Mexico’s greenhouse gas (GHG) emissions from the electricity sector were 126.6 MtCO_2eq, corresponding to 19.0% of total national emissions (INECC, 2015). This is due to the fact that around 80% of the Mexico’s electricity generation comes from fossil fuels (SENER, 2018a), with NGCC plants accounting for the largest proportion (∼50%) and they are expected to increase their participation in the long term (SENER, 2019; SENER, 2021). Therefore, technologies to reduce GHG emissions, especially CO₂, from current and future NGCCs are of great interest to the country.

Carbon capture and storage (CCS) is a key technological tool for CO₂ abatement in NGCC plants. One of the most promising CCS technologies is post-combustion carbon capture (PCC), which consists of separating the CO₂ contained in flue gases from the stack through a chemical process with absorption-desorption cycles using an amine-based solvent as carbon captor material. In general, PCC technology is preferred over other CCS technologies (e.g., pre-combustion and oxy-fuel combustion) mainly because it can be coupled to new and existing plants with minor modifications (Figueroa et al., 2008; Freeman and Bhown, 2011; Sanchez Fernandez et al., 2014; Pan et al., 2016) and its higher technology maturity level (Wilcox, 2012; Oh et al., 2018; Feron et al., 2019; Finney et al., 2019).

Another alternative for CO₂ mitigation in NGCC plants is the use of blending of natural gas with blue hydrogen (bH₂) or green hydrogen (gH₂). Especially, gH₂ makes sense in the Mexican context, as there are vast renewable energy and water resources available. The Federal Comission of Electricity, a state-owned utility company, has realized this local advantage and, in early 2022, announced the implementation of a pilot project for the use of blending of natural gas with gH₂ in a gas turbine (CFE, 2022). The purpose of this is to generate knowledge about the challenges and opportunities of using gH₂ in electricity generation and, based on the results, to advance towards commercial scale-up.

Due to the relevant role that CCS and hydrogen could play in the decarbonization of the Mexican electricity sector in the long term, Díaz-Herrera et al. (2021) reported the comparison between PCC technology and the use of bH₂ and gH₂ in existing NGCCs. The techno-economic analysis considers fuel costs, capital expenditure, operating cost, and the plant capacity factor. The results show that the NGCC equipped with PCC (NGCC + PCC) is a much more economical alternative to reduce CO₂ emissions ($140.4/tCO₂) than the use of blendings of bH₂ ($256.9/tCO₂) and gH₂ ($435.8/tCO₂). Although PCC technology is an excellent alternative to mitigate CO₂ emissions in NGCC plants, its high cost is the critical barrier to its commercial-scale deployment (Wilcox, 2012; DOE/NETL, 2010; Rubin et al., 2015; Irlam, 2017; Muhammad et al., 2020; Liu, 2020).

One strategy that several countries have adopted to promote the early deployment of carbon capture technologies in industrial processes that are difficult to decarbonize, such as an NGCC, is the application of a carbon tax (Shirmohammadi et al., 2020). For example, in 1991 Norway was one of the first countries in the world to introduce a carbon tax to reduce emissions from upstream oil and gas operations. This initiative prompted the development of the Sleipner project in 1996, the first large-scale carbon capture and storage (CCS) project in the world (Equinor, 2019). If we look into the North American region, in 2008, the United States implemented a carbon tax credit called 45Q, which provided $10/tCO₂ stored via enhanced oil recovery (EOR) and $20/tCO₂ stored in geologic formations (GCCSI, 2020). Later, in 2018, the 45Q tax credit was reformed as part of the Bipartisan Budget Act, increasing its value to $35/tCO₂ for EOR projects and $50/tCO₂ for geological storage (Jones and Sherlock, 2021). Recently, in 2022, the US Inflation Reduction Act has increased the value of the 45Q tax credit to $60/tCO₂ for EOR purposes and $85/tCO₂ for geological purposes (Bipartisan Policy Center, 2022). Canada introduced an investment tax credit for CCS deployment of $319 million over 7 years from 2021 (Government of Canada, 2021). So far, Mexico still does not have a solid regulatory framework for the financing of CCS projects; however, the recent United States-Mexico-Canada Agreement is expected to be a driver for the insertion and development of the first CCS projects in the country.

Assuming that Mexico acquires the same level of commitment in the fight against climate change as its partners in the short term and, considering that the CAC for a NGCC + PCC plant in the local context is $140.4/tCO₂, this value is 1.7 times higher than the 45Q tax credit for geological storage. Therefore, a substantial increase in the tax credit would have to be necessary to implement the first commercial project in the country. This could lead to a significant increase in electricity costs, which would impact the economy of the population. Nevertheless, a positive approach to the fact that PCC technology is still in the research and development phase is that it can reduce its costs through the learning-by-doing effect, so it could be successfully developed in the next few years.

