Techno-Economic Assessment of On-Site Production of Biomethane, Bioenergy, and Fertilizer from Small-Scale Anaerobic Digestion of Jabuticaba By-Product

: Bioenergy recovery from biomass by-products is a promising approach for the circular bioeconomy transition. However, the management of agri-food by-products in stand-alone treatment facilities is a challenge for the low-capacity food processing industry. In this study, the techno-economic assessment of a small-scale anaerobic digestion process was evaluated for the management of jabuticaba by-product and the production of biomethane, electricity, heat, and fertilizer. The process was simulated for a treatment capacity of 782.2 m 3 y − 1 jabuticaba peel, considering the experimental methane production of 42.31 L CH 4 kg − 1 TVS. The results of the scaled-up simulated process demonstrated the production of biomethane (13,960.17 m 3 y − 1 ), electricity (61.76 MWh y − 1 ), heat (197.62 GJ y − 1 ), and fertilizer (211.47 t y − 1 ). Economic analysis revealed that the process for biomethane recovery from biogas is not proﬁtable, with a net margin of − 19.58% and an internal rate of return of − 1.77%. However, biogas application in a heat and power unit can improve project feasibility, with a net margin of 33.03%, an internal rate of return of 13.14%, and a payback of 5.03 years. In conclusion, the application of small-scale anaerobic digestion can prevent the wrongful open-air disposal of jabuticaba by-products, with the generation of renewable energy and biofertilizer supporting the green economy toward the transition to a circular economy.


Introduction
The production of renewable energy based on sustainable techniques has been a challenge for the reduction of greenhouse gases and climate change [1]. Recently, there are governmental incentives to reduce the use of non-renewable energy sources, especially petroleum-based energy, which has expanded the research trends on the production of renewable energy based on biomass, wind, solar, geothermal, oceanic, and hydroelectric sources [2]. However, knowledge advancement of renewable energy production revealed that, in some cases, there are several implementation problems [3], including the demand for large areas, the limitation and dependence on adverse climate conditions, environmental damage in areas of occupation, and high implementation costs [4]. Currently, incentives for renewable energy can be observed in Europe, where the consequences of sanctions applied in the conflict between Ukraine and Russia decreased the supply of natural gas [5,6]. From an environmental perspective, the decrease in the use of natural gas can be an advantage, especially because natural gas emits greenhouse gases and causes disturbances in the ecosystem [7]. However, alternatives for renewable energy production should be better investigated for long-term energy supply [8].
A promising alternative for the consolidation of renewable energy is applying anaerobic digestion (AD) technology [9,10], which can be a sustainable process to produce biogas.

Recovery of Biomethane, Bioenergy, and Fertilizer
The experimental biogas production and composition are presented in Figure 1 considering semi-continuous AD in a laboratory-scale reactor of 4.3 L. Experimental biogas production at the laboratory scale and other parameters were established by da Rosa et al. [31]. The results demonstrate that the AD reactor oscillated the methane content during the 50 d, which is expected for the digestion of lignocellulosic biomass. Stabilization occurred after approximately 20 d, when the methane content increased and reached 50%. After 50 d of semi-continuous AD, the accumulated biogas and biomethane volumes reached 49.6 and 14.93 L, respectively. The experimental methane yield obtained at the end of the process was 42.31 L kg −1 TVS (total volatile solids). This experimental methane yield was used as a basis for the scale-up and economic analysis.
in the jabuticaba processing industry, subsidizing decision-making processes towards the adoption of the waste management process.

