Integration of microgrids in chemical industries with hydrogen as a byproduct: Styrene production Case Study

International Journal of Hydrogen Energy 59 (2024) 947–957 Available online 13 February 2024 0360-3199/© 2024 The Authors. Published by Elsevier Ltd on behalf of Hydrogen Energy Publications LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Integration of microgrids in chemical industries with hydrogen as a byproduct: Styrene production Case Study Sergio J. Navas b , * , G.M. Cabello Gonz´ alez b , Alan Pino a , F.J. Pino b a AICIA, Escuela T´ ecnica Superior de Ingeniería, Camino de los Descubrimientos s/n, 41092, Sevilla, Spain b Dpto. de Ingeniería Energ´ etica, Escuela T´ ecnica Superior de Ingeniería, Universidad de Sevilla, Camino de los Descubrimientos s/n, 41092, Sevilla, Spain ARTICLE INFO Keywords: Chemical process Simulation Solar energy Microgrids Hydrogen ABSTRACT The chemical industry serves as a global economic backbone and it is an intensive consumer of conventional energy. Due to the depletion of fossil fuels and the emission of greenhouse gases, it is necessary to analyze energy supply solutions based on renewable energy sources in this industrial sector. Unlike other sectors, such as residential or service industries, which have been thoroughly analyzed by the scientific community, the use of renewable energies in the chemical industry remains comparatively less examined by the scientific community. This article studies the use of an energy supply system based on photovoltaic technology or a PEM fuel cell for a styrene production industry, analyzing the integration of energy storage systems such as batteries, as well as different uses for the surplus hydrogen produced by the facility. The most interesting conclusions of the article are: (1) the renewable microgrid considered is viable both technically and economically with a discounted payback period between 5.4 and 6.5 years, using batteries as an energy storage system; and (2) the use of hydrogen as energy storage system for a styrene industry is not yet a viable option from an economic point of view. 1. Introduction The chemical and petrochemical sectors stand as integral components of the global economy, generating trillions of dollars in revenue every year. Nevertheless, these industries are characterized by exhibiting high energy consumptions, whether in the form of thermal or electrical energy [1]. Thermal energy is mainly expended in the separation units, such as distillation columns. Conversely, electrical energy powers auxiliary equipment, like pumps and compressors [2]. Traditionally, these energy requirements have been met through fossil fuel based electricity from the primary electrical grid or via the utilization of diesel microgrids. That way, the sector mainly relies on fossil fuel, being one of the major contributors to global industrial CO 2 emissions. The latest Intergovernmental Panel on Climate Change (IPCC) report on climate change [3] underscores the role of anthropogenic emissions in accelerating climate change and affect the entire Earth’s climate system, where only strong and sustained reductions in emissions of greenhouse gases could mitigate these effects. Hence, there is an imperative need to explore innovative and sustainable solutions. Addressing this challenge involves multiple aspects. On one hand, it involves the development of new chemical routes that make use of renewable raw materials and, on the other hand, it requires the integration of renewable energy into the chemical industrial process [4]. A necessary step forward in a decarbonized chemical industry involves replacing the conventionally fossil fuels derived heat and power with electrical energy derived from renewable sources [5,6]. In this sense, Spain has immense potential to incorporate renewable energy into its industrial sector. With abundant solar and wind resources, the country can leverage these clean energy sources to power various industrial processes. By transitioning to renewable energy, Spain can reduce carbon emissions, enhance energy security, and create new job opportunities, contributing to a sustainable and low-carbon future for its industries [7]. While numerous studies exist on hybrid solar energy systems with hydrogen and electrical energy storage for residential houses [8] or even entire cities [9], assessing the integration of solar-hydrogen systems from both technical and economic perspectives, there is a scarcity of research in the realm of chemical industries. A review from 2011 examining the use of solar energy in the industrial sector [10] comprehensively explores its applications, developments, and future proyection. It highlights how incorporating solar energy can enhance both the quality and the quantity of industrial products while mitigating greenhouse gas emissions. Solar thermal and photovoltaics (PV) systems * Corresponding author. E-mail address: [email protected] (S.J. Navas). Contents lists available at ScienceDirect International Journal of Hydrogen Energy journal homepage: www.elsevier.com/locate/he https://doi.org/10.1016/j.ijhydene.2024.01.290 Received 23 October 2023; Received in revised form 8 January 2024; Accepted 24 January 2024 International Journal of Hydrogen Energy 59 (2024) 947–957 948 are both suitable for various industrial process applications, but the overall efficiency of the system depends on appropriate integration and proper design of solar collectors. Large-scale solar thermal systems are economically viable due to the use of stationary collectors [11], while PV systems are reliable substitutes for power in buildings and industrial processes, especially in remote regions. Recent studies have assessed the integration of solar energy within the chemical industry. For example, a study evaluated the potential of using solar energy for the ammonia synthesis by Haber-Bosch process, aiming to reduce or eliminate the use of fossil fuels in the process [12]. The study