪ A steep cost to prevent exogenous retirement of generation beyond existing capacity.
▪ A steep cost to prevent exogenous building of capacity beyond build limits.
▪ A steep cost to 'surplus' energy, a model relaxation to prevent infeasibility conditions on the
optimisation.

B4 Policy Constraints
OpenCEM offers the following policy constraints for scenarios:

▪ NEM wide renewable energy targets as a minimum ratio of total generation per year.
▪ NEM wide renewable energy targets as minimum generation in GWh per year.
▪ Region wide renewable energy targets as a minimum ratio of total generation per region per year.
▪ NEM wide maximum emissions, specified in MT per year.
▪ NEM wide emission costs, specified in A$/kg

These constraints either penalise or enforce hard limits that influence capacity and dispatch decisions in simulations. Model decisions seek to find the least total cost of running the system under any combination of specified policies.

B5 Operating Constraints

B5.1 Transmission constraints
Electricity transmission between NEM planning regions is modelled using a linear pipeline or "truck route" model for the topology shown below.

S774Doc-676563417-363 / v0.2 153 CST Value Proposition
Figure B-2: NEM transmission between planning regions from AEMO

Pipeline transmission constraints ensure that at each dispatch hour, the amount of energy transmitted between any two zones is less than the prescribed thermal limits for that link. Transmission capacity expansion decisions can increase the thermal limit on a given link (bidirectionally) at a cost in A$/MW/km defined for each link. The decision to upgrade a link is made simultaneously with all the other capacity expansion decisions, considering the trade-off between imports/exports of energy and local generation/storage.

All links have separate forward and reverse thermal limits with initial values as described in Roam (2013), except for new links which start with an initial capacity of zero. All links by default assume a 2% transmission loss in each direction, except for inter-regional links which also incorporate AEMO proportioning factors on the applicable direction of the link. Transmission configuration options can be found in the ZONE_INTERCONS dictionary contained in the const module.

B5.2 Operating reserve constraint
The operating reserve constraint defines a margin of minimum available capacity at each hour for each
NEM region. Reserve operating capacity is defined as the sum of:

▪ Not dispatched capacity from flexible generators (quick start)
▪ Non dispatched but committed capacity from non-flexible generators (spinning reserve)
▪ Non dispatched capacity from storage and hybrid devices, provided stored energy is available (quick
start)

S774Doc-676563417-363 / v0.2 154 CST Value Proposition
The constraint is enforced at each hour of dispatch, and the reserve operating capacity must be greater or equal than a percentage of region demand for that hour. By default, operating reserves are set at
7.5 pct of demand.

B5.3 Unserved energy constraint
In addition to unserved energy costs, capacity expansion calculations must ensure that unserved energy in each region does not exceed 0.002% of demand.

B6 Proxy Model for the WA SWIS
ITP adapted OpenCEM, to approximately mimic the SWIS by modifying input assumptions currently used to test the potential for CST in the NEM The following modifications were performed on the model:

▪ Based SWIS on NEM NSW region, and removed interconnectors and other NEM regions (i.e., SA, TAS,
QLD and VIC)
▪ NSW intra-regional links have been maintained and allowed to upgrade automatically.
▪ Scaled down demand and its components (e.g., BTM solar, EV, ESS) by a factor of 3.7 based on
historical data for NSW and the SWIS for the previous 12 months.
▪ Loaded list of existing generators for the SWIS and their planned retirement dates from publicly
available data (as a result NSW hydro is also excluded
▪ Included Draft 2023 IASR Step Change assumptions about technology cost projections for new
generation and storage projects.
▪ Included a proportion of ISP 2022 Step Change annual emissions, to set a comparable emissions
target for the SWIS.
▪ Included working assumptions for CST costs as produced by Fichtner and OpenCEM constraints for
optimal dispatch of CST, for a configuration with 20 hours of storage and a collector multiple of 3.88.
▪ Performance traces for Utility Scale PV, Rooftop PV, Wind remain as per ISP 2022 data and CST
collector only resource traces as generated by ITP for OpenCEM.

These assumptions are a limited and basic approximation of the SWIS.

B7 OpenCEM vs ISP 2022
The AEMO ISP 2022 assumptions workbook and published traces were used as inputs to OpenCEM, including:

▪ Existing fleet across the NEM, including capacity, performance, fuel costs and retirements.
▪ Committed and announced projects.
▪ Demand assumptions (OPSO, EV, rooftop PV, BTM BESS, etc.)
▪ New technology entrant types, including future build costs, fuel costs, traces, etc.
▪ Policies, including LRET, VRET, QRET, TRET and Emissions pathways.
▪ Base scenario assumptions align with the ISP 2022 Step Change Scenario
▪ Coordinated DER storage uptake predicted by other modelling.
▪ The year-by-year emissions budget.