Recovery of Biomethane, Bioenergy, and Fertilizer
The experimental biogas production and composition are presented in Figure 1 considering semi-continuous AD in a laboratory-scale reactor of 4.3 L. Experimental biogas production at the laboratory scale and other parameters were established by da Rosa et al. [31]. The results demonstrate that the AD reactor oscillated the methane content during the 50 d, which is expected for the digestion of lignocellulosic biomass. Stabilization occurred after approximately 20 d, when the methane content increased and reached 50%. After 50 d of semi-continuous AD, the accumulated biogas and biomethane volumes reached 49.6 and 14.93 L, respectively. The experimental methane yield obtained at the end of the process was 42.31 L kg −1 TVS (total volatile solids). This experimental methane yield was used as a basis for the scale-up and economic analysis.   Table 1 considering the established results obtained by da Rosa et al. [31]. Evaluating the scale-up AD process, Scenario 1 considered the recovery of biomethane through the purification of biogas and injection into the gas network to replace the use of natural gas. In this case, the production of biomethane reached 13,960.17 m 3 y −1 in the scaled-up simulated process. For Scenario 2, the biogas produced was converted into electricity (61.76 MWh y −1 ) and heat (197.62 GJ y −1 ) considering the application of biogas in a combined heat and power unit. In Scenario 3, the biogas was upgraded into electricity (61.76 MWh y −1 ) in a generator without heat recovery. In all the scenarios studied, fertilizer production from the digestate was considered (211.47 t y −1 ). By evaluating the data obtained, the process that was designed can be an advantage for the jabuticaba processing industry, where the by-products can be converted into biomethane, electricity, heat, and fertilizer, thus presenting value-added products to reduce energy consumption and recover natural and sustainable fertilizer for agricultural application. In addition, other important advantages of using bioenergy produced by the AD process are the environmental, social, and health benefits due to the burning of a cleaner fuel, which can reduce deforestation by replacing the use of wood while also creating new sources of energy and fertilizer generation [32,33].  Figure 2 presents the flow diagram for the small-scale AD of jabuticaba agro-industrial by-product and the recovery of biomethane, electricity, heat, and fertilizer. The results obtained for the simulated scenarios are presented in Table 1 considering the established results obtained by da Rosa et al. [31]. Evaluating the scale-up AD process, Scenario 1 considered the recovery of biomethane through the purification of biogas and injection into the gas network to replace the use of natural gas. In this case, the production of biomethane reached 13,960.17 m 3 y −1 in the scaled-up simulated process. For Scenario 2, the biogas produced was converted into electricity (61.76 MWh y −1 ) and heat (197.62 GJ y −1 ) considering the application of biogas in a combined heat and power unit. In Scenario 3, the biogas was upgraded into electricity (61.76 MWh y −1 ) in a generator without heat recovery. In all the scenarios studied, fertilizer production from the digestate was considered (211.47 t y −1 ). By evaluating the data obtained, the process that was designed can be an advantage for the jabuticaba processing industry, where the by-products can be converted into biomethane, electricity, heat, and fertilizer, thus presenting value-added products to reduce energy consumption and recover natural and sustainable fertilizer for agricultural application. In addition, other important advantages of using bioenergy produced by the AD process are the environmental, social, and health benefits due to the burning of a cleaner fuel, which can reduce deforestation by replacing the use of wood while also creating new sources of energy and fertilizer generation [32,33].

Economic Analysis
It is important to evaluate the small-scale AD of food industry by-products, and economic analysis provides important insights for verifying project feasibility for further implementation [34,35]. Process simulation can be a positive approach to identifying obstacles in the implementation of the project, which can avoid difficulties during the implementation process associated with project execution [36]. In this simulation, the annual feedstock demand (jabuticaba by-product) for the AD process was equivalent for the three scenarios. The scaled-up process was simulated with a flow of jabuticaba byproduct equal to 782.2 m 3 y −1 . For AD, it is necessary to control the pH of the reactor, and in this case, 9.67 kg NaOH y −1 was necessary. Operational labor was considered the same for the studied scenarios. The amount of water in the process can be considered reused water, which is produced in the AD process, decreasing the costs for the company. For the simulated process, it was estimated that 294.13 m 3 y −1 of water could be reused in the AD process. The annual sales of biomethane, electricity, heat, and fertilizer can be observed in Table 2. Scenario 2 presented the highest revenues (12,192.30 USD y −1 ), followed by Scenario 3 (10,565.89 USD y −1 ) and Scenario 1 (5211.29 USD y −1 ).