demonstrates the efficient production of ammonia from renewable energy sources through three designed plants using a thermochemical model. The design provides a distributed, efficient, and sustainable ammonia production that integrates renewable energy sources such as solar/wind electricity, biogas from anaerobic digestion, and syngas from biomass gasification. Such integration reduces CO 2 emissions and costs, increases supply security, and allows for the storage of renewable energy into an energy vector with low emissions and high energy density. The production process consumes around 14–15 kW/kg NH 3 and has no discernible impact on primary energy consumption. The study of Panchenko et al. [13] reviews and analyzes contemporary methods and technologies for utilizing solar energy in anaerobic digestion systems. The findings indicate widespread use of solar radiation, particularly through solar thermal collectors for heating reactor substrates and photovoltaic modules for powering low-energy equipment and control systems. While research on the hybrid use of solar radiation for both heat and electricity supply is less extensive, it holds significant potential due to the overall efficiency of solar photovoltaic thermal modules. Concentrator solar systems, capable of supporting thermal processes and simultaneous electricity generation, are highlighted, especially when combined with matrix photovoltaic converters. The integration of solar converters in anaerobic bioconversion systems enhances economic efficiency by reducing the reliance on produced gas for internal power needs, making these systems more cost-effective and operationally efficient. Additionally, the use of green hydrogen, produced through solar-powered electrolysis, improves biogas quality, transforms carbon dioxide into biomethane, and reduces greenhouse gas emissions. Likewise, Chen et al. [14] compared the environmental and economic performance of three methanol production routes in China: conventional coal, solar energy coupled with coal gasification, and solar energy-biomass assisted CO 2 hydrogenation. The results show that the methanol production routes incorporating solar energy have significantly smaller environmental impacts compared to conventional coal methods. However, their methanol production cost is about 3–5 times higher. While the solar energy-biomass route yields negative greenhouse gas emissions, it is not currently economically feasible due to high costs. To render it economically viable, the development of highly efficient photovoltaic electricity generation and biomass integrated gasification combined cycle systems, coupled with a reduction in capital costs, is imperative. Regarding methanol production, Svitnic et al. [15] focused on the design of a renewable methanol production system using the FluxMax approach. The design incorporated waste-heat utilization and considered various subsystems such as generation, storage, and chemical production. The research identified a cost-effective process configuration that harnessed parabolic troughs for process heat, thermal energy storage for managing daily fluctuations, and waste-heat utilization technologies like steam turbines and heat pumps. Although the calculated costs of these designs did not currently match market prices, they provided valuable insights for future research in the line of enhancing the competitiveness of renewable methanol production. Also, in the renewable methanol production framework, the research of Van Antwerpen et al. [16] introduces an open-source cost framework, encompassing a farm-to-gate model of power to metanol, incorporating various factors such as renewable power generation, electrolyser configurations, and balancing technologies. By applying this framework to case studies in Australian and international locations, the study estimates indicative levelized costs of methanol (LCMeOH), providing insights into the economic viability of power to methanol pathways. The open-source tool is designed to support the assessment, optimization, and deployment of commercial-scale methanol projects worldwide. The estimated LCMeOH for power to methanol, including CO 2 sourced from industrial flue gas and direct air capture, is 1360 $/tMeOH and 1710 $/tMeOH, respectively. Also in the line of introducing renewable energies in the industry, Hwangbo et al. [17] proposed a mathematical model called eRELU to develop electricity networks based on renewable energy in large petrochemical industries in South Korea. The model comprises a deterministic component for optimizing utility networks and a stochastic component for constructing clean electricity networks, integrating variable renewable energy and energy storage systems. By incorporating carbon capture and storage systems, the model successfully captures a significant amount of greenhouse gases from utility networks, leading to a 16% reduction in total costs and the mitigation of 114 tCO 2 /d compared to the base case. Finally, Igogo et al. [18] discussed the potential for renewable energy integration in the mining industry, focusing on iron ore extraction and gold production. While both activities demand significant amounts of electricity, iron and steel production processes also rely on fossil fuels for heat and feedstock, therefore, technical challenges need to be addressed, including feedstock replacement, process heat demand, and renewable energy variability. They show biomass and hydrogen as potential alternatives, but found that sustainable production at the scale required by mining is challenging. Anyway, hybrid systems and energy storage can help mitigate renewable energy variability. Among the chemical industries, the styrene industry is a relevant candidate for the inclusion of renewable energies. Styrene, an important monomer and platform chemical, plays a pivotal role in various industries, serving as the precursor for the production of numerous polymers types including polystyrene, foamed