Negative Emissions Technologies (NETS): Feasibility Study - Technical Appendices

Technical annex for study that estimates the maximum Negative Emissions Technologies (NETs) potential achievable in Scotland, 2030 - 2050.


Appendix 2. LCOC analysis

Data sources

Disregarded sites

The following sites were disregarded due to a lack of data in the literature, the site is fossil fuel powered, and/or the site is located on an island. In the instance that the site is an island, the capture carbon would have to be shipped and then transported by road to one pf our chosen CO2 injection points, which is unrealistic given that the site will offer low CO2 capture potentials.

Table 40: A list of sites disregarded from our analysis
Site Owner Type of Technology Reason
North British Distillery Lothian Distillers Grain whisky Already deploys CCS
Caol Ila Diageo Malt whisky Island based
Laphroaig Beam Suntory Malt whisky
Bunnahabhain Burn Stewart Distillers (Distell International) Malt whisky
Highland Park The Edrington Group Malt whisky
Jura Whyte & Mackay (Emperador) Malt whisky
Lagavulin Diageo Malt whisky
Bowmore Beam Suntory Malt whisky
Bruichladdich Rémy Cointreau Malt whisky
Scapa Chivas Brothers Ltd. (Pernod Ricard) Malt whisky
Ardbeg The Glenmorangie Co. (LVMH) Malt whisky
Arran Isle of Arran Distillers Malt whisky
Tobermory Burn Stewart Distillers (Distell International) Malt whisky
Lagg Isle of Arran Distillers Malt whisky
Ardnahoe Hunter Laing & Co. Malt whisky
Kilchoman Kilchoman Distillery Co. Malt whisky
Harris Isle of Harris Distillers Ltd. Malt whisky
Isle of Raasay R&B Distillers Malt whisky
Abhainn Dearg Mark Tayburn Malt whisky
Lerwick Energy Recovery Plant Shetland Islands Council BECCS EfW (Heat only)
Western Isles Integrated Waste Management Facility Western Isles Waste Management BECCS AD (CHP)
Pulp Mill House[150] Pulp-tec BECCS Industry (pulp) Lack of data
Cullen[151] Robert Cullen Ltd BECCS Industry (pulp)
Sapphire Mill Fourstones Paper Mill Co Ltd BECCS Industry (paper towel) A gas-powered site. No potential for NETs

BECCS Biomethane

The following sites considered were identified using the NNFCC database and REPD.

Table 41: BECCS Biomethane data point summary. Extracted from the literature and used in the calculations for the NETs Model
Plant Owner Operational Status Type of Technology Gross Electrical Capacity (MWe) Feedstock Capacity (t/year) Biomethane produced (m3/hr) Data source
Portgordon Maltings Beyside Grissan Energy Operational Biomethane grid injection & CHP 5 837,500 800 NNFCC
Brae of Pert Farm Qila Energy Operational Biomethane grid injection & CHP 0.25 35,000 550 NNFCC
Charlesfield Industrial Estate Charlesfield First Operational Biomethane grid injection & CHP 0.249 24,995 550 NNFCC
Cumbernauld AD Shanks Operational Biomethane grid injection & CHP 3.6 100,000 495 NNFCC
Downiehills Farm Buchan Biogas Operational Biomethane grid injection & CHP 0.5 55,000 550 NNFCC
Girvan Distillery Grissan Energy Operational Biomethane grid injection & CHP 7.2 300,000 2,750 NNFCC
Glenfiddich Distillery William Grant & Sons Operational Biomethane grid injection 3.5 80,000 2,000* NNFCC
Hatton Farm AD Grissan Energy Operational Biomethane grid injection & CHP 0.5 38,000 450 NNFCC
Inchdairnie Farm Qila Energy Operational Biomethane grid injection & CHP 2 40,000 500 NNFCC
Invergordon Distillery Whyte & Mackay Operational Biomethane grid injection & CHP 0.25 36,500 500 NNFCC
Keithick Farm Keithick Biogas Operational Biomethane grid injection & CHP 0.249 36,000 605 NNFCC
Lockerbie Creamery Lockerbie Biogas Ltd Operational Biomethane grid injection 0.5 98,250 768 NNFCC
Morayhill AD Qila Energy Operational Biomethane grid injection & CHP 0.25 40,000 495 NNFCC
Peacehill Farm TD Forster & Son Operational Biomethane grid injection & CHP 0.237 30,450 550 NNFCC
Rosskeen Farm Qila Energy Operational Biomethane grid injection & CHP 0.25 36,000 450 NNFCC
Savock Farm Qila Energy Operational Biomethane grid injection & CHP 0.25 40,000 600 NNFCC
Tambowie Farm Tambowie Biogas Operational Biomethane grid injection & CHP 0.973 24,000 220 NNFCC
TECA AD Aberdeen City Council Operational Biomethane grid injection & CHP 0.35 81,012 425 NNFCC
Portgordon Maltings – Anaerobic Digestion Facility Grissan Engineering Services Limited Permission Granted Biomethane 0 0 2,000** REPD
Lockerbie Creamery – Anaerobic Digester Lockerbie Biogas Limited Permission Granted Biomethane 0 0 916*** REPD
Mains Of Boquhan - Anaerobic Digester Facility Grahams Family Dairy Permission Granted Biomethane 0 21,500 274**** REPD
Millerhill AD Biogen Operational BECCS Biomethane (grid injection & CHP) 1.5 35,000 445.6 NNFCC

* Regarding the Glenfiddich Distillery site, it is split into two installations. The first consists of two 11.5m high by 30m diameter reactors with a 3.5 MW biogas-fuelled CHP[152], and the second installation includes two 14.7m high by 28m diameter reactors[153]. Assuming a standard medium-sized anaerobic digestor tank produces around 500m3/hr of biomethane, the estimated biomethane production potential is approximately 2000 m3/hr.

** The information provided in the REPD Database regarding the operational capacity of the Portgordon Maltings site was insufficient. To obtain more comprehensive details, we conducted a thorough examination of the planning application submitted to the council.[154] Our investigation revealed that the project had been divided into two phases, which were represented separately in the REPD. The complete project consists of four primary digester plant tanks, each measuring 28m in diameter and 16.9m in height, along with three smaller feedstock tanks measuring 22.5m in diameter and 10m in height. The primary source of feedstock for the project will be waste from the malting plant and nearby distilleries. Considering that both project phases are located on the same site, we have chosen to merge them. Based on the assumption that an average biomethane tanker has a capacity of 500 m3/hr, we have determined a total site capacity of 2,000 m3/year.

*** The information provided in the REPD Database regarding the operational capacity of the Lockerbie Creamery site was insufficient. To obtain more comprehensive details, we conducted a thorough examination of the planning application submitted to the council.[155] The application for the project exhibited several variations, and according to the SEPA permit, the installation will comprise of two 1.5MWth input natural gas CHP units and inject around 54,965 MWh of biomethane into the grid. This injection rate is equivalent to ~5,496,500 m3/year, which we have utilised in our analysis.

Mathematical equation

**** The information provided in the REPD Database regarding the operational capacity of the Mains of Boquhan site was insufficient. To obtain more comprehensive details, we conducted a thorough examination of the planning application submitted to the council.[156] According to the application, the site is designated as an AD Upgrading facility utilizing the water scrubbing method. The proposed feedstock inputs for the facility include whey (20,000t), sludge from milk pasteurization (500t), manure (500t), and grass silage (500t), totalling 21,500 tonnes per year. Since the provided data only pertains to feedstock inputs, it was necessary to employ calculations specific to BECCS AD Upgrading to determine the biomethane production capacity and associated costs. This approach differs from the methodology used for other BECCS Biomethane sites.

BECCS Power

The following sites considered were identified using the REPD.

Table 42: BECCS Power data point summary. Extracted from the literature and used in the calculations for the NETs Model
Plant Owner Operational Status Type of Technology Gross Electrical Capacity (MWe) Data source
Markinch Biomass CHP Plant RWE Operational BECCS Power (CHP) 65 REPD
Stevens Croft E.ON Operational BECCS Power 50.4 REPD
Westfield Biomass Power Station EPR Scotland Operational BECCS Power 12.5 REPD
Speyside Biomass CHP Plant Speyside Renewable Energy Partnership Operational BECCS Power (CHP) 12.5 REPD
Rothes Bio-Plant Scottish Bio-Power Operational BECCS Power (CHP) 8.3 REPD
Sustainable Power and Research Campus University of St Andrews Operational BECCS Power (CHP) 6.5 REPD
Acharn Forest Killin Biomass Plant Northern Energy Developments Operational BECCS Power (CHP) 5.6 REPD
Diageo Biomass Energy Project Diageo Operational BECCS Power 5.5 REPD
Harbour Road Glennon Brothers Troon Operational BECCS Power (CHP) 2.6 REPD
Gleneagles Hotel Biomass Boiler Plant Room AMP Energy Services Limited Operational BECCS Power 1.2 REPD
Macphie of Glenbervie Macphie Ltd. Operational BECCS Power 1.2 REPD
Co-Op, Polwarth Street - Biomass boilers Gold Energy Limited Permission Granted BECCS Power 0.44 REPD
Hillhead Of Coldwells, Longhaven - Biomass Boilers Private Developer Permission Granted BECCS Power 0.26 REPD
Little Broomfield - Biomass boiler Private Developer Permission Granted BECCS Power 0.21 REPD

BECCS Industry

The sites considered were identified using the REPD, HNPD, SPRI, and CHPQA databases. The Morayhill Mill site was identified through the REPD, but site-specific data was obtained through stakeholder engagement.