Economic Analysis
It is important to evaluate the small-scale AD of food industry by-products, and economic analysis provides important insights for verifying project feasibility for further implementation [34,35]. Process simulation can be a positive approach to identifying obstacles in the implementation of the project, which can avoid difficulties during the implementation process associated with project execution [36]. In this simulation, the annual feedstock demand (jabuticaba by-product) for the AD process was equivalent for the three scenarios. The scaled-up process was simulated with a flow of jabuticaba byproduct equal to 782.2 m 3 y −1 . For AD, it is necessary to control the pH of the reactor, and in this case, 9.67 kg NaOH y −1 was necessary. Operational labor was considered the same for the studied scenarios. The amount of water in the process can be considered reused water, which is produced in the AD process, decreasing the costs for the company. For the simulated process, it was estimated that 294.13 m 3 y −1 of water could be reused in the AD process. The annual sales of biomethane, electricity, heat, and fertilizer can be observed in Table 2. Scenario 2 presented the highest revenues (12,192.30 USD y −1 ), followed by Scenario 3 (10,565.89 USD y −1 ) and Scenario 1 (5211.29 USD y −1 ). Table 2. Annual sales of biomethane, electricity, thermal energy, and fertilizer from the small-scale AD of jabuticaba by-product.  Table 3 presents the significant costs of the process, and Figure 3 shows cost discrimination over the cost of manufacturing (COM). The sum of fixed capital investment (FCI), cost of operational labor (COL), cost of utilities (CUT), cost of waste treatment (CWT), and cost of raw material (CRM) represents the total annual cost for carrying out the AD of jabuticaba by-product. In general, the results showed that the management of jabuticaba by-product with the simulated small-scale AD process can be considered a low-cost process for properly disposing waste from industrial processing [37]. The total cost was higher for Scenario 2 (9034.76 USD y −1 ), especially because this scenario presented high FCI (4990.80 USD y −1 ). In addition, FCI had the highest cost in all scenarios, representing 50.88% for Scenario 1, 55.24% for Scenario 2, and 53.55% for Scenario 3. High FCI may be associated with the high implementation cost of the biogas purification equipment, the heat and power unit, and the power generator. CRM (162.94 USD y −1 ) and COL (2160.00 USD y −1 ) were the same for all scenarios since the processes were simulated with the same flow of jabuticaba by-product; consequently, the same amount of water, inoculum, and NaOH for pH neutralization is necessary in the process. CUT was higher for Scenario 1 (1605.59 USD y −1 ), while Scenario 2 and Scenario 3 required lower expenditures (1285.52 USD y −1 ), which can be associated with the electricity demand for operating the process of biogas purification in Scenario 1, a cost that is not necessary in the other scenarios. In this study, CWT was considered the same for all three scenarios (435.50 USD y −1 ) since the same amount of digestate was upgraded into fertilizer. CWT consisted of the cost associated with the construction process for upgrading the digestate into fertilizer. Cost discrimination over the COM was FCI > COL > CUT > CWT > CRM. FCI represented up to 55% of total costs, followed by COL (23.91-24.81%) and CUT (14.23-18.07%). CWT and CRM represented the lowest contribution, lower than 5% in all scenarios. According to Fernando-Foncillas and Varrone [38], installation and labor costs represent more than 50% of the operating costs in the AD process. Vinardell et al. [39] demonstrated that the waste treatment and disposal fee represented the highest cost contribution, showing a notable impact on the net cost. Finally, Sillero et al. [40] found that the highest costs of the AD process were FCI and the CUT.  Table 3 presents the significant costs of the process, and Figure 3 shows cost discrimination over the cost of manufacturing (COM). The sum of fixed capital investment (FCI), cost of operational labor (COL), cost of utilities (CUT), cost of waste treatment (CWT), and cost of raw material (CRM) represents the total annual cost for carrying out the AD of jabuticaba by-product. In general, the results showed that the management of jabuticaba by-product with the simulated small-scale AD process can be considered a lowcost process for properly disposing waste from industrial processing [37]. The total cost was higher for Scenario 2 (9034.76 USD y −1 ), especially because this scenario presented high FCI (4990.80 USD y −1 ). In addition, FCI had the highest cost in all scenarios, representing 50.88% for Scenario 1, 55.24% for Scenario 2, and 53.55% for Scenario 3. High FCI may be associated with the high implementation cost of the biogas purification equipment, the heat and power unit, and the power generator. CRM (162.94 USD y −1 ) and COL (2160.00 USD y −1 ) were the same for all scenarios since the processes were simulated with the same flow of jabuticaba by-product; consequently, the same amount of water, inoculum, and NaOH for pH neutralization is necessary in the process. CUT was higher for Scenario 1 (1605.59 USD y −1 ), while Scenario 2 and Scenario 3 required lower expenditures (1285.52 USD y −1 ), which can be associated with the electricity demand for operating the process of biogas purification in Scenario 1, a cost that is not necessary in the other scenarios. In this study, CWT was considered the same for all three scenarios (435.50 USD y −1 ) since the same amount of digestate was upgraded into fertilizer. CWT consisted of the cost associated with the construction process for upgrading the digestate into fertilizer. Cost discrimination over the COM was FCI > COL > CUT > CWT > CRM. FCI represented up to 55% of total costs, followed by COL (23.91-24.81%) and CUT (14.23-18.07%). CWT and CRM represented the lowest contribution, lower than 5% in all scenarios. According to Fernando-Foncillas and Varrone [38], installation and labor costs represent more than 50% of the operating costs in the AD process. Vinardell et al. [39] demonstrated that the waste treatment and disposal fee represented the highest cost contribution, showing a notable impact on the net cost. Finally, Sillero et al. [40] found that the highest costs of the AD process were FCI and the CUT.

