Aiming to reduce emissions related to self-consumption, which currently relies on its own thermopower, this study assessed the implementation of an onsite photovoltaic system for complementing the internal supply-demand. In this context, the authors considered all visually available areas, including rooftops and ground portions. The polygons identification and measurement applied to the Google Earth Pro mapping tool, version 7.3.2.5776.

For system sizing, this paper used version 2018.11.11 of the System Advisor Model (SAM), developed by the National Renewable Energy Laboratory (NREL) under the United States Department of Energy. SAM is a simulation software that allows technical-economic modelling of renewable energy power systems, including photovoltaics. The estimations applied the "Photovoltaic (detailed)" calculation mode through the "LCOE Calculator" option, which provides performance and Levelized Cost of Energy (LCOE) results for specific equipment [11].

Among the equipment available on SAM's database, this study considered the following selection criteria: array configuration technical requirements for maximizing area exploitation, such as nominal power and voltage; well-known manufacturers. Thus, the selected references were: Canadian Solar Inc. CS6U-330P module, with 17,57% of efficiency and 330 Wdc of nominal power; SMA America SC250U (480V) inverter, with 250 kWac of nominal power and 96,8% of electric current conversion efficiency [11].

SAM's database was also used for solar resource estimation by selecting the specific plant location. Thereby, the irradiance data applied refers to a point centered on its largest building. The calculation considered default values of alternate and direct current losses, and disregarded shading parameters. Other information concerning system configuration relied on adaptations from [12], such as: module azimuth, ground coverage ratio (related to tilt and distancing) and direct to alternate current ratio.

Regarding economic-financial parameters, inflation and the exchange rate between Brazilian Reais (BRL - R$) and American Dollars (USD - US$) corresponded to those on Focus Report, the market report issued by [13]. Additionally, this study considered the fixed Weighted Average Cost of Capital (WACC) of Brazilian power generation [14] as a proxy for the discount rate. As for capital and operational costs of photovoltaic technology, the estimations applied adapted data from [15], responsible for subsidizing Brazil's energy planning through official studies.

Then, for comparing the estimated photovoltaic output with the equivalent portion of thermopower consumption to be replaced, 2019's self-consumption was estimated based on Termope's most recent public data. In this sense, it was assumed the exact relation between self-consumption and power generation informed by [16]. Finally, for estimating the Marginal Abatement Cost (MAC), the plant's capital costs were considered already covered, so the photovoltaic LCOE was directly compared with the plant's Variable Unit Cost (VUC) by calculating their gap per mitigated emissions.

To effectively reduce GHG emissions from burning natural gas, alternative combustion sources were studied by the combination of cleaner and chemically compatible fuels.

The first alternative assessed was biogas originated in landfills. Every year, solid urban waste generation continues to increase in Brazil and worldwide, and its inadequate disposal has negative impacts on the environment and public health. Therefore, the use of biogas resulting from the decomposition of these residues appears as a relevant alternative. Beyond adding value to those, it can incorporate revenue by obtaining carbon credits and electricity generation.

Biogas is a gas with low calorific value when compared to other combustible gases. In the case of Termope, it could be used in the secondary boiler or, in small proportions, combined with natural gas to feed the turbine. As it is a poor gas and presents different impurities, it is necessary to assess the composition of landfill biogas and adapt the equipment to not compromise their structure or impair the plant's efficiency. It is key, however, to achieve a balanced trade-off between emissions reduction and efficiency required. To proceed with biogas use, landfills within the Termope's region have been mapped, among which the closest ones are located at distances between 40-50 km.

The second alternative of fuel mix analyzed corresponds to the use of synthesis gas through biomass gasification, a technology known as a Biomass Integrated Gasification Combined-Cycle (BIGCC). Although it has not yet reached a high technological maturity, it is a promising technology because of its strong potential of environmental performance, reducing emissions of CO₂, SO_x, NO_x and particulate materials [17]. In addition to the advantage of being a carbon-neutral fuel, its use in BIGCC systems with CCS offers the possibility of generating energy with a negative CO₂ balance [18]. Thus, for using this alternative, regional supply availability was analyzed.

The third alternative refers to hydrogen gas (H₂), since it is a potential fuel for a decarbonized future when generated from renewable sources. However, it is necessary to overcome significant technical and economic barriers [19]. For using hydrogen gas in a potential combination with natural gas to supply the plant's demand, some considerations were made for the analysis. To avoid significant changes in the structure of a natural gas-fired thermopower plant, the amount of injectable hydrogen gas in the mixed fuel is limited according to the equipment specifications, varying in volumetric percentage up to 12% [20]. This study considered a 10% volumetric hydrogen ratio in relation to natural gas.

For dimensioning this alternative, energy and water requirements of the electrolysis process to produce 10% in volumetric percentage of H₂ were calculated according to the average monthly consumption of natural gas at the plant. PEM electrolyser was established as the technology for producing H₂, as it is the most suitable when powered by an intermittent source, such as wind or solar. For this purpose, the electric energy required by the electrolysis was estimated by the equipment's efficiency [21], according to eq. (1).