Figure B-3 compares the uptake in capacity between the AEMO and OpenCEM results.

S774Doc-676563417-363 / v0.2 155 CST Value Proposition
Figure B-3: Capacity expansion results by technology from AEMO ISP 2022 (top) and OpenCEM using AEMO
assumptions (bottom)

It can be seen that there is strong qualitative agreement between the ISP and OpenCEM results, suggesting that within the limitations of such cost minimising capacity expansion models, both models are predicting essentially the same optimal mix of generation for a given set of input assumptions.
Effectively they are vindicating each other although this is not to say that it proves that such a combination is truly optimal.

Comparing the two, it can be observed that:

▪ The maroon and beige wedges that represent the uptake of DER BESS is an input assumption not an
outcome of either of the models. This trajectory is included by AEMO based on modelling produced
by Green Energy Markets17. So has been taken as an input also for the OpenCEM modelling.

S774Doc-676563417-363 / v0.2 156 CST Value Proposition
▪ The overall level of installed capacity in total and for each technology is close but is not exactly equal.
On the other hand, the amount of energy dispatched is such that the level of demand in each year,
which is taken as an input from AEMO, is exactly met in both cases. The differences in allocations of
capacity result from the different algorithms interacting with the nonlinear sensitivities to technology
costs and performance and well within the uncertainty within which the results of these models
should be interpreted.
▪ Neither the ISP results nor the OpenCEM modelling with the same assumptions indicate any uptake of
CST.

Table B-2 lists the differences in predicted optimal installed capacities of key technologies. Where there is no difference, this is a reflection that for that technology the installed capacity is a shared input assumption.

Table B-2: Comparison of ISP2022 and OpenCEM installed capacities for 2050.

ISP Step Change 2050 OpenCEM 2050
Technology Difference [%]
Capacity [MW] Capacity [MW]

Black Coal 0 0 0

Brown Coal 0 0 0

Mid-merit Gas 0 0 0

Peaking Gas + Liquids 9,640 3,232 -200

Hydro 7,056 7,643 8

Utility Scale Storage 15,778 14,101 -11

Coordinated DER Storage 30,637 30,637 0

Distributed Storage 14,447 14,447 0

Solar Thermal 0 0 0

Offshore Wind 0 0 0

Wind 70,473 71,880 2

Utility Scale PV 70,250 90,935 13

Distributed PV 68,593 68,593 0

Biomass (New Entrant) 0 1,348 100

S774Doc-676563417-363 / v0.2 157 CST Value Proposition
Figure B-4: Dispatched energy results by technology from AEMO ISP 2022 (top) and OpenCEM using AEMO
assumptions (bottom).

Figure B-4 compares the dispatched energy between the ISP 2022 results and OpenCEM. The results are qualitatively similar but do show some levels of difference. OpenCEM uses the OPSO_MODELLING18 variant of the ISP 2022 Step Change as the Net Demand for the NEM. Modelling results below show that demand is satisfied by each year by a mix of generation output. Excess generation is needed to cover transmission losses, auxiliary loads, and round-trip efficiency of storage. The difference in dispatched energy in Figure B-4 between the ISP results and OpenCEM reflects different outcomes predicted for charging and discharging storage and from losses in transmission and storage round trip efficiencies.
Table B-3 shows that the total utility scale generation in each year, is slightly larger than the net demand that is to be met, with the difference accounted for by transmission and storage losses.

S774Doc-676563417-363 / v0.2 158 CST Value Proposition
Table B-3: Comparison of net demand and total utility generation predicted by OpenCEM

ISP 2022 Step Change Net Demand
Year OpenCEM Utility Scale Generation [GWh]
[GWh]

2023 180,749 193,745

2030 184,272 200,292

2040 216,910 242,152

2050 261,865 288,548

In the context of an investigation of the value proposition for CST, the absence of any CST uptake in the
ISP 2022 modelling raises some fundamental questions. From the CST community's point of view there appeared to be some fundamental questions around the validity of the AEMO ISP input assumptions in
2022. These were:

▪ The 2022 capital cost estimate for the CST configuration chosen appeared anecdotally to be too high,
giving it a harder cost benefit hurdle to uptake.
▪ Only a single configuration of a CST plant (12 hrs storage with SM 2.4) was included in the technology
options, so missing the possibility that even with the same underlying cost model, other
configurations (such as longer storage duration) may satisfy the cost benefit analysis better.
▪ The cost reduction trajectory applied to CST by AEMO appeared to be much slower than other
published analysis suggested it would be and certainly slower than assumed for other technologies,
so the relative competitiveness predicted for CST declined over time.
▪ The dispatch of CST was based on an assumed generation trace in a similar manner to wind dispatch.
I. e. it was not modelled with strategically optimised dispatch as was done with gas turbines for
example and as a real CST plant would attempt to operate.