Cost of Manufacturing
Measurement of the COM is one of the main parameters used to determine project feasibility [41]. Table 4 presents the COM of biomethane, electricity, thermal energy, and fertilizer for the scenarios studied. In Scenario 1, the COM of biomethane and fertilizer was 0.72 USD m −3 and 47.20 USD t −1 , respectively. For Scenario 2, the COM of electricity (157.58 USD MWh −1 ), heat (0.049 USD MJ −1 ), and fertilizer (46.02 USD t −1 ) production were determined. In addition, Scenario 3 showed a marginal increase in the COM of electricity (155.96 USD MWh −1 ) compared to Scenario 2 (157.58 USD MWh −1 ). The COM of fertilizer in Scenario 3 was 45.54 USD t −1 , similar to the other scenarios. In previous studies, Sillero et al. [40] demonstrated that the economic analysis of a temperature-phase anaerobic co-digestion of sewage sludge, wine vinasse, and poultry manure revealed that the COM of electricity (84.99 USD MWh −1 ), heat (0.019 USD MJ −1 ), and fertilizer (30.91 USD t −1 ) was lower than that in the present study, which can be associated with the scale of the anaerobic reactor. In addition, Sganzerla et al. [42] demonstrated that the lowest COM was obtained for the AD process operated with a higher processing capacity. Although the COM for biomethane, electricity, heat, and fertilizer was higher compared to the literature, the cost benefits of the small-scale AD of jabuticaba by-product should be observed, as well as the annual sales and profitability indicators. In addition, this study is the first approach in determining the economic assessment of small-scale AD of jabuticaba by-product, and future optimization can be better elucidated to increase the methane yield of this process, which will consequently improve bioenergy recovery.  Figure 4 shows cash flow and Table 5 presents the results of profitability indicators for the scenarios studied. Cash flow was assessed by deducting depreciation, interest rates, and income taxes over the investment period [43]. The results obtained demonstrated that the current assets were negative in the first four years, indicating an increase according to yearly revenues. Regarding the profitability indicators, Scenario 1 presented the lowest gross margin (24.61%) when compared with Scenario 2 (67.78%) and Scenario 3 (62.82%). This fact demonstrates that the direct costs of the project are expensive when compared with the revenues. This fact was confirmed in the net margin, which measures the profit generated after revenue. The process of producing biomethane is not profitable since it presents a negative net margin. In addition, the highest return on investment (ROI) (18.52%) and internal rate of return (IRR) (13.14%) were obtained for Scenario 2, demonstrating that, for small-scale AD processes, the application of a heat and power unit to upgrade biogas into electricity and heat is the best option. Scenario 2 resulted in the lowest payback (5.03 years) and the highest net present value (NPV) (USD 49,953.98) after 10 years of operation.  Sensitivity analysis ensures robustness and the best cost-effectiveness for the systems to be installed [44]. Sensitivity analysis was carried out for all scenarios to verify the influence of the market price of biomethane, electricity, heat, and fertilizer and the influence of costs (FCI, CRM, COL, and CUT) on NPV and IRR. The sensitivity analysis results (±30%) can be seen in Figure 5. CRM did not show significant differences in the scenarios studied in terms of price variation. The other indicators (CUT, COL, and FCI) were more sensitive, with COL and FCI being the most affected in all scenarios, both being strongly affected by a sensitivity of ±30%. Scenario 2 was the most beneficial, where the FCI decrease to 30% promoted an IRR of almost 20%. Checking the project's profitability in different market conditions is a decisive factor in the practical implementation of the project [40]. Variations in the market price of biomethane, electricity, heat, and fertilizer are expected. For Scenario 1, a positive variation of 30% in the market price of biomethane promoted an increase in NPV of almost USD 5000, and a decrease of 30% would cause a reduction of more than USD 10,000. For Scenario 2 and Scenario 3, electricity market price was the most affected parameter, where the increase of 30% in Scenario 2 and Scenario 3 increased IRR by 17%. Sensitivity analysis becomes a tool of great importance for the implementation of processes as sales prices suffer several variations throughout the year. With the performance of sensitivity analysis, the chances of a project being successful are increased.

