Resources, Conservation & Recycling 199 (2023) 107287 Contents lists available at ScienceDirect Resources, Conservation & Recycling journal homepage: www.elsevier.com/locate/resconrec Full length article Resource recovery from desalination, the case of small islands Mar Palmeros Parada a, *, Serena Randazzo b, Gonzalo Gamboa c, Rodoula Ktori d, Britte Bouchaut e, Andrea Cipolina b, Giorgio Micale b, Dimitrios Xevgenos e a Water Management faculty, Delft University of Technology, Stevinweg 1, 26278ZN Delft,The Netherlands. b Dipartimento di Ingegneria (DI), Università degli Studi di Palermo, viale delle Scienze Ed.6, 90128 Palermo, PA, Italy c Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona. Edifici ICTA-ICP (Z), Campus UAB. 08193, Bellaterra (Cerdanyola del Vallès), Spain d Applied Sciences faculty, Delft University of Technology, Jaffalaan 5, 2628 BX Delft,The Netherlands. e Technology, Policy & Management faculty, Delft University of Technology, Jaffalaan 5, 2628 BX Delft,The Netherlands. A R T I C L E I N F O A B S T R A C T Keywords: This work explores resource recovery coupled to seawater desalination in small islands. As small islands depend Desalination on seawater desalination for water access, they make an excellent ground for exploring the trade-offs associated Circular economy to resource recovery, like potential economic gains, energy use, and environmental impacts. Here, we investi­ Resource recovery gated these tensions in the context of Lampedusa, in Italy. We then developed and evaluated scenarios for the Responsible innovation Sustainable design recovery of additional water, Mg, and other resources from brines, to identify if and how resource recovery is an interesting approach for the island vis-à-vis these tensions. We have found that the potential to increase water production with water recovery from brine is an interesting alternative for small islands, especially when har­ nessing waste heat. However, while some technologies offer possibilities for recovering additional resources, in places like small islands the potential benefits from additional recovery do not seem to justify the costs to the local system. 1. Introduction energy availability and costs. In this paper we discuss the case of desalination in the specific Seawater desalination (SWD) has been gaining prominence in water context of small islands, which often face water scarcity due to climatic scarce areas to satisfy water demand. However, as water demands are and topographical characteristics that do not allow for storage (Falkland met, questions over the impacts of desalination emerge: it requires vast and Custodio, 1991). Recently, SWD has gained prominence for pro­ amounts of energy with associated costs and greenhouse gas emissions, ducing freshwater in small islands, becoming the main or only source of and it produces a brine effluent that can affect aquatic eco-systems freshwater in some cases (Gómez-Gotor et al., 2018; Rossi, 2014). In this (Missimer and Maliva, 2018). In this context, resource recovery ap­ context, small islands where SWD takes such a prominent role for society proaches have been suggested as a means to reduce these impacts and – providing water access – make an excellent ground for exploring the contribute to circular economy objectives. Besides table salt (NaCl), tensions mentioned above. That is, when desalination is so critical for SWD brines are rich in Ca and Mg salts, the latter having being identified water access, is resource recovery desirable, and how? as a critical raw material in the European Union, for example (Magnus To investigate this question, we took as case study the demonstration Gislev and Milan Groho, 2018). However, questions remain about the of a resource recovery system in the island of Lampedusa, in Italy. The potential trade-offs that resource recovery in the context of seawater pilot-plant is based on a battery of technologies for the recovery of desalination brings (Palmeros Parada et al., 2022). For example, water, Ca, Mg and other chemicals from SWD brines, and was developed recovering resources implies additional processing steps that can in­ within the WATER MINING H2020 project. The technologies in the crease energy and capital costs, or bring additional economic benefits pilot-plant have been tested in previous R&D projects, and in this work depending on the resources to recover. To reduce the impacts of energy they were implemented within the local power plant to use waste heat use, some have proposed integration to alternative energy sources, on-site. Our analysis takes a Responsible Innovation perspective, which however these can bring additional issues, such as land use or renewable aims to make the innovation process more anticipatory, reflexive, and * Corresponding author. E-mail address: m.d.m.palmerosparada@tudelft.nl (M. Palmeros Parada). https://doi.org/10.1016/j.resconrec.2023.107287 Received 4 July 2023; Received in revised form 2 October 2023; Accepted 23 October 2023 Available online 26 October 2023 0921-3449/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). M. Palmeros Parada et al. Resources, Conservation & Recycling 199 (2023) 107287 responsive by promoting participation of stakeholders (Marques Postal to recover other salts (e.g. Mg(OH)2 and Ca(OH)2 from the MF-PFR; et al., 2020). To engage in the innovation process, we followed a “con­ Na2SO4 from the EFC) and chemicals (HCl, NaOH from the EDBM) text-sensitive design approach”. This approach is based on Value Sen­ from the NF concentrate rich in divalent ions. sitive Design (VSD), a design methodology to proactively integrate societal values in the design of technologies. In contrast to VSD, which is 2.2. Researcher and stakeholder roles typically applied for the design of technological products and often with direct contact with end users, in this work we use elements of this In this work, ‘researchers’ refer to the scientists within the project approach as a Responsible Innovation exercise. That is, the aim of this responsible of the research design and coordination for this work, with work is to use the design of a resource recovery system to gain first in­ experience on VSD and participatory approaches. ‘Technical partners’ sights on its societal implications in the context of small islands, with are researchers within the project responsible for the design and testing explicit recognition of emerging tensions between societal values of the technologies. Researchers designed surveys and interviews with around it. stakeholders, and technical partners conducted them in Italian. ‘Project partners’ refer to other project partners not directly involved with this 2. Methodology work or the technological system in question. Stakeholders are referred to a broad category of organizations and individuals, especially those The context-sensitive design approach incorporates elements of VSD beyond the project who may affect or be affected by the WM (Davis and Nathan, 2015; Friedman et al., 2017), sustainable design and technologies. participatory assessments (Barron et al., 2021; Gamboa et al., 2016; Palmeros Parada et al., 2020, 2018). Two participatory phases were 2.3. Main activities followed: Phase 1 (months 1–10), when societal (i.e. stakeholder values, value tensions and uncertainties) and technical aspects (i.e. design 2.3.1. Phase 1 variables and evaluation indicators) around the resource recovery sys­ A literature review was conducted during the first year of the project, tem are identified; and phase 2 (months 11–35), when technical sce­ with the aim to identify social values around resource recovery from narios for the full scale implementation of the system were developed desalination brines and support the development of surveys and inter­ and brought to stakeholders. In the next subsections we describe the case view guides. The results of the review have been published elsewhere study, the different roles in the project, the conducted activities along (Palmeros Parada et al., 2022) and are not elaborated here. the two project phases (see Fig. 1), and the data analysis method. Interviews were carried out to identify stakeholder values. For this, identified stakeholder organizations were invited for interviews. While 2.1. Case study the interviews followed this guide, they remained flexible to attend to any emerging topic. Interviews were recorded and transcribed and, Lampedusa is a small island of about 6000 inhabitants located in the when needed, translated to English. Transcripts of the interviews are southern Italy, between Sicily and northern Africa. Lampedusa suffers available by contacting the corresponding author. water scarcity and for a long time it depended on water imports. VSD Meetings with technical partners and researchers were held Freshwater is currently produced by an in-situ reverse osmosis desali­ during the first year of the project. These meetings had the aim to nation plant (Trapanese and Frazitta, 2018), which in small islands can familiarize researchers and project partners with their different back­ constitute around 30% of the total electricity use (Giudici et al., 2019). grounds, and to support the identification of technical and societal as­ The island is not connected to the Italian energy grid, and relies on a pects around the system. boat service to deliver diesel to a local power plant. As result, power and First stakeholder meeting (SM1) was held online at the end of the potable water production are more expensive than in the mainland, and first year of the project to present to stakeholders the demonstration have relied on public incentives to equalize costs to users with the system in the case study, and to open a discussion around the identified mainland (Curto et al., 2020). An increasing population in the summer values and tensions. (reaching more than 50,000 tourists (Tsalidis et al., 2023)) leads to increased electricity and water consumption, so water supplements by 2.3.2. Phase 2 boat are sometimes needed (Trapanese and Frazitta, 2018). Technical scenarios were developed to explore different ways of Fig. 2 shows the resource recovery system at the pilot-plant, which implementing the WM technologies at large scale. The scenarios explore uses waste heat from the local power plant where the pilot-plant is different technical configurations to respond to the identified value located. Seawater or seawater brine pass through a nanofiltration unit tensions directly (e.g. varying product qualities that result in different (NF) that splits the stream into a permeate rich in NaCl and a retentate costs) or indirectly (i.e. as basis to elicit a discussion about value ten­ where divalent salts, such as Mg and Ca, are concentrated. From there sions with stakeholders). In this way, the scenarios are a means to on, the system has two sides: the concentration side where the waste explore different technology development pathways, allowing to antic­ heat from the power plant is used to obtain pure NaCl (i.e. table salt) and ipate potential impacts and explore the identified tensions more water from the NF permeate stream by using the Multiple Effect Distil­ concretely. Although in this work all identified tensions are presented, lation (MED) and Evaporative Crystallization (EC); and a separation side here we only develop technical scenarios for one of the presented Fig. 1. Main activities for the context-sensitive design of a resource recovery system for SWD. 2 M. Palmeros Parada et al. Resources, Conservation & Recycling 199 (2023) 107287 Fig. 2. Diagram of the resource recovery process within the Water Mining project, indicating recovered resources: salts (green), chemicals (yellow) and water (blue). The concentration side is the processing train at the top; the separation side is the processing train at the bottom. tensions due to the scope of this work and time constraints. Neverthe­ identified from aspects of the technology or ways to implement it that less, recommendations for further research and implementation are can contribute and oppose several values at the same time. At the end of given for the other two tensions in Section 4. Phase 1, the identified values and value tensions were reviewed by all Scenario development and evaluation: A desk study to develop, VSD researchers and technical partners. These values and value tensions model and simulate technical scenarios was carried out in an interdis­ were the basis to construct the