Table 43: BECCS Industry data point summary. Extracted from the literature and used in the calculations for the NETs Modell
Plant Owner Operational Status Type of Technology Gross Electrical Capacity (MWe) Data source
Caledonian Papermill Caledonian Paper Operational BECCS Industry CHP (paper - coated magazine) 26 REPD and CHPQA
Cowie Biomass Facility Norbord Operational BECCS Industry CHP (Particle & MDF) 15 REPD and CHPQA
Invergordon Pellet Mill Balcas Operational BECCS Industry CHP (wood pellets) 5 REPD and CHPQA
Morayhill Mill Norbord Operational BECCS Industry (Oriented Strand Board) 100* Stakeholder engagement
Barony Road, Auchinleck Egger Barony LTD (particle) Operational BECCS Industry CHP (Chipboard and wood recycling) 5.5 HNPD
Dunbar Cement Tarmac Operational BECCS Industry (Cement) N/A SPRI and websites[114], [108]

Please note that the REPD mistakenly states that the following sites are not CHPs: Caledonian Papermill, Cowie Biomass Facility, Invergordon Pellet Mill, and Barony Road, Auchinleck. We understand that these sites are CHPs based off previous CHPQA submissions and additional information found in the literature.

No data on the heat and power usages of the Dunbar Cement site could be found in the literature; therefore, the carbon capture capacity of the site was determining using data submitted to SEPA under the SPRI Database. This in turn was used to determine the costs. The only data that was found was the production of 867 t,clinker per day[114] and that the site aims to use a fuel mixture that consists of 45% SRF[108].

Caledonian Paper Mill

This paper mill has been in production since April 1989 and has the capacity to produce 250,000 tonnes of lightweight coated paper (LWC) specifically designed for printing magazines, catalogues, and brochures[157]. The mill operates a 26 MWe CHP plant that exclusively uses 100% biomass as fuel, derived from both virgin and recycled sources, including solids sourced from a primary effluent treatment plant.

Cowie Biomass Facility

The site produces Caberfloor[158], a specially processed and compressed woodchip material. The site operates both a large biomass boiler that produces steam as well as two high-pressure natural-gas turbines to produce hot exhaust gas and to supply on-site power. The SPRI database indicates emissions with a biomass content of 72.4%, but it is unclear whether this includes sources outside of the CHP plant.

Invergordon Pellet Mill

This site produces wood pellets[159]. Again, there is limited data on site heat and power demands, with the planning application[160] and REPD mentioning that a biomass fired CHP is present onsite.

Morayhill Mill

This site manufactures Oriented Strand Board (OSB) using timber chips sourced from nearby sawmills. Any timber residue from the plant is used to fuel a biomass boiler. This includes bark stripped from the logs at the start of the manufacturing process, wood dust extracted from various production processes around the plant, along with any timber residue and non-specification timber flakes. The burner generates heat for use in the drying and curing stages in board production. The SPRI database confirmed that emissions from this site are 100% biogenic.

After some stakeholder engagement we learnt that there are two biomass burners present onsite, one 57 MWth and the other 43 MWth.

Egger Barony

This site manufactures approximately 400,000 m3 of raw chipboard per annum[161], which can then either be used in its raw form or be upgraded for use in the furniture and interior design markets or building market. In 2021 Egger stated that they wish the Barony plant to be powered 100% through a new biomass CHP (5.5 MWe output) and generate hot gas to be used in drying wood material[162]. This CHP has now been completed and features in the Heat Networks Planning Database[163], where heat is sold to an industrial customer nearby consisting of three buildings. The SPRI database confirmed that emissions from this site are 100% biogenic.

Dunbar Cement

The site produces 867 t/d of clinker with a heat consumption of 3.26 MJ/kg[114].

Dunbar cement plant by agreeing a contract with leading Scottish resource management company, Hamilton Waste and Recycling, to begin using Solid Recovered Fuel (SRF) at the plant[164]. Combined with other waste-derived fuels, this new supply of SRF at Dunbar will support our aim to replace up to 45% of its traditional fossil-based fuels with alternatives which are fully or partially classed as carbon neutral.

Disregarded sites

Sapphire Mill was initially considered, since it manufactures paper towels. However, upon further investigation, we found that the site sources heat and power demands via natural gas. This was reflected in the SPRI database which shows no mention of biogenic emissions.

The sites Pulp Mill House and Robert Cullen Ltd were all considered due their work in manufacturing moulded pulp. However, since no information was available on their site operations then they were removed from the analysis.

BECCS AD

The sites considered were identified using the NNFCC database and REPD.

Table 44: BECCS AD data point summary. Extracted from the literature and used in the calculations for the NETs Model
Plant Owner Operational Status Type of Technology Gross Electrical Capacity (MWe) Feedstock Capacity (t/year) Data source
Rainton Farm D Finlay & Son Operational BECCS AD (CHP) 0.025 2,500 NNFCC
Loanhead Farm N Poett Operational BECCS AD (CHP) 0.05 2,000 NNFCC
Carterhaugh Farm BQ Farming Partnership Operational BECCS AD (CHP) 0.195 2,000 NNFCC
Genoch Mains Farm Mr J McIntosh Operational BECCS AD (CHP) 0.225 17,500 NNFCC
Genoch Mains Farm (Extension) Mr J McIntosh Operational BECCS AD (CHP) 0.237 17,500 NNFCC
Kirkton Farm Kirkton Farm Operational BECCS AD (CHP) 0.475 2,000 NNFCC
Wester Clockeasy Farm AGTEC Operational BECCS AD (CHP) 0.125 5,000 NNFCC
Dronley Farm AD Dronley Farming Ltd Operational BECCS AD (CHP) 0.0623 3,000 NNFCC
East Reston Farm AD RH & DH Allan Operational BECCS AD (CHP) 0.076 3,500 NNFCC
Mains of Fortrie AD D Bartlet & Son Operational BECCS AD (CHP) 0.076 3,500 NNFCC
Old Ballikinrain House AD M Percy Ltd Operational BECCS AD (CHP) 0.076 3,500 NNFCC
Forthar Farm AD J&C Wilson Operational BECCS AD (CHP) 0.1275 5,000 NNFCC
Meinside AD Mein Farming Ltd Operational BECCS AD (CHP) 0.088 4,500 NNFCC
Baltier Farm Baltier Farm Operational BECCS AD (CHP) 0.5 10,000 NNFCC
Girvan Road AD AGTEC Operational BECCS AD (CHP) 0.098 7,000 NNFCC
Lemington Farm AD Greenshields Agri Ltd Operational BECCS AD (CHP) 0.18 8,000 NNFCC
Mayfield Farm PALL Operational BECCS AD (CHP) 0.2 7,000 NNFCC
Balmangan Farm Mathers Dairy Utensils Operational BECCS AD (CHP) 0.124 5,500 NNFCC
Crofthead farm W Callander Operational BECCS AD (CHP) 0.124 3,000 NNFCC
Slacks Farm D Kincaid Operational BECCS AD (CHP) 0.124 3,000 NNFCC
Standingstone Farm Mathers Dairy Utensils Operational BECCS AD (CHP) 0.124 5,500 NNFCC
East Denside Farm M Forbes Operational BECCS AD (CHP) 0.243 5000 NNFCC
Knockrivoch Farm Knockrivoch Farm Operational BECCS AD (Heat only) 0.15* 480 NNFCC
East Knockbrex AD Iain Service & Co Ltd Operational BECCS AD (CHP) 0.154 12,800 NNFCC
Littleton Farm (2) Mathers Dairy Utensils Operational BECCS AD (CHP) 0.19 5,500 NNFCC
Harpers Transport AD Harpers Transport Operational BECCS AD (CHP) 0.197 10,000 NNFCC
Ignis Wick AD Ignis Wick Ltd Operational BECCS AD (CHP) 0.197 6,000 NNFCC
Balmachie Farm AD JF Lascelles Operational BECCS AD (CHP) 0.086 4,000 NNFCC
Slains Park Farm J Forbes Operational BECCS AD (CHP) 0.399 8,000 NNFCC
Standhill Farm JG Shanks & Son Operational BECCS AD (CHP) 0.185 11,000 NNFCC
Woodside Farm AGTEC Operational BECCS AD (CHP) 0.1792 5,228 NNFCC
Balmenach Distillery Inver House Distillers Operational BECCS AD (CHP) 0.25 5,000 NNFCC
Auchencheyne AD Auchencheyne Ltd Operational BECCS AD (CHP) 0.1275 5,000 NNFCC
Girvan Mains Farm AB Young Operational BECCS AD (CHP) 0.238 8,000 NNFCC
Allerbeck Farm Wyseby Hill Ltd Operational BECCS AD (CHP) 0.093 5,770 NNFCC
Camieston Farm AD Camieston Renewables Ltd Operational BECCS AD (CHP) 0.485 18,000 NNFCC
Kinknockie Farm Yorston & Sinclair Operational BECCS AD (CHP) 0.457 9,500 NNFCC
Gask Farm J Rennie & Son Operational BECCS AD (CHP) 0.46 15,000 NNFCC
Broadwigg Farm N Forsyth & Son Operational BECCS AD (CHP) 0.465 28,000 NNFCC
North British Distillery AD North British Distillery Operational BECCS AD (CHP) 0.479 9,855 NNFCC
Bendochy Farm ET Bioenergy Operational BECCS AD (CHP) 0.44 9,500 NNFCC
Claylands Farm Strathendrick Biogas Operational BECCS AD (CHP) 0.499 30,000 NNFCC
Dailuaine Distillery Diageo Operational BECCS AD (CHP) 0.5 15,000 NNFCC
Edge Farm Composting GP Green Recycling Operational BECCS AD (CHP) 0.5 12,000 NNFCC
Glendullan Distillery Diageo Operational BECCS AD (CHP) 0.5 15,000 NNFCC
Levenseat Recycling facility Levenseat Operational BECCS AD (CHP) 0.5 25,000 NNFCC
Pure Malt Products Pure Malt Products Operational BECCS AD (CHP) 0.5 25,000 NNFCC
Roseisle Speyside Whisky Distillery Diageo Operational BECCS AD (CHP) 0.5 47,450 NNFCC
Charlesfield Farm Hoddom & Kinmount Estates Operational BECCS AD (CHP) 0.475 11,200 NNFCC
Glenmorangie Distillery Glenmorangie Operational BECCS AD (Heat only) 940 0 NNFCC
Wester Alves Farm Wester Alves Biogas Operational BECCS AD (CHP) 0.8 25,000 NNFCC
Wester Kerrowgair Farm Qila Energy Operational BECCS AD (CHP) 0.45 20,650 NNFCC
GSK Irvine GlaxoSmithKline Operational BECCS AD (CHP) 0.98 10,000 NNFCC
Deerdykes Composting and Organics Recycling Facility Scottish Water Horizons Operational BECCS AD (CHP) 1 30,000 NNFCC
Auchentoshan Distillery Morrison Bowmore Distillers Operational BECCS AD (CHP) 0.5 20,000 NNFCC
West Roucan Farm J Cunnigham-Jardine Operational BECCS AD (CHP) 0.95 20,000 NNFCC
Lochhead Landfill (Dry-AD) Fife Council Operational BECCS AD (CHP) 1.14 45,000 NNFCC
Binn Farm AD TEG Biogas Operational BECCS AD (CHP) 1.4 30,000 NNFCC
Barkip AD SSE Operational BECCS AD (CHP) 2.2 75,000 NNFCC
Charlesfield Industrial Estate (2) Iona Capital Operational BECCS AD (CHP) 3 36,000 NNFCC
Glenfiddich Distillery AD (Extension) William Grant and Sons Distillers Permission Granted BECCS AD (CHP) 2 0 REPD
Skeddoway Farm RM Brown & Son Operational BECCS AD (CHP) 2 0 NNFCC
Energen biogas Cumbernauld Bio Capital Limited Operational BECCS AD (CHP) 2.4 0 REPD
Balmcassie Commercial Park - Anaerobic digestion plant Brewdog Limited Under Construction BECCS AD (CHP) 3.5 0 REPD
Academy Road - Energy Centre & Anaerobic Digestion Facility Grissan Engineering Services Limited Permission Granted BECCS AD (CHP) 4 0 REPD
Glasgow Renewable Energy and Recycling Centre Viridor Operational BECCS AD (CHP) 4 100,000 NNFCC
Cameron Bridge Distillery Diageo Operational BECCS AD (CHP) 5.5 90,000 NNFCC