Re-running OpenCEM with a range of plausibly lower CST CAPEX models and testing other configurations, however, did not result in predicted uptake of CST at all. Further investigation revealed that what appeared to be an unrealistically low CAPEX model for battery systems appeared to be locking out CST and also to a large degree Pumped Hydro systems from uptake. The battery model appeared to be both fundamentally low, based on the 2 hour battery cost, but also have an unrealistically low incremental increase for BESS of duration.

S774Doc-676563417-363 / v0.2 159 CST Value Proposition
Figure B-5: Comparison of dispatchable renewable LCoEs using cost data from ISP 2022

Comparison of dispatchable renewable LCoEs using cost data from ISP 2022 Figure B-5 shows an analysis of integrated system LCoEs based on the methods of ITP (2018b) with the 2022 ISP cost assumptions adopted. This indicates that a PV battery system or a PV PHES system would have the same LCoE as a
CST system even at 12 hours duration. In ITP’s previous analysis the breakeven duration was just 2 hours duration, with a battery based solution very much more costly as duration was increased. If the 2022 ISP assumptions were an accurate representation, then it is easy to see why no CST uptake would ever be expected. Cost reduction trajectories into the future make the cost comparison for CST even less favourable going forward.

Rather that investigate the 2022 ISP situation further, the publication of AEMO’s draft 2023 IASR in
December 2022, provided an opportunity to re-visit analysis with more up to date numbers for other technologies. It transpires that there are considerable changes between the two data sets. One of the most striking and pertinent changes is that the underlying cost model for BESS shows a significant increase and most importantly the dependence on cost increase on storage duration seems much more realistic.

Table B-4 compares the 2022 and draft 2023 IASR cost models for key relevant technologies to illustrate these changes.

Table B-4: AEMO draft 2023 IASR cost assumptions for ISP 2024 compared to ISP 2022

AEMO 2023 for 2022-23 AEMO 2020-21 Increase
Technology
[A$/kWe] [A$/kWe] [%]

Battery storage (1 hour) 931 818 114

Battery storage (2 hours) 1,346 1,097 123

Battery storage (4 hours) 2,184 1,746 125

S774Doc-676563417-363 / v0.2 160 CST Value Proposition
AEMO 2023 for 2022-23 AEMO 2020-21 Increase
Technology
[A$/kWe] [A$/kWe] [%]

Battery storage (8 hours) 3,880 3,078 126

Biomass (small scale) 7,825 7,534 104

CCGT with CCS 4,354 4,559 96

Gas combined cycle 1,766 1,808 98

Gas open cycle (large) 943 905 104

Gas open cycle (small) 1,499 1,519 99

Pumped Hydro 8hrs 3,138 2,525 124

Pumped Hydro 24hrs 4,404 3,543 124

Pumped Hydro 48hrs 6,616 5,323 124 solar PV Large scale 1,572 1,561 101

Solar thermal (15 hours) 8,265 7,685 108

Wind - offshore (fixed) 5,682 5,985 95

Wind - onshore 2,642 2,023 131

In addition to the changes to cost models, the draft AEMO draft 2023 IASR assumptions also include a number of other important changes:

▪ The assumed configuration for the single CST option is changed from 12 hour to 15 hour SM2.4
▪ Redefine zones – different number
▪ Different list of technologies
▪ Update transmission augmentation options
▪ Update emissions pathways
▪ VPP coordinated DER now a feature…
▪ Demand forecasts
▪ Generation traces
▪ Redetermination of cost reduction trajectories for all technologies
▪ Format of workbook changed, so upload re-programmed.

S774Doc-676563417-363 / v0.2 161 CST Value Proposition
Figure B-6: Comparison of dispatchable renewable LCoEs using AEMO draft cost data for ISP2024 with addition of
new CST cost model

Checking the comparison using the dispatchable LCoE analysis in Figure B-6 shows that at 12 hours duration, the Concentrating Solar option is considerably cheaper than battery solutions and on par with pumped hydro. Even with these adjustments, questions remain as to if the new cost model for BESS in
AEMO's draft is not still a bit on the low side.

