1. Introduction
According to the report from the Intergovernmental Panel on Climate Change in 2018 [
1], humans must reduce anthropogenic CO
2 emission levels by 45% from 2010 to 2030 and reach zero emissions by 2050 to limit global warming to 1.5 °C. The Paris Agreement from 2015 has set a goal for preventing global temperature increases by 2 °C, relative to pre-industrial levels, and seeks to limit temperature increases to 1.5 °C. In this agreement, Brazil pledged to reduce its GHG (greenhouse gas) emissions by 37% by 2030 and 43% by 2050, relative to 2005. Recently, Brazil reinforced its participation in reducing emissions to zero by 2050 at the 2021 Climate Summit.
Carbon capture and storage (CCS) systems and negative emission technologies (NETs) will be essential in meeting this target [
2]. CCS systems are already available in the market; however, they are still expensive [
3]. A complete CCS system can constitute 80% of the total cost of a plant, including capture, transportation, and storage [
4]. A report released by the Global CCS Institute [
5] presented different strategies for mitigating global warming and pointed out that bio-energy with carbon capture and storage (BECCS) technologies are crucial.
BECCS technologies refer to the integration of CCS systems with bioenergy-based systems, including biomass-fueled boilers and furnaces, biogas upgrading facilities, and ethanol plants. Biomass, as a renewable energy source, is considered carbon-neutral throughout its lifecycle [
6]. Therefore, BECCS is viewed as the most viable approach for achieving negative emissions. This is especially true when compared to the application of CCS to fossil fuel-based systems, which can transform them into negative emission technologies at a cost of up to USD 1000 per tonne of CO
2 [
7].
The main limitation, and what keeps the BECCS systems away from economic feasibility, is the energy penalty associated with its operation, as well as CO
2 compression, transportation, and storage processes [
8]. Therefore, the tradeoff between energy efficiency and CO
2 capture is key to assessing the technical and economic feasibility of these systems. Fajardy et al. [
9] emphasize that biomass residues are a more attractive option economically, since the energy allocated for planting can also be used for other purposes by diversifying the products’ portfolio, like ethanol production from sugarcane. Sugarcane presents one of the highest efficiencies in converting solar energy into biochemical energy via photosynthesis [
10], and it is the main biomass feedstock for energy in Brazil, accounting for 11.7 GW (406 thermoelectric plants) of installed capacity.
In fact, sugarcane represents one of the most important energy sources in the world, being widely used for bioethanol production and presenting a self-sustainable energy processing, often using sugarcane bagasse as a renewable solid fuel to simultaneously produce steam for process, bioethanol, and surplus electricity [
11]. Moreover, the sugarcane processing sector is widely used for producing sugar and many other inputs for the food industry [
12], and since sugarcane biomass has been also highlighted as a sustainable source of renewable hydrogen [
13], its thermal cracking has proven to be a valuable way to obtain this energy vector [
14].
Despite being a renewable resource, the sugarcane production chain has various environmental impacts, depending on the agricultural practices employed. These practices need to be properly managed to make sugarcane a more sustainable feedstock. A study focused on South Africa by Pryor et al. [
15] showed that green cane harvesting could reduce energy inputs by 4% and greenhouse gas (GHG) emissions by 16%. However, mechanization leads to soil compaction and stool damage, resulting in lower yields and increased energy consumption and GHG emissions. Also, the proper use of sugarcane residues for energy production can increase the process efficiency even further [
16].
Based on production records for 36 billion liters of ethanol in 2019, a potential capture of 44.77 tons of CO
2/year is estimated from the fermentation stage in the ethanol production process. For annual sugarcane production at 665.1 Mt, 246.1 Mt of CO
2/year can potentially be avoided via BECCS systems [
17].
Among the available technologies for CCS systems, post-combustion is the most promising carbon capture method [
6], given the relative ease of retrofitting existing thermal plants. In this process, CO
2 is removed from chemical absorption, which is the most widespread technique, given its technological maturity and potential for short-term applications [
18], besides being applicable to sources of CO
2 between 3 and 20% in the gaseous mixture [
19].