*Heat only site so units are kWth

BECCS Fermentation

This data from whisky distilleries was taken from Whisky Invest Direct[165], [166] and for beer producing sites from the Scottish Carbon Capture Storage (SCCS)[92].

Table 45: BECCS Fermentation data point summary. Extracted from the literature and used in the calculations for the NETs Model
Plant Owner Operational Status Type of Technology Alcohol production Capacity (MLPA) Data source
Cameronbridge Diageo Operational Grain whisky 110 Whisky Invest Direct
Girvan William Grant & Sons Operational Grain whisky 110 Whisky Invest Direct
Invergordon Whyte & MacKay Operational Grain whisky 36 Whisky Invest Direct
Strathclyde Chivas Brothers Operational Grain whisky 39 Whisky Invest Direct
Starlaw/Glen Turner Distillery La Martiniquaise Operational Grain whisky 25 Whisky Invest Direct
Loch Lomond (Grain) Loch Lomond Group Operational Grain whisky 18 Whisky Invest Direct
Reivers Mossburn Distillery Co. Operational Grain whisky 0.1 Whisky Invest Direct
Glenlivet Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 21 Whisky Invest Direct
Glenfiddich William Grant & Sons Operational Malt whisky 21 Whisky Invest Direct
Macallan The Edrington Group Operational Malt whisky 15 Whisky Invest Direct
Ailsa Bay William Grant & Sons Operational Malt whisky 12 Whisky Invest Direct
Glen Ord Diageo Operational Malt whisky 11.5 Whisky Invest Direct
Roseisle Diageo Operational Malt whisky 10.8 Whisky Invest Direct
Dalmunach Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 10.5 Whisky Invest Direct
Teaninich Diageo Operational Malt whisky 10.2 Whisky Invest Direct
Glenmorangie The Glenmorangie Co. (LVMH) Operational Malt whisky 6.5 Whisky Invest Direct
Glen Grant Campari Group Operational Malt whisky 6.1 Whisky Invest Direct
Glen Moray Glen Turner (La Martiniquaise) Operational Malt whisky 6 Whisky Invest Direct
Dufftown Diageo Operational Malt whisky 5.9 Whisky Invest Direct
Miltonduff Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 5.8 Whisky Invest Direct
Glen Keith Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 5.7 Whisky Invest Direct
Auchroisk Diageo Operational Malt whisky 5.7 Whisky Invest Direct
Balvenie William Grant & Sons Operational Malt whisky 5.6 Whisky Invest Direct
Glenrothes The Edrington Group Operational Malt whisky 5.5 Whisky Invest Direct
Tomatin Tomatin Distillery Co. Operational Malt whisky 5 Whisky Invest Direct
Ardmore Beam Suntory Operational Malt whisky 4.9 Whisky Invest Direct
Tormore Elixir Distillers Operational Malt whisky 4.9 Whisky Invest Direct
Dailuaine Diageo Operational Malt whisky 4.9 Whisky Invest Direct
Loch Lomond (Malt) Loch Lomond Group Operational Malt whisky 4.75 Whisky Invest Direct
Longmorn Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 4.7 Whisky Invest Direct
Clynelish Diageo Operational Malt whisky 4.7 Whisky Invest Direct
Allt-a-Bhainne Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 4.5 Whisky Invest Direct
Braeval Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 4.5 Whisky Invest Direct
Kininvie William Grant & Sons Operational Malt whisky 4.4 Whisky Invest Direct
Glenburgie Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 4.3 Whisky Invest Direct
Dalmore Whyte & Mackay (Emperador) Operational Malt whisky 4.3 Whisky Invest Direct
Speyburn Inver House Distillers (Thai Beverages plc) Operational Malt whisky 4.2 Whisky Invest Direct
Craigellachie John Dewar & Sons (Bacardi) Operational Malt whisky 4.2 Whisky Invest Direct
Tamnavulin Whyte & Mackay (Emperador) Operational Malt whisky 4.2 Whisky Invest Direct
Glentauchers Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 4.1 Whisky Invest Direct
Royal Brackla John Dewar & Sons (Bacardi) Operational Malt whisky 4.1 Whisky Invest Direct
Tamdhu Ian Macleod Distillers Operational Malt whisky 4 Whisky Invest Direct
Glenfarclas J. & G. Grant Operational Malt whisky 4 Whisky Invest Direct
Glenallachie The Glenallachie Distillers Co. Operational Malt whisky 4 Whisky Invest Direct
Aberlour Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 3.8 Whisky Invest Direct
Mortlach Diageo Operational Malt whisky 3.8 Whisky Invest Direct
Linkwood Diageo Operational Malt whisky 3.7 Whisky Invest Direct
Benrinnes Diageo Operational Malt whisky 3.6 Whisky Invest Direct
Glendullan Diageo Operational Malt whisky 3.6 Whisky Invest Direct
Macduff [Glen Deveron] John Dewar & Sons (Bacardi) Operational Malt whisky 3.4 Whisky Invest Direct
Tomintoul Angus Dundee Distillers Operational Malt whisky 3.3 Whisky Invest Direct
Cardhu Diageo Operational Malt whisky 3.3 Whisky Invest Direct
Aberfeldy John Dewar & Sons (Bacardi) Operational Malt whisky 3.3 Whisky Invest Direct
Laphroaig Beam Suntory Operational Malt whisky 3.275 Whisky Invest Direct
Inchgower Diageo Operational Malt whisky 3.2 Whisky Invest Direct
Aultmore John Dewar & Sons (Bacardi) Operational Malt whisky 3.2 Whisky Invest Direct
Talisker Diageo Operational Malt whisky 3 Whisky Invest Direct
Tullibardine Picard Vins & Spiriteaux Operational Malt whisky 2.9 Whisky Invest Direct
BenRiach Benriach Distillery Co. (Brown Forman) Operational Malt whisky 2.8 Whisky Invest Direct
Glenlossie Diageo Operational Malt whisky 2.8 Whisky Invest Direct
Balmenach Inver House Distillers (Thai Beverages plc) Operational Malt whisky 2.8 Whisky Invest Direct
Deanston Burn Stewart Distillers (Distell International) Operational Malt whisky 2.7 Whisky Invest Direct
Glen Elgin Diageo Operational Malt whisky 2.6 Whisky Invest Direct
Mannochmore Diageo Operational Malt whisky 2.6 Whisky Invest Direct
Strathisla Chivas Brothers Ltd. (Pernod Ricard) Operational Malt whisky 2.5 Whisky Invest Direct
Blair Athol Diageo Operational Malt whisky 2.5 Whisky Invest Direct
Glenkinchie Diageo Operational Malt whisky 2.5 Whisky Invest Direct
Fettercairn Whyte & Mackay (Emperador) Operational Malt whisky 2.3 Whisky Invest Direct
Cragganmore Diageo Operational Malt whisky 2.2 Whisky Invest Direct
Dalwhinnie Diageo Operational Malt whisky 2.2 Whisky Invest Direct
Auchentoshan Beam Suntory Operational Malt whisky 2.15 Whisky Invest Direct
Ben Nevis Ben Nevis Distillery Ltd (Nikka, Asahi Breweries) Operational Malt whisky 2 Whisky Invest Direct
Strathmill Diageo Operational Malt whisky 2 Whisky Invest Direct
Inchdairnie Inchdairnie Distillery Ltd Operational Malt whisky 2 Whisky Invest Direct
Glendronach Benriach Distillery Co. (Brown Forman) Operational Malt whisky 1.8 Whisky Invest Direct
Balblair Inver House Distillers (Thai Beverages plc) Operational Malt whisky 1.8 Whisky Invest Direct
Knockdhu [AnCnoc] Inver House Distillers (Thai Beverages plc) Operational Malt whisky 1.8 Whisky Invest Direct
Borders The Three Stills Co. Ltd Operational Malt whisky 1.8 Whisky Invest Direct
Glen Spey Diageo Operational Malt whisky 1.6 Whisky Invest Direct
Bladnoch Bladnoch Distillery Ltd Operational Malt whisky 1.5 Whisky Invest Direct
Glencadam Angus Dundee Distillers Operational Malt whisky 1.4 Whisky Invest Direct
Knockando Diageo Operational Malt whisky 1.4 Whisky Invest Direct
Pulteney Inver House Distillers (Thai Beverages plc) Operational Malt whisky 1.4 Whisky Invest Direct
Glen Garioch Beam Suntory Operational Malt whisky 1.3 Whisky Invest Direct
Glengoyne Ian Macleod Distillers Operational Malt whisky 1.1 Whisky Invest Direct
Glenglassaugh Benriach Distillery Co. (Brown Forman) Operational Malt whisky 1 Whisky Invest Direct
Ardross Greenwood Distillers Operational Malt whisky 1 Whisky Invest Direct
Oban Diageo Operational Malt whisky 0.8 Whisky Invest Direct
Brora/Clynelish Distillery Diageo Operational Malt whisky 0.8 Whisky Invest Direct
Falkirk Falkirk Distilling Co. Operational Malt whisky 0.75 Whisky Invest Direct
Glengyle J & A Mitchell Operational Malt whisky 0.75 Whisky Invest Direct
Springbank J & A Mitchell Operational Malt whisky 0.75 Whisky Invest Direct
Glen Scotia Loch Lomond Group Operational Malt whisky 0.75 Whisky Invest Direct
Aberargie The Perth Distilling Co. Operational Malt whisky 0.75 Whisky Invest Direct
Burn o'Bennie Mike Bain & Liam Pennycook Operational Malt whisky 0.69 Whisky Invest Direct
Speyside Speyside Distillers Co. Operational Malt whisky 0.6 Whisky Invest Direct
Royal Lochnagar Diageo Operational Malt whisky 0.5 Whisky Invest Direct
Benromach Gordon & MacPhail Operational Malt whisky 0.5 Whisky Invest Direct
Bonnington John Crabbie & Co. Operational Malt whisky 0.5 Whisky Invest Direct
Glenturret Lalique Group Operational Malt whisky 0.5 Whisky Invest Direct
Clydeside Morrison Glasgow Distillers Operational Malt whisky 0.5 Whisky Invest Direct
Torabhaig Mossburn Distillers Operational Malt whisky 0.5 Whisky Invest Direct
Ardnamurchan Adelphi Distillery Ltd Operational Malt whisky 0.45 Whisky Invest Direct
Glasgow The Glasgow Distillery Co. Operational Malt whisky 0.44 Whisky Invest Direct
Eden Mill Paul Miller Operational Malt whisky 0.3 Whisky Invest Direct
Edradour No.2 Signatory Vintage Scotch Whisky Co. Ltd Operational Malt whisky 0.27 Whisky Invest Direct
Annandale Annandale Distillery Co. Operational Malt whisky 0.26 Whisky Invest Direct
Arbikie Arbikie Distilling Ltd Operational Malt whisky 0.25 Whisky Invest Direct
Holyrood Holyrood Distillery Ltd. Operational Malt whisky 0.25 Whisky Invest Direct
Lindores Abbey The Lindores Distilling Co. Operational Malt whisky 0.25 Whisky Invest Direct
Kingsbarns Wemyss Vintage Malts Operational Malt whisky 0.205 Whisky Invest Direct
Lone Wolf Brewdog plc. Operational Malt whisky 0.2 Whisky Invest Direct
Lochlea Lochlea Distilling Co. Operational Malt whisky 0.18 Whisky Invest Direct
Wolfburn Aurora Brewing Ltd Operational Malt whisky 0.175 Whisky Invest Direct
GlenWyvis GlenWyvis Distillery Ltd Operational Malt whisky 0.15 Whisky Invest Direct
Strathearn Douglas Laing Operational Malt whisky 0.14 Whisky Invest Direct
Edradour Signatory Vintage Scotch Whisky Co. Ltd Operational Malt whisky 0.135 Whisky Invest Direct
Nc'nean Drimnin Distillery Co. Operational Malt whisky 0.1 Whisky Invest Direct
Ballindalloch MacPherson-Grant Operational Malt whisky 0.1 Whisky Invest Direct
Daftmill Francis Cuthbert Operational Malt whisky 0.065 Whisky Invest Direct
Dornoch Phil & Simon Thompson Operational Malt whisky 0.025 Whisky Invest Direct
Tennent Caledonian Wellpark Brewery Operational Beer production 8.36 SCCS
Belhaven Belhaven Operational Beer production 0.51 SCCS

