Life cycle assessment of repurposing abandoned onshore oil and gas wells for geothermal power generation

The annual global growth rate for geothermal power generation between 2021 and 2030 is targeted to be 13 % to meet net-zero emissions by 2050. Repurposing abandoned oil and gas wells (AOGWs) presents a strategic alternative to boost geothermal power by minimising the drilling requirements. This study performed the first cradle-to-grave life cycle assessment to evaluate the environmental performance of three options for geothermal power generation from repurposed oil and gas wells: i) two completely AOGWs (R-GEO double ); ii) a single completely AOGW (R-GEO single ); iii) two semi-AOGWs (R-GEO semi - still in operation but with high water-cut). Their results are then compared

The annual global growth rate for geothermal power generation between 2021 and 2030 is targeted to be 13 % to meet net-zero emissions by 2050.Repurposing abandoned oil and gas wells (AOGWs) presents a strategic alternative to boost geothermal power by minimising the drilling requirements.This study performed the first cradle-to-grave life cycle assessment to evaluate the environmental performance of three options for geothermal power generation from repurposed oil and gas wells: i) two completely AOGWs (R-GEO double ); ii) a single completely AOGW (R-GEO single ); iii) two semi-AOGWs (R-GEO semi -still in operation but with high water-cut).
Their results are then compared with a business-as-usual geothermal power plant (GEO bau ).All 18 impact categories of the ReCiPe 2016 midpoint methodology plus cumulative energy demand have been analysed in detail, with background data from the Ecoinvent v3.8 database.R-GEO semi is deemed the most promising repurposed system, exhibiting the lowest values in 11 impact categories.Specifically, R-GEO semi produces 34 %, 23 %, and 14 % less CO 2 eq./kWh when compared to GEO bau , R-GEO double , and R-GEO single , respectively.Conversely, R-GEO double performed the worst in 12 impact categories, and the second worst in the rest of the indicators.Meanwhile, GEO bau achieved the lowest impacts in nine categories when compared with repurposed systems, indicating the reduction of drilling and construction activities cannot always guarantee the mitigation of all environmental impacts.Sensitivity analyses showed that a longer lifetime could lower environmental impacts, but increasing annual power generation is constrained by site-specific factors.A 'breakeven' point analysis revealed that 85 % of repurposed systems' impact indicators could match GEO bau if their lifetime reaches 30 years, but this remains uncertain.The findings of this study will be of interest to national and local governments developing future policies aimed at renewable energy transformation from oil and gas industries.

Introduction
Geothermal power generation has experienced steady growth in the past decade, with global annual output rising from 68,467 GWh in 2012 to 92,047 GWh in 2019 (International Renewable Energy Agency, 2021).However, geothermal power generation is still in arrears of its promised contribution towards the net-zero emissions target for 2050.The International Energy Agency (IEA) projected that a minimum of 13 % annual growth of geothermal power from 2021 to 2030 is required to satisfy the renewable electricity generation demand in alignment with the commitment to achieving net-zero emissions (International Energy Agency, 2021).Such significant growth in demand leads to unprecedented opportunities for the implementation of geothermal power plants (GPPs), considering it is a renewable, reliable and low-waste option for electricity production (International Renewable Energy Agency, 2017).
Previously, geothermal power generation required medium-high temperature hydrothermal reservoirs, which are typically located in tectonically active regions (e.g., Iceland, Philippines, and New Zealand) (International Renewable Energy Agency, 2021).However, nowadays, enhanced geothermal systems (EGS) allow power generation at lowmedium temperatures via the organic Rankine cycle (ORC) after stimulation of hot dry rocks, which can benefit tectonically inactive areas (Huenges, 2016).Despite the technical feasibility of EGS, financial barriers remain a major concern, as geothermal exploration is considered a high-risk activity with high capital investment (International Renewable Energy Agency, 2017;Prodi, 2014).More specifically, high investment is needed before reliably attesting the energy exploitability potential.Furthermore, the exploration and drilling processes are considered the most cost-intensive stage in geothermal projects (Bu et al., 2011).Consequently, geothermal power generation projects heavily rely on support from government policies or the availability of public funding (International Renewable Energy Agency, 2017).In addition to financial risks, geothermal applications entail drilling activities that require diesel consumption, particularly during the drilling of deep wells.These fossil fuel-intensive operations consequently contribute to climate change (Li et al., 2023).
Therefore, the imperative to reduce capital costs and the consumption of fossil fuels in the construction of GPPs is apparent.To achieve this requirement, adopting the 'repurpose' principle from the 'R-strategies' in the circular economy framework presents a promising approach to steer the future utilisation of geothermal resources.Specifically, repurposing encourages the use of discarded products or their parts in various products for different purposes (Li et al., 2023;Potting et al., 2017).In this line, adapting alternative underground infrastructures that possess sufficient heat for geothermal power generation should be targeted.
Utilising abandoned wells has the potential to reduce total geothermal capital costs by 42-95 % due to the presence of existing well bores (Tester et al., 2007).Particularly, as suggested by Li et al. (2023), oil and gas wells that are abandoned could be a potential alternative geothermal supply resource, but their environmental sustainability assessment is yet to be performed.More than 29 million oil and gas wells are currently abandoned due to high water-cut (i.e., the percentage of water content in the total fluids production of the oil and gas well) that hinders their commercial value for fossil fuel production (Mehmood et al., 2019;Groom, 2020).Taking only abandoned oil and gas wells in the state of Texas (USA) as an example, >180,000 out of a total of 364,000 wells are deeper than 3 km and have high bottom temperatures (in the range of 125-175 • C) (Michaelides, 2016).Therefore, abandoned oil and gas wells potentially provide abundant opportunities for reliable wellbore integrity for geothermal energy extraction (Mehmood et al., 2019) and, hence, electric power production.
Besides reducing the capital costs of GPPs, concerns about the contribution to the environmental impacts of electricity generation also support the concept of repurposing abandoned oil and gas wells.As a rule, the drilling processes in GPP contribute to over 80 % of its climate change potential (CCP) due to high diesel consumption (Pratiwi et al., 2018;Lacirignola and Blanc, 2013).Moreover, Menberg et al. (2016) and Paulillo et al. (2020) highlighted that diesel combustion would also incur high emissions of sulphur dioxide (SO 2 ) and nitrogen oxides (NO x ), which result in high acidification potential and eutrophication potential.Notably, existing literature on the cradle-to-grave environmental performance of GPPs provided insufficient assessments of end-of-life treatments and tended to focus on conventional geothermal power generation (Karlsdottir et al., 2020;Paulillo et al., 2020).To the authors' knowledge, no studies have investigated the complete life cycle environmental performance of repurposing abandoned oil and gas wells at different stages of their lifetime (i.e., completely abandoned or semiabandoned) for geothermal power generation.
This study performs for the first time the environmental life cycle assessment (LCA) of repurposed abandoned oil and gas wells for geothermal power generation.Importantly, this study would provide insights of interest to policymakers seeking to develop strategies for the renewable energy transition, as well as oil and gas companies exploring opportunities in renewable energy production.More specifically, this study assesses three options: R-GEO single (a single completely abandoned well feeding an ORC), R-GEO double (two completely abandoned wells feeding one ORC), and R-GEO semi (two semi-abandoned wells feeding one ORC).These options are then compared to a business-as-usual GPP (GEO bau ) which plays its role as a benchmark.To unfold the analysis, Section 2 details the methodology to conduct the LCA and the brief technical descriptions of each assessed system.Results and discussions of the environmental impacts are presented in Section 3, including the sensitivity analysis for the impactful factors.The conclusions can be found in Section 4.