In the literature, Dubois and Thomas [
20], analyzed three different post-combustion chemical absorption configurations and obtained specific energy consumption at 2.39 GJ/ tCO
2 in the solvent regeneration for a mixture of MDEA 10% + PZ 30%. Bougie et al. [
21] demonstrated that mixtures of MEA with other solvents like glycol monomethyl ether (DEGMEE) increased CO
2 absorption and reduced energy consumption by up to 78%, compared to traditional MEA at 30%. Li et al. [
22] used aqueous ammonia to minimize energy consumption when capturing CO
2. The results indicated potential reductions of 3.3% in plant energy penalty efficiency compared to conventional MEA. Even though other solvents and mixtures may provide better results from an energy point of view and have high corrosion rates [
23], MEA is the most used alternative for removing CO
2, mostly due to its costs [
24].
Post-combustion technology based on MEA was evaluated for a BECCS system placed in the Brazilian sugarcane sector by [
25], and although the energy penalty varied from 43% to 52%, investing in a BECCS system was placed as a better investment in comparison to a natural gas-based power plant. BECCS investments would be lower, and negative emissions might be achieved.
Therefore, several works on chemical absorption focus on the performance of pilot plants and models/simulations to find process improvements. In this work, the main objective is to investigate the technical feasibility of BECCS systems for use in the sugar energy sector using carbon capture technologies from chemical absorption under different Rankine cycle configurations. Multi-objective optimization will be performed using the metaheuristic Lichtenberg algorithm based on a thermodynamic cycle developed in the Aspen Plus® V11.
3. Parametrical Optimization Methodology
The technical and thermodynamic evaluation of the BECCS system involved four stages: (1) simulation of biomass combustion; (2) simulation of the Rankine cycle configurations; (3) simulation of the CCS system; and (4) parametric optimization of the BECCS system. The four steps are shown in the flowchart in
Figure 5.
The thermodynamic problem in question can be statistically analyzed using variance analysis. An optimized matrix of experiments can be generated for the problem using the design of experiments. One must define the input variables—which are the variation intervals of each in the thermodynamic cycle—and the response variables.
After analyzing the cycle parameters, a multi-objective optimization can be performed to find the non-dominated solutions to the problem. All non-dominated solutions are optimal, as are those for which it is not possible to improve an objective without negatively affecting another objective. The set of these solutions is called the Pareto front. Meta-heuristics can better handle complex optimization problems where classical methods have limitations, as well as having the ability to handle optimization problems that do not have explicit objective functions. This approach is particularly useful for simultaneously assessing conflicting goals, such as the maximum cogeneration net power and minimum CCS energy penalty.
The Lichtenberg algorithm [
46], will be applied for this. This meta-heuristic model was inspired by lightning and Lichtenberg figures, and examples of its application can be found in [
47]. Also, the same metaheuristic model has already been validated for other renewable energy systems, such as steam reforming systems [
48].
For optimization, one must define the search domain, i.e., the variation ranges for each variable, which are the same as in the design of experiments. So, the parameter optimizer must be adjusted. The following parameters were chosen based on the recommendations from Pereira et al. [
46]: Pop = 100; Niter = 100; Rc = 200; Np = 106; S = 1; ref = 0.4; and M = 0.
A relevant indicator of the stripper is the specific consumption of thermal energy per mass of captured CO
2 (GJ/tCO
2 or MJ/kgCO
2), which varies between 3.5 and 7.4 GJ/tCO
2 (
Table 5). It is of global interest that this indicator be as low as possible to reduce the plant’s energy penalty. In this sense, two objective functions were considered in this optimization: maximizing CO
2 absorption in the CCS system and maximizing the net electrical power (
) (or minimizing the energy penalty). These indicators are the most influential in determining the technical and economic feasibility of BECCS systems.
The following design variables were selected for this study: boiler outlet temperature, reheating temperature, turbine outlet pressure, and the pinch point in the economizer and regenerators. It is important to point out that in order to avoid the algorithm losing much of its efficiency, increasing the total number of simulations, and consuming more computational resources, only the design parameters for the cogeneration cycle were considered in the optimization process.
Appendix A summarizes the input variables in the optimization cycle.