multicriteria structure presented in the ciplinary team of researchers and technical partners. The process started following section. with the analysis of how identified tensions could be explored through technical scenarios and their evaluation. Scenarios were then developed 2.4.2. From social values to indicators taking as basis the Water-Mining system being demonstrated in Lamp­ Indicators were identified to allow the comparison of performance in edusa, and exploring four main variables: (1) process and technology, relation to identified values. In this case, the technical scenarios are the (2) product and by-products, (3) scale and supply chain, (4) raw mate­ alternatives to be evaluated and compared under a set of criteria derived rials and utilities (Palmeros Parada et al., 2018). The scenario devel­ from identified social values. The information is structured in a m x n opment relied on data obtained from the demonstration system as well multicriteria impact matrix (i.e., with m rows and n columns), where n is as from literature, and was brought for discussion with technical part­ the number of alternatives and m the number of criteria. The element gij ners in month 15 and with project partners in month 25. Main consid­ of the matrix is the criterion score of alternative i under criterion j. This erations are included in Appendix 1, and the “Water Mining Scenario” information was used to foster discussion among stakeholders about the comprises the original set-up of the Water-Mining system. The descrip­ trade-offs in resource recovery from desalination brines in small islands. tion and design parameters of the demonstration system in Lampedusa To populate the impact matrix, we defined alternatives as explained can be found in Xevgenos et al. (2022), and technical details are avail­ in Section 2.3.2 Phase 2. The criteria were defined following Gamboa able at Xevgenos et al. (2022). Further analyses on bench-scale tests are et al. (2020, 2016). Social values, concerns and expectations were reported in Culcasi et al. (2022) and performance of individual process translated into a set of multi-dimensional criteria, and indicators rep­ steps at pilot scale in (Cassaro et al., 2023; Herrero-Gonzalez et al., 2023; resenting different views present in society. Criteria are the elements Morgante et al., 2022). Additionally, a separate manuscript is now being used within a specific narrative to describe a system: a description of an prepared to report the results from the integrated demonstration in observable relevant quality. For example, the assertion that “water Lampedusa island. desalination technologies are highly energy intensive” contains a value Second stakeholder meeting (SM2): A second online meeting with judgment, which is used to identify “energy consumption” as a criterion stakeholders took place on month 29, where the identified tensions and within it. To perform a quantitative characterization of the system under the technical scenarios were presented to stakeholders with aid of the study, indicators were identified as a means of representing a criterion, indicators identified in Phase 1. During the meeting participants were as an attribute of the system. They can be defined as the image of an divided into three groups and were asked to discuss which scenario attribute, formalized in terms of a specific measurement (Gallopín, G., would be the most suitable for Lampedusa. The groups were pre-selected 1997) For example, the “amount of MJ used as energy input to desali­ seeking to have stakeholders with different backgrounds in each group nate water” can be used as the indicator for the criterion “energy con­ and based on a list of confirmed participants. The aim was that, by sumption”. The value of the indicator provides information about the having participants from different backgrounds, they would have to condition and/or the trend of the attribute of the system. debate their choice and make explicit their judgements and assumptions. 3. Results and discussion 2.4. Data analysis 3.1. Stakeholder interviews and SM1 2.4.1. Value analysis From the interviews and the first stakeholder meeting an initial Values were identified from the surveys and interview outcomes, as identification was made of how stakeholders respond to the proposed well as from project documents. All documents and transcripts were resource recovery system in the island context. Particularly, stake­ analyzed with open coding, focused on objectives, expectations, hopes, holders stated the importance of desalination plants for covering water concerns around the project and case study. Coded segments were needs in the islands and expressed that projects that improve the per­ contrasted to each other to form value categories vis-à-vis the literature formance of such plants are positive. At the same time they showed review findings. This was done iteratively as new documents were concerns regarding the illegal extraction of underground and seawater retrieved and analyzed. Based on this analysis, value tensions were by some consumers, worsening the global availability and putting the 3 M. Palmeros Parada et al. Resources, Conservation & Recycling 199 (2023) 107287 system at risk, since seawater may enter into the aquifers and pollute thermal expansion (Beccali et al., 2020), even if in practice it has been them. While some stakeholders see the potential of extracting new re­ slower than expected as discussed by some stakeholders. sources from desalination as something positive, especially when it Another long-term issue that arises is that as drinking water needs comes to avoiding brine discharge and when it could economically are mostly covered by SWD already, and the effects on water con­ benefit the final consumer, some considered this a threat to their busi­ sumption when providing extra water are unknown. There is a potential ness or doubted its economic feasibility. Additionally, the high con­ to increase water consumption due to higher efficiency in the provision sumption of energy was seen as a prominent issue to tackle, and they of water, as discussed, for example, around irrigation in the agricultural considered the use of waste heat as an opportunity to reduce the global sector (Sears et al., 2018). In the case of Lampedusa, the effect could be footprint of the plant. through tourism expansion, which is a prominent economic activity in These findings are generally in agreement with the Water Mining the island, and lead to larger environmental impacts even at higher ef­ project, and little disagreement emerged from the gathered data. This ficiencies (i.e. what is commonly referred to as a ‘rebound effect’ result can be explained by the limited participation of local stakeholders (Makov and Font Vivanco, 2018)). despite efforts to engage them. An overview of participants can be found in Appendix 2, where it can be seen that, e.g., no representatives of the 3.2.3. Tension 3, with the ownership of water, recovered resources and local administration nor environmental organizations attended the ac­ technology tivities). Therefore, for the value analysis, these results were contrasted This tension is related to the ownership of seawater, of desalinated with the findings from the literature review as mentioned in Section 2.4. water, and of the technology developed with public funds. From the review of the literature, debates regarding the ownership and manage­ 3.2. Value tensions ment of water were identified (i.e. who should be benefitted and who is responsible for the management of resource recovery?) (Palmeros Par­ 3.2.1. Tension 1, between water and energy security, and sustainability ada et al., 2022). As this case study relates to seawater as a common The investigated system brings a tension between the values of water good, and is being partly developed with public funding, the question security, energy security and sustainability. Particularly, the system about who owns, manages and benefits from implementing this system would increase water availability and self-resilience, but would come at becomes relevant. For example, cost associated to the process of a cost on energy security by increasing the need for energy imports. resource recovery can be distributed differently across the recovered Additionally, impacts on different sustainability aspects are expected, resources, or be compensated through public support. In this way, how such as reducing brine discharge into the sea, increasing GHG emissions costs are distributed will affect the competitiveness of resource recovery due to higher energy use, and an uncertain impact on the cost of water. and the affordability of water in a water scarce region, or the public Although a large part of the energy demand would be provided by waste budget allocated to it. heat from the local power plant, the system would nevertheless require electrical energy to, e.g., operate membrane equipment and pump 3.3. Performance indicators streams through the system. Considering that even if only a small frac­ tion of energy demands is not covered by waste heat, this demand would Indicators are considered here as the technical translation of social add to the energy requirements of the island and increase energy imports values, concerns and expectations. The next Table 1 presents the criteria and the associated GHG emissions under the current energy system (see and indicators derived from the social values presented in the previous Tension 2 for a discussion on renewable energy sources (RES)). The section. impact of that additional energy on overall GHG emissions vis-à-vis impacts on water security, therefore needs to be investigated. With 3.4. Technical scenarios regards to impacts on the cost of water, there are uncertainties on the impact of resource recovery process on the production cost of water, and Considering the identified tensions and the scope of the WM system, on how much (and how) the recovered salts can compensate the cost of technical scenarios were developed around Tension 1, between water the process. In a context where public support to keep water or energy security, resource security and sustainability. Therefore, while all sce­ cost low is uncertain, this is a very relevant question to the stakeholders narios aim to use waste heat to increase water recovery from seawater in this case study. As these aspects can be explored quantitatively and reduce brine discharge (compared to a typical SWD), they do so through techno-economics and carbon foot-printing, it was decided to base the development of technical scenarios on this tension, and to bring Table 1 to discussions with stakeholders in Phase 2 of the project. Selected indicators for CS1 and associated criteria and values. Value Criteria Indicators Unit Directiona 3.2.2. Tension 2, between efficiency and long-term sustainability 3 Integration with waste heat is being proposed for the sake of energy Water security Water Volume of Hm / ↑ availability water year efficiency (i.e. less primary energy is needed for obtaining a given production amount of water when using waste heat), as an aspect contributing to Sustainability - Recovery Amount of % ↑ the overall sustainability of desalination and power production. How­ Resource Efficiency recovered ever, as the waste heat has fossil origins (i.e. diesel transported by boat), security minerals a risk of promoting a fossil energy lock-in was identified. That is, Sustainability - Energy Electrical kWhel/ ↓ Energy security consumption Energy year investing in equipment integrated with fossil resources could deter or consumption slow-down the adoption of renewable energies, as noted for other sec­ Sustainability - Energy Thermal Energy kWhth/ ↓ tors, e.g. (Janipour et al., 2020). When this issue was brought up to Energy security consumption consumption year discussion with stakeholders, it was pointed out that in Lampedusa there Sustainability - CO2 emissions Tons of CO2 tCO2eq/ ↓ Climate Change equivalent year are limited areas for Renewable Energy Sources (RES), leading to ex­ Mitigation pectations that fossil resources are the only large-scale energy alterna­ Economic viability Investment CapEx €/year ↓ tive in the short to medium term. This situation has also been discussed of the plant costs in the literature, when RES potential is limited by landscape protection Economic viability Operational OpEx € ↓ laws in addition to the surface limitation in small Italian islands (Giudici of the plant costs et al., 2019). Nevertheless, an expansion of RES in Lampedusa is already a Direction has two options: For maximizing higher is better, and for mini­ under consideration in local planning (with, e.g., photovoltaics and solar mizing lower is better. 