plastics, styrene-modified polyesters, and ABS, among others. Today, global styrene production reaches approximately 30 million tons per year, with a predominant reliance on petroleum-derived ethylbenzene [19]. The production of styrene involves the dehydrogenation of ethylbenzene, a reaction characterized by its endothermic nature (124 kJ/mol) and high energy activation (152 kJ/mol), resulting in high operating temperatures for the reactor (around 600 ◦C). The considerable heat generated, along Nomenclature CAPEX capital expenditure CCS carbon capture and storage CO 2 carbon dioxide DPP discounted payback period EUA emission unit allowance H 2 hydrogen IRR internal rate of return LHV lower heating value NG natural gas NPV net present value O&M operation and maintenance OPEX operating expenses PEM proton exchange membrane PSA pressure swing adsorption PV photovoltaic SOC state of charge SPP simple payback period TOU time of use VAT value added tax S.J. Navas et al. International Journal of Hydrogen Energy 59 (2024) 947–957 949 with the necessary energy to produce the steam that accompanies ethylbenzene into the reactor to lower its partial pressure, as well as the posterior separation and purification of styrene, renders this industry a substantial energy consumer, especially in the form of heat. A simplified diagram of the styrene production process can be later found in Fig. 1 in section 2.1. To satisfy the whole energy demand of a styrene chemical industry, the implementation of a low-carbon microgrid, based on renewable energy, is proposed. This kind of system has the potential to enable small or medium-sized industries to achieve energy cost saving, increase energy supply reliability, and meet renewable energy targets, thereby contributing to the reduction of greenhouse gas emissions [20]. The two renewable energy sources under consideration are a photovoltaic system with or without batteries and a hydrogen fuel cell, both designated to provide electricity. Photovoltaic technologies provide an extended service life, low operation costs, and have experienced a considerable reduction in module prices during the last years. The utilization of hydrogen for storing electrical energy from renewable sources has been widely assessed [21,22]. Specifically, the use of hydrogen as an energy vector is the key point of this study, with its potential role as either a reactant or a byproduct contingent upon the specific chemical industry evaluated. Traditionally, surplus hydrogen has been sold and utilized in other industries, such as ammonia synthesis or other hydrogenation processes, where it serves as a raw material. However, hydrogen is a valuable fuel, and its use as an energy vector can be beneficial for the own facility to cover the electrical or heat demand. The design of the microgrid coupled with an styrene facility will be based on previous studies [23,24] carried out in the laboratory named HyLab, situated at the Engineering School of the University of Seville. These studies have developed simulation models of the different equipment in the proposed microgrid that have been validated experimentally [24]. Therefore, the main objective of this paper is to explore the feasibility of integrating a microgrid based on renewable energy, utilizing the produced hydrogen produced as an energy vector. The main scientific novelties of this study can be summarized as follows: - As revealed in the literature review, there is no similar renewable energy integration approaches for the styrene industry. This paper presents a pioneering approach to bridge the research gap in utilizing the excess hydrogen generated by chemical industries directly as energy vector for the own industry. - The study contributes with a comprehensive modeling and analysis of the chemical process involved in styrene production, including the energy integration process to determine the load profile, energy consumption, and hydrogen by-product stream to estimate its potential utilization as an energy source. - Technical-economic evaluation of implementing of a microgrid with photovoltaic solar energy in a chemical industry characterized by high electrical consumption and continuous 24 h operations. To the best of our knowledge, there are no reference studies, mainly due to the intermittence and variability of the solar resource, requiring energy storage systems that increase the cost of the solar system. - Economic analysis of the utilization of the hydrogen available in the byproduct stream for various scenarios: purification and bulk sale to other customers, partial use in a hydrogen fuel cell for electricity generation, and use to enrich the gas consumed in the boiler for heat supply. - Scenario analysis to evaluate the economic performance of the proposed cases for current energy prices at the industry location and hypothetical scenarios with reduced energy price scenarios to establish economic feasibility limits. That way, as above-mentioned, this study introduces a groundbreaking approach to address the research gap in directly utilizing excess hydrogen generated by chemical industries as an energy vector for internal use, particularly focusing on the styrene industry, where similar renewable energy integration approaches are notably absent in the literature. The study includes a comprehensive modeling and Fig. 1. Process flow diagram of the styrene facility. S.J. Navas et al. International Journal of Hydrogen Energy 59 (2024) 947–957 950 analysis of the styrene production process, assessing the potential utilization of the hydrogen by-product as an energy source. Additionally, it conducts a technical-economic evaluation of implementing a solarpowered microgrid in a 24-h operational chemical industry, presenting novel insights given the solar resource’s intermittence and variability. Consequently, the first step of this work involves the study of the base case, namely, the detailed energetic study (electrical and thermal energy