4. Results
As was mentioned before, the main technical barrier of BECCS systems is the energy penalty associated with CO
2 capture. For this reason, the reheating cycle with three regenerators was optimized using the net power of the cogeneration cycle as an objective function to evaluate the energy penalty associated with the CO
2 capture system. The results (
Table 8) showed that net electrical power was 62.82 MWe, representing an energy efficiency of 31%, and emissions were equal to 1300 gCO
2/kWh.
The results show that all the configurations of the thermal system provided a perfect negative correlation between the objective functions for the operating range of the evaluated design variables. For the cogeneration cycles that discarded steam reheating (REG1, REG2, and REG3), the steam temperature at the boiler outlet had the greatest influence on the thermodynamic performance of the system. The higher the temperature of the steam at the turbine input, the greater the enthalpy variation during steam expansion in the equipment. Furthermore, high vaporization temperatures ensured that the steam exiting the last turbine stage remained saturated with pressure parameters close to the lower 250 kPa limit for meeting the plant’s process steam quality conditions.
Raising temperatures close to 520 °C resulted in reduced steam mass flow in the cycle, generating less thermal energy for meeting CO2 absorption. A lower temperature at the boiler’s outlet increased steam availability for the processes. Under these operating conditions, higher pressures at the last turbine stage are needed to ensure that steam is saturated for alcohol and CCS production processes, leading to decreased power generation in the cogeneration cycle.
The vaporization pressure and pinch point are important design parameters. Similar to the evaporator temperature, the vaporization pressure is directly proportional to the enthalpy variation during steam expansion in the turbine, promoting power generation. The higher the pinch point, the higher the exhaust gas temperature at the inlet of the heat exchanger, favoring energy generation for the CCS process; however, this decreases the mass flow for the working fluid, decreasing power generation in the power cycle.
For configurations with steam reheating (Reheat1, Reheat2, and Reheat3), the results show that both the vaporization pressure and the low-pressure turbine discharge pressure influenced the thermal system the most, generating more electrical power or heat for downstream plant processes. On the other hand, the vaporization temperature had little influence on the evaluated objectives compared to the configuration without reheating. We also observed that there was a greater interaction between input parameters for the configurations with reheating, although they had little influence on the results when compared to the operating pressures in the cogeneration cycle.
Figure 6 shows the results of the multi-objective optimization for each evaluated configuration. As was mentioned above, we observed that, for each evaluated configuration, there was a negative linear correlation in which an increase in CO
2 capture capacity led to reduced electrical power generation in the cycle. The heat demand for stripping is critical for releasing CO
2 from the solvent, enabling its subsequent capture and separation. This demand encompasses sensible heat, which is needed to raise the solution’s temperature; desorption heat, which is responsible for breaking the chemical bonds between CO
2 and the solution; and latent heat, which is essential for evaporating the solution’s water content. Therefore, a higher CO
2 capture capacity requires an increased availability of heat in the Rankine cycle for CO
2 capture purposes. Consequently, a greater capture rate would lead to reduced water vapor available for power generation, resulting in a decrease in power output, commonly referred to as the energy penalty.
Considering CO
2 absorption, the REG1 configuration was the thermodynamic cycle with the highest heat availability for the CCS system, at approximately 0.304 MtCO
2/yr, representing a 70.6% CO
2 capture percentage from nearby exhaust gases; however, it was limited to 16 MWe of net electric power generation (
Figure 6a). The Reheat3 configuration was the technological option with the greatest capacity for generating electrical power (47.8 MWe) and for capturing CO
2, at 0.156 MtCO
2/yr (
Figure 6f). The rest of the configurations showed electricity generation values and CO
2 capture levels within intermediate ranges between the two previously mentioned configurations.
The REG1 configuration is the least complex alternative (fewest devices), i.e., it is the least expensive thermodynamic cycle in terms of installation and maintenance. ReHeat3 is the opposite. Therefore, the more complex cogeneration cycle configurations that produced the same amount of electrical power and captured CO
2 were excluded from this analysis. Furthermore, the BECCS system showed lower specific emissions relative to the Reheat3 cycle without CCS (1300 gCO
2/kWh), discarding any set of optimal solutions above this restriction.
Figure 7 shows the set of optimal solutions for all evaluated configurations.