Sensitivity Analysis
Sensitivity analysis ensures robustness and the best cost-effectiveness for the systems to be installed [44]. Sensitivity analysis was carried out for all scenarios to verify the influence of the market price of biomethane, electricity, heat, and fertilizer and the influence of costs (FCI, CRM, COL, and CUT) on NPV and IRR. The sensitivity analysis results (±30%) can be seen in Figure 5. CRM did not show significant differences in the scenarios studied in terms of price variation. The other indicators (CUT, COL, and FCI) were more sensitive, with COL and FCI being the most affected in all scenarios, both being strongly affected by a sensitivity of ±30%. Scenario 2 was the most beneficial, where the FCI decrease to 30% promoted an IRR of almost 20%. Checking the project's profitability in different market conditions is a decisive factor in the practical implementation of the project [40]. Variations in the market price of biomethane, electricity, heat, and fertilizer are expected. For Scenario 1, a positive variation of 30% in the market price of biomethane promoted an increase in NPV of almost USD 5000, and a decrease of 30% would cause a reduction of more than USD 10,000. For Scenario 2 and Scenario 3, electricity market price was the most affected parameter, where the increase of 30% in Scenario 2 and Scenario 3 increased IRR by 17%. Sensitivity analysis becomes a tool of great importance for the implementation of processes as sales prices suffer several variations throughout the year. With the performance of sensitivity analysis, the chances of a project being successful are increased.