4 M. Palmeros Parada et al. Resources, Conservation & Recycling 199 (2023) 107287 differently: Scenario 1 maximizes water and resource recovery as in the technology and recovered resources, research organizations in the en­ demonstration project and integrated to a reverse osmosis plant, Sce­ gineering and policy domains, and engineering companies (see Appen­ nario 2 does the same but as a standalone system (without RO), Scenario dix 2). Participants, who were divided in three discussion groups, largely 3 focuses only on water recovery to keep energy requirements low, and agreed on their preference for a water recovery scenario (Scenario 3) in Scenario 4 balances water and resource recovery with electricity re­ the context of Lampedusa. In their discussions, they referred to energy quirements. The scenarios are summarized in Table 2, and block consumption, efficiency and availability, technology reliability and de­ schemes of each scenario are included in Appendix 3. mand for recovered products in the deliberation. Particularly, two par­ ticipants of the first discussion group argued that from an energy 3.4.1. Technical scenario performance efficiency point of view, Scenario 1 is the best one. They also referred to The estimated performance of the technical scenarios is summarized reverse osmosis (part of Scenario 1) as the safest option in terms of it in Table 3. Scenario 3 on water recovery performs yields the most water being a technology with known performance. Participants in the second and performs better under most of the indicators except economic group additionally referred to the advantages of Scenario 1 in terms of margin and thermal energy requirements. Scenarios 1, 2 and 4 require economics, having a higher revenue potential. Nevertheless, low CO2 higher amounts of electricity and imply large CO2 emissions than sce­ emissions, as well as plant footprint and having Lampedusa as location, nario 3. Nevertheless, due to the recovery of multiple valuable products determined the preference for Scenario 3 in all groups. For the latter (high resource efficiency) with high revenue potential, their economic issue, participants in all groups discussed that a desalination plant in the margin are much higher than for Scenario 3, which has a negative small island of Lampedusa would not be a preferred alternative margin. The differences in water recovery are related to the consider­ considering the local energy constraints and the limited local industrial ation of evaporation ponds to reduce energy demand in Scenario 4, and activities (i.e. no chemical demand). When reflecting over the perspec­ the internal use of produced water within the systems to recover tives of stakeholders who were not present in the workshop (i.e. chemicals in the EDBM unit (Scenarios 1, 2 and 4). Despite that sce­ reflecting on what the perspectives of, e.g., civil servants and environ­ narios 1 and 2 have similar designs, they perform differently in eco­ mental protection agencies would be), they responded that they would nomics due to the different equipment capacities associated the volume likely prefer plants with an ‘environmentally friendly’ approach that reduction with an initial RO step. Scenario 4 focused on the recovery of relies on a circular system, implicitly referring to local production and Magnesium yields the highest economic margin despite showing the use. They then suggested it as a potential attractive point for the island. highest OPEX and CAPEX too. In this scenario, the potential for recov­ Lastly, Scenario 1 with the largest resource recovery potential and in­ ering and selling magnesium (Mg) and chemicals compensates for these tegrated with RO, and Scenario 4 with Magnesium recovery were dis­ high investments, resulting in a profitable scenario with a significant cussed as a desirably system for other locations without the energy, land, economic margin. The production rate of chemicals (NaOH, HCl) is and/or chemicals demand characteristics of Lampedusa. higher than the internal demand, allowing for additional profits. Despite the integration of fewer technologies and the use of evaporation ponds 4. Discussion and conclusions for NaCl recovery, Scenario 4 exhibits values for electrical energy con­ sumption and CO2 emissions in the same range as in Scenarios 1 and 2. The presented work shows an exploration of the societal desirability This is primarily attributed to the upscaling of energy-intensive tech­ of resource recovery for small islands, taking as case study Lampedusa. It nologies, such as EDBM. was found that resource recovery scenarios outperform substantially the In relative terms, scenario 3 yields about 33 m3 per each kWhel water focus scenario in terms of economics, however they lead to lower consumed, whereas scenarios 1, 2 and 4 deliver about 7 m3/kWhel; water recoveries, additional energy requirements, carbon emissions, and likewise, Scenario 3 yields about 17 m3 of recovered water for each they carry market uncertainties associated to the recovered resources. emitted tCO2eq, whereas scenarios 1, 2 and 4 deliver about 5 times less. That is, water, energy, and land are critical resources in Lampedusa and In economic terms, while Scenario 3 yields between 0.6 to 3 times more a system focused on water recovery, like in Scenario 3, yields the most water per invested euro than the other scenarios, it yields the lowest water resources per amount of energy spent, emitted emissions, and profitability since it has less revenues from additional resources. How­ required capital. This result, as well as the discussions with stakeholders ever, it must be noted that the estimated economic margin carries large who pointed to a limited industrial demand for recovered resources, uncertainties as the prices of recovered resources depend on achievable indicate that a water recovery focus is most suitable for Lampedusa. qualities and on market characteristics under uncertainty. Thus, based on this work, a