Methodology
The following sections describe the different stages of the attributional LCA, guided by the requirements and recommendations of ISO 14040 (ISO (International Organization for Standardization), 2006a) and ISO 14044 (ISO (International Organization for Standardization), 2006b).In addition, technical descriptions of each power generation system are briefly provided.

Goal and scope definition
The goal of this LCA study is to assess the life cycle environmental performance of the three repurposed systems and compare their results with GEO bau .The three repurposed systems are R-GEO double , R-GEOsingle , and R-GEO semi .The selection and configurations of the repurposed systems can be found in S1-S3 of Supplementary Information (SI).To ensure comparability, the results for all environmental impact indicators are allocated to the functional unit "1 kWh of electricity generated".
In this study, the lifetime for the business-as-usual GPP (GEO bau ) is assumed to be 30 years, and 15 years for R-GEO double , R-GEO single , and R-GEO semi (Paulillo et al., 2019;Menberg et al., 2021;Colucci et al., 2021;Wang et al., 2020).This difference is due to the repurposed systems previously being oil producing wells (Liu et al., 2018).Oil production typically exceeds 50 years, however declines sharply towards end-oflife, reducing oil company revenue.(Darko, 2014).An interview conducted with an Innovation Project Manager who worked in the oil industry provided an example that a high water cut may occur in the 35th year of a 55-year oil production plant.Based on Liu et al. (2018), the simulated results suggested the lifetime of 10-20 years is viable for repurposed systems before revenue decline of the oil company starts.Therefore, this study has chosen the midpoint of 15 years for the lifetime of R-GEO semi .For the repurposed completely AOGWs' systems (both R-GEO double and R-GEO single ), their complete abandonment introduces the greatest uncertainty in determining a technically viable lifetime as no publicly reported pilot projects exist in Europe.To simplify the evaluation and align with the expectations of widely deploying the repurposed systems, the same lifetime as R-GEO semi has been chosen for R-GEO double and R-GEO single .These lifetime uncertainties are later explored via sensitivity analysis.
The life cycle stages include the exploration, construction, operation, and end-of-life, as shown in Fig. 1.Noticeably, the exploration stage is excluded for repurposed systems, as it is assumed that all necessary geographical data and files have already been documented by the oil and gas industries, and no additional exploration (i.e., exploratory well drilling) is required.The construction stage is further divided into underground and aboveground activities.The operation stage encompasses electricity generation, maintenance, and fluid leakage.Lastly, the end-of-life stage covers well-abandonment activities, including cementplugging and well-closure, as well as material recycling, incineration, and landfilling where appropriate.Transportation is considered in all life cycle stages, including 500 km on average for raw materials and fuel during construction and operation stages, and 100 km for dismantled materials to treatment at end-of-life stage.

Inventory analysis
The foreground data applied in the life cycle inventory (LCI) analysis were sourced from open literature sources: the GEOENVI (Pratiwi et al., 2018;Douziech et al., 2021), and Multidisciplinary and multi-context demonstration of Enhanced Geothermal Systems exploration and Exploitation Techniques and potentials (MEET) (Multi-sites EGS Demonstration, 2021) projects.In particular, GEOENVI was the first project to assess environmental impacts and risks for deep geothermal energy in Europe, which elaborated detailed inventory data for LCA models.The MEET project documented the pilot plant for repurposed systems in France as detailed in S3 of SI, which provided the fundamental system configurations for this study.Background data have been sourced from the Ecoinvent v3.8 database (Ecoinvent, 2022).The detailed background process selection and inventory data estimation can be found in SI (S4-S6).

Overview of the assessed options
Technically, GEO bau is the hypothetical Rittershoffen EGS power plant which serves as the business-as-usual case in this study, with an installed capacity of 4.1 MW e , and a system efficiency of 11.6 % (Pratiwi et al., 2018).The repurposed abandoned oil and gas well options represent pilot-scale systems with an installed capacity of 20 kW e , located at the Chaunoy oil field in France.R-GEO double , R-GEO single , R-GEO semi have production flow rates of 5.6 kg/s.Notably, R-GEO semi repurposes semi-abandoned oil and gas wells with a 95 % water cut and therefore does not require the unplugging, re-casing and re-cementing activities needed by the other repurposed systems.All repurposed systems share similar technical conditions, such as annual working hours (8500 h), production and reinjection temperature differences (13.6 • C), system efficiency (4.2 %), and well depths (2.5 km per well).More detailed technical descriptions and the justifications for the system selection can be found in S3 of SI.The foreground inventory data for all systems are shown in Table 1.

Main assumptions
Considering that repurposed systems are not yet commercially deployed and lack standardised processes to retrofit the completely abandoned wells, the following assumptions have been made to fill data gaps and adapt the inventory dataset applied in this study: • The unplugging process includes cement removal via drilling.Diesel is applied to drive the drilling rig.• The repurposed completely AOGWs are assumed to only require full re-casing and re-cementing to meet the geothermal power generation target.The old casing and cementing of the abandoned wells would not be removed.No further retrofitting processes (e.g., side-drilling or creation of perforation zones) are considered necessary.

Exploration stage
GEO bau is the only option that includes exploration for seismic investigation (Pratiwi et al., 2018).Shallow exploratory wells are drilled at this stage to confirm the technical feasibility for geothermal exploitation before up-scaled drilling begins.Diesel fuel is used for machinedriven purposes during exploration, and no other well-construction materials such as cement and steel are required.