Where, Elt: energy required by the electrolyser (kWh); PCI: hydrogen lower heating value (kWh/kg); ProdH ₂ : H₂ production (kg); 𝜂: electrolysis efficiency.

It was adopted 3 kWh/Hm³ for hydrogen lower heating value and 60% for the efficiency of electrolysis process by PEM, according to [22]. As for water consumption, according to [23], it was estimated that approximately 9 liters of water would be needed to produce 1 kg of hydrogen by the electrolysis process.

The electrolysis plant values presented in [23] was used to estimate investment cost. In this sense, the prices related to the renewable plant energy - necessary to feed the system - were not considered, and neither the prices related to the H₂ gas transport and storage.

Finally, the MAC of this measure was estimated, assuming a lifespan of 20 years for the electrolysis plant and an approximate discount rate by the fixed WACC of Brazilian electric power generation [14].

Carbon capture and storage represents an important mitigation option to deal with climate change. It is one of the few technical solutions capable of capturing up to 90% of CO₂ emissions produced by the use of fossil fuels in electricity generation and industrial processes. Although CCS technology has a great potential to drop greenhouse gases emissions in the long term, it still faces economic and technological barriers [24,25].

A post-combustion system is the most recommended method for CO₂ capture in a thermopower plant already in operation, such as Termope, since it can be add-on to the power plant. Therefore, this route is easier to be implemented in existing plants and provides a better potential for short-term applications. CO₂ is captured in the exhaust from the combustion process, subsequently, the gas is absorbed by an adequate solvent. Thus, the CO₂ is compressed and transported by pipelines to appropriate geological formations for storage [26].

The Oil & Gas sector is one of the pioneers of developing and implementing CCS projects, using the captured CO₂ for enhanced oil recovery (EOR) [27]. CO₂ injection assists in increasing the formation pressure, directing the oil to producing wells. Depending on reservoir parameters, CO₂ blends with the oil into a single-phase at miscible conditions, improving oil mobility and allowing easier flow through the formation [28,29].

The Northeastern Brazilian oil basins only correspond to 7% of oil production in the country and are mainly composed of mature fields and shallow reservoirs, being attractive to EOR implementation and representing an option for lower injection cost due to its lower formations' depth [30,31].

The first stage of the technical feasibility analysis consisted of selecting onshore and mature oil fields that benefit from receiving CO₂ injection from Potiguar, Recôncavo and Sergipe-Alagoas (Northeastern Brazilian) basins, or at least those that were not rejected on the data analysis. Thus, the Executive Summary of Development Plan was used to verify some reservoir characteristics, such as porosity and permeability, original oil in place (OOIP), and oil API degree [32].

This study only considered the miscible CO₂ injection since it provides a greater oil recovery [33]. Typical values of favorable characteristics for the CO₂-EOR application are shown in [34]. These parameters are required but not sufficient for a reservoir to be appropriate to use CO₂ injection. However, some data are not available for public consultation in most of the fields analyzed in this study, thus parameters such as temperature, pressure and reservoir depth, and oil viscosity were not considered. Concerning the depth parameter, shallower reservoirs can be selected as long as they do not contain drinking water in their proximity. The reservoir formation must have a good seal to prevent CO₂ contamination in adjacent formations [34]. Oil fields that do not have OOIP information were discarded.

Using the collected data of oil fields suitable for receiving CO₂ injection, the storage capacity of each field was calculated. Subsequently, the total quantity of CO₂ that Northeastern oil fields can store was estimated. This analysis was based on the methodology adopted by [35], which considers that CO₂-EOR can provide an increase of 10% of the original oil in place in current oil production [33].

Additionally, the storage capacity is related to the ratio between the mass of CO₂ stored per unit mass of oil produced. According to [33], the storage efficiency of the Recôncavo basin corresponds to 0.18 tCO₂/bbl. Since this study was based on the same region of this basin, the same value of storage efficiency was adopted. Using ºAPI data from each of the selected oil fields, the median density was calculated and presented 848.1 kg/m³. These values correspond to a storage of 1.33 tCO₂ per tonne of oil produced. Thus, eq. (2) was used to estimate the oil fields' storage capacity given by the sum of the individual capacity of each field [35]. The results are presented in Section 3.2.3.

Where, QCO _2i : CO₂ storage capacity of the field i in million metric tonnes; p: percentage of the OOIP that can be additionally produce using CO₂-EOR in %; OOIP _i : original oil in place of the field i in million metric tonnes; mCO ₂ /mOil: ratio of mass of CO₂ stored per unit mass of oil produced.

Afterward, hotspots were defined for each of the three basins mentioned above, which means that the field with the most significant storage capacity in each basin was considered the delivery point of CO₂. Using a georeferencing tool, the distance between the hotspots and the thermopower plant was calculated to estimate the pipeline length.

Finally, a cost analysis of the CCS technology implementation in the thermopower plant was made using the IECM tool, version 11.2 [36]. This model estimates the costs of implementing the CO₂ capture plant, and the pipeline construction costs. The EOR costs were not considered in this study, as it would not be attributed to the thermopower plant but to the operator responsible for the oil fields.