BECCS EfW/ACT

The sites considered were identified using the REPD and relevant websites/blog posts.

Table 46: BECCS EfW/ACT data point summary. Extracted from the literature and used in the calculations for the NETs Model
Plant Owner Operational Status Type of Technology Gross Electrical Capacity (MWe) Fuel source Data source
Thainstone Energy Park Project ERF Agile Energy Recovery Application Submitted BECCS EfW 35 MSW REPD
Dunbar EfW (previously Oxwellmains EfW) Viridor Operational BECCS EfW 25.6 MSW REPD
Westfield (former Opencast Coal Mine) Brockwell Energy Under Construction BECCS EfW 23.7 MSW Website[167] Blog post[168]
CalaChem Fine Chemicals (Grangemouth) - Earlsgate Energy Centre Brockwell Energy, Covanta and Green Investment Group Under Construction BECCS EfW (CHP) 21.5 MSW REPD
South Clyde Energy Centre Fortum (formerly Peel Environmental) Under Construction BECCS EfW 20 MSW REPD
Oldhall Industrial Estate Dover Yard Permission Granted BECCS EfW 15 MSW REPD
Millerhill EfW FCC Environment Operational BECCS EfW 12.5 MSW REPD
Barr Killoch Energy Recovery Park Barr Environmental Limited Application Submitted BECCS EfW 12 RDF REPD
Ness Energy Project Aberdeen / Aberdeenshire / Moray Councils Under Construction BECCS EfW (CHP) 11.1 MSW REPD
Baldovie Industrial Estate (Forties Road) MVV Environment Under Construction BECCS EfW (CHP) 10 MSW REPD
Baldovie Dundee Energy Recycling Operational BECCS EfW 8.3 MSW REPD
Binn Farm EfW Binn Group Permission Granted BECCS EfW 7.3 MSW REPD
Polmont landfill site EfW NPL Group Planning application submitted BECCS EfW 7.4 MSW News paper[169]
Drumgray Energy Recovery Centre (DERC) FCC Planning granted BECCS EfW (CHP) 25.5 MSW Blog post[170]
Charlesfield Biomass CHP Plant Charlesfield First LLP & Biogas Power Operational BECCS EfW ACT (CHP) 10 MSW REPD
Coatbridge Material Recovery and Renewable Energy Facility Golder Associates (UK) Ltd/ Shore Energy Under Construction BECCS EfW ACT (CHP) 25 MSW REPD
Levenseat Waste Management Facility Levenseat Permission Granted BECCS EfW ACT (CHP) 17 RDF REPD
Levenseat EfW Levenseat Operational BECCS EfW ACT 12.5 RDF REPD
Glasgow Renewable Energy and Recycling Centre (ACT) Viridor Operational BECCS EfW ACT (CHP) 10 MSW REPD
Achnabreck Northern Energy Developments Permission Granted BECCS Power ACT (CHP) 5.5 Wood pellets REPD
Binn Eco Park SITA UK/Binn Group Permission Granted BECCS EfW ACT 4.6 RDF REPD
Avondale Quarry (Pilot) Grangemouth Generation Ltd Operational BECCS EfW ACT 2 MSW REPD

Carbon Capture Calculations

The following section provides an overview of the modelling parameters and methodology used to estimate the CO2 capture potential of each NET.

BECCS Biomethane

We utilised a simple mass balance to determine the CO2 capture potential of a biomethane facility (see Figure 19 below). The methodology used was as follows:

1.) Since we know the amount of biomethane produced hourly, taken from the NNFCC or estimated using the REPD, then we can estimate the annual biomethane production rate by assuming standard operating hours of 6,000 hr/year (this is equivalent to a utilisation factor of 68%).

2.) The amount of methane directed to the upgrading facility, from the anaerobic digestor, can be determined by assuming that 4.7% of the methane entering the upgrader is lost to the surroundings[171].

3.) The amount of biogas and CO2 directed towards the upgrading facility can be determined by assuming a biogas composition of 55:45 methane to CO2 (on a volumetric basis)[171].

4.) By assuming that the biogas exiting the anaerobic digestor is under normal conditions (i.e., at 25 degC and 1 atm) then the mass of CO2 entering the CO2 capture unit can be determined using a density of 1.795 kg/m3.

5.) The CO2 production potentials determining using this mass balance method were compared to mass balance benchmarks found in a LCA paper (0.00161 tCO2/m3,biomethane)[132]. The values were found to be very close to one another and hence confirms our assumptions and calculations are valid.

6.) The capture potential was finally determined by applying a CO2 capture efficiency of 95%.

Figure 19: A simple mass balance of a BECCS biomethane reference facility
This is a flowchart demonstrating the process of carbon capture from a BECCS biomethane facility.

The comparison of CO2 production potentials through our mass balance method and that of the LCA mass balance benchmarks are shown below in Table 47. The values are very close, deviating by +- 0.002 MtCO2/year, validating our assumptions and calculations.