Construction stage
The activities implemented in the construction stage are divided into two categories (underground and aboveground), as shown in Fig. 1.Inventory for drilling activities in GEO bau includes the diesel consumed for powering the building machine as well as materials used in specific drilling activities (i.e., casing, cementing, and drilling mud).In well cementing, cement slurry is applied as a mix of chemicals, such as bentonite, cement, and silica, while drilling mud is applied to facilitate the drilling and cementing activities (Russo et al., 2014).A series of chemicals (e.g., hydrochloric acid, potassium chloride) and hydraulic stimulations were then used in GEO bau for the injection well to enhance the well productivity.The production pump used is a line shaft pump with 39 stainless steel stages and impellers, submerged at a depth of 460 m and held in place by a steel tube.A series of surface pumps and valves are used to ensure proper fluid circulation and achieve the desired flow rate (Pratiwi et al., 2018).The ORC unit in GEO bau , with an installed capacity of 4.1 MW e , includes an aboveground heat exchanger, turbine, generator, and cooling fan (i.e., air condenser system).
The repurposed completely AOGWs require unplugging processes to open the well, which involves breaking and removing the cement plugs since the wells are entirely closed.The amount of diesel required for this process is estimated based on the amount of cement plugs implemented in the well-abandonment procedure proposed by Pratiwi et al. (2018).The main difference between R-GEO double and R-GEO single is the number of wells considered (two wells for R-GEO double and one for R-GEO single ).This results in differences in the unplugging process, heat exchanger selection, pipe length, and well-abandonment process, as outlined in Table 1.More detailed estimation descriptions can be found in S1.2 and S4 of SI.In the case of R-GEO semi , no oil-water separation process is required in the production process and, therefore, only pipes, valves, and ORC unit are required to be installed for the geothermal power generation (Karlsdottir et al., 2020).

Operation stage
All the electricity production is carried out at this stage.Meanwhile, maintenance of the machinery is undertaken regularly, taking up to 260 h per year (Pratiwi et al., 2018).In detail, maintenance of the heat exchanger is carried out annually through water blasting to remove scaling build-up within the tubes.Sodium chloride and lubricating oil are applied for lubrication purposes.R134a is utilised as the refrigerant in all options as it exhibits superior performance and is particularly suitable for repurposed power systems compared to other refrigerants such as R600a, R600, propylene, R290, or R143 (Kurnia et al., 2021;Nian and Cheng, 2018).The annual leakage rate for the refrigerant is 0.5 % as suggested by Pratiwi et al. (2018), which may be attributed to corrosion, operational stresses, and mechanical failures.During maintenance, geothermal fluid circulation is stopped, and it is introduced into a separator that continuously releases dissolved CO 2 during this period (Pratiwi et al., 2018).The electricity consumption during the operation stage is associated with the use of pumps.Fig. 1.System boundary and life cycle stages for a business-as-usual geothermal power plant (GEO bau ), repurposed two completely abandoned oil and gas wells for geothermal power generation (R-GEO double ), repurposed single completely abandoned oil and gas well for geothermal power generation (R-GEO single ), repurposed semi-abandoned oil and gas wells for geothermal power generation (R-GEO semi ) from the "cradle to the grave" approach.R-GEO double and R-GEO single are the geothermal power generation systems from the repurposed completely abandoned oil and gas wells.*Electricity is used for daily on-site activities.

Table 1
Inventory inputs for main materials and energy consumption of the four options assessed (GEO bau , R-GEO double , R-GEO single , R-GEO semi ) for 1 kWh electricity.2.2.6.End-of-life stage At the end-of-life stage, the well is plugged to prevent the contamination of adjacent aquifers by using cement to form four plugs at intervals throughout the well.There are no process differences in the wellabandonment among GEO bau , R-GEO double and R-GEO semi despite GEO bau having 30 % more well-depth in total.This is because the well abandonment procedures are standardised and regulated, regardless of well depth.The vertical thickness of each cement plug is 50 m, as reported by Pratiwi et al. (2018).

Life cycle impact assessment
This study considers the ReCiPe 2016 v1.1 methodology from the hierarchist perspective (Huijbregts et al., 2017) to calculate the environmental impacts.ReCiPe integrates some of the most current and scientifically advanced life cycle impact assessment models (e.g., climate change, toxicity), covering a comprehensive scope of both elemental flows and land transformations, and undergoes regular updates and expanded scope in new versions demonstrating an active commitment to incorporating the latest science and continuously improving the methodology over time.The Sphera LCA for Experts software (version 10.7) (Sphera, 2023) was used to model the foreground inventory process, and the Ecoinvent v3.8 database was used for background inventory (Blanc et al., 2020) (cut-off approach).The potential environmental impacts are quantified and grouped by considering emission and resource extraction-related processes.All 18 impacts from ReCiPe 2016 v1.1 have been considered: climate change (CCP), fine particular matter formation potential (FPMFP), fossil depletion potential (FDP), freshwater consumption potential (FCP), freshwater ecotoxicity potential (FETP), freshwater eutrophication potential (FEP), human toxicity potential cancer effect (HTP, cancer); human toxicity potential non-cancer effect (HTP, non-cancer), ionising radiation potential (IRP), land use potential (LUP), marine ecotoxicity potential (METP), marine eutrophication potential (MEP), metal depletion potential (MDP), photochemical ozone formation, ecosystems potential (POFEP), photochemical ozone formation, human health potential (POFHP), stratospheric ozone depletion potential (SODP), terrestrial acidification potential (TAP), and terrestrial ecotoxicity potential (TETP).Additionally, cumulative energy demand (CED) from primary energy is also calculated to assess the dependence on renewable and non-renewable energy consumption at each life cycle stage for each option.

Interpretation
Interpretation aims to discuss the obtained results for all impact categories and perform contribution analysis among the options assessed.The study includes two sensitivity analyses (lifetime of the plants and annual electricity production) to evaluate the robustness of the results (Hauschild et al., 2018).GEO bau plays a benchmark role to show the variations of the environmental performance due to the above parameters.

Sensitivity analysis of plant lifetime
The lifetime of repurposed power systems may vary depending on the quality of the oil and gas wells or the high-water-cut time point.Therefore, a sensitivity analysis was conducted for each 20 kW e repurposed system to investigate lifetimes from 10 to 20 years.This analysis aimed to identify the 'breakeven' point for the repurposed systems, where their environmental impacts are equal to or lower than those of GEO bau .

Sensitivity analysis of annual electricity production
There is a large difference in annual electricity production between repurposed power systems and GEO bau as outlined in Section 2.2.1.Hence, a sensitivity analysis was conducted to investigate the impacts of increasing the annual electricity production capacity of repurposed systems on their environmental performance.In this regard, system efficiency is crucial as it depends on site-specific production flow rates and temperature differences between production and reinjection as illustrated in S5 of SI.
However, these substantially lower electricity outputs are disadvantageous to repurposed systems' environmental performance (per kWh).As part of the goal of this study stated in Section 2.1, exploring the breakeven points is crucial to assess the feasibility of widely implementing repurposed systems.Consequently, this sensitivity analysis seeks to determine the minimum system efficiency required for each environmental impact indicator to enable every repurposed system to be environmentally competitive compared to GEO bau .