demands) of the selected chemical process (a 115.9 kmol/h styrene production facility in Tarragona, Spain) that may be suitable for the integration of a microgrid based on renewable energies. Then, certain scenarios will be proposed to meet the required demand [25]. Finally, the results will be discussed to determine what kind of microgrid could be recommended for this particular process from an economic point of view. 2. Model description and simulation 2.1. Styrene plant The chemical plant considered in this study is a styrene production plant that generates hydrogen as a byproduct. The process flow diagram for the styrene plant is depicted in Fig. 1 and was based on the one presented in Annex B3 in Ref. [26]. An explanation of the styrene production process will be as followed. The incoming stream of fresh ethylbenzene is combined with recycled ethylbenzene, heated, and then mixed with superheated steam. Steam, which is an inert in the reaction, provides the energy required to drive the reaction and influences the equilibrium of the reaction of styrene production (Equation (1)). After that, the reactant enters two adiabatic reactors with intermediate heating. The resulting products of the reaction are cooled, producing steam, and then directed to a three-phase separator whree light gases (mainly hydrogen), organic liquid, and water are separated. The organic stream is first introduced into a distillation column, where benzene and toluene (byproducts resulting from side reactions, Equation (2) and Equation (3), respectively) are extracted as the distillate product, and then introduced into a second distillation column where the styrene is extracted as the bottom product and ethylbenzene as a distillate product, that is recycled back into the reaction process. C6H5C2H5↔C6H5C2H3+H2(Equation 1) C6H5C2H5→C6H6+C2H4(Equation 2) C6H5C2H5+H2→C6H5CH3+CH4(Equation 3) Kinetics were implemented as a simple power law for all reactions (Equations (4)–(7)). Concentration units are partial pressures in bar, and reaction rates have units of kmol/m 3 s. The values of the kinetic parameters are presented in Table 1. R1F=pEBk1Fe−E1F/RT (Equation 4) R1R=pSpHk1Re−E1R/RT (Equation 5) R2=pEBk2e−E2/RT (Equation 6) R3=pEBpHk3e−E3/RT (Equation 7) The simulation of the chemical plant evaluated in this study was carried out using Aspen Hysys. The styrene plant has been simulated for a target production of 12080 kg/h with 99.9% purity. In Table 2, the conditions and main composition of the main streams are presented, each stream number corresponding to the one shown in Fig. 1. After completing the simulation, including thermal energy integration between the different consumptions and productions, the plant exhibited net thermal and electric power consumptions of 55770 kW and 253.3 kW, respectively. The electric power associated to each component can be seen in Table 3. Whereas, the thermal consumption it is only associated to the steam current production due to the other thermal consumptions being satisfied by the leftover steam current or the steam generated in the cooling system after the second reactor. 2.2. PV and battery system The simulation of the photovoltaic energy production, battery charge/discharge cycles, and electricity consumption and injection from/to the grid was conducted employing TRNSYS 18 [27]. This commercial software is widely used to simulate renewable energy systems, including PV and solar thermal [28]. Not only documentation and components’ model are continuously updated by the developer team [29], but also there are several studies performed by independent researchers validating the performance of the components models [30], and quantifying the uncertainty of its results [31]. Type194 from TRNSYS components library was used to model the photovoltaic array. It calculates the power output at the maximum power point taking into account the meteorological data, which was supplied employing Type15 from a Meteonorm TMY file of the location [32]. Losses due to the operating temperature of the cell were considered in the PV model. In the case that the microgrid system proposed included energy storage, a battery pack was modeled employing Type47a. A 90% constant charging efficiency was assumed. In addition, a component modeling a delayed output device (Type150) was included to save the state of charge (SOC) of the batteries of the current simulation time step to be used in the next one. The simulation time step was 15 min. Moreover, a charge controller was modeled based on equations that compare the instantaneous electricity demand from the industrial facilities with the PV generation. In the event of a PV power surplus, the batteries are charged until they reach a SOC of 90%. Once this SOC threshold is attained, power is injected into the grid. Conversely, in the case of power deficiency (i.e., PV power <electric demand), batteries are used until they reach a lower threshold of 30%. Below this limit, electricity is purchased from the grid. If no battery were considered in the proposed microgrid, the charge and discharge conditions were ignored. In addition, the hours of the year were classified into 6 time bands employing Type9, which reads an input file with the hour classification. Therefore, the energy excess and deficit were calculated and aggregated for the different period. These time bands correspond to the six electricity tariffs currently in force in Spain (Time of Use rates) to be considered in the economic analysis. Finally, integrator components Table 1 Kinetic parameters. k E, kJ/kmol Equation (1) Forward 42000 90826 Equation (1) Reverse 210 −32652 Equation (2) 300000000 207840 Equation (3) 42000 112370 Table 2 Stream conditions and composition. Conditions Stream Number 1 2 3 4 5 6 7 Temperature, ◦C 136 800 610.2 570.5 203 124.7 124.3 Pressure, kPa 200 4202 170 120 200 65 200 Component Flowrates, kmol/h Water 0 4016 3000 3000 5.9 0 0 Ethylbenzene 121 0 228 110.5 1.2 108.2 0.1 Styrene 0 0 