Table 47: Fermentation CO2 production rates comparison. Calculated using our mass balance method, compared against an LCA paper.
Site CO2 generated from upgrading (Mt/year)
General Mass Balance LCA Mass Balance
Portgordon Maltings Beyside 0.007 0.008
Brae of Pert Farm 0.005 0.005
Charlesfield Industrial Estate 0.005 0.005
Cumbernauld AD 0.005 0.005
Downiehills Farm 0.005 0.005
Girvan Distillery 0.025 0.027
Glenfiddich Distillery 0.018 0.019
Hatton Farm AD 0.004 0.004
Inchdairnie Farm 0.005 0.005
Invergordon Distillery 0.005 0.005
Keithick Farm 0.006 0.006
Lockerbie Creamery 0.007 0.007
Morayhill AD 0.005 0.005
Peacehill Farm 0.005 0.005
Rosskeen Farm 0.004 0.004
Savock Farm 0.006 0.006
Tambowie Farm 0.002 0.002
TECA AD 0.004 0.004
Portgordon Maltings - Anaerobic Digestion Facility, Phase 2 0.018 0.019
Lockerbie Creamery - Anaerobic Digester 0.008 0.009

BECCS Power and Industry (Wood)

As described in the Data Sources section, all the BECCS Industry (Wood) sites considered in our study employ biomass boilers or CHPs on-site. These units are the focus for potential CCS implementation, enabling negative emissions similar to BECCS Power sites. Consequently, our carbon and cost calculations will adopt the same methodology and utilise identical parameters for BECCS Power and Industry (Wood).

To determine the CO2 capture potential of a power only, CHP or heat only site, we carried out the following:

1.) The fuel input rate going into the CHP or power plant was determined by back-calculating from the gross electrical capacity, quoted in the REPD, using an assumed electrical efficiency of 38.7% for power only sites and 25% for CHPs[46]. If the site is heat only then a heat efficiency of 80% is used.

2.) The CO2 production rate was then determined by applying a conversion factor of 0.35 kgCO2/kWh,fuel using the BEIS greenhouse gas reporting conversion factors[93], taken from the ‘Outside of Scopes’ tab.

3.) The CO2 capture potential of the site was then determined using an assumed capture efficiency of 90%.

An example calculation for Caledonian Papermill is provided in Box 1 below.

Box 1: CO2 capture potential, BECCS Power calculation

Caledonian Papermill (26 MWe CHP).

Mathematical equation
Mathematical equation

To ensure the validity of our assumptions and data sources, the CO2 production potential of the BECCS Power and Industry (Wood) sites were compared to values from the SPRI Database[172]. The comparison revealed a general alignment between the values, as demonstrated in below. It is worth noting that data for most BECCS Power sites was unavailable in the SPRI Database due to their low gross capacities. This further supports the reasonableness of our assumptions and data sources.

Table 48: BECCS Power and Industry (wood) CO2 production rates comparison. Compared using two data sources: our own calculations based on REPD data and the values quoted in the SPRI database.
Site NET REPD (MtCO2/year) SPRI (MtCO2/year)
Markinch Biomass CHP Plant BECCS Power (CHP) 0.717444 0.371911
Stevens Croft BECCS Power 0.359363721 0.388518964
Westfield Biomass Power Station BECCS Power 0.089127907 0.108348
Caledonian Papermill BECCS Industry CHP (paper - coated magazine) 0.2869776 0.3012053
Cowie Biomass Facility BECCS Industry CHP (Particle & MDF) 0.165564 0.264579705
Invergordon Pellet Mill BECCS Industry CHP (wood pellets) 0.055188 N/A
Morayhill Mill BECCS Industry (Oriented Strand Board) 0.344925 0.19257725
Barony Road, Auchinleck BECCS Industry CHP (Chipboard and wood recycling) 0.0607068 0.020513208

BECCS Cement

Due to limited data availability for calculating the CO2 capture potential, we relied on the CO2 emissions rate provided by the SPRI database to estimate our own CO2 capture potential. To determine the portion of emissions considered biogenic, we assumed that approximately 40% of cement emissions result from the combustion of fossil fuels[173]. Additionally, considering that Dunbar Cement intends to utilise a fuel mix composed of 45% RDF/SRF waste, with a biogenic content of 17%[174], we determined the quantity of emissions classified as biogenic.

BECCS AD

For the AD sites, we assume that all generated biogas is converted into biomethane, with a small portion being utilised by an onsite CHP to meet onsite requirements. With these assumptions, we can determine the maximum CO2 capture potential for each site.

To determine the CO2 capture potential, we carried out the following:

1.) The feedstock input data, provided via the NNFCC Database, was utilised to determine the biogas production rate using a feedstock to biogas mass balance benchmark taken from an LCA paper (0.17 kg,biogas/kg,feedstock).

2.) The volume of biogas produced is then determined using an assumed biogas density of 1.2 kgm-3.

3.) The CO2 and biomethane production rate are calculated using the assumed biogas composition of 55:45 methane to CO2 on a volumetric basis. The mass of CO2 produced can then be determined using the density of CO2 under standard conditions (1.795 kgm-3).

4.) Finally, the CO2 capture potential is determined using a CO2 capture efficiency of 95%.

Six of the sites listed in Table 44 lacked data on feedstock input rates, making it difficult to calculate their carbon capture potential. Consequently, additional research was required. Among them, five sites had gross power capacities available in either the REPD or NNFCC databases. For these sites, a similar methodology to that used for BECCS Power/Industry (Wood) could be applied, although with different efficiencies, utilisation factors, and emission conversion factors. As for the remaining heat-only site (‘Glenmorangie Distillery’), its planning application documents indicated a biogas production rate of 8000 m3/day, enabling the determination of its CO2 capture potential using the same methodology as BECCS Biomethane.

An example calculation for the ‘Glenfiddich Distillery AD (Extension)’ site is shown below.

Box 2: AD CO2 capture potential example

The annual electricity production potential is calculated using our utilisation factor of 68%, derived from the assumption that a standard AD site operates at 6000hr/year.

Mathematical equation

Determine biogas fuel input rate going into the CHP by assuming an CHP electrical efficiency of 25%. The volumetric rate of this input can then be determined by assuming a biogas energy density of 26 MJ/m3.

Mathematical equation

To determine biomethane production and CO2 capture potential, we assume that all biogas is now upgraded. Using a typical biogas composition of 55:45 methane and CO2 (volume basis), that we exhibit a 5% biomethane losses during upgrading, and a CO2 capture potential of 95%.

Mathematical equation
Mathematical equation

BECCS Fermentation

For the brewery sites, we have estimated CO2 capture potentials as well as costs by assuming that the sites operate like industrial bioethanol plants.

To determine the CO2 capture potential, we carried out the following:

1.) The litres of pure alcohol (LPA) produced by each site is provided[92]. We can utilise a CO2 conversion factor of 754.7 tonnes per Ml of alcohol to determine the quantity of CO2 produced.

2.) The CO2 capture potential can then be determined using a capture efficiency of 90%

BECCS EfW/ACT

For EfW/ACT sites, we can apply the same methodology as BECCS Power and Industry (Wood), but with variations in efficiencies, utilisation factors, and emission conversion factors. Another possible approach is to directly calculate CO2 capture potentials by gathering waste input rates from literature. However, we chose not to use this method in order to maintain consistency in our methodology, as well as to reduce errors and discrepancies in assumptions and data sources, enabling a more accurate and meaningful comparison of different NETs on a like-for-like basis.

To determine the CO2 capture potential, we carried out the following:

1.) Firstly, we calculate the waste input rate. For this purpose, we employed a back-calculation approach utilising the electrical gross capacities provided in the REPD data and assuming a plant utilisation factor of 85%[175]. The power-only sites utilised an electrical capacity of 22%, calculated based off data sourced from AECOM and a MSW energy capacity of 10 MJ/kg[77], while an assumed value of 15% was used for CHP sites and 80% heat efficiency for heat only sites.

2.) Based on the waste input rate we can determine the CO2 production potential using conversion factors. Depending on the choice of waste used, which can be either MSW or RDF/SFR, then the CO2 capture potential will change. This is due to three factors: 1.) SRF/RDF is a more energy dense fuel, 2.) MSW is a more carbon intense fuel, and 3.) SRF/RDF has a higher biogenic carbon content (on a mass basis).

3.) A standard 90% capture rate is then employed to determine CO2 capture potential.

An example below for the ‘Thainstone Energy Park Project ERF’ site is provided in Box 3.

Box 3: BECCS EfW example CO2 capture calculation

Determine EfW power efficiency based off AECOM data:

  • Waste input = 350,000 t/year
  • MSW LHV = 10 MJ/kg (taken from IEAGHG)
  • Net power output (pre-CCS) = 25MWe
  • Convert net to gross efficiency using a standard industrial scaler of 1.11
Mathematical equation
Mathematical equation

Determine waste input using electrical efficiency

Mathematical equation

Convert waste input into units of mass using the waste energy densities (10 MJ/kg for MSW and 13 MJ/kg for SFR/RDF)

Mathematical equation

Determine CO2 captured

Mathematical equation

To double check our calculations, we compared our determined values for waste input against that quoted in the literature (see Table 49 below).

Table 49: Waste input rate comparison of EfW/ACT sites. Comparing REPD data (utilised in our calculations) against literature values
Site REPD feedstock input (t/year) Literature feedstock input (t/year) Reference
Thainstone Energy Park Project ERF 422,414 200,000 [176]l
Dunbar EfW (previously Oxwellmains EfW) 308,966 300,000 [177]
Westfield (former Opencast Coal Mine) 286,034 200,000 [178]
CalaChem Fine Chemicals (Grangemouth) - Earlsgate Energy Centre 384,214 162,000 [179]
South Clyde Energy Centre 241,379 350,000 [180]
Oldhall Industrial Estate 181,034 180,000 [181]
Millerhill EfW 150,862 152,500 [182]
Barr Killoch Energy Recovery Park 111,406 166,000 [183]
Ness Energy Project 198,361 150,000 [184]
Baldovie Industrial Estate (Forties Road) 178,704 110,000 [185]
Baldovie 100,172 90,000 [185]
Binn Farm EfW 88,103 85,000 [186]
Lerwick Energy Recovery Plant 19,549 26,000 [187]
Polmont landfill site EfW 89,310 150,000 [188]
Drumgray Energy Recovery Centre (DERC) 455,695 300,000 [189]
Charlesfield Biomass CHP Plant 178704 70,000 [190]
Coatbridge Material Recovery and Renewable Energy Facility 446760 160,000 [191]
Levenseat Waste Management Facility 233689.8462 315,000 [192]
Levenseat EfW 116047.7454 215,000 [192]
Glasgow Renewable Energy and Recycling Centre (ACT) 178704 222,000 [193]
Achnabreck 38413.95275 N/A N/A
Binn Eco Park 42705.57029 60,000 [194]
Avondale Quarry (Pilot) 24137.93103 N/A N/A

Please note that for the ‘Achnabreck’ site, we could not utilise the above calculations, since this site plans on gasifying wood. Again, we used the REPD method to determine CO2 production potential, using the same CO2 conversion factor as BECCS Power/Industry (Wood). The feedstock input rate was then calculated using a wood pellet energy density of 4.8 kWh/kg, using that the fuel has a moisture content of 10%[195].