S774Doc-676563417-363 / v0.2 162 CST Value Proposition
Appendix C Sensitivity Analysis of Uptake in grid
connected Systems

C1 Impact of Configuration
For a given nameplate electrical capacity, CST plants can be configured with various amounts of thermal storage and solar field sizes (measured with the Solar Multiple). A project developer for a specific project will consider the various sources of revenue likely to be available and explore which combination of storage duration and SM offers the highest likely rate of return. This may vary from project to project depending on timing and location. For capacity expansion modelling full configuration optimisation at each time period is computationally impractical. Instead, one or more specific configurations needs to be included in the overall menu of technology choices.

Investigations looked at the issue in two ways; using one configuration at a time and determining the level of uptake in each case and allowing multiple configuration options and examining which is favoured in uptake. Figure C-1 and Figure C-2 show the results from a series of runs where a range of single configurations were tested for uptake first exploring storage hours (using a SM in each case that would minimise LCoE), then exploring Solar Multiple for a fixed 20h duration.

Figure C-1: Comparison of dispatchable renewable LCoEs using AEMO draft cost data for ISP2024 with addition of
new CST cost model

S774Doc-676563417-363 / v0.2 163 CST Value Proposition
In Figure C-1 it is seen that the predicted uptake capacity is zero when no storage is included but rises rapidly to around 5 GWe as storage duration is increased to around 20 hours, it then peaks at 30 hours. In terms of contribution to the electricity system, the volume of electricity dispatched is of more significance, this also peaks at around 30 hours duration. The driving metric behind the optimisation is the total net present cost (NPC) of system operation to 2050, this shows a progressive decline with duration with an apparent flat minimum at around 25 hours duration. This is consistent with other investigations of including a range of durations as technology options, in which OpenCEM exclusively favoured longer duration options.

This investigation of storage duration was carried out with the Solar Multiple in each case set by a formula derived such that the configuration was one at which the lowest LCoE would be obtained for that duration.

Figure C-2: Variation of Solar Multiple for CST with 20hour storage duration

Examining the variation of SM for a fixed storage duration of 20 hours gives the results in Figure C-2. In this case the uptake increases rapidly as SM is increased to 2 and then appears to peak at around 3.8, which is close to the value that offers minimum LCoE. Examining the amount of electricity dispatched suggests that this continues to grow but levels off as SM is increased further to around 5. The results however do show a level of scatter around the general trend. This is interpreted as showing the nonlinear nature of the optimised technology choice mixes and the uncertainty around the convergence to final solution. NPC declines to a flat minimum also at around SM 5. This is a surprisingly large value for SM, in excess of the minimum LCoE value of 3.88.

S774Doc-676563417-363 / v0.2 164 CST Value Proposition
An investigation with Multiple CST duration and SM configurations in the menu of technologies also results in longer durations and higher SM being chosen exclusively.

C2 Sensitivity to Cost and other Factors
A range of parameters that can be expected to influence the uptake of CST have been tested in turn.

Figure C-3 shows the changes in capacity uptake and dispatched energy between the ISP 2022 assumptions and the draft 2023 IASR assumptions. No CST capacity is introduced but extra wind replaces
PV and battery storage capacity. Under the ISP 2022 assumptions no CST is built in any year.

Figure C-3: Change in Capacity (left) and dispatched energy (right), ISP2022 vs Draft ISP2023 assumptions

Figure C-4 shows the changes in capacity uptake and dispatched energy between the AEMO draft 2023
IASR and adopting the same assumptions but changing to the new Fichtner CST cost model (with configuration unchanged at 15 hr SM 2.4). It is seen that 2 GW of CST are built and that this displaces a considerable amount of PV and wind but interestingly very little other storage. The dispatch changes mirror the capacity changes.

Figure C-4: Change in Capacity and dispatched energy from introducing new cost model vs draft ISP2024

Figure C-5 shows the changes in capacity uptake and dispatched energy between the base case and that of a 10% CAPEX reduction for CST. A further uptake of 3GW MW of CST results displacing almost 6GW of
PV, Wind, 8h battery and 24h PHES. The change in dispatched energy mirrors this.

S774Doc-676563417-363 / v0.2 165 CST Value Proposition
Figure C-5: Change in Capacity and dispatched energy from a 10% CST CAPEX reduction 20h SM 3.88.

Figure C-6 in turn shows the changes in capacity uptake and dispatched energy between the base case and that of a 10% CAPEX increase for CST. The changes are a mirrored reversal of the impact of a cost decrease. However, the capacity and dispatch decrease from CST is considerably smaller than the increase on a cost reduction. This nonlinearity appears to indicate the high value that can be attributed to having some CST in the system even at higher cost, but with the marginal value per MW then declining as more is built.