potential higher profitability from recovered resources, like in Scenarios 1, and 4, does not seem to compensate for the 3.4.2. Stakeholder discussion in SM2 additional energy requirements in Lampedusa and the lack of local de­ Participating stakeholders to SM2 included potential users of mand for the resources. Newer technologies, such as membrane distil­ lation have the potential to lower the energy and emission impacts of Table 2 desalination (Abdel-Karim et al., 2021), however the comparison across Description of technical scenarios. scenarios would likely yield a similar conclusion for Lampedusa when factoring in the recovery and demand of salts and chemicals. Scenario Technologies Products This finding holds in the current Lampedusa context, where the 1 Integrated RO, NF, MED, ThCryst, Ca(OH)2, HCl, Ice, Mg availability of RES is limited and its development has been slower than scenario MFPR, EFC, EDBM (OH)2, NaCl, NaOH, Na2SO4, Water anticipated. Nevertheless, RES remains in long term regional plans 2 Water mining NF, MED, ThCryst, Ca(OH)2, HCl, Ice, Mg (Beccali et al., 2020). The eventual availability of RES would not lead, scenario MFPR, EFC, EDBM (OH)2, NaCl, NaOH, nevertheless, to an automatic desirability for a wider resource recovery Na2SO4, Water from brine in the island, even with higher water recovery and profit­ 3 Water recovery NF, MED, ThCryst Water, Mixed salts ability. If in the current state, intensive resource recovery is not desir­ scenario 4 Magnesium NF, MED, EvPond, Ca(OH)2, HCl, Mg(OH)2, able due to a large energy use with associated emissions, as well as recovery scenario MFPR, EDBM NaCl (lower purity), NaOH, reliability and energy security implications, the availability of RES Water brings additional questions around priorities for energy use. That is, EDBM: Electrodialysis with bipolar membranes; EFC: Eutectic freeze crystalli­ what would be the most desirable uses of limited RES, considering the zation; EvPond: Evaporation Pond; MED: Multi-effect distillation; MFPR: Plug- trade-off between its negative impacts (e.g. burdens on landscape or flow reactor; NF: Nanofiltration; RO: Reverse Osmosis, ThCryst: Thermal economic requirements) and the benefits of its possible uses, like crystallizer. house-hold electricity, transportation, water supply and sanitation, and 5 M. Palmeros Parada et al. Resources, Conservation & Recycling 199 (2023) 107287 Table 3 Summary of results of the evaluation of technical scenarios in CS1. Indicator Unit Directiona Scenario 1: RO-SWD Scenario 2: water Scenario 3: water Scenario 4: Mg integration mining recovery recovery Water production 1000m3/year ↑ 100 690 102 509 122 674 57 735 Resource efficiency % ↑ 96.19 92.06 94.41 79.64 Electricity consumption GWh/year ↓ 13.39 13.21 3.79 12.42 Thermal energy GWh/year ↓ 223.76 349.52 403.24 280.58 consumption CO2 emissionsb kton/year ↓ 26 516 26 152 7 500 24 582 OPEX M€/year ↓ 5.07 7.29 6.09 9.16 CAPEX M€ ↓ 32.11 40.11 24.55 37.77 Economic marginc €/m3 distillate ↑ 19.29 11.32 -5.38 42.10 water a Direction has two options: For maximizing higher is better, and for minimizing lower is better. bEmissions are those associated to energy use within the process and transportation of brine. c Economic margin represents the yearly potential revenues minus annualized production costs per amount of recovered water. Revenues are estimated for already marketed resources, considering average prices. industrial resources? These questions bring a component of environ­ concentrated solar power plants and flue-gas from power plants. mental justice that should be further explored, especially if economic Therefore, the design of the thermal equipment (MED and crystallizer) and environmental costs are taken by the public (e.g. (Levenda et al., should have in perspective the type of thermal RES in regional plans. 2021)). While Tensions 2 and 3 remained beyond the scope of this work, The presented results, nevertheless, indicate that in contexts where some recommendations for future study are extracted from this work. RES and demand for recovered chemicals are readily available, the re­ Particularly, it has been found that rebound effects can be avoided covery of additional resources like Mg(OH)2, HCl and NaOH would be an through policy and water management measures (Freire-González, interesting development, especially when coupled to RO SWD as in 2021). Possibilities to consider are, e.g., allocating the extra water to Scenario 1. The latter is emphasized by a possible use of HCl and NaOH users during peak demand only, and the rest of the time to substitute within the desalination process. Such a location could be, for instance, desalinated water capacity. However, the feasibility of this or other larger Mediterranean islands or desertic areas that also face water policy measures depends on contracts and arrangements between local scarcity but that have more alternatives for renewable energy sources, entities, and their impacts should be investigated in more detail. To and where there is infrastructure to use or trade the recovered resources, explore Tension 3 around the ownership of resources and technology, a like for plastic production in which Mg(OH)2 could be used as fire possible approach could be to combine the economic evaluation of retardant (Rothon and Hornsby, 2014). However, a more detail analysis technical scenarios like presented here, and ‘what-if’ analyses that would be necessary to identify how different sustainability aspects explore the impacts of, e.g. incentives, distribution of costs, and elicit would be affected in that context, and thus identify the most suitable stakeholders responses. Such an exploration would be helpful to iden­ configuration. A priori, targeting the recovery of only water, HCl and tify, with concrete examples, what are desirable policies or acceptability NaOH in Lampedusa could be interesting for the local desalination plant. factors for recovered resources (e.g. willingness to pay, perceived However, to recover these chemicals there is a need to pre-process the required subsidies…). streams before the EDBM equipment, which would turn the process into Overall, the presented results give an ex-ante estimation of the trade- something