Previous Laboratory-Scale Experiments
The laboratory-scale AD reactor was set up and operated in a semi-continuous mode to manage the jabuticaba by-product [31]. Table 6 summarizes the operating parameters and experimental results for the laboratory-scale AD reactor and the production of biomethane, electricity, heat, and fertilizer. The laboratory-scale reactor had a total volume of 4.3 L, with a working volume of 3.32 L for the substrate and 0.98 L for the headspace. The biogas produced was collected in a Tedlar bag, which was used to measure methane composition. The AD process was operated under mesophilic (36 °C) and methanogenic (pH between 7 and 8.5) conditions with a hydraulic retention time (HRT) of 33.2 d, organic loading rate (OLR) of 4.32 gCOD L −1 d −1 , and volatile solids loading rate (VSR) of 1.47 gTVS L −1 d −1 . For this, the feed of the laboratory-scale process was composed of 69.58 g substrate d −1 (14.58 g jabuticaba by-product and 55 mL water). Stirring was applied for 10 min d −1 (5 min before sampling and 5 min after NaOH addition) at

Previous Laboratory-Scale Experiments
The laboratory-scale AD reactor was set up and operated in a semi-continuous mode to manage the jabuticaba by-product [31]. Table 6 summarizes the operating parameters and experimental results for the laboratory-scale AD reactor and the production of biomethane, electricity, heat, and fertilizer. The laboratory-scale reactor had a total volume of 4.3 L, with a working volume of 3.32 L for the substrate and 0.98 L for the headspace. The biogas produced was collected in a Tedlar bag, which was used to measure methane composition. The AD process was operated under mesophilic (36 • C) and methanogenic (pH between 7 and 8.5) conditions with a hydraulic retention time (HRT) of 33.2 d, organic loading rate (OLR) of 4.32 g COD L −1 d −1 , and volatile solids loading rate (VSR) of 1.47 g TVS L −1 d −1 . For this, the feed of the laboratory-scale process was composed of 69.58 g substrate d −1 (14.58 g jabuticaba by-product and 55 mL water). Stirring was applied for 10 min d −1 (5 min before sampling and 5 min after NaOH addition) at approximately 350 rpm. The process was operated in semi-continuous mode for 50 d. The experimental results revealed a production of 49.6 L of biogas, which was composed of approximately 30% methane. The methane volume was 14.93 L and the accumulated methane yield was 42.31 L CH 4 kg −1 TVS. Finally, the productivity of methane was 4.29 L CH 4 kg −1 substrate considering the initial mass to start the process and the mass of feed in the process. Table 6. Summary of the operating parameters and experimental results for the laboratory-scale AD reactor and the production of biomethane, electricity, thermal energy, and fertilizer.