We compared our CO2 production values to the benchmark values provided by Tolvik[196], using a waste to CO2 benchmark of 0.992 kg CO2/kg waste. Our calculations using the REPD method closely aligned with the Tolvik benchmark (refer to Table 50), reinforcing our confidence in our assumptions and methodology. The only site which had significant deviation were: Barr Killoch Energy Recovery Park, Levenseat Waste Management Facility, Levenseat EfW, and Binn Eco Park. and which is planning to burn RDF/SFR fuel instead of MSW. Please note that the Achnabreck site gasifies wood pellets - not waste, and hence cannot be compared against the Tolvik benchmark.

Table 50: EfW/ACT CO2 production potential comparison. Compared using our REPD method and Tolvik benchmarks
Site NET CO2 production (Mt/year) Difference (%)
REPD Method Tolvik benchmark
Thainstone Energy Park Project ERF BECCS EfW 0.423 0.419 1%
Dunbar EfW (previously Oxwellmains EfW) BECCS EfW 0.309 0.306 1%
Westfield (former Opencast Coal Mine) BECCS EfW 0.286 0.284 1%
CalaChem Fine Chemicals (Grangemouth) - Earlsgate Energy Centre BECCS EfW (CHP) 0.384 0.381 1%
South Clyde Energy Centre BECCS EfW 0.242 0.239 1%
Oldhall Industrial Estate BECCS EfW 0.181 0.180 1%
Millerhill EfW BECCS EfW 0.151 0.150 1%
Barr Killoch Energy Recovery Park BECCS EfW 0.124 0.111 11%
Ness Energy Project BECCS EfW (CHP) 0.198 0.197 1%
Baldovie Industrial Estate (Forties Road) BECCS EfW (CHP) 0.179 0.177 1%
Baldovie BECCS EfW 0.100 0.099 1%
Binn Farm EfW BECCS EfW 0.088 0.087 1%
Lerwick Energy Recovery Plant BECCS EfW (Heat only) 0.020 0.019 1%
Polmont landfill site EfW BECCS EfW 0.089 0.089 1%
Charlesfield Biomass CHP Plant BECCS EfW ACT (CHP) 0.179 0.177 1%
Coatbridge Material Recovery and Renewable Energy Facility BECCS EfW ACT (CHP) 0.447 0.443 1%
Levenseat Waste Management Facility BECCS EfW ACT (CHP) 0.261 0.232 11%
Levenseat EfW BECCS EfW ACT 0.130 0.115 11%
Glasgow Renewable Energy and Recycling Centre (ACT) BECCS EfW ACT (CHP) 0.179 0.177 1%
Achnabreck BECCS Power ACT (CHP) 0.098 N/A N/A
Binn Eco Park BECCS EfW ACT 0.048 0.042 11%
Avondale Quarry (Pilot) BECCS EfW ACT 0.024 0.024 1%

Table 51 on the subsequent pages provides summary of the NETs parameters used in the analysis.

Summary of NETs parameters

Table 51: A breakdown in parameters used to model NETs carbon capture potential
NET Gross Power Efficiency (power only) Net Power Efficiency (CHP) Net Heat Efficiency (CHP) Gross Heat efficiency (heat only) Utilisation Factor Conversion Factor (kg/kWh) Lifespan (years) CO2 Capture efficiency Biogenic content of carbon (%mass) Reference
BECCS Biomethane N/A N/A N/A N/A Not used
68% Assumed
N/A* Not used
20 [197]
95% Assumed
100% Assumed
BECCS Power & Industry (Wood) 38.7% BEIS, Wood (2018)[45]
25% Assumed
37.5%
80% Assumed
90% Assumed
0.35 BEIS, GHG Reporting[93]**
25 BEIS, Wood (2018)[45]
90% BEIS, Wood (2018)[45]
100% Assumed
BECCS Cement N/A N/A N/A N/A N/A N/A Not used
30 IEA
90% BEIS, Wood (2018)[45]
3%***
BECCS Fermentation N/A N/A N/A N/A N/A Not used
754.7***** SCCS Paper[92]
30 l
90% Assumed
100% Assumed
BECCS EfW/ACT 22%****** AECOM[175]
15% Assumed
31%
80% Assumed
85% AECOM[175]
0.36 (MSW) 0.31 (RDF/SFR) IEAGHG[77]
20 AECOM[175]
90% BEIS, Wood (2018)[45]
50% (MSW) 17% (RDF/SFR) IEAGHG[77]and IEA

*Mass balance used instead

**The greenhouse gas reporting conversion factors used were taken from the ‘Outside of Scopes’ tab.

***We understand that 40% of cement emissions are from combustion, with Dunbar Cement aiming to utilise 45% SFR in their fuel mix which has a biogenic content of 17%.

****This corresponds to the amount of biogas output per tonne of feedstock anaerobically digested. The units are t/t.

*****Units of tonnes of CO2 produced per mega litre of alcohol produced (MLA)

******See calculations above in EfW section to see how this efficiency is derived.

LCOC Methodology

The Levelised Cost of Carbon (LCOC) will serve as a filtering mechanism for potential NETs sites. It supports the determination of whether these sites should be included in subsequent pathway analysis or not; the higher the LCOC the less economically attractive the site.

Sites located on islands were excluded from further analysis due to the complexities involved in transporting the CO2 to a suitable storage location.

LCOC calculation of existing sites
Mathematical equation
LCOC of future sites
Mathematical equation
Table 52: A breakdown in the definitions of the various parameters used to determine LCOC
Variable Meaning Units
LCOCexisting This figure serves as a summary of the economic viability of a particular Negative Emission Technology (NET), acting as an indicator to aid in the selection process among different NETs options. £/tCO2,captured
CAPEX This figure accounts for the investment cost associated with purchasing, installing, and commissioning plant equipment.
Capital Recovery Factor (CRF) This figure represents the fraction of the initial capital investment that needs to be recovered each year to cover the cost of the investment. It is used to annualise the investment cost. N/A, this is a dimensionless quantity
OPEXfix This figure accounts for expenses that remain relatively constant regardless of the level of production. This figure is typically in proportion with the capital expenditure. £M/year
OPEXvar This figure accounts for expenses that vary in proportion to the level of production. For industrial sites this cost is typically heavily linked to energy costs. £M/year
Ctransport This figure represents the cost associated with transporting the CO2 to the designated storage site(s). In this report, we have made the assumption that the CO2 can be transported either entirely by truck to St Peterhead, or partially by truck to an injection point, where it is then injected into a pipeline. £M/year
Cstorage This figure represents the cost of storing the CO2 in the North Sea. In this report, our assumption is based on the utilisation of depleted oil and gas wells, which aligns with the proposed approach of the Acorn project. £M/year
Crevenue This figure represents the gain in revenue associated with selling heat and power from BECCS Power, BECCS Industry and BECCS EfW sites. It's important to note that the analysis does not include revenues derived from the sale of biomethane for future BECCS Biomethane and AD Upgrading sites. £M/year
CO2,captured This figure depicts the quantity of CO2 captured by the NETs site. Please note that this is not the same as the negative emission potential, which does not account for emissions from fossil sources. MtCO2/year

Cost Analysis

Capital Expenditure (CAPEX)

In estimating the CAPEX, we employed the widely used sixth-tenths rule, which is a method for approximating costs. According to this rule, the cost of a project can be estimated by taking the cost of a comparable completed project and scaling it by the exponent 0.6, based on the capacity of the reference plant. This scaling can be done by considering various factors such as CO2 capture potential, heat or power production, biomethane production, or any other relevant parameter. An example calculation is provided below.

Mathematical equation

It's important to note that this method provides a quick and rough estimate of costs. However, it should be used with caution, as it may not encompass all the unique aspects and complexities of each specific project. It's worth mentioning that we do not intend to conduct a detailed cost analysis for each NETs site in our current scope, making this method sufficient for our purposes.

Operational Expenditure (OPEX)

The OPEX is divided into two categories: Fixed costs and Variable costs. Fixed OPEX encompasses ongoing expenses that tend to remain consistent regardless of the production or operational level. Examples include salaries, benefits, equipment maintenance, repairs, and insurance. On the other hand, Variable OPEX includes costs that fluctuate based on the level of production or operation. These costs can include raw materials, fuel, maintenance supplies, and energy consumption.

For existing sites, we assumed that the fixed OPEX would amount to 5% of the CAPEX (a typical industrial benchmark), while the variable OPEX would account for increases in fuel/electricity usage alone as this is typically the main source of variable OPEX cost. We could account for additional variable costs; however, this would to a complex cost analysis for each site which is outside of the scope of this project. Although cost benchmarks for Variable and Fixed OPEX were available in the literature, we chose not to utilise them due to significant discrepancies in assumptions and parameters included in the costs, depending on the specific NETs being discussed. As a result, the costs for existing NETs would not be comparable on a like-for-like basis. The exception was the ‘Dunbar Cement’ plant, where site-specific data on heat and power demands was unavailable in the literature, and so we had to rely on benchmarks to determine Variable OPEX.