0 116.7 0.8 115.9 115.9 Hydrogen 0 0 0 116 116 0 0 S.J. Navas et al. International Journal of Hydrogen Energy 59 (2024) 947–957 951 calculated hourly and monthly values for the variables of interest in order to generate the output file that was used in the economic analysis. Fig. 2 depicts the diagram of the TRNSYS model. On the other hand, to model the hydrogen consumption and electricity production of a fuel cell, the nominal values declared by the manufacturer have been considered. This choice is informed by the continuous operation of the fuel cell at its nominal load, according to the 24/7 operation of the factory. Therefore, the fuel cell does not experience power ramps up or down or operation at partial load. Similarly, when modelling a boiler that operates with natural gas or other fuels, the nominal efficiency, fuel consumption and heat (or steam) production was assumed to be constant, due to the continuous consumption of the plant. 3. Microgrid solutions This section outlines various microgrid designs evaluated for their integration into the styrene chemical plant. These designs are compared to a base case that does not incorporate a microgrid to assess if it is economically viable for their implementation and, if so, to determine the amortization period. An overview of the characteristics and microgrid components that define each case is presented in Table 4 and Fig. 3. 3.1. Base case The base case does not include the use of a microgrid. Therefore, the electric and thermal energy demand of the plant is procured by external sources, entailing payments to an electricity and a natural gas company, respectively. The hydrogen produced as a byproduct undergoes purification through a Pressure Swing Adsorption (PSA) system and then sold as pure hydrogen. This case is considered to be the actual way the proposed chemical industry satisfies its energy demands and manages the hydrogen current that is produced as byproduct. Regarding, CO 2 emissions, a conversion factor between energy from natural gas and CO 2 emission of 0.252 tonCO 2 /MWh is considered in this study [33]. However, CO 2 emissions related to electricity generation are not calculated herein since they strongly depend on the electric mix of the location and the distribution company and contract. In addition, the economic benefits for generating carbon-free electricity are perceived by the electricity supply company and not the styrene industry analyzed and, thereby, it does not affect in the cash flow of the industry. As a benchmark, the average CO 2 emission intensity of the Spanish electric mix in 2022 was 0.16 tonCO 2 /MWh [34]. 3.2. Case 1 Case 1 assesses the inclusion of a PV system to provide a share of the electric demand. The PV system presents an installed capacity of 612 kWp. This power was selected considering that the ratio between the annual energy produced by the PV system and the annual electric energy demanded during solar hours should be approximately 1. 3.3. Case 2 Case 2 involves the incorporation of a lead-acid battery storage system alongside the PV system. The power installed in this case is 1240 kWp, aiming to maintain a ratio of approximately 1 between the annual energy produced by the PV system and the annual electric energy demand. The battery system has a capacity of 3000 kWh to keep the ratio between annual energy excesses and defects, and the energy produced by the PV system around 10%. Table 3 Electric power consumption of equipment. Equipment Consumption, kW Recycled Ethylbenzene pump 0.933 Hydrogen compressor 192.7 Purge compressor 1.050 Styrene product pump 0.800 Water pump 2.677 Energy integration pumps 55.122 Fig. 2. Simulation deck of the PV and battery system model in TRNSYS. Table 4 Characteristics and microgrid components of each case. Purified H 2 sale H 2 Combustion PV Batteries Fuel Cell Base Case ✓ x x x x Case 1 ✓ x ✓ x x Case 2 ✓ x ✓ ✓ x Case 3 ✓ x x x ✓ Case 4 x ✓ x x x S.J. Navas et al. International Journal of Hydrogen Energy 59 (2024) 947–957 952 3.4. Case 3 Case 3 analyzes the use of a fuel cell to produce the electricity required by the styrene plant utilizing a fraction of the purified hydrogen. After the purification process, an available hydrogen stream of approximately 163.64 kg/h is obtained. The production of 253.3 kW requires a hydrogen flow rate of 15.2 kg/h; therefore, the rest of the stream is sold as in the base case. 3.5. Case 4 Case 4 studies the use of the hydrogen byproduct stream, before its purification process, as fuel for the gas boiler that produces the thermal energy required by the styrene plant. Therefore, the boiler fuel corresponds to an enriched natural gas, characterized by a higher lower heating value (LHV). This application of the byproduct hydrogen stream results in a reduction of 14.6% of natural gas consumption and 10.6 % decrease in carbon dioxide emissions. 4. Techno-economical assessement For the techno-economic analysis, it is necessary to establish all the boundary conditions to compare the four case studies. This section indicates the values employed to perform the analysis. For instance, the methodology for energy calculations, the economic metrics to compare the results, the period of analysis and, most importantly, the costs associated with investment (or CAPEX) and operation and maintenance (or OPEX). 