Parameters
Unit Results

Process Design Simulation and Scenarios
Based on previous laboratory-scale experiments, a simulated process was conducted to elucidate the industrial application of AD for biomethane, bioenergy, and fertilizer production from the management of jabuticaba by-product. The process was designed and simulated considering a small-scale AD process with a treatment capacity of 9 m 3 d −1 . This flow was adopted since the jabuticaba processing industry is considered a low-capacity industry in Brazil. The simulated process consisted of the collection and homogenization of jabuticaba peel (782.2 m 3 y −1 ) in an equalization tank, with the addition of water and NaOH to control the pH of the process. In the start-up of the AD reactor, the substrate was composed of jabuticaba peel (72.14 m 3 ), inoculum (126.09 m 3 ), and water (101.77 m 3 ). The inoculum is necessary only once since the process operates in continuous mode. Hence, the total flow of the substrate in the process is 789.1 m 3 y −1 . The volume of the AD reactor was calculated according to Equation (1).
where HRT is the hydraulic retention time based on the laboratory-scale experiments (33.2 d) and Q is the flow of substrate considering the treatment capacity of the simulated process (9 m 3 d −1 ). After AD, the biogas produced was stored in a bag for further conversion into biomethane or bioenergy. For the recovery of biomethane, a purification system was adopted to remove carbon dioxide, hydrogen sulfide, water vapor, and other impurities. The assumptions adopted for biogas purification were based on the literature [45]. In addition, the biogas produced can be upgraded into bioenergy (electricity and heat). In a common generator, biogas is converted into electricity, while a heat and power unit can be used for the combined conversion of biogas into electricity and heat. The potential for electricity and heat was estimated according to Equations (2) and (3).
where V biogas is the volume of biogas produced in the AD process (m 3 ), LCV CH 4 is the lower calorific value of methane (35.59 MJ m −3 ), C m is methane composition (%), η e is engine efficiency for electricity (assumed as 40%), η h is engine efficiency for heat (assumed as 50%), and CF is the conversion factor from MJ to MWh (1 MWh = 3600 MJ). The values for engine efficiency were obtained from commercial heat and power units available in the market [46]. The digestate obtained was stored in a tank and submitted to a dehydrator to separate the water and concentrate the nutrients. For this, it was considered a mass biodegradation of 33% after AD and a mass reduction of 60% in the dehydration process operated with a maximum dehydration rate of approximately 3 m 3 d −1 . The final moisture content of the fertilizer was assumed to be 70% [47]. The water generated after dehydration was considered as reuse water, which means that it can be applied in the AD process or used in the industry for cleaning. The assumptions adopted for the digestate upgrade into fertilizer were based on the literature [40].
Hence, based on the process designed, the following scenarios were studied in the small-scale AD process: Scenario 1, recovery of biomethane via biogas purification and injection into the gas grid to replace the use of natural gas; Scenario 2, recovery of bioenergy (electricity and heat) using biogas in a combined heat and power unit; and Scenario 3, recovery of electricity using biogas in a generator. In all scenarios, the use of digestate as fertilizer was considered. These scenarios were adopted to estimate the better application of biogas in the small-scale AD process of jabuticaba by-product.

General Assumption
Previous studies addressed a conceptual model for economic analysis of the AD process [40,42,48]. Economic analysis was conducted to compare the costs and benefits of the simulated scenarios. Capital investment, operating costs, COM, revenues, profitability indicators, and sensibility analysis were considered. Table 7 summarizes the general parameters used for the economic analysis. In general, the values were considered based on Brazilian economic indicators.

Itemized Cost Estimation and Cost of Manufacturing
The itemized costs were estimated by evaluating the FCI, CRM, COL, CUT, and CWT. FCI was related to the implementation of the small-scale AD process. Equipment installation costs (Table 8) were collected from current market prices. The CRM of jabuticaba peel was estimated on-site at 0.2 USD m −3 . The CRM for water and NaOH was estimated as 0.2 USD m −3 and 0.53 USD kg −1 , respectively [48]. COL was assumed as 3 USD h −1 worked and a demand of 1 worker at 2 h d −1 [42]. The CUT considered in this study included electricity (0.15 USD kWh −1 ) and heat (0.0082 USD MJ −1 ) for the process, both of which were estimated considering current Brazilian market prices. CWT was established as the capital necessary for implementing the process for digestate upgrade into fertilizer.
where COM is the cost of manufacturing, FCI is fixed capital investment, COL is the cost of operational labor, CUT is the cost of utilities, CWT is the cost of waste treatment, and CRM is the cost of raw material.

Profitability Analysis
Profitability analysis was determined by evaluating the gross margin, net margin, IRR, ROI, payback, and NPV. The gross margin was calculated by the difference between revenue and the cost of goods sold, while net margin is the gross margin considering operating expenses. NPV is the difference between the present value of cash inflows and outflows over a period, calculated according to Equation (5). IRR is the discount rate that equals the NPV to zero and can be calculated using Equation (6).
where FC t is the cash flow in period t, t is the period in which the money will be invested, n is the useful life of the project, i is the cost of capital, and I 0 is the initial investment. ROI is used for capital budgeting and to evaluate the performance of an investment project. ROI was calculated according to Equation (7).