Figure C-6: Change in Capacity and dispatched energy from a 10% CST CAPEX increase 20h SM 3.88

Considering the questions that remain around the cost model for battery systems, examination of the scenario of higher battery costs is of interest, results are shown in Figure C-7. The result is less BESS but more CSP and wind.

Figure C-7: Change in Capacity and dispatched energy from increasing battery costs by 10%

S774Doc-676563417-363 / v0.2 166 CST Value Proposition
C3 Managing Construction Trajectories
The OpenCEM model does not make allowances for the realistic limits on the rate at which construction can ramp up. The default results see a sudden jump to 4GW of installed capacity in a single iteration from
2045 to 2050. Evidence from the large uptake of CST in Spain when the FIT policy was in place suggests that once the revenue available is sufficient to finance a plant, a CAGR of 40%/year is realistically achievable. If the whole global history of CST is examined since the mid-1980s, across many stop-start national policies, a CAGR of between 15 and 20% is the best fit to historical deployment.

Figure C-8 illustrates the impact of achieving 5.5 GW by 2050 using the range of plausible CAGR values.
Table C- 1 shows the impact that each of these would have on the reduction in NPC over the no CST case and the impact on emissions.

Figure C-8: Capacity trajectories to 4GW by 2050 as a function of CAGR.

S774Doc-676563417-363 / v0.2 167 CST Value Proposition
Table C- 1: Impact of CST uptake scenarios on emissions and system NPC

Scenario Emissions [MT] System NPC [A$]

ISP 2023 Forced No CST 966.70 A$253,665,370,233.89

20h no forced trajectory 957.62 A$247,332,762,701.82

20h 25% CAGR 953.48 A$247,804,495,756.30

20h 30% CAGR 955.97 A$247,594,574,376.34

20h 35% CAGR 956.29 A$247,538,072,612.10

20h 40% CAGR 956.35 A$247,490,306,329.21

There is an NPC penalty in building over an extended period rather than ‘’overnight’’. The penalty is least for a 40% CAGR however a slower initial growth and earlier start to deployment would maximise the chances for establishing a sustainable Australian CST industry. The penalty can be attributed to the higher capital cost of systems installed earlier on the global cost reduction trajectory. There is virtually no change in emissions from these cases as the impact of earlier installation of CST is less of other renewables rather than less emissions from fossil generators.

C4 Impact of optimal Dispatch Modelling

CST plants are highly dispatchable and have the flexibility to support the grid during periods of high demand, low VRE resource as well as regular operation. But as the grid transitions away from conventional generation and predictable electricity demand patterns, it becomes a challenge to predict the operation of grid integrated VRE generation, dispatchable generation and storage. Therefore, CST is modelled as a hybrid between a VRE generator and a storage technology that can charge from a resource-constrained "behind the meter” generator and can discharge to the grid as required to satisfying electricity demand in concert with dispatch options from other technologies.

Another approach, as adopted for example by AEMO for the ISP, is to pre-calculate the performance of a plant from a detailed model that considers energy collection, storage charge/discharge and electricity dispatch to achieve high levels of electricity production from the plant or other utility maximisation strategy. The pre-calculated results are then fed to a capacity expansion or integrated model approach as a resource trace, comparable to those employed to simulate PV and Wind generation.

OpenCEM was set up to simulate both operation strategies. Traces for pre-calculated CST operation have been made available by AEMO as part of the ISP 2022, for a plant configured to have a solar multiple of
2.4 and 8 hours of storage. One simulation employed a collector only trace (as published by ITP for
OpenCEM) and a CST option with the configuration, allowing the OpenCEM dispatch engine to manage collection, storage, and dispatch decisions for plant at each planning zone. Another simulation used pre- calculated traces across NEM regions, (selecting the REZ adjacent to a planning zone with a trace

S774Doc-676563417-363 / v0.2 168 CST Value Proposition
featuring the highest capacity factor) limiting hourly CST dispatch decisions to the pre-calculated performance of a plant. Other assumptions are the same between simulations and to other simulations in this study (e.g.,draft IASR 2023 CAPEX assumptions, Fichtner cost model for CST, Step change demand and so forth).

Figure C-9: Change in Capacity and dispatched energy fixed vs optimal dispatch strategy

Results show that when optimal dispatch is employed, approximately 350 MW of CST capacity is deployed in the NEM in 2050, in line with other simulations in this study, but no CST is deployed when using the pre-calculated traces.

The ability to adapt to a less predictable set of supply-demand gaps afforded by employing an optimal dispatch strategy seems to be a more robust method to estimate the value of CSP in grid connected systems and should be considered in future planning and integration studies.