4.1. Metrics Commonly, in energy projects, the cost per unit of energy is the most relevant parameter for investment decisions. However, in industrial applications, the amortization of a project must be adjusted to the investment horizon of the company. Therefore, the choice may prioritize lower development effort and shorter payback time over options that optimizes long-term costs. Consequently, two economic metrics have been employed for this analysis: the Net Present Value (NPV) and the Discounted Payback Period (DPP). The NPV measures the difference between the present value of cash inflows and the present values of cash outflows over a period of time. It is employed to analyze the profitability of an investment or project. If the NPV is positive at the end of the period, the project is considered as profitable. The Payback Period or Simple Payback Period (SPP) is the amount of time (in years) it takes for the initial cost of a project to equal the expected cash flow value. It is used to estimate the profitability of a project., however, it does not consider the time value of money. Therefore, a better option is the DPP, which considers the discounted values of expected cash flows and, therefore, a discount rate must be provided. A project is profitable when the resulting DPP is less than the useful life of the project. The initial cash flow is considered negative as it is an investment, but the cash flows of the following periods is positive as they include de company profits. Therefore, the DPP will occur when the negative cumulative discounted cash flows become positive. The Internal Rate of Return (IRR) represents the discount rate that makes the NPV of an investment equal to zero. Essentially, the IRR is the rate at which the cash inflows from an investment equal the cash outflows, taking into account the time value of money. Higher IRRs in investments are generally considered more attractive because they offer a higher potential return. For the present study, a discount rate of r =5% and a total analysis period of 25 years are assumed. 4.2. Market conditions In order to properly evaluate the economic performance of each Fig. 3. Simplified microgrid diagram for the different cases. S.J. Navas et al. International Journal of Hydrogen Energy 59 (2024) 947–957 953 proposed energy supply option, the investment and Operations and Maintenance (O&M) cost assumptions must be established in detail. The reference values used in the study are presented here. 4.2.1. PV system The cost of PV systems is highly dependent on the cost of the PV module (between 30 and 40% of the total cost). An increase in PV module prices was seen between 2020 and 2021 due to increased demand but, by the end of 2022, their cost dropped to the levels of the beginning of 2020 thanks to increased production by manufacturers [35]. The total installed cost includes the PV modules, the inverter, the balance of plant, and installation costs. For small-scale PV systems, such as residential setups, the average installed system price without value-added tax (VAT) is around 1100 € /kWp [36]. For utility scale systems (>10 MW) prices for 2021 were around 790 € /kWp [37] for the total installed cost. However, the system sizes analyzed in this work are well below utility scale, thus the price reference for small systems is assumed. Additionally, an annual operation and maintenance cost equivalent to 3% of the initial investment cost is considered. In addition, the land rental cost is included in the study. For this purpose, the current cost per unit area of undeveloped flat land in the vicinity of the plant was analyzed. The assumed land price is 2.63 € /m 2 -year, based on 11 sample price announcements consulted in January 2023, removing 3 outliers (avg.:0.219 € /m 2 -month, max. 0.500 € /m 2 -month, min. 0.125 € /m 2 -month, std. dev.: 0.1325 € /m 2 -month). A coverage ratio of 2.5 was used, i.e. the land area is 2.5 times the area of the PV modules. The expected lifetime of the PV system is 25 years, then it should be replaced by a new system. 4.2.2. Batteries The battery pack considered in the analysis is of the lead-acid type. Given the stationary nature of the application, there are no space constraints for the batteries. Besides, lead-acid batteries have a significant lower cost when compared to Li-ion, which Capital Expenditure (CAPEX) was around 800 € /kWh [37] in 2022. For the analysis, the investment cost of lead-acid batteries is assumed to be 100 € /kWh, based on a quote received in 2022 from a local supplier in Spain. The expected lifetime of the lead-acid batteries is 14 years, after which they should be replaced with a new battery pack. 4.2.3. Fuel cell The fuel technology assumed in this analysis is of Proton Exchange Membrane (PEM) type. The investment cost has been calculated as the average between the values of 2020 and the target of 2024 provided by the Clean Hydrogen Partnership [38]. Therefore, the CAPEX of the fuel cell is 1550 € /kWe. The operation and maintenance cost, averaged over the first 10 years of the system, is considered to be at the 2020 level, standing at 0.05 € /kWh. This cost includes potential stack replacement, but excludes the fuel cost (H 2 ). The electrical efficiency of the fuel cell is assumed to be 50% over the lower heating value (LHV). 4.2.4. Hydrogen burner The LHV of hydrogen surpasses that of natural gas, and its combustion does not release greenhouse gases into the environment, hence being attractive to be used as a heat source. However, using pore hydrogen in a burner/boiler presents a technical challenge, being only viable under ultra-lean conditions due to issues as blow-off and flash back. A feasible limit of 15–20% of H 2 can be mixed with natural gas, increasing the heating value of the mix. The main modification required is to include a mixing manifold at the feeder of the natural gas (NG) burner [39]. For the present analysis, it is interesting to assess the benefits of injecting the byproduct stream of the process to the feeder of the boiler. Since the components of the byproduct stream are hydrocarbons, containing H 2 , it can be used in the combustion chamber directly, since this stream represents less than 15% of the gas mix. That way, no investment cost is assumed. In addition, the H 2 available in the process stream should not be purified, eliminating associated cost. The nominal efficiency of the boiler using either natural gas or enriched natural gas is assumed constant at 85% based on the fuel LHV. 4.2.5. Price of natural gas, electricity, hydrogen and carbon dioxide The impact of the cost of fuel and