Results and discussion
This section analyses in detail the impacts and hot spots for all options considered in this study.The results of the sensitivity analysis are commented upon afterwards.

Environmental assessment
As illustrated in Fig. 2, R-GEO semi gives the best environmental performance in 11 of the 20 impact categories and the second best in the remaining nine.On the other hand, R-GEO double exhibits the poorest environmental performance in 12 of the 20 impact categories.This is due to its higher unplugging process requirements as compared to R-GEO single , as well as its significantly lower annual electricity production and shorter lifetime when compared to GEO bau .Noticeably, GEO bau performed the best in nine impact categories (although not for climate change), principally due to its significantly larger installed capacity (i.e., 4.1 MW e ) and lifetime electricity generation compared to the repurposed systems.
Regarding the life cycle stages, construction is the main contributor to 95 % of environmental impacts for all four options.In GEO bau , drilling is predominant, responsible for over 50 % of impacts in 15 of the 20 categories.However, with fewer drilling needs in repurposed systems, ORC unit construction surpasses drilling, contributing to over 56 % of N.B.N/A implies 'Not Applicable'.
a GEO bau is the business-as-usual geothermal power plant; R-GEO double and R-GEO single are repurposed completely abandoned oil and gas wells, two wells and single well system, respectively; R-GEO semi is semi-abandoned oil and gas wells for geothermal power generation.b Composition of drilling mud: 1 kg of drilling mud consists of 38 % water, 11 % bentonite, 10 % calcium carbonate, 8 % carboxymethyl cellulose, 27 % inorganic chemicals, 3 % sodium chloride, 1 % citric acid, 1 % soda ash, and 1 % sodium hydroxide (Douziech et al., 2021).c Anti-corrosion agent is continuously applied in the geothermal fluid, where 1 kg additive includes 0.3 kg ethylene glycol, 0.1 kg glycerine, 0.05 kg ammonium chloride (Douziech et al., 2021), and the rest 0.55 kg for water.d Amount of the refrigerant R134a is estimated according to the reported fluid quantity per circuit (Multi-sites EGS Demonstration, 2021).impacts in all categories, apart from SODP.SODP deviates due to the leakage of organic working fluid (R134a) during operation.The study also highlights the positive impact of recycling metals, such as steel, aluminium, and copper at the end-of-life stage, which mitigates a number of environmental impacts (notably FPMFP, FETP, HTP and METP) associated with GPPs.

Climate change potential (CCP)
GEO bau exhibits the highest climate change potential (CCP) of 10.9 g CO 2 eq./kWh.R-GEO double , R-GEO single , and R-GEO semi have an impact of 9.3 g, 8.4 g and 7.2 g CO 2 eq./kWh (i.e., 15 %, 23 %, and 34 % less), respectively.The lower impact is mainly due to the reduction of drilling activities, which implies diesel burning.This finding highlights repurposing as an effective strategy for lowering CCP, regardless of the low electricity generation during their lifetime.
Fig. 2 (a) indicates the construction stage as a significant contributor, accounting for over 70 % of the total CCP for all options.In the R-GEO double , R-GEO single , and R-GEO semi, the reduced drilling activity leaves the ORC unit as the major contributor, responsible for 65 %, 73 %, and 97 % in the construction stage, respectively.The end-of-life stage positively affects the total CCP in GEO bau , R-GEO double , and R-GEO single , largely due to the environmental benefits from recycling significant steel quantities from production and injection pumps.

Air pollution (SODP, POFEP, POFHP, FPMFP)
As shown in Fig. 2 (b), the estimated stratospheric ozone depletion potential (SODP) in GEO bau is calculated as 17.3 μg CFC-11 eq./kWh.R-GEO double , R-GEO single , and R-GEO semi have an impact of 9.5 μg, 9.2 μg, and 9.0 μg CFC-11 eq./kWh (i.e., 45 %, 46 %, and 48 % less than GEO bau) , respectively.Drilling during construction stage significantly influences SODP, due to the use of chlorofluorocarbons and hydrochlorofluorocarbons in upstream material production.The remaining contribution stems from volatile organic compounds and NOx emissions that occur due to diesel combustion during construction activities.The reduction in drilling and construction activities significantly decreases SODP from R-GEO double , R-GEO single , to R-GEO semi .In the case of the repurposed options, the operational stages have surpassed the construction stage and become the primary focal point.This shift is due to the inevitable minor leakage of the organic working fluid (i.e., R134a), which is a hydrofluorocarbon, during the operation stage (Pratiwi et al., 2018).Photochemical ozone formation (ecosystems) potential (POFEP) and photochemical ozone formation (human health) potential (POFHP) produce similar results in each option, as shown in Fig. 2 (b).In all cases, the repurposed systems are preferable to GEO bau : 0.018-0.064g NO x eq./kWh vs 0.064 g in the case of POFEP, and 0.017-0.022g NO x eq./ kWh vs 0.062 g in the case of POFHP.In GEO bau , drilling activities in the construction stage are the most significant contributor (90 %) due to NO x and volatile organic compounds emitted from diesel engines as a result of incomplete combustion (Link et al., 2016).For R-GEO double , R-GEO single and R-GEO semi , the construction of a 20 kW e ORC unit is the main contributor to emissions at the construction stage (over 64 %).
According to Fig. 2 (a), GEO bau 's fine particulate matter formation potential (FPMFP) is 20.9 mg PM2.5 eq./kWh, primarily from diesel combustion during drilling.In R-GEO double , R-GEO single and R-GEO semi , FPMFP resulted in 16.5 mg, 15.1 mg, and 12.5 mg PM 2.5 eq./kWh (i.e., 21 %, 28 %, 40 % less compared to GEO bau , respectively).The construction of a 20 kW e ORC unit accounts for >71 % of the construction stage emissions for the repurposed systems due to dust arising from raw material handling (e.g., coke, quicklime and iron) during steel production.