C5 Other Scenarios
The Snowy 2.0 pumped hydro system is being built using taxpayer funds as a result of a unilateral federal government decision. It remains a controversial project and its cost effectiveness is very much in question. Whilst it is almost certain to be completed, Figure C-10 examines the impact of it not being built compared to its scheduled completion in 2026 (as it was while these results were prepared). In early years it is seen to simply result in more dispatch of gas fired generation. After that, in the absence of
Snowy 2.0, more wind, other pumped hydro and ultimately more CST (1.24 GW) is predicted.

Figure C-10: Change in Capacity (left) and dispatched energy (right) removing Snowy 2.0, CST at 20h SM 3.88

In the ISP modelling, coordinated DER, being household battery systems that can be centrally dispatched, are separately modelled for uptake based on assumptions around consumer behaviour. This uptake assumption has also been used as the default here. It is an interesting question to consider what happens if they are left out and the capacity expansion model can then decide to deploy extra short duration

S774Doc-676563417-363 / v0.2 169 CST Value Proposition
batteries or not. Figure C-11 shows the results of this test case. It is seen that their removal results in an uptake of BESS and PHES and an extra 1.5 GW of CST and a significant decrease in PV uptake. This is mirrored in the changes in dispatched energy.

Figure C-11: Change in Capacity and dispatched energy when coordinated DER batteries are not forced in.

Going beyond this, the results in Figure C-12, result from removal of the assumed DER batteries whilst holding emissions constant at the lower level that their presence results in. In this case there is an increase in wind and PV uptake predicted a total uptake of 2.54 GW of CST from the base case and only from 2040 onwards, some uptake of BESS that is smaller than the assumed DER systems that are removed.

Figure C-12: Change in Capacity (left) and dispatched energy (right) when coordinated DER batteries are not forced
in, but emissions are held constant.

It would appear from this that from a societal perspective, the assumed short duration DER household batteries are not an optimal choice.

OpenCEM is working with an assumed emissions reduction trajectory that goes to zero by 2050. The scenario of zero emissions by 2040 is examined in Figure C-13. The main impact is a more rapid uptake of wind and PV in earlier years. Plus, BESS and PHES and ultimately and extra 2.36 GW of CST capacity.

S774Doc-676563417-363 / v0.2 170 CST Value Proposition
Figure C-13: Change in Capacity (left) and dispatched energy (right) from zero emissions by 2040 trajectory

S774Doc-676563417-363 / v0.2 171 CST Value Proposition
Appendix D Green Fuels Data

D1 Green Fuels Data

D1.1 Overview of Green Fuels
Table D-1 shows a general overview of the inputs for the methanol synthesis process.

Table D-1: A general overview of the methanol synthesis process, including DAC, normalised to 1 kg of methanol
(Bellotti, Rivarolo, & Magistri, 2020)

Parameter Unit Value

Carbon Dioxide Production (DAC)

Electricity demand kWh/kgMeOH 1.022

Thermal energy demand kWht/kgMeOH 3.212

Air (dry) demand kgair/kgMeOH 2,330

Specific carbon dioxide output kgCO2/kgMeOH 1.46

Methanol Synthesis

Electricity demand kWh/kgMeOH 0.236

Thermal energy demand kWhth/kgMeOH 1.85

Hydrogen demand kgH2/kgMeOH 0.2

Methanol output kgMeOH 1

Carbon dioxide demand kgCO2/kgMeOH 1.46

Process efficiency (including DAC) % 56.6

Operating temperature °C 210

Operating pressure bar 80

Overall Process (Including Electrolysis)

Process efficiency % 34.6

S774Doc-676563417-363 / v0.2 172 CST Value Proposition
A technical overview of the ammonia synthesis process, including the ASU, is provided in Table D-2.

Table D-2: An overview of the ammonia synthesis process, including the ASU, with values normalised to 1 kg of
ammonia (Gilbert, Alexander, Thornley, & Brammer, 2013; Smith, Hill, & Torrente-Murciano, 2019)

Parameter Unit Value

Nitrogen Production (ASU)

Electricity demand kWh/kgNH3 90

Air (dry) demand kgair/kgNH3 1.095

Nitrogen output kgN2/kgNH3 0.827

Oxygen output kgO2/kgNH3 0.253

Ammonia Synthesis (Haber-Bosch Process)

Electricity demand kWh/kgNH3 0.51

Hydrogen demand kgH2/kgNH3 0.179

Nitrogen demand kgN2/kgNH3 0.827

Ammonia output kgNH3 1

Process efficiency % 78.3

Operating temperature °C 440

Operating pressure bar 250

Overall Process (Including Electrolysis and Air
Separation)

Process efficiency % 47.3

D1.2 Electrolyser Key Performance Indicators
Key performance indicators allow important properties of electrolysers to be compared, demonstrating significant advantages of the employment of one type over the others. The KPIs of the three electrolyser technologies are summarised in Table D-3.