electricity is particularly relevant in economic analysis. Globally, the price of fossil fuels shows considerable variability depending on supply, demand, and geopolitical conflicts. The global price of natural gas has increased since the restrictions on Russian gas imports due to the invasion of Ukraine in February 2022. This has, in turn, led to a substantial increase in the wholesale electricity market price in most parts of Europe. Therefore, the price of electricity in Spain has also increased substantially. Industrial districts are present in most cities of Spain, often supplied with natural gas at very competitive cost. Despite the potential for industrial solar thermal energy, it has not proven attractive due to prevailing natural gas prices ranging between 18 and 24 € /MWh [40]. Since October 1, 2021 a new classification for natural gas consumers applies, which impacts in the distribution fees. Eleven categories were created depending on the annual consumption, ranging from under 5000 to over 500000000 kWh/year. The industry under this study is classified in the RL.10 category (150000000 <Consumption ≤500000000 kWh/year). The Ministry of "Ecological Transition and the Demographic Challenge" established indexed values for the natural gas consumers, which varies every three months if the wholesale market experiences a ±2% variation compared to the previous period. The values for the RL.10 category corresponding to the first trimester 2023 are: fixed term of 27292.74 € /month, and variable term of 0.0552 € /kWh [41]. For the analysis, two natural gas tariff was considered: the current one of 0.0552 € /kWh (high price) and a hypothetical one at 0.02 € /kWh (low price), reflecting a scenario where fossil fuel prices decrease to levels similar to 2020 and before. Moreover, a specific tax for natural gas of 0.00234 € /kWh must be added [42]. Finally, the boiler efficiency is assumed to be 80%. Large industries can access to more favorable electricity rates than smaller consumers due to economies of scale. There are two options when contracting power with the distribution companies: (a) selecting a fixed-rate offered by the distribution company, which changes every trimester according to the distribution fees established by the Government, with the price per unit of energy is fixed for a specified period (commonly for 12 months); or (b) opting for a variable tariff indexed to the wholesale market values plus the distribution fees. For either case, specific rates, including all the additional fees, are not listed since the different distribution companies can slightly modify them to offer better rates and compete among them. In addition, six-time bands (time-of-use or TOU) are considered in the Spanish electricity supply (P1 to P6), which depend on the hour of the day, the day of the week and the month of the year. Table 5 details the distribution of the TOU periods along the year. Two tariffs have been selected for this study, listed in Table 6. The first one (A) is an actual tariff offered by a company for the first trimester 2023 (commercial name 6.1TD) and represents the high-price scenario [43]. The second one (B) corresponds to the average wholesale market price for 2022 plus the distribution fees established by the government [44] and represents the low-price scenario. For both tariffs, the rate of the contracted power is assumed to be the same, and belongs to the one in Tariff A. In addition, there is a 0.5% of specific tax to electricity, which is not included in the rates of Table 3. Furthermore, since 2019, users in Spain with self-generation capabilities can inject the surplus production to the grid and receive economic benefit. Compensation for selling electricity is expressed as a percentage of the purchased electricity price. The regulation does not fix this percentage, hence, distribution companies establish it. In 2021, certain companies paid prosumers between 44% and 83%, depending on the month and the TOU period. However, for 2022, the percentages ranged between 23% and 49%. For the present analysis, it is assumed that surplus electricity S.J. Navas et al. International Journal of Hydrogen Energy 59 (2024) 947–957 954 will be sold to the grid at a 25% of the purchase price of the respective TOU band. Hydrogen, usually from non-renewable sources (gray hydrogen), is traded commodity within the chemical sector. The styrene industry is assumed to sell 99% pure hydrogen on the wholesale market at a price of 6.20 € /kgH 2 , while the purification cost is 1.39 € /kg (on the by-product stream) [45]. The recovery rate has been assumed to be 70% of the hydrogen present in the stream. A 2% annual reduction of the prices is assumed for the first 17 years, with cost remaining constant thereafter. Activities emitting CO 2 are subject fees. Consequently, there is a market for offsetting these emissions through the trade of greenhouse gas emission allowances representing 1 ton of CO 2 that has not been emitted into the atmosphere. For the economic analysis, a price for Emission Unit Allowance (EUA) of 80.87 € /tonCO 2 is employed, corresponding to the 2022 average price. The price has exponentially increased from 4.45 € /tonCO 2 in 2013 to 86.88 € /tonCO 2 in the first trimester of 2023 [46]. A 10% annual increase of the prices is assumed for the first 7 years, with costs stabilizing thereafter. Furthermore, fossil fuels long-term price projections show an increase for the next years. A 3.3% annual increase is expected for natural gas and a 4.2% for oil and liquefied petroleum gas [47]. Since the Spanish electricity generation mix is still considerably dependent on natural gas, a 3% annual increase for both electricity and natural gas price is assumed for this study. Although the study considers the selected values of increment/decrease of price of energy, H 2 and CO 2 , the economic calculation tool developed allows for the modification these values to analyze more conservative or risky scenarios. 