When evaluating future sites, it was necessary to consider the overall costs associated with both installing and operating a NETs site. To accomplish this, cost benchmarks were employed in their entirety and then scaled up using the sixth-tenths rule. This approach ensured that the complete range of costs were considered for accurate cost estimation. Although this approach may lead to potential cost overestimation, it is not a concern since our objective does not necessitate comparing future sites on a like-for-like basis.

We tested our assumption of fixed OPEX being equivalent to 5% CAPEX, for existing sites, by comparing the respective costs to industrial benchmarks. The overall estimation was reasonably close, as shown below in Table 53. Please note that all cost benchmarks have been scaled to 2023 costs via inflation and converted to pounds sterling.

Table 53: OPEX and CAPEX comparison used in modelling. A comparison between the assumption of fixed OPEX for existing projects, estimated to be equivalent to 5% of CAPEX, and the fixed OPEX benchmarks cited in the literature
NET 5% assumption (£/tCO2) Cost benchmark
BECCS EfW (combustion and ACT) 16.1 15.7[175]
BECCS Biomethane 11.4 7[197]
BECCS AD Upgrading 263.3* N/A, not broken down into fixed and variable OPEX[171]
BECCS Fermentation 2.3 3.1[198]
BECCS Power and BECCS Industry (Wood) 4.1 3.3[45]
BECCS Industry (Cement) 0.0041** 0.0037[175]**

*Please note that AD Upgrading is in units of £/m3/year

**Please note that BECCS Industry (Cement) is in units of £/tclinker/day

BECCS Biomethane

For existing sites, the costs associated with CCS installation encompassed CO2 liquefaction only. This is because biomethane sites already separate out CO2 into a pure stream during the biogas upgrading process, and hence the CO2 capture is more efficient, less energy intensive, and cheaper. The resulting liquefication process takes the capture carbon and removes water and impurities, where the clean CO2 stream is then compressed to high pressures in preparation for transport.

CO2 Liquefication

The initial step of liquefication is to compress the gas to the desired pressure (circa 130 bar) and help it reach its critical temperature. Water is then removed by condensation to prevent hydration and the gas is subsequently cooled (e.g., using a set of heat exchangers or expansion cooling) to transition the gas into a liquid. An impurity removal unit is used to remove impurities when the delivered CO2 needs to meet a high purity requirement.

CAPEX

To determine the CAPEX of liquefication a benchmark was taken from a 2022 techno-economic analysis paper investigating the costs associated with CO2 capture for AD Upgrading sites. The paper considers a reference case of an AD Upgrading facility producing 4400 tCO2 at a CAPEX of 1MEUR.

To determine the CAPEX for future sites, we also have to account for the costs associated with constructing the AD Upgrading and AD plant. These costs were taken from a techno-economic analysis paper that investigates biomethane upgrading via the water scrubbing method, which accounts for investment costs in constructing the biogas plant, silage pit, AD upgrading plant, gas grid connection, and CNG service station. For simplicity, we have assumed the water scrubbing technique is used in all biogas AD Upgrading sites in Scotland, so that this paper can be applied to all biomethane sites. This assumption is reasonable, given the fact that around one third of AD Upgrading sites in the UK use water scrubbing technology, according to the EBA Statistical Report of 2020. However, an improvement opportunity of the analysis is to consider membrane separation costs in the analysis instead.

OPEX

As previously mentioned, the Fixed OPEX for existing sites was taken to be 5%. This assumption was also applied to future sites since the CCS OPEX benchmarks from the 2022 AD Upgrading paper was not clearly broken down.

The Variable OPEX for existing sites was calculated by determining the power demands of CO2 liquefication (i.e., powering the compressors and chiller). This method was also used for new sites, again due to an unclear breakdown in CCS OPEX from the 2022 Upgrading paper. The liquefication power demands were taken from a 2019 paper which details a thorough breakdown in cooling and compression power requirements, with compression requiring 12.5 MW and cooling 40.48 MW in order to capture 1 MtCO2/year. We are assuming to compress CO2 to 130 MPa in preparation for transport, which is a standard industrial benchmark.

An example calculation is shown in Box 4 below for the ‘Portgordon Maltings Beyside’ site.

Box 4: Example Variable OPEX calculation process

We use the liquefication power demand of 52.98 MW, in order to capture 1 MtCO2/year, and convert to MJ/kg. We assume standard operating hours of 6000 hr/year for an AD Upgrading site.

Mathematical equation

To convert to MJ/year, we multiply the energy demand by the CO2 capture potential and scale up using the sixth tenths rule.

Mathematical equation

To determine the cost of this power demand, an electricity price of 14.6 p/kWh was used based off Ofgem’s wholesale electricity price. This final cost is taken as the Variable OPEX.

Mathematical equation

When considering future sites, the Variable OPEX associated with the AD and Upgrading plants was taken from a 2018 paper. The total OPEX was quoted as a single figure, including both fixed and variable OPEX, which accounts for maintenance and overheads, electricity demands, thermal demands, feedstock and disposal costs, plant OPEX, depreciation, and gate fee(s).

Revenue

We did not include revenue sources from biomethane in our analysis because they fall outside the mass/energy balance boundary that we established.

BECCS Power and Industry Wood

The costs for BECCS Power and BECCS Industry (Wood) were taken from the same source. This is because BECCS Industry (Wood) sites either have biomass boilers or biomass powered CHPs onsite that meet onsite demands (as described earlier in Data Sources). The reference is a 2018 BEIS paper (BEIS Wood 2018). This paper examines the costs associated with installing CCS on a 498 MWe bioenergy power station capturing circa 4.2 MtCO2/year.

CAPEX

The sixth tenths rule is applied, where the costs associated with constructing the bioenergy plant and installing the CCS equipment are £813.7M and £322M respectively.

OPEX

Fixed OPEX was taken to be 5% of CAPEX.

To determine Variable OPEX of existing sites, we had to determine the impact installing CCS would have on the NETs power export of a site. In particular, installing CCS requires heat demands of circa 3.4 MJ/kg,CO2, which are sourced by extracting low-pressure steam from the turbine at circa 3 bar. As a result, the overall efficiency of the site decreases by approximately 5%. In our analysis, we have used these drops in NETs efficiency to determine losses in revenue, which is equivalent to an increase in Variable OPEX.

If a site is in fact a CHP, then we have to account for the impacts on both power and heat export. To make this analysis simpler, and avoid the need to undertake multivariable optimisation, we assumed that the power export of a site will remain the same, but the heat export potential is impacted by the CCS heat requirements. This results in NETs power efficiencies remaining constant and heat efficiencies dropping. An electrical efficiency of 25% was assumed for CHP sites, based off CHPQA knowledge, and the resulting heat efficiency (pre-CCS) was determined using a z ratio of 3.5. The price of heat was taken to be 4 p/kWh.

The assumed low price of heat is justified by the fact that we are considering the heat price at the export point rather than the price at which heat is sold directly to the customer, such as 12p/kWh through a third party. In this scenario, the generator sells heat to a third party at 4p/kWh. The third party is responsible for constructing and operating the heat network, which incurs significant costs. Subsequently, the third-party charges the customer for the heat supplied.

Box 5 below is a summary of how Variable OPEX was calculated for “Markinch Biomass CHP Plant”

Box 5: Further example of variable OPEX calculation process

Using the assumed NETs power efficiency of 25%, we can calculate the heat efficiency using the CHP zed ratio of 3.5. This results in a NETs heat efficiency of ~37.5%.

Mathematical equation

Taking the feedstock input rate, we can determine the heat export potential pre-CCS.

Mathematical equation

Using a standard CCS heat demand of 3.4 MJ/kg, CO2, we can determine the resulting heat demand of CCS using the annual CO2 capture potential. This enables us to check whether the site can meet CCS demands without the need for the installation of an additional boiler.

Mathematical equation
Mathematical equation

Using a heat price of 4p/kWh, we can determine the variable OPEX.

Mathematical equation

For the Morayhill Mill industrial site, which is a heat only site, we have assumed that the gross boiler efficiency is 80%. The same methodology to CHP sites is then applied, where a heat efficiency post-CCS was found to be 50%.

For future sites, we directly applied the bioenergy and CCS benchmarks quoted in the 2018 BEIS paper to determine fixed and variable OPEX. One key thing to note is that the benchmarks used do not account for fuel usage within the Variable OPEX benchmark. Therefore, a wood pellet price of £24/MWh was applied and scaled up according to the feedstock demands of the site.

Revenue

For future sites, the same methodology described above was applied to determine the revenue stream potential of each site, using the expected heat and/or power export potential post-CCS and multiplying it by the respective energy costs (14.6 p/kWh and 4 p/kWh for power and heat respectively).

BECCS Cement

The only site associated with BECCS Cement is Dunbar Cement, which is already operational. As there are no other cement sites expected to be constructed in Scotland, that we have only considered the costs associated with installing and operating CCS.

CAPEX

The investment cost of CCS was taken from the reference site used as the cost benchmark assumed a clinker production rate of 1 Mt/year and a carbon capture potential of 0.8 MtCO2/year, as well as utilising some waste as a fuel feedstock. This last point is ideal for our calculations of Dunbar cement, which plans to implement a 45% blend of SRF/RDF into their fuel feedstock.

The investment cost of CCS, obtained from the AECOM Next Generation Carbon Capture Technology techno economic analysis paper, for a reference site producing 1 Mt/year of clinker and capturing 0.8 MtCO2/year is £192.5M. This reference site also utilises waste as a fuel feedstock, which aligns with our calculations for Dunbar cement due to the site aiming to incorporate a 45% blend of SRF/RDF into their fuel feedstock. The capital expenditure covers EPC and Project Development costs.

OPEX

The Fixed OPEX was taken to be 5% of the CCS Capex.

As there was no data available on the heat and power usages of the Dunbar Cement site in the literature, then the Variable OPEX was determined by applying cost benchmarks directly, despite the site being operational. This does go against the methodology applied to all other existing NETs sites; however, no other options were available to us. This will lead to the LCOC of the Dunbar Cement site being overestimated compared to other existing industrial sites. An improvement point for this work is that the Variable OPEX of the Dunbar Cement site is recalculated based on potential increases in heat and power demands for the site, as well as the impact CCS has on potential losses in revenue sales from clinker, heat and power. The variable OPEX was taken to be £72.7M/year for the reference site.