S774Doc-676563417-363 / v0.2 173 CST Value Proposition
Table D-3: Summary of the KPIs of the electrolyser technologies considered in this project.

Parameter Unit ALK PEM SOEC Comments

Technological - Mature Commercial Commercial ▪ Although ALK has been used for over 100 maturity (less years, recent developments promise
mature improvements in plant flexibility.
than ALK ▪ SOEC technology is relevant for this study
and PEM) due to its high heat requirements, however,
has a lower Technology Readiness Level
(TRL) compared to PEM and ALK

Average kWh/kgH2 55 55 37 (el) ▪ When comparing different technologies and specific 6 (th) vendors, it is crucial to check the system energy boundaries and process parameters for consumption which the specific energy consumption is
specified.
▪ Fichtner suggests specifying the boundary
conditions for reference values (e.g.,
hydrogen at 30 bar and 20°C, which enables
the inclusion of compression if required).

Lower % 20 5 5 ▪ Concerning large-scale plant dynamic dynamic range, the lower dynamic range of range individual electrolysis systems is not a
dominating factor due to the typical
modular approach.
▪ Ultimately, a dynamic range between 0%
and 100% can be considered.
▪ Interaction with downstream systems (e.g.,
compression) are more relevant.

System - seconds milli- seconds ▪ All systems provide sufficient response response seconds times.

Cold start min 5 - 15 < 15 min >60 ▪ Apart from cold start times, the hot stand-
by capability allows for significant faster
response.

Stack Thousand 50 - 90 30 - 80 10 - 30 ▪ Stack lifetime refers to the degradation of lifetime hours stack efficiency and is typically a trade-off
between maintenance/replacement costs
and OPEX.
▪ Degradation is influenced by plant
operation modes (higher degradation for

S774Doc-676563417-363 / v0.2 174 CST Value Proposition
Parameter Unit ALK PEM SOEC Comments

frequent start/stop operation) and water
quality.

From this information, the key advantages and disadvantages of each technology can be summarised, as is presented in Table D-4.

Table D-4: Key advantages and disadvantages of the three electrolyser technologies considered in this project.

Electrolysis
Advantages Disadvantages
Technology

▪ Proven technology ▪ Relatively poor system response time

ALK ▪ Long stack lifetimes ▪ High maintenance

▪ Comparably low CAPEX

▪ Proven technology ▪ Relatively high specific energy
consumption (low efficiency)
▪ Fast response times, allowing for
PEM
dynamic operation ▪ Relatively high CAPEX

▪ Fast start-up

▪ Lower technology readiness level (TRL)
compared to PEM and ALK
▪ Party substitutes electricity demand ▪ Limited number of manufacturers
by thermal energy - high synergies
▪ Relatively poor system response time;
SOEC with CST
low operational flexibility
▪ Low specific energy consumption
▪ High cold start-up time due to high
(high efficiency)
operational temperatures required

▪ High CAPEX compared to PEM and ALK

The three types of electrolysers can be graphically classified according to their component sizes and technological maturity, as shown in Figure D-1.

S774Doc-676563417-363 / v0.2 175 CST Value Proposition
Technological maturity
ALK
PEM

SOEC

Typical equipment size

Figure D-1: The relative positioning of the three types of electrolyser technologies with respect to technological
maturity and physical equipment size.

In addition to the advantages and disadvantages discussed, other important factors to consider when assessing electrolysis technologies include the availability of raw manufacturing materials, and capital and operational costs. These additional KPIs are discussed in below.

Availability of materials
For the production of ALK and PEM electrolysers, key materials are required where both availability and accessibility must be considered to account for the anticipated significant increase in production over the next decade. For ALK, nickel is a key material that is used to resist the highly caustic environment of the cells, which is fortunately forecasted to be readily available for enhanced electrolyser production with limited bottlenecking (excluding the Ukraine crises, as Russia is one of the main nickel producers).