like Scenarios 1 and 4. offs involved in resource recovery from SWD, and the perspective of A possible argument in favor of intensive resource recovery from stakeholders on what is desirable for the small island of Lampedusa. desalination brines in the Lampedusa context could be additional water While resource recovery technologies offer many possibilities for recovery with the potential benefits from the export of recovered re­ recovering resources with high potential economic benefit, a simple sources. However, such potential economic benefits carry large un­ resource recovery system with a focus on additional water recovery from certainties, especially if markets for recovered resources need to be SWD brines seems to be the most suitable alternative for places like developed as has been discussed elsewhere (Kehrein et al., 2020), and small islands (i.e. with limited chemical demand, limited land to would need to be further studied, as well as how they would contrast implement RES, and dependence on fossil fuel imports). vis-à-vis overall environmental impacts, including energy for export. Even more, such a scenario would challenge a circular economy that CRediT authorship contribution statement benefits from short supply chains, bringing users and producers closer (Kiss et al., 2019). Some authors have discussed the role of ports in a Mar Palmeros Parada: Conceptualization, Formal analysis, Inves­ ‘global circular economy’, where they allow to take advantage of tigation, Methodology, Writing – original draft. Serena Randazzo: existing infrastructure and industrial symbiosis to bring benefits to local Investigation, Writing – review & editing. Gonzalo Gamboa: Formal economies (Gallaud and Laperche, 2016), however for small islands like analysis, Investigation, Methodology, Writing – original draft. Rodoula Lampedusa the availability for infrastructure and connectedness seems Ktori: Formal analysis, Investigation, Methodology, Writing – original limited. draft. Britte Bouchaut: Investigation, Methodology, Writing – review & In any case, flexibility has to be designed into the energy integration editing. Andrea Cipolina: Funding acquisition, Project administration, approach, to avoid lock-ins with fossil resources as discussed for tension Writing – review & editing. Giorgio Micale: Funding acquisition, 2. Particularly, a flexible integration between thermal equipment and Project administration, Writing – review & editing. Dimitrios Xevge­ waste heat is recommended for large scale implementation. That is, heat nos: Funding acquisition, Project administration, Writing – review & exchange equipment for the resource recovery system should be suitable editing. or easily adaptable to take heat at the conditions available at an existing power plant and heat deliverable from renewable sources in the plan­ Declaration of Competing Interest ning for Lampedusa (e.g. at certain temperature and pressure). For the resource recovery system as in the discussed in the research project, the The authors declare that they have no known competing financial ideal would be to have thermal energy directly from RES. However, the interests or personal relationships that could have appeared to influence type of heat delivered can vary largely between, e.g., solar collectors, the work reported in this paper. 6 M. Palmeros Parada et al. Resources, Conservation & Recycling 199 (2023) 107287 Data availability Gamboa, G., Mingorría, S., Scheidel, A., 2020. The meaning of poverty matters: trade-offs in poverty reduction programmes. Ecol. Econ. 169, 106450 https://doi.org/ 10.1016/j.ecolecon.2019.106450. Data will be made available on request. Giudici, F., Castelletti, A., Garofalo, E., Giuliani, M., Maier, H.R., 2019. Dynamic, multi- objective optimal design and operation of water-energy systems for small, off-grid islands. Appl. Energy 250, 605–616. https://doi.org/10.1016/j. apenergy.2019.05.084. Acknowledgements Gómez-Gotor, A., Del Río-Gamero, B., Prieto Prado, I., Casañas, A., 2018. The history of desalination in the Canary islands. Desalination 428, 86–107. https://doi.org/ This work was carried out within the WATER MINING (Next gener­ 10.1016/j.desal.2017.10.051. Herrero-Gonzalez, M., López, J., Virruso, G., Cassaro, C., Tamburini, A., Cipollina, A., ation water-smart management systems: large scale demonstrations for Cortina, J.L., Ibañez, R., Micale, G., 2023. Analysis of operational parameters in acid a circular economy and society) project. This project has received and base production using an electrodialysis with bipolar membranes pilot plant. funding from the European Union’s Horizon 2020 research and inno­ Membranes 13, 200. https://doi.org/10.3390/membranes13020200. Janipour, Z., de Nooij, R., Scholten, P., Huijbregts, M.A.J., de Coninck, H., 2020. What vation programme under grant agreement No 869474. We extend our are sources of carbon lock-in in energy-intensive industry? A case study into Dutch gratitude to the participating stakeholders and to Ada Agata Hlebowicz chemicals production. Energy Res. Soc. Sci. 60, 101320 https://doi.org/10.1016/j. for her data analysis support in the early stages of this work. erss.2019.101320. Kehrein, P., Loosdrecht, M.van, Osseweijer, P., Garfí, M., Dewulf, J., Posada, J., 2020. A critical review of resource recovery from municipal wastewater treatment plants – Supplementary materials market supply potentials, technologies and bottlenecks. Environ. Sci. Water Res. Technol. 6, 877–910. https://doi.org/10.1039/C9EW00905A. Kiss, K., Ruszkai, C., Takács-György, K., 2019. Examination of short supply chains based Supplementary material associated with this article can be found, in on circular economy and sustainability aspects. Resources 8, 161. https://doi.org/ the online version, at doi:10.1016/j.resconrec.2023.107287. 