The BECCS system was able to obtain a maximum capture of 0.224 Mt/year for REG1 and a CO2 capture rate from exhaust gases close to 51.9%; however, it was limited to electrical power generation, which was 33.46 MWe. This was true up to 29.36 MWe, as less than 62.82 MWe was generated by the ReHeat3 configuration without CCS (14.49% penalty on the plant’s electrical efficiency). On the other hand, Reheat3 (with CCS) resulted in more electrical power generation (47.80 MWe) and had a lesser penalty for electrical efficiency (7.41%); however, it had a minimum CO2 capture rate of 36.3%, which was emitted by the plant.
Figure 8 shows three scenarios for percentages of CO
2 capture, as well as the respective electrical power required for compression at the plant. To compress the CO
2 generated from fermentation, approximately 2 MWe are needed. Considering the limiting scenario at 90% CO
2 capture from exhaust gases, 8.24 MWe would be needed to compress the CO
2 generated by the plant to its maximum capacity, which would be equivalent to 1.48 MWe of power for 0.1 Mt/year of CO
2 captured at the plant.
Figure 9 shows a reduction in the net electric power generated by the BECCS system from the CO
2 compression system in the plant. The minimum capture point for CO
2 showed a reduction of 8.27% in the net electric power generated (52.11 to 47.8 MWe), while the maximum capture penalty was 13.67% (38.76 to 33.46 MWe). For a theoretical scenario for a configuration with a greater CO
2 capture capacity, one could capture up to 88.73% of all the CO
2 generated at the plant; however, one would need to consume all the electrical power generated to meet the power demands of the compression system.
In this sense, the proposed carbon capture and storage (CCS) approaches have shown the potential to enhance the sustainability of sugarcane-derived bioethanol by further reducing its carbon footprint. Additionally, since CO2 can serve as an input for various industrial processes and biofuel production, such as in Fischer–Tropsch synthesis, carbon capture could foster a greater integration between the sugarcane industry and other market sectors, advancing a circular and renewable economy.
5. Conclusions
This paper evaluated different bio-energy system configurations integrated with post-combustion chemical absorption (MEA) CO2 capture technology. This work differs due to its approach of capturing carbon not only from the CO2 of the fermentation process but also from the combustion of bagasse and sugarcane straw, in addition to considering the heat required to supply the ethanol production process in the plant, which globally implies a high thermal demand to be managed from the extractions of steam turbines.
The parametric analyses showed that it is challenging to define the best combination of pressure and temperature parameters, given that the objectives were conflicting (electrical power generation and CO2 capture). Furthermore, of the evaluated configurations, different parameters with stronger influences were found for each configuration. Thus, we must use multi-objective and stochastic optimization methods to define the correct operational parameters and tradeoffs between generated electrical power and CO2 capture.
From a power generation and carbon capture perspective, the results showed a tradeoff for all the evaluated configurations of the BECCS system. The REG1 configuration resulted in the highest (51.9%) carbon capture with a 14.49% penalty on electrical efficiency (10.49% on the plant’s cogeneration efficiency); therefore, it cannot capture all the CO2 generated by the plant (theoretical limitation of 88.7% where all generated electricity would be used to compress the captured CO2). The second analysis using (gCO2/kWh) indicators showed that CO2 capture is more expensive as more power must be used to capture the same amount of CO2 in terms of mass, since less electrical power is generated and larger tons of CO2 need to be compressed. CO2 capture from 51.9% (0.224 Mt/year) would result in emission rates above 1300 g/kWh, which are higher than the plant’s operating emissions with reheating and three regenerators without CCS. On the other hand, the Reheat3 configuration showed the best ratio at 1155 g/kWh (145 g less per kWh generated), with an even smaller penalty on the plant’s electrical efficiency (7.41%); however, it was limited to a minimum capture level of 36.6% for all the CO2 emitted at the plant.
The scenarios allowed us to reach reasonable results, where the BECCS system technically partially resulted in negative CO2 emissions. This is a plant typical to the Brazilian sugarcane industry, with large demands for suppressed steam from ethanol and sugar production processes (115.5 MW). To capture 90% of all generated CO2 from the bagasse and chaff combustion process, 198.4 MW would be needed, and 72% more heat would be destined to a secondary plant process. Future studies are needed to validate the operating ranges and configurations studied in this paper from an economic standpoint.