ROI (%) =
Annual net profit Total capital investment (7) Payback is the period in years required to recover the original investment, calculated according to Equation (8).

Sensitivity Analysis
Sensitivity analysis was assessed to evaluate the relevance of the input parameters on the economic performance of the scenarios studied. The profitability variables modeled in the cash flow were NPV and IRR. A tornado diagram was made with a variation of ±30% to better comprehend the effect of FCI, CRM, COL, CUT, and the market price of biomethane, electricity, heat, and fertilizer over the IRR and NPV of the process.

Limitations and Future Prospects
The small-scale AD process with on-site production of biomethane, electricity, heat, and fertilizer was demonstrated as a suitable alternative for the management of jabuticaba by-products. However, some limitations should be highlighted to provide robust conclusions and future prospects for process implementation. The first limitation is associated with the demand for automated processes with real-time control of operational parameters and methane measurement. This can support the stable operational performance of the process and lead to a consequent decrease in the amount of operational labor required. In addition, an increase in methane content during continuous operation can be achieved by applying process control and optimization, which will affect the project's feasibility. In this study, laboratory-scale results were obtained in a standard stirred tank reactor. However, the design and optimization of novel reactors with different configurations can be an alter-native means of improving biomass degradability and biomethane production. AD is a biological process, and analysis of microbial community dynamics should be conducted to determine the interactions between the substrate, operational performance, and biogas production. Correlation between these variables is a promising approach to increasing methane content in biogas and a project's feasibility. Nonetheless, a decrease in the COM of biomethane is essential to compete with the low market price of natural gas. The production of biogas with low carbon dioxide content and a low amount of impurities (hydrogen sulfide, siloxanes, and water vapor) can increase the lifetime of the heat and power unit and decrease the cost of biogas purification into biomethane. Another essential aspect that should be considered is the supply and availability of jabuticaba by-products in Brazil over the years. The seasonal and regional production of jabuticaba can be a challenge for application of the AD process, and this should be overcome by the large-scale production and processing of jabuticaba. Finally, the avoided greenhouse gas emissions resulting from application of small-scale AD can be further considered to decrease the carbon footprint of the industrial sector, which is the main environmental benefit. This can be investigated in the future by a life cycle assessment of the process. For instance, the digestate generated after AD can be used as a sustainable fertilizer for agricultural applications, replacing mineral fertilizers in jabuticaba crops and closing the life cycle.

Conclusions
This study demonstrated the feasibility of small-scale AD in the on-site production of biomethane, bioenergy, and biofertilizer for the management of jabuticaba by-product. The project was simulated for the management of 782.2 m 3 jabuticaba peel y −1 in a continuous anaerobic reactor of 300 m 3 . The fixed capital investment of the process ranged between USD 49,563 and 54,263, with annual operation costs of up to USD 9034.76. The annual production of biomethane (13,960.18 m 3 y −1 ), electricity (61.76 MWh y −1 ), heat (197.62 GJ y −1 ), and fertilizer (211.47 t y −1 ) demonstrated that the jabuticaba by-product can be converted into value-added products and bioenergy. Based on the revenues established, the most profitable scenario was obtained for the process with biogas upgrading in a heat and power unit, with a gross margin of 67.78%, net margin of 33.03%, ROI of 18.53%, IRR of 13.14%, payback of 5.03 y, and NPV of USD 49,953.98. Hence, the application of a heat and power unit for biogas upgrading into electricity and heat was demonstrated as a better option when compared with biogas purification for biomethane and biogas conversion into electricity in a common generator without heat recovery. This condition was achieved due to the high efficiency of the heat and power unit with co-generation of electricity and heat. In conclusion, the application of AD can prevent the wrongful open-air disposal of jabuticaba by-products, with the generation of renewable energy and biofertilizer supporting the green economy toward the transition to a circular economy.