10.3390/resources8040161. Levenda, A.M., Behrsin, I., Disano, F., 2021. Renewable energy for whom? A global References systematic review of the environmental justice implications of renewable energy technologies. Energy Res. Soc. Sci. 71, 101837 https://doi.org/10.1016/j. erss.2020.101837. Abdel-Karim, A., Leaper, S., Skuse, C., Zaragoza, G., Gryta, M., Gorgojo, P., 2021. Magnus, Gislev, Groho, Milan, 2018. Report on critical raw materials and the circular Membrane cleaning and pretreatments in membrane distillation – a review. Chem. economy. Directorate-general for internal market, industry. Entrepr. SMEs Eng. J. 422, 129696 https://doi.org/10.1016/j.cej.2021.129696. (European Commission). Barron, L.J.R., Andonegi, A., Gamboa, G., Garmendia, E., García, O., Aldai, N., Makov, T., Font Vivanco, D., 2018. Does the circular economy grow the pie? The case of Aldezabal, A., 2021. Sustainability assessment of pasture-based dairy sheep systems: rebound effects from smartphone reuse. Front. Energy Res. 6, 39. https://doi.org/ a multidisciplinary and multiscale approach. Sustainability 13, 3994. https://doi. 10.3389/fenrg.2018.00039. org/10.3390/su13073994. Marques Postal, A., Benatti, G., Palmeros Parada, M., Asveld, L., Osseweijer, P., Da Beccali, M., Bonomolo, M., Di Pietra, B., Leone, G., Martorana, F., 2020. Solar and heat Silveira, J.M.F.J., 2020. The role of participation in the responsible innovation pump systems for domestic hot water production on a small island: the case study of framework for biofuels projects: can it be assessed? Sustainability 12, 10581. Lampedusa. Appl. Sci. 10, 5968. https://doi.org/10.3390/app10175968. https://doi.org/10.3390/su122410581. Cassaro, C., Virruso, G., Culcasi, A., Cipollina, A., Tamburini, A., Micale, G., 2023. Missimer, T.M., Maliva, R.G., 2018. Environmental issues in seawater reverse osmosis Electrodialysis with bipolar membranes for the sustainable production of chemicals desalination: intakes and outfalls. Desalin. Rev. Res. Develop. Desalin. 434, from seawater brines at pilot plant scale. ACS Sustain. Chem. Eng. 11, 2989–3000. 198–215. https://doi.org/10.1016/j.desal.2017.07.012. https://doi.org/10.1021/acssuschemeng.2c06636. Morgante, C., Vassallo, F., Battaglia, G., Cipollina, A., Vicari, F., Tamburini, A., Culcasi, A., Ktori, R., Pellegrino, A., Rodriguez-Pascual, M., van Loosdrecht, M.C.M., Micale, G., 2022. Influence of operational strategies for the recovery of magnesium Tamburini, A., Cipollina, A., Xevgenos, D., Micale, G., 2022. Towards sustainable hydroxide from brines at a pilot scale. Ind. Eng. Chem. Res. 61, 15355–15368. production of minerals and chemicals through seawater brine treatment using https://doi.org/10.1021/acs.iecr.2c02935. Eutectic freeze crystallization and Electrodialysis with bipolar membranes. J. Clean. Palmeros Parada, M., Asveld, L., Osseweijer, P., Posada, J.A., 2020. Integrating value Prod. 368, 133143 https://doi.org/10.1016/j.jclepro.2022.133143. considerations in the decision making for the design of biorefineries. Sci. Eng. Ethics. Curto, D., Favuzza, S., Franzitta, V., Musca, R., Navarro Navia, M.A., Zizzo, G., 2020. https://doi.org/10.1007/s11948-020-00251-z. Evaluation of the optimal renewable electricity mix for Lampedusa island: the Palmeros Parada, M., Asveld, L., Osseweijer, P., Posada, J.A., 2018. Setting the design adoption of a technical and economical methodology. J. Clean. Prod. 263, 121404 space of biorefineries through sustainability values, a practical approach. Biofuels, https://doi.org/10.1016/j.jclepro.2020.121404. Bioprod. Bioref. 12, 29–44. https://doi.org/10.1002/bbb.1819. Davis, J., Nathan, L.P., 2015. Value Sensitive Design: Applications, Adaptations, and Palmeros Parada, M., Kehrein, P., Xevgenos, D., Asveld, L., Osseweijer, P., 2022. Societal Critiques. In: van den Hoven, J., Vermaas, E.P., van de Poel, I. (Eds.), Handbook of values, tensions and uncertainties in resource recovery from wastewaters. J. Environ. Ethics, Values, and Technological Design: Sources, Theory, Values and Application Manage. 319, 115759 https://doi.org/10.1016/j.jenvman.2022.115759. Domains. Springer Netherlands, Dordrecht, pp. 11–40. Rossi, R., 2014. MSc Thesis. Lund University. Falkland, A., Custodio, E., 1991. Hydrology and Water Resources of Small islands: a Rothon, R., Hornsby, P., 2014. Chapter 9 - Fire retardant fillers for polymers. In: Practical guide, Studies and Reports in Hydrology. UNESCO, Paris. Papaspyrides, C.D., Kiliaris, P. (Eds.), Polymer Green Flame Retardants. Elsevier, Freire-González, J., 2021. Governing Jevons’ Paradox: policies and systemic alternatives Amsterdam, pp. 289–321. https://doi.org/10.1016/B978-0-444-53808-6.00009-3. to avoid the rebound effect. Energy Res. Soc. Sci. 72, 101893 https://doi.org/ Sears, L., Caparelli, J., Lee, C., Pan, D., Strandberg, G., Vuu, L., Lin Lawell, C.-Y.C., 2018. 10.1016/j.erss.2020.101893. Jevons’ paradox and efficient irrigation technology. Sustainability 10, 1590. https:// Friedman, B., David, G.H., Alan, B., 2017. A survey of value sensitive design methods. doi.org/10.3390/su10051590. Foundations and Trends® in Human–Computer Interaction 11, 63–125. https://doi. Trapanese, M., Frazitta, V., 2018. Desalination in small islands: the case study of org/10.1561/1100000015. Lampedusa (Italy). In: OCEANS 2018 MTS/IEEE Charleston. Presented at the Gallaud, D., Laperche, B., 2016. Circular Economy, Industrial Ecology and Short Supply OCEANS 2018 MTS/IEEE Charleston, pp. 1–7. https://doi.org/10.1109/ Chain: Towards Sustainable Territories. John Wiley & Sons, Hoboken, UNITED OCEANS.2018.8604484. STATES. Incorporated. Tsalidis, G.A., Xevgenos, D., Ktori, R., Krishnan, A., Posada, J.A., 2023. Social life cycle Gallopín, G., 1997. Indicators and their use: information for decision-making. In: assessment of a desalination and resource recovery plant on a remote island: analysis Moldan, B., Billharz, S. (Eds.), Sustainability Indicators. Report of the Project on of generic and site-specific perspectives. Sustain. Prod. Consump. 37, 412–423. Indicators of Sustainable Development. Scope, pp. 111–112, 58. https://doi.org/10.1016/j.spc.2023.03.017. Gamboa, G., Kovacic, Z., Di Masso, M., Mingorría, S., Gomiero, T., Rivera-Ferré, M., Xevgenos, D., Diamantidou, D., Cipollina, A., Trezzi, A., Xenogianni, C., Alarcon, D., Giampietro, M., 2016. The complexity of food systems: defining relevant attributes Sapoutzi, D., Ktori, R., Rodrigues Pascual, M., 2022. Deliverable 3.3 “Demo systems and indicators for the evaluation of food supply chains in Spain. Sustainability 8, used in CS1 and CS2. Water-Mining Project. 515. https://doi.org/10.3390/su8060515. 7