Water and soil pollution (FEP, MEP, TAP)
GEO bau displays a total freshwater eutrophication potential (FEP) of 2.6 mg P eq./kWh as shown in Fig. 2 (a).In comparison, R-GEO double , R-GEO single , and R-GEO semi show an increase in FEP, resulting in values of 3.7 mg, 3.3 mg, and 2.7 mg P eq./kWh, respectively.Among all options, the construction stage is the key contributor to FEP, where drilling activities contributed 55 % to the construction stage in GEO bau .For R-GEO double , R-GEO single and R-GEO semi , the construction of the 20 kW e ORC unit becomes the main hotspot, contributing approximately 63 %, 69 %, and 97 %, respectively.In more detail, the significant contributor to the construction of a 20 kW e ORC unit is the production process of low-alloyed steel from background data.The iron ore mining process can result in the release of phosphorous compounds, which may be washed into nearby water bodies.The results for repurposed systems benefit from the substantial reduction of drilling and construction activities, but this is more than offset by increased low-alloyed steel consumption for the ORC unit construction: 1.9 g/kWh for the 20 kW e ORC unit compared to 0.1 g/kWh in the 4.1 MW e ORC unit in GEO bau .In other words, higher absolute consumption of steel in GEO bau is offset by substantially higher electrical output, meaning FEP is sensitive to the differences in the plant lifetime and the annual power generation capacity.Moreover, a lower FEP in R-GEO single compared to R-GEO double is due to less low-alloyed steel included in the ORC construction.This is due to the downhole heat exchanger production process does not require low-alloyed steel according to its background data.
The marine eutrophication potential (MEP) for GEO bau , R-GEO double , R-GEO single , and R-GEO semi is estimated at 0.33 mg, 0.50 mg, 0.47 mg, and 0.43 mg N eq./kWh, respectively.For GEO bau , drilling activities at the construction stage play a vital role in the contribution of MEP, which can be further attributed to steel production from background data.Specifically, the steelmaking process releases NO x emissions due to the combustion of fossil fuels for energy.These emissions can subsequently convert to nitrate in the environment through dry deposition and precipitation when they come into contact with water bodies.In marine environments, nitrogen compounds act as the limiting nutrient and contribute to the nutrient load in water bodies, thereby facilitating eutrophication.For repurposed systems, the construction of a 20 kW e ORC unit is the main contributor in the construction stage as well as the total MEP.In particular, the steel production process involved in ORC unit construction is the main contributor, which applied the same reason as mentioned above.The MEP performances of repurposed systems are again hindered by their lower capacities for total electricity generation, similar to the cause of FEP.
The total terrestrial acidification potential (TAP) in GEO bau , R-GEO double , R-GEO single , and R-GEO semi is calculated as 42 mg, 27 mg, 24 mg, and 20 mg SO 2 eq./kWh, respectively, as shown in Fig. 2 (b).The construction stage is the prominent hotspot, contributing 99 % of the TAP in GEO bau .The steel recycling at the end-of-life stage mitigates the total TAP by 5 %.The most significant contributor from the construction stage is diesel consumption in drilling activities (66 %).Compared to GEO bau , the TAP in the construction stage for R-GEO double , R-GEO single , and R-GEO semi decreased by 36 %, 43 %, and 52 %, respectively, due to the elimination of drilling activities.Specifically, low-alloyed steel production in the construction of 20 kW e ORC unit construction is the main contributor to TAP in repurposed systems.The results show that the impact of diesel on TAP outweighs that of low-alloyed steel, with reductions in drilling having a greater effect than the differences in electricity generation between GEO bau and repurposed systems.

Ecotoxicity (TETP, FETP, METP)
As shown in Fig. 2 (b), the total terrestrial ecotoxicity potential (TETP) for GEO bau , R-GEO double , R-GEO single , and R-GEO semi is calculated as 49 g, 84 g, 79 g, and 59 g 1,4-Dichlorobenzene (DB) eq./kWh, respectively.Compared to GEO bau , R-GEO double , R-GEO single , and R-GEO semi exhibit respective increases of 71 %, 61 %, and 20 %.The construction stage is still the main hotspot in all options.In GEO bau , well drilling, construction of production pumps and 4.1 MW e ORC units in the construction stage contribute approximately 94 % to the TETP at the construction stage.Diesel use in drilling and steel and copper production are the main TETP contributors.In particular, the management of raw materials in steel production, encompassing the utilisation of iron ore and coke and the release of heavy metals to the terrestrial systems (e.g., arsenic, cadmium, chromium), plays a significant role.Concurrently, the extraction and processing stages of copper production, such as blasting at the mining stage, contribute to the TETP as a result of the release of toxic substances (e.g., lead, copper, zinc and arsenic) that can contaminate the soil.Since no well drilling processes are needed in R-GEO semi , TETP is mainly attributed to the steel production processes in ORC unit construction as explained above: the electricity production is relatively low, thus the impacts of the ORC unit construction are disproportionately increased.The results suggest TETP performance is more tied to electricity generation capacity than drilling and construction reductions.
The total freshwater ecotoxicity potential (FETP) in GEO bau , R-GEO double , R-GEO single , and R-GEO semi is 0.4 g 1,4-DB eq./kWh, 0.6 g, 0.6 g, and 0.4 g, respectively.For all four options, the construction stage is the main hotspot in contributing to the FETP, contributing >95 %, primarily due to the steel and copper production processes (e.g., construction of injection and production pumps, ORC unit) which lead to the release of heavy metals and toxic substances such as iron, chromium, nickel, and zinc to the aquatic environment.At the end-of-life stage, the recycling of steel and copper can mitigate the FETP somewhat (see negative values in Fig. 2 (a)).Regarding the copper production process, the mining and ore processing, smelting and refining possess the risks of releasing heavy metals such as cadmium, arsenic, and zinc into the aquatic environment.As a result, R-GEO semi has the lowest amount of steel to be recycled hence is the option that the least benefitted from steel recycling.However, the FETP of R-GEO semi is 1 % lower than GEO bau , highlighting the benefits of reduced drilling and construction over recycling.
In assessing the marine ecotoxicity potential (METP), GEO bau registers a value of 0.5 g 1,4-DB eq./kWh.In contrast, the METP values for R-GEO double , R-GEO single , and R-GEO semi are 0.8 g, 0.7 g, and 0.6 g 1,4-DB eq./kWh, representing increases of 46 %, 36 %, and 5 %, respectively compared to GEO bau .Similar to FETP, METP primarily arises from the leaching of heavy metals such as lead, cadmium, chromium, arsenic, mercury and nickel during metal ore extraction, as well as polycyclic aromatic hydrocarbons emitted in steelmaking.For both FETP and METP, it is important to note that, regardless of recycling credits somewhat mitigating the impacts of high raw material consumption, the 'R-strategies' hierarchy displayed in the circular economy framework (Potting et al., 2017) emphasises the fact that reducing consumption is preferable to recycling.