5. Results The main results of the analysis include the economic metrics established for the comparison. Each scenario represents a unique set of assumptions and variables that affect the cash flow of a project, investment, or business venture over time of the particular microgrid solution relative to the base case. The objective is to compare different technological solutions under current and future market conditions. While certain parameters have been varied in order to observe their impact in the financial performance, it is not within the objectives of this study to find the optimum solution, perform a sensitivity analysis, or to propose a hybrid alternative including more than one option. The CO 2 emissions related to every case are 153892 tonCO 2 /year, except for case 4 that are 137540 tonCO 2 /year, a 10.6 % less than in the other cases. In case 1, 2, and 3 the two electricity rates described in section 4.2 are employed to understand the long-term impact of the energy cost. In addition, since nowadays the PEM fuel cell and the use of hydrogen to generate electricity are too expensive, for case 3 it was also calculated the current price of purified hydrogen at which the solution would be cost-effective. This results in a reduced hydrogen price of 3.96 € /kg instead of the actual 6.2 € /kg. Furthermore, when employing the hydrogen and other by-products to enrich the natural gas stream of the burner (case 4) the two prices of natural gas declared in section 4.2 are considered. The following figures plot the cumulative cash flows for the three different technological solutions. For instance, Fig. 4 presents the photovoltaic option with and without batteries (case 2 and 1 respectively), Fig. 5 presents the PEM fuel cell option, and Fig. 6 depicts the enriched natural gas option. The cumulative cash flow is the sum of all cash inflows and outflows that have occurred up to a specific point in time. By presenting this data graphically, it becomes easier to analyze and compare the performance of each scenario, identify trends, and make informed decisions. From Fig. 4 it can be inferred that, regardless of the use of batteries or the considered price electricity, a photovoltaic system proves profitable (positive NPV). Furthermore, the payback periods range between 5.4 and 6.5 years, which are short enough to be attractive for an industry. The systems that include a battery pack require a higher investment but also offer higher profitability. On the other hand, the use of a PEM fuel cell to self-generate electricity is not profitable, leading to a 25-year NPV of −3.5 and −4.2 million euros for expensive and cheap electricity scenarios, respectively (Fig. 5). This is mainly due to the elevated O&M cost of the fuel cell, with stack degradation being a significant operating costs. The conventional use of hydrogen, such as selling it to other companies, proves more profitable. A reduction of 36% in the current hydrogen price from 3.96 euros/kg instead of 6.2 euros/kg could yield a positive NPV. However, the payback period would be 25 years, diminishing its economic attractiveness. Finally, if the by-product stream were burned directly in the boiler without any modifications to the boiler itself, the inversion would be profitable from the first year (Fig. 6). This is attributed to the high price of natural gas, the substantial CO 2 emission fees, and the limited utility of selling purified hydrogen wholesale considering the current cost of the pressure swing adsorption process. The summary of economic results is listed in Table 7. The NPV, DPP and IRR were calculated for the 25-year period. When there is no DPP value, it means that the investment is not amortized within the period of analysis. If its value is 0, it means that the project is profitable from the start. 6. Conclusions The results obtained in this paper lead to a series of conclusions that are described as follows. Firstly, the incorporation of a microgrid to produce part of the electricity demanded by the styrene plant is profitable with a DPP between 5.4 and 6.5 years. This range depends on the price of the electricity considered and on the inclusion of lead-acid batteries in the system. In addition, it seems clear that the inclusion of lead-acid batteries results in a higher NPV with mostly the same DPP than in the case without them. Table 5 Spanish electricity supply time of use for industrial users. Table 6 Rates extracted for an industrial electricity contract. Time-of-use bands Contracted power price, € /kW/year Energy price, € /kWh Tariff A Tariff B P1 30.55 0.237 0.216 P2 25.91 0.209 0.190 P3 14.92 0.192 0.173 P4 12.10 0.183 0.164 P5 3.94 0.161 0.146 P6 2.11 0.175 0.162 S.J. Navas et al. International Journal of Hydrogen Energy 59 (2024) 947–957 955 Secondly, the use of the purified hydrogen stream to produce electricity by means of a PEM fuel cell is not economically viable at this moment. The sale price of the hydrogen should decrease by around 36% to have a DPP of less than 25 years. For that reason, it is not recommended to be implemented in the microgrid. Finally, the option of using the by-product hydrogen stream before purification for thermal energy production appears to be a profitable choice, with a DPP of 0. This is due to the fact that there is no need of investment in the steam boiler because the by-product stream results in less than 15% of the gas mix, and in turns, it presents a high saving in natural gas consumption and CO 2 emissions. Overall, it can be concluded that the inclusion of renewable energy sources in chemical industries may be not only technically viable, but also profitable in some cases, as demonstrated in this paper. Specifically, Fig. 4. Cumulative cash flow for photovoltaic with (case 2) and without (case 1) batteries cases. Fig. 5. Cumulative cash flow for the PEM fuel cell case (case 3). Fig. 6. Cumulative cash flow for the enriched natural gas burner case (case 4). S.J. Navas et al.