Concerning PEM electrolysers, titanium, platinum, and iridium are key materials for production, with platinum and iridium being two of the scarcest metals on Earth. Currently, approximately 1 gram of platinum is used per kilowatt in the production of PEM electrolyser stacks (IRENA, 2020). Considering a precious metal utilisation of 1 mg/cm² and a power density of 5,000 W/m² (2 A/cm²; 2.5 V), 200 kg of platinum or iridium are required for the production of 1 GW of PEM electrolysers, which is equivalent to
0.1% and 3% of the world’s platinum and iridium production, respectively. Considering a planned reduction in noble metal utilisation for electrolyser production and the potential for material recycling, no supply bottlenecks are currently forecasted.

However, the market situation for these raw materials must also be considered. The supply of critical materials in electrolysers is mostly dominated by only a few nations, with more than 90% of the world’s platinum production originating from three countries (71% South Africa; 13% Russia, and 7% Zimbabwe) and more than 90% of the world’s iridium production from only two countries (92% South Africa and 5 %
Zimbabwe) (Eynard, Georgitzikis, Wittmer, & Latunussa, 2020). Furthermore, price volatility for these rare
Earth metals is also relatively high, particularly for iridium (IRENA, 2020), which will likely result in significant price jumps for these materials as global electrolyser production rates boom.

S774Doc-676563417-363 / v0.2 176 CST Value Proposition
Capital costs
When considering and comparing capital expenditure (CAPEX) for electrolyser systems, it is important to define system boundaries and process parameters precisely, as these often differ for different vendors, especially for different technology options, such as for pressurised and ambient systems. Therefore, it is suggested to define a baseline scenario where the included subsystems, such as water treatment, rectifiers and transformers, gas upgrading, and hydrogen compression, are clearly defined with boundary conditions fixed (including output pressure of the system, input voltage, and output hydrogen purity).
Preferably, the system should be included in the scope of the OEM to reduce the number of interfaces required and to enable the OEM to optimise the system according to electrolyser-specific requirements.

Investment costs for electrolyser systems are currently in the range of US$ 1,000-1,700 /kW, with ALK at the lower end of the scale and PEM towards the high end. Almost all studies and vendors expect the price of electrolysers to reduce significantly, aligning with anticipated rapid uptake of the technologies, as well as the implementation of R&D developments. With ALK being the most mature technology, prices are expected to reduce less than what is expected for PEM. For PEM electrolysers, the greatest cost reduction potential is at the stack level, where automated production and reduction in rare and expensive materials is expected to have a major contribution (IRENA, 2020). Assembly costs will thereby benefit most from the economies of scale, with stack costs for large-scale production assumed to be dominated by the membrane, and therefore platinum, iridium, titanium, and gold. Looking at the electrolyser system as a whole, the balance of plant (BOP) contributes to more than half of the overall costs, with the power supply having the largest contribution. It is also expected that these costs will benefit less from economies of scale, with the cost fraction of BOP increasing further to approximately 70% (IRENA, 2020).
This thereby increases the focus to the identification of cost reduction potential for such components, especially the power supply and rectification equipment.

Operational costs
Operational and maintenance expenditure (OPEX) includes planned and unplanned maintenance costs, but excludes the cost of electricity, and is often given as a percentage of the investment costs.
Independent of the electrolyser technology, OPEX is typically in the range of 2 - 5% per annum of the
CAPEX, and is usually proportional to system size, as maintenance costs scale down with system size.
Despite this, the development in electrolysis technology is often closely linked to the CAPEX. For example, thicker membranes are mechanically stronger and have a longer lifetime, however, also increase the resistance to transport, decreasing efficiency. The system efficiency, measured in kWh/kg of hydrogen gas, is a result of the individual efficiencies of the cell, the stack, as well as the BOP. At the cell level, the cell voltage is the determining factor for system performance, where cell voltage is inversely proportional to cell efficiency. ALK and PEM electrolysers typically operate with approximate current densities of
0.2 - 0.8 A/cm2 and 2.0 A/cm², respectively. R&D efforts aim to increase current densities of electrolysers whilst decreasing diaphragm thicknesses, to achieve greater power densities and therefore stack efficiencies.

As for capital costs, the BOP for the electrolyser system and the system boundary conditions must be carefully considered when assessing and comparing the OPEX of competing technology types, especially when including electricity costs. For example, an electrolyser operating at atmospheric pressure requires additional compression stages to supply hydrogen at the typical pressure for ammonia plants of 30 bar, thereby requiring an additional electricity demand of approximately 2.4 kWh/kg of hydrogen produced.

S774Doc-676563417-363 / v0.2 177 CST Value Proposition
Fichtner Australia Pty Ltd

Level 28, 31 Market Street
Sydney NSW 2000
Australia
+61 419 999 300
info@fichtner.com.au
www.fichtner.com.au

S774Doc-676563417-363 / v0.2 178 CST Value Proposition