Resource depletion (FDP, MDP, FCP)
As per Fig. 2 (a), GEO bau has the highest FDP compared to the other options, resulting in 3.0 g oil eq./kWh.Diesel combustion in drilling activities contributes 95 % to the FDP during the construction stage.Compared to GEO bau , R-GEO double , R-GEO single , and R-GEO semi have a result of 3.0 g, 2.7 g, 2.3 g oil eq./kWh, (i.e., reduced by 2 %, 11 %, and 22 %, respectively).In repurposed systems, the 20 kW e ORC plant construction contributes 70 %, 77 %, and 98 % to the FDP during the construction stage.The low-alloyed steel required for the 20 kW e ORC is the key contributor due to upstream fossil fuel-powered equipment for material extraction and handling.
The total metal depletion potential (MDP) value of GEO bau , R-GEOdouble , R-GEO single , and R-GEO semi is calculated as 0.17 g, 0.38 g, 0.35 g, and 0.28 g Cu eq./kWh, respectively, as shown in Fig. 2 (b).In all options, the construction stage remains dominant, driven by steel production for drilling, pump construction in GEO bau , R-GEO double , R-GEO single , and the 20 kW e ORC unit in repurposed systems.Moreover, the variation of MDP is more sensitive to the total electricity generation compared to the reduced drilling and construction activities, which can explain the higher results of repurposed systems.The total freshwater consumption potential (FCP) of GEO bau results in 55 cm 3 water/kWh.Compared to GEO bau , R-GEO double , R-GEO single , and R-GEO semi have an impact of 84 cm 3 , 77 cm 3 , and 68 cm 3 water/ kWh (i.e., 51 %, 40 %, and 23 % more, respectively).The increase in repurposed systems is due to low electricity generation, overshadowing the benefits of reduced drilling and construction on FCP.The construction stage is the main hotspot across all options, contributing to the FCP of 98 % for GEO bau , 97 % for R-GEO double, 97 % for R-GEO single and 90 % for R-GEO semi .Specifically, drilling activities in GEO bau contributed 72 % towards the FCP at the construction stage.The primary cause of the impact is the production of drilling mud, particularly the use of inorganic chemicals, which are intended to regulate pH levels.For all repurposed systems, the construction of a 20 kW e ORC unit is the main contributor to FCP.Again, the low-alloyed steel production process is mainly responsible for the contribution, due to the water-consumptionintensive activities (e.g., dust suppression, ore processing, and slurry transportation) involved in the raw material extraction and handling processes.

Human health (HTP, cancer; HTP, non-cancer; IRP)
As per Fig. 2 (a), the total human health potential, cancer (HTP, cancer) of GEO bau is 2.3 g 1,4-DB eq./kWh.R-GEO double , R-GEO single , and R-GEO semi have an impact of 7.8 g, 7.6 g, and 6.8 g 1,4-DB eq./kWh (i.e., 237 %, 226 %, and 193 % more, respectively).Across all four options, the construction stage is again the main contributor to HTP, cancer, due to emissions from the steel production chain including mining and blast furnaces.Reduced construction in repurposed systems, especially R-GEO semi , should lower such impacts, but low electricity generation offsets these benefits.
The total human health potential, non-cancer (HTP, non-cancer) of GEO bau , R-GEO double , R-GEO single , and R-GEO semi are calculated as 10.7 g 1,4-DB eq./kWh, 13.2 g, 12.2 g, and 9.2 g 1,4-DB eq./kWh, respectively.Steel production remains the main contributor, releasing toxins like lead and mercury that pose inhalation, ingestion, and dermal contact risks.Furthermore, blast furnaces release SO 2 and NO x , which may contribute to respiratory and cardiovascular diseases.As shown in Fig. 2 (a), the end-of-life stage of all four options greatly reduces the total HTP, noncancer, via the avoided production of virgin steel.
The total ionizing radiation potential (IRP) of GEO bau results in 0.5 Bq Co-60 eq. to air.Compared to GEO bau , repurposed systems have an impact of 1.1 Bq Co-60, 1.1 Bq Co-60, and 1.0 Bq Co-60 eq. to air (i.e., 138 %, 123 %, and 117 % more, respectively).The construction stage is again the main hotspot for all four options.For GEO bau , the medium voltage electricity supply required during the drilling activities is the main contributor, mostly arising from upstream processes in the nuclear power life cycle.This is particularly relevant for this study as the plant is based in France, which has a high percentage of nuclear power in its electricity mix (nearly 70 %) (World Nuclear Association, 2023).Electricity is also the main contributor to IRP for the repurposed systems, with most power demand arising in the ORC construction stage.Again, the variation of total IRP is more sensitive to the total electricity generation rather than the reduced drilling and construction activities.

Land use potential (LUP)
As shown in Fig. 2 (a), the total land use potential (LUP) value of GEO bau is 0.2 annual crops eq.year/kWh.Compared to GEO bau , R-GEO double , R-GEO single , and R-GEO semi are 5 %, 13 % and 29 % lower, respectively.The construction stage is the main hotspot across all four options, where drilling activities in GEO bau play a vital role.In particular, the drilling mud applied during drilling activities is the main contributor as one of its constituents is carboxymethyl cellulose (CMC) powder (for viscosity control purposes), produced from wood pulp and therefore requiring forest operations such as tree harvesting and land clearing.For repurposed systems, the LUP stems from the low-alloyed steel in the 20 kW e ORC unit, linked to upstream mining of iron ore, coal, and limestone.Regarding Fig. 2 (a), the minimisation of drilling activities in repurposed systems unconditionally benefits the LUP, which implies the effectiveness is not restricted by the significantly low total J. Li et al. electricity generation.Overall, the reduced requirement in drilling activities can reduce land use compared to the business-as-usual GPP.

Cumulative energy demand (CED)
According to Fig. 2 (b), the total CEDs of R-GEO double and R-GEO single are higher than GEO bau , but the percentage of non-renewable primary energy dependence in R-GEO double , R-GEO single , and R-GEO semi is lower (i.e., 93.0 %, 92.7 %, 92.9 %, respectively, compared to 95.4 % in GEO bau ).The unplugging process (i.e., breaking the plugged cement, re-casing and re-cementing), construction of 20 kW e ORC unit and production pumps are the main non-renewable energy-dependent activities due to the application of diesel alongside the steel production processes (as blast furnaces rely on fossil fuels) in R-GEO double and R-GEO single .Additionally, the construction of a downhole heat exchanger has a higher non-renewable energy demand than the conventional heat exchanger used in the ORC unit .Furthermore, the amount of renewable energy applied in repurposed systems has increased compared to GEObau .The minimisation of construction activities, especially the Fig. 3. Sensitivity analysis of the various plant's lifetime (10, 15 and 20 years) for repurposed systems compared to the business-as-usual geothermal power plant (GEO bau ) (30 years).The performance of GEO bau serves as the benchmark, illustrating the gaps of environmental performance between GEO bau and repurposed systems while varying the lifetime.Nomenclature for repurposed systems and impact categories can be referred to Fig. 2. underground part, could effectively reduce the dependence on primary energy input, as shown in R-GEO semi .

Sensitivity analyses
As discussed above, in cases where the repurposed systems have higher impacts than GEO bau , this is due to their shorter lifetime (15 years) and low installed capacity (20 kW e capacity; 125,120 kWh per year).Therefore, these two parameters are investigated in the following sensitivity analysis.