The model developed with the SAM tool considered a total land area of 25,042 m², enabling the consolidation of a 2.84 MW photovoltaic power system. Its composition included 8,586 modules and nine inverters by the specifications previously mentioned.

The simulation results indicated a capacity factor of 20% and an average annual output of 4,963 MWh, considering the first year of operation as a reference. Such energy amount represented around 8% of Termope's self-consumption. In terms of GHG, the mitigation potential corresponded to 1,491 tCO2_e/year, accounting for 0.14% of the total emissions estimated by this study for 2019.

The capital and operation costs obtained for the proposed system were, respectively, 3.1 MUS$ and 13,835 US$/year. After levelizing these figures to the present value, the resulting LCOE was 0.05 US$/kWh. Therefore, the calculated MAC for this mitigation option was 48.10 US$/tCO_2e.year.

Although its contribution may seem of minor importance, this was considered an appropriate option since this kind of system has a short-term implementation of approximately one year [15]. Furthermore, this solution may serve as a good practice reference, encouraging the dissemination of similar projects within corporations. Hence, this is a measure capable of collaborating with renewable energy expansion and climate change mitigation.

For fuel mix alternatives, the feasibility of each option was verified. In the case of landfill biogas, it was observed that the regional generation capacity is much lower than the plant's fuel demand. Also, biogas production related to the decomposition of organic matter from urban solid waste decreases considerably over the years, reducing supply throughout the landfill lifespan. Therefore, the use of this type of fuel does not represent a feasible alternative, mainly due to its lack of availability in the long run.

For the second alternative, the BIGCC use analysis, a deficit of biomass was observed in the region. According to [49], Pernambuco presents the worst biomass supply performance among the states in the Northeast region, with demand almost seven times greater than the legal supply. Due to the lack of supply, using a combined-cycle with integrated biomass gasification was discarded as an option.

Regarding the use of hydrogen gas, calculations of energy and water consumption were performed so that a volumetric proportion of 10% H₂ was met. Using as a reference the volume of gas, for a monthly average of approximately 50 Mm³ of natural gas, 5 Mm³ of H₂ is required. According to the calculation parameters mentioned in Section 2, the electricity required by the electrolysis was estimated at 24,892 MWh/month, and the consumption of approximately 4 million liters of water per month is also required. With these requirements, an electrolyser of approximately 35 MWe is needed.

It was assumed that the entry of such technology might occur up to 2030, with an estimated capital cost of 52.5 MUS$ this year. The implementation of this technology could allow a 9.8% emission reduction, and with this, the obtained MAC for this measure was 41.46 US$/tCO_2e.year.

Considering 381 Brazilian oil fields in production, of which 219 are onshore and in Northeastern Brazil, 90 were selected in the scope of this study as they are reasonable to receive CO₂ injection. These fields' total estimated storage capacity was 143 MtCO₂, considering a stock of 1.33 tCO₂ per ton of oil produced (Table 2).

Table 2

Source: The authors.

Using a georeferencing tool, the fields that presented the greater storage capacity from each basin were selected as hotspots for this study: Canto do Amaro, from Potiguar Basin, Água Grande, from Recôncavo Basin, and Pilar, from Sergipe-Alagoas Basin. Fig. 2 shows the locations of Termope and the hotspots. The distance between Termope and Canto do Amaro field is 438 km, and between Termope and Pilar and Água Grande fields is 557 km. Then, it was established the construction of only one pipeline, due to the large-scale gains of this type of construction [50].

Figure 2 Source: The authors.

Potiguar Basin presented an estimated storage capacity of 39.4 MtCO₂, Recôncavo Basin demonstrated a storage capacity of 91.8 MtCO₂, and Sergipe-Alagoas Basin indicated a capacity of 11.9 MtCO₂. Hence, only the pipeline's construction connecting Termope to Pilar and Água Grande fields would be necessary since these fields can provide a storage capacity of 103.7 MtCO₂, providing sufficient storage volume to mitigate Termope's emissions in the long term. Supposing the volume of CO_2e emissions provoked by Termope, the chosen pipeline would afford a lifetime of approximately 90 years to mitigate the combined-cycle natural gas-fired thermopower plant emissions or even other emission sources located around it.

IECM tool configuration selected amines as a solvent for capturing the CO₂ from Termope. Results showed an Equivalent Annual Cost of 64.95 MUS$/year to implement the capture plant and to build the pipeline, considering a power plant lifetime of 30 years. In this sense, Termope’s LCOE would be increased by US$ 64.90/MWh, while the MAC was estimated at US$ 179.51/tCO_2e.year.

CCS is an incipient technology, and there are no projects on a commercial scale yet, only pilots, thus this mitigation measure has significantly high costs, which tend to decline due to learning by doing [51,52]. However, this alternative is attractive for a carbon-neutral future, as it is one of the only mitigation technologies capable of substantially contributing to the reduction of CO₂ emissions in the long term, whether from energy sources or industrial processes [53,54].