Due to the lack of available data on the heat and power usages of the Dunbar Cement site, the Variable OPEX was determined by directly applying cost benchmarks, despite the site being operational. This deviates from the methodology used for other existing sites, but no other options were available. The Variable OPEX for the reference site was £72.7M/year. An improvement for this study is to recalculate the Variable OPEX of the Dunbar Cement site considering the impacts CCS will have on heat and power demands, as well as potential revenue losses from clinker, heat, and power sales.

BECCS AD

For existing sites, the costs associated with BECCS AD were taken to be that associated with constructing and operating an AD Upgrading and liquefication plant. As for future sites, the costs associated with constructing and operating an AD facility also had to be accounted for. The calculations, as well as the cost benchmarks used, are discussion previously in the BECCS Biomethane cost section.

BECCS Fermentation

For existing sites, the costs associated with BECCS Fermentation were taken to be that associated with constructing and operating the CCS plant. As potential future sites were not listed in the literature, please see the Data Sources section, then costs associated with future sites (e.g., cost of building a whisky distillery or brewery) were not investigated into. Similar to BECCS Biomethane sites, the mechanism for capturing carbon will be similar to the CO2 liquefication process, since the stream of CO2 exiting the distillery is pure. In our case, we used a 2014 paper published by the US Department of Energy which investigated the cost of capturing carbon from a bioethanol plant, which is a reasonable approximation to a whisky distillery and/or brewery. The paper in question considers a reference bioethanol plant that produces 50 Mgal/year of ethanol and 0.145045 MtCO2/year.

CAPEX

The investment cost associated with capturing 0.145045 MtCO2/year is £7.846M. This cost was then scaled up accordingly using sixth tenths rule and the alcohol production rate.

OPEX

The fixed OPEX was again taken to be 5% of CCS Capex.

Similar to BECCS Biomethane, the electricity requirements associated with processing this pure CO2 are similar to that of CO2 liquefication. Regarding this 2014 bioethanol paper, a power demand of 1.9 MWh/hr is needed to capture the carbon, which is equivalent to 14147.4 MWh/year based off the fact that the reference bioethanol plant operates at a utilisation factor of 85%. Now assuming a standard carbon capture efficiency of 90%, we can achieve a power demand of 109.9 GWh/MtCO2.

Please see Box 6 below for a breakdown in this calculation.

Box 6: Example of fixed OPEX calculation for BECCS fermentation, power demand

A power demand of 1.9 MWh/hr is needed to capture the carbon, which is equivalent to 14147.4 MWh/year based off the fact that the reference bioethanol plant operates at a utilisation factor of 85%.

Mathematical equation

Assuming a standard carbon capture efficiency of 90%, we can achieve a power demand of 109.9 GWh/MtCO2

Mathematical equation

Using the sixth tenths rule, we can then scale up the power demand depending on the alcohol production rate. The final cost associated with power usage can then be calculated using an electricity price of 14.6 p/kWh.

Please see the calculations in Box 7 relating to ‘Cameronbridge’ grain whisky distillery

Box 7: Example of fixed OPEX calculation, cost implication

The CCS power requirement is scaled using the sixth-tenths rule

Mathematical equation

The power requirement is costed

Mathematical equation

BECCS EfW/ACT

The paper used to source our CCS cost benchmarks is from the AECOM Next Generation Carbon Capture Technology paper, which quotes a reference site that processes 350,000 t/year of waste, has a gross power output 29 MWe and captures 0.3 MtCO2/year. Costs associated with constructing and operating a EfW facility were taken from a Catapult Energy Systems paper that focussed on UK deployment of EfW facilities, considering a ‘Core EfW Plant’ where 350,000 t/year of waste is burnt under a gross capacity of 32MWe.

CAPEX

The investment cost of capturing 0.3 MtCO2/year from a EfW site is quoted to be £96.8M, which is then scaled accordingly using sixth tenths and the CO2 capture potential of each site. When considering future sites, the investment cost associated with constructing an EfW facility is taken to be £224M, when not scaled for inflation.

OPEX

For existing sites, the fixed OPEX is again taken to be 5% of CAPEX.

The methodology behind calculating Variable OPEX for existing EfW/ACT sites is similar to that of BECCS Power and Industry (Wood). The key difference is the choice around the electrical and heat efficiencies used, which are calculated below using parameters described in the AECOM paper.

The NETs power efficiency pre and post CCS is 19% and 11% for power only sites

Given the fact that the reference case consumed 350,000 t/year of MSW and that the site operates 7446 hr/year, then we can determine the fuel energy input rate based on the fact that MSW has an energy density of 10 MJ/kg.

Mathematical equation

We know that the gross power is 29 MWe, NETs power output (pre-CCS) is 25 MWe, and NETs power output (post-CCS) is 14 MWe. Furthermore, we know that the Therefore, we can determine the power efficiencies.

Mathematical equation
Mathematical equation
Mathematical equation

The choice of NETs power and heat efficiencies (pre-CCS) of a EfW CHP was determined based off CHPQA data. The power and heat efficiencies were taken to be 15% and 31% respectively.

The NETs heat efficiency of a EfW CHP was determined to be 0.4% once CCS was installed (i.e., all heat output from the CHP is utilised onsite to meet CCS demands). An example calculation for the “Charlesfield Biomass CHP Plant” site is shown in Box 8.

Box 8: BECCS EfW OPEX calculation example

Taking the feedstock input rate, we can determine the heat export potential pre-CCS.

Mathematical equation

Using a standard CCS heat demand of 3.4 MJ/kg,CO2, we can determine the resulting heat demand of CCS using the annual CO2 capture potential. This enables us to check whether the site can meet CCS demands without the need for the installation of an additional boiler.

Mathematical equation
Mathematical equation

Using a heat price of 4p/kWh, we can determine the variable OPEX.

Mathematical equation

For the only heat only site, Lerwick Energy Recovery Plant, the heat efficiency pre-CCS was taken to be 76.7%. This is based off calculations using data from the Tolvik 2022 EfW statistics paper, where the site is labelled as Gremista. The total heat export from this site was 49 GWth in 2022, the waste input rate 23,000 t/year, and the energy density of the fuel was assumed to be 10 MJ/kg (MSW equivalent). This calculated heat efficiency is reasonable compared to industrial benchmarks of a typical boiler efficiency being ~80%.

For future sites, we directly applied the EfW plant and CCS benchmarks quoted in the AECOM and Catapult Energy Systems papers. This includes the proposed Achnabreck site, which plans on gasifying wood pellets in a CHP. The reasoning behind why we used these benchmarks is because the technology is the same and hence costs are unlikely to differ significantly if we switch the biomass source. However, it is still worth noting that an improvement point to this work is to update the Achnabreak cost calculations so that they are more closely aligned to woody biomass gasification, since the fuel density and gasification efficiencies will differ compared to waste, which in turn will impact the variable OPEX costs slightly.

CO2 Transport

The costs associated with carbon transport were determined in a separate Excel tool compared to the LCOC Tool.

In this case, the X and Y coordinates identified during the literature review of each site were taken and mapped against potential CO2 injection points across Scotland using GIS Mapping.

The key CO2 injection points considered were all along the St Fergus gas pipeline, which is proposed to be upgraded to enable CO2 transport cross country as part of the Acorn project. The injection points were Bathgate, Kirriemuir, Garlogie, and Peterhead. The distance between all four injection points and the proposed NETs site were calculated in km using GIS Mapping by assuming that all sites will initially transport CO2 by truck to an injection point unless they are physically located next to one of the four proposed injection points. The shortest possible road distance was then selected.

The onshore pipeline distances from each of the four injection points to the St Peterhead gas terminal was then calculated using GIS mapping and are shown in Table 13 of the main report.

The final offshore pipeline distance from the Peterhead to the Acorn site is 80km for all sites.

The resulting costs associated with each mode of travel: road, onshore pipeline, and offshore pipeline, are then calculated using the benchmarks outlined in Table 20, page 36 following benchmarks taken from the IEAGHG EfW-CCS paper. The benchmarks can then be used to scale linearly scale up transport costs based on CO2 production capacity and transport distance.

CO2 Storage

Storage costs were taken directly from the IEAGHG EfW CCS paper, where the high-end costs were utilised in order to ensure that costs are conservative. As these costs are quoted in a £/tCO2 basis, with no reference to plant size, then costs are linearly scaled up based on CO2 capture capacity.

Inflation

To ensure all cost benchmarks are within the correct format, they are adjusted for inflation and converted to pounds sterling using indexes from the World Bank and exchange rates from the OECD. Please note that inflation indexes were only available up to 2021. An improvement point of this analysis is to utilise updated inflationary figures, when they are available, to determine the difference in costs based on the fact that the British economy is going through a period of very high inflation.

Table 54: Figures used to convert costs into British Pounds Sterling and scaled for inflation
Parameter 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
GDP deflator index (2010 =100) to 2021 for US 91.86 93.77 95.52 97.19 99.01 100.00 101.00 102.92 105.40 107.29 108.69 113.57
GDP deflator index (2021 =100) to 2021 for US 0.81 0.83 0.84 0.86 0.87 0.88 0.89 0.91 0.93 0.94 0.96 1.00
Exchange rate USD to GBP 0.65 0.62 0.63 0.64 0.61 0.65 0.74 0.78 0.75 0.78 0.78 0.73
Exchange rate USD to Euro 0.75 0.72 0.78 0.75 0.75 0.90 0.90 0.89 0.85 0.89 0.88 0.85
Exchange rate Euros to GBP 0.86 0.87 0.81 0.85 0.81 0.73 0.82 0.88 0.89 0.88 0.89 0.86

Contact

Email: NETs@gov.scot

Back to top