Plant lifetime
Fig. 3 depicts the sensitivity of each repurposed system's results to lifetime when to 10, 15 and 20 years are considered.GEO bau with a 30 years lifetime is also shown as a benchmark.The extension of the lifetime to 20 years cannot make all environmental impact indicators in repurposed systems perform better than GEO bau .In detail, nine indicators (e.g., FCP, FEP, and IRP) in R-GEO double , seven indicators (e.g., FCP, HTP cancer, IRP) in R-GEO single and three indicators (HTP cancer, IRP, MDP) in R-GEO semi still perform worse than GEO bau even in the best case of a life extension to 20 years.
Fig. 4 shows the breakeven lifetime required for each impact category of the repurposed systems to make their performance environmentally competitive to GEO bau .Eight impact categories (CCP, FPMFP, FDP, LUP, POFEP, POFHP, SODP, TAP) do not necessitate improvement based on increasing lifetime for all repurposed systems.In other words, these impact categories achieve the equivalent environmental performance to GEO bau in 15 years or less.There are another nine impact categories (FCP, FETP, FEP, HTP non-cancer, METP, MEP, TETP, CED non-renewable, CED renewable) for which repurposed systems could achieve competitiveness within a typical lifetime of GPP (i.e., 30 years).The breakeven lifetimes for MDP and IRP are 29-34 years, while HTP cancer requires the longest lifetime to reach breakeven: 50 years, 49 years, and 33 years for R-GEO double , R-GEO single and R-GEO semi , respectively.
While the above analysis estimates the theoretical minimum required lifetime for each repurposed system, the technical feasibility to achieve these lifetimes requires further research.Technically, pilot plants (i.e., 20 kW e repurposed systems) are not designed for long-term power production but for testing parameters and technical feasibility.
The extension of a longer lifetime may face several technical challenges such as corrosion of equipment, uncertainties of the geographical conditions of abandoned oil and gas wells, insufficient flow rate and reinjection rate.At this point, a clear prediction or recommendation cannot be provided, but the identification of the 'breakeven' lifetimes aim to provide useful targets for further experimentation and scale-up.

Annual electricity production
The productivity of the repurposed systems in terms of annual electricity production is primarily determined by the efficiency of the ORC unit in converting heat to electricity, as an important element of the system efficiency.In this study, the ORC unit of the repurposed systems designed at the Chaunoy site has an installed capacity of 20 kW e which, according to theoretical calculations based on its specifications and an exchanged heat of 350 kW (Multi-sites EGS Demonstration, 2021), could achieve a theoretical system efficiency of 5.7 %.If the system consistently runs at this efficiency over its 15-year lifetime, it could theoretically generate 2,553,469.4kWh of electricity (Multi-sites EGS Demonstration, 2021).However, the actual system efficiency has been measured at only 4.2 % due to subpar turbine performance (Multi-sites EGS Demonstration, 2021), yielding an estimated 1,876,800 kWh of electricity.
Fig. 5 depicts the minimum required system efficiency for the repurposed systems to achieve equivalent performance to GEO bau , establishing the 'breakeven' point while maintaining the 15-year operational lifetime.The current system efficiency of the repurposed systems (4.2 %) serves as the reference point for determining the necessary improvement required to reach these 'breakeven' points.
Fig. 5 illustrates that eight impact categories (CCP, FPMFP, FDP, LUP, POFEP, POFHP, SODP, TAP) do not necessitate an increase in system efficiency, which is consistent with the finding in Section 3.2.1.Despite the low system efficiency, the performance of these impact categories suggests that adopting the circular economy strategy (repurpose), which involves minimising underground drilling and construction activities by utilising abandoned oil and gas wells, can effectively compensate for the limitations of system efficiency.Three impact categories, HTP cancer, IRP and MDP, require almost or more than doubling the current system efficiencies for all repurposed systems.Among them, HTP cancer has the highest system efficiency requirements for R-GEO double (14.2 %), R-GEO single (13.7 %), and R-GEO semi (9.2 %) to Fig. 4. Minimum repurposed system to enable their impact categories perform equivalently as GEO bau .Nomenclature for technologies and impact categories can be referred to Fig. 2. reach breakeven.
Overall, the breakeven efficiencies shown in Fig. 5 range from 1.1 % to 14.2 % and, therefore, do not exceed the typical 10 % to 20 % found in ORC-based GPPs (DiPippo, 2004;Saleh et al., 2007;Li et al., 2011).However, real-world efficiency is constrained by site specific parameters.The fixed heat exchange specific to the Chaunoy site dictates a maximum of 5.7 %.When the system efficiency reaches this level, a total of ten indicators for R-GEO double , 13 for R-GEO single , and 17 for R-GEO semi are as competitive as those of GEO bau .In terms of this sensitivity analysis, the widespread deployment of repurposed systems, especially R-GEO double , cannot be promoted unless their low electricity generation capability is addressed.Strategies that could improve the overall system efficiency include: i) increasing the production flow rate; ii) increasing the temperature difference between the production and reinjection temperature especially by lowering the cold loop temperature; iii) selecting appropriate organic working fluid; and iv) optimising the heat exchanger design (DiPippo, 2004).These parameters are discussed in turn below.
Production flow rate: At Chaunoy, the production flow rate is very low (5.6 kg/s) compared to the Ritterhsoffen plant (82.5 kg/s).Increasing the production flow rate at a specific site may involve incorporating a pump with higher absorbed power or retrofitting the internal structure of the oil and gas well using techniques such as side drilling or creating perforation zones.However, it should be noted that increasing the production flow rate may improve heat transfer but also result in higher pumping power consumption and additional stress on the thermal and mechanical aspects of ORC components (such as the turbine) (Hirschberg and Wiemer, 2015), which may negatively affect the system efficiency.Retrofitting the internal structure of the oil and gas well, through side drilling or additional perforation zones, is beyond the scope of this study as we assumed that the repurposed systems do not involve such modifications.
Temperature difference: In this study, the high temperature in the cold loop of the ORC is due to the ambient temperature at the test point being higher than the design point (20 • C) (Multi-sites EGS Demonstration, 2021).This reduces the amount of exchanged heat converted to electricity.This reduced the amount of exchanged heat converted to electricity.Moreover, the assumed re-casing and re-cementing of the completely AOGWs was built upon the old casing and cementing without removal.This process may have further hindered the transferable heat from the well.However, the degradation condition of the old casing and cementing cannot be measured quantitatively at this point due to a lack of data.Therefore, when selecting potential oil and gas wells for geothermal power generation, the ambient temperature conditions of the site should be rigorously considered.Sites with cooler climates should be preferred over those with hotter climates.Organic working fluid: The organic working fluid (i.e., refrigerant) in this study is R134a, but more comparative studies are needed to justify different ORC setups.For instance, studies (Baral, 2019;Fontalvo et al., 2017) suggested that R245fa is more effective than R13a in conventional GPP (e.g., GEO bau ).This is because R245fa has a higher critical temperature, enabling a greater temperature difference between the hot and cold loops and resulting in higher efficiency.However, other factors such as environmental impacts (e.g., CCP), heat transfer characteristics, and safety considerations should also be included and optimised when selecting the organic working fluid.
Heat exchanger design: Optimisation of the heat exchanger should be undertaken to identify construction materials with high thermal conductivity, optimising the geometry of the heat exchanger, promoting the internal turbulent flow pattern, and minimising the fouling (Quoilin et al., 2013).
The new insights generated from the above suggested studies should be tested practically and analysed with respect to their impacts on the environmental sustainability performance of repurposed systems.

Comparison of results with literature
In order to establish a more precise understanding of the environmental performance of repurposed systems, the life cycle environmental impacts obtained in GEO bau , R-GEO double , R-GEO single , and R-GEO semi have been compared with other ORC-based GPPs from the literature (Li et al., 2023;Hirschberg et al., 2015;Tomasini-Montenegro et al., 2017;Ahmadi et al., 2020).The limited volume of comparative literature is due to the inconsistent system boundaries applied in the previous LCA studies as well as differences in impact assessment methodologies (Li et al., 2023).Only four commonly assessed indicators (CCP, FEP, MEP, SODP) with the same system boundaries were identified.No comparison with the results of repurposed systems is possible as this is the first study that performs LCAs of these alternatives.Therefore, Fig. 6 is plotted by using the obtained ranges of commonly assessed environmental impacts from the literature as benchmarks to compare with the performances of the evaluated geothermal power systems in this study.
According to Fig. 6, the performance of climate change potential (CCP), freshwater eutrophication potential (FEP) and marine eutrophication potential (MEP) in the geothermal power generation systems in this study are all lower than the collected minimum values from the literature.The main reason is this study's inclusion of recycling (e.g., of steel, copper, aluminium) at the end-of-life stage, which mitigated the corresponding environmental impacts, whereas the referred literature did not explicitly include this (Li et al., 2023;Hirschberg et al., 2015;Tomasini-Montenegro et al., 2017;Ahmadi et al., 2020).Stratospheric ozone depletion potential (SODP) shows a different trend, with all assessed options falling within the collected value ranges.Despite the results from repurposed systems being higher than GEO bau due to their low annual electricity generation (i.e., 125,120 kWh for repurposed systems), they are considerably lower than the maximum SODP value (93.2 μg CFC-11 eq./kWh) gathered from the literature.
Overall, when comparing the environmental performance of GPPs with literature values, there is a wide range of variations.These differences can be primarily attributed to site-specific factors such as power plant size, selection of input materials, and energy sources used for infrastructure construction.Importantly, Fig. 6 demonstrates that the repurposed systems in this study exhibit improvements in these four environmental impact categories compared to a broader range of ORC-based GPPs.None of these impact categories surpass the maximum values reported in the literature.These improvements indicate that the total electricity generation capability of the repurposed systems studied here does not act as an absolute barrier that would hinder their environmental performance when compared to the literature values.Therefore, repurposed systems can be viewed as an option that effectively enhances the environmental sustainability of geothermal power generation compared to ORC-based GPPs.However, it is crucial to conduct further comparisons with additional environmental impact categories (e.g., HTP cancer and non-cancer, IRP, FETP, METP) in order to provide further justification for this conclusion.

Conclusions
This study has assessed the effectiveness of repurposing abandoned oil and gas wells for geothermal power generation as a strategy to reduce drilling activities and improve the life cycle environmental performance of geothermal power plants (GPPs).Four comparable options were analysed from cradle to grave: i) a business-as-usual GPP (GEO bau ), ii) a repurposed system using two completely abandoned oil and gas wells (R-GEO double ), iii) a single-well system (R-GEO single ), and iv) a repurposed semi-abandoned two well system (R-GEO semi ).
The life cycle assessment reveals that repurposed systems do not consistently show reductions in all environmental impact categories compared to GEO bau .However, reduced drilling in repurposed systems effectively mitigates climate change potential.GEO bau performs best in nine out of the 20 impact categories considered, including freshwater eutrophication potential, marine eutrophication potential, and ionizing radiation potential.Among repurposed systems, R-GEO semi shows the highest potential for broad adoption.This conclusion is derived from its minimised underground drilling and construction activities, resulting in the lowest values in 11 impact categories, including fine particulate matter formation, fossil depletion and land use.Importantly, all of the rest of the indicators performed the second-lowest.However, R-GEOdouble is potentially considered the least favourable option, as it has the highest values in 12 impact categories and none of the indicators as the lowest.
R-GEO double and R-GEO single are penalised more by their low annual electricity generation and the impacts from the repurposing processes Fig. 6.Comparisons of life cycle environmental impacts between repurposed systems assessed in this study and the obtained the range of results of organic Rankine cycle based geothermal power plants from the literature (Li et al., 2023;Hirschberg et al., 2015;Tomasini-Montenegro et al., 2017;Ahmadi et al., 2020) Nomenclature for each geothermal power generation system and impact categories can be referred to Fig. 2.

Fig. 2 .
Fig. 2. Life cycle environmental impacts of different geothermal power generation systems.The impacts are presented per kWh of electricity generated.a a GEO bau : business-as-usual geothermal power plant; R-GEO double : repurposed two completely abandoned oil and gas wells for geothermal power production; R-GEO single : repurposed single completely abandoned oil and gas well for geothermal power production; R-GEO semi : repurposed semi-abandoned oil and gas wells for geothermal power production.The values are scaled for fitting purposes, and the actual values could be obtained by multiplying the factor provided in each bracket.CCP: climate change potential (including biogenic carbon); FPMFP: fine particulate matter formation potential; FDP: fossil depletion potential; FCP: freshwater consumption potential; FETP: freshwater ecotoxicity potential; FEP: freshwater eutrophication potential; HTP, cancer: human toxicity potential, cancer; HTP, noncancer: human toxicity potential, non-cancer; IRP: ionising radiation potential; LUP: land use potential; METP: marine ecotoxicity potential; MEP: marine eutrophication potential; MDP: metal depletion potential; POFEP: photochemical ozone formation, ecosystems potential; POFHP: photochemical ozone formation, human health potential; SODP: stratospheric ozone depletion potential; TAP: terrestrial acidification potential; TETP: terrestrial ecotoxicity potential; CED non-renewable: cumulative energy demand from non-renewable energies; CED renewable: cumulative energy demand from renewable energies.

Fig. 5 .
Fig. 5. Sensitivity analysis of the minimum required organic Rankine cycle efficiency to make the environmental performance of repurposed systems equivalent to GEO bau .Nomenclature for technologies and impact categories can be referred to Fig. 2.