As with other types of technology, it is expected that the first generation of NGCC plants using PCC technology (FOAK plants) can be significantly more costly than later or advanced generations, which are referred to as NOAK plants. This has a favorable economic impact on the technology since the carbon tax could be very high in the early years for FOAK plants and progressively decrease until NOAK plants are economically competitive in the market on their own. A clear example of this was what happened with renewable energy in Mexico. In 2014, the Mexican government implemented clean energy certificates (CECs), a legal instrument promoting investment and development of solar and wind FOAK plants on a commercial scale (DOF, 2014). The CECs were gradually reduced until their cancellation in 2022, once renewable energies reduced their costs to the point of being economically competitive (Forbes, 2022). A very interesting aspect to analyze is the extent to which NOAK plants could reduce their costs or whether they will need to be permanently subsidized to ensure their operation during their useful life. Therefore, the cost projection of the NOAK plants is a key element since it could support the planning of the portfolio of strategies for the energy transition not only in Mexico, but also in other countries that depend on natural gas for their electricity generation (e.g., Japan, the United States, and Canada).

Very few research publications focused on cost projection for FOAK and NOAK NGCC plants equipped with PCC are available in the literature. One of them was published by Rubin et al. (2007), who performed a sensitivity analysis varying different technic-economic parameters for estimating NOAK NGCC power plants equipped with CO₂ capture systems. In the case of NGCC + PCC plants, the results show that LCOE of a NOAK plant could be between 12% and 20.4% cheaper compared to a FOAK one. Additionally, (Irlam, 2017) reported the costs of NOAK plants for different industries (power, cement, iron, and steel). In the case of NGCCs, they conclude that the CAC for a FOAK and NOAK plant is $89 and $43/tCO₂, respectively. This represents a cost reduction equal to 51.7%. Although both studies explain in-depth the economic assumptions that were considered to arrive at these results, they did not consider the fact that technology costs could also be constrained by thermodynamic principles. This is especially important to analyze, as the PCC plant currently demands a lot of energy for its operation, technological improvements will make it more and more efficient, which in turn will increase its economic performance, here a key question arises: to what extent can the NOAK NGCC + PCC plant be really efficient?

A key indicator to respond the above question is the second-law efficiency for the CO₂ separation process ( η 2 n d ), since it allows us to know the thermodynamic limits of the PCC technology and to recognize which sub-processes present the best areas of opportunity for energy savings (Wilcox, 2012). Sanchez Fernandez et al. (2014) reported that the net efficiencies of conventional NGCC and coal-fired power plants with PCC are reduced by 8.4% and 11.7% points, respectively. In the case of NGCC + PCC plants, Bolland and Undrum (2003) mentioned that the work required for solvent regeneration ( W r e g e n ) represents the highest energy penalty, with a loss of 4.5% points in the power plant efficiency, while the penalty associated with the use of mechanical power and CO₂ compression is 1.8% and 2% points, respectively. In addition, W r e g e n has a much lower technological maturity than the mechanical power and CO₂ compression processes, which means, in theory, a larger energy-saving opportunity window. Therefore, several studies have focused on innovative energy-saving pathways.

Oh et al. (2018) focused on heat integration and the design process for the energy penalty reduction in a coal-fired power plant integrated with an amine-based PCC plant. Simulation results show that the net efficiency of the plant is reduced by 9.7% points (3.1% points less than the conventional PCC technology). Saleh et al. (2019) proposed a novel conceptual NGCC configuration integrated with a lithium-based PCC technology, which shows a reduction in its net efficiency equal to 9.2% points, being this value close to the energy penalty of the amine-based PCC technology published elsewhere (Sanchez Fernandez et al., 2014; Díaz-Herrera et al., 2021). Other studies reported the use of thermal solar energy to compensate for the energy penalty of the PCC plant (Li et al., 2012; Lambert et al., 2014; Wang et al., 2015; Liu et al., 2017; Shirmohammadi et al., 2021). A very interesting study on this subject is the one published by Zhai and co-workers (Zhai et al., 2018), who mentioned that a coal-fired power plant with a solar-assisted PCC plant shows a higher LCOE than the conventional PCC technology, mainly because of the increase in the CAPEX associated with the solar farm. Other authors evaluated the energy performance of novel power plant configurations; for example, Lindqvist et al. (2014) notified that an NGCC + EGR has a lower energy penalty than the conventional NGCC plant (7.6% vs. 8.7% points), mainly because its higher CO₂ concentration in the flue gas.

Capture level is another strategy to reduce the energy penalty of the power plant and reduce the CAPEX of the PCC (Hildebrand and Herzog, 2009; Wilcox et al., 2017; Feron et al., 2019). This consists of capturing an amount of CO₂ at lower rates or partially capturing it to reduce the cost. Several studies focus on the technical and economic implications of the use of the capture level in the PCC plant; one of these is published by Wilcox et al. (2017), who mentioned that there is a direct relationship between the carbon capture level and the minimum thermodynamic work for the CO₂ separation ( W min ). Normann et al. (2017) presented a techno-economic analysis of the capture level of a PCC plant considering key design parameters, such as gas flow, load hours, and CO₂ concentration. For a capture level of 22.5%, the CAC is 80 €/tCO₂, representing an 11.1% cost reduction compared to the 90% capture level. Díaz-Herrera et al. (2020) performed an economic analysis of the capture level design for an NGCC + PCC plant using novel configurations. At 90% capture, the CAC for conventional NGCC + PCC is 117.7 $/tCO₂. This cost could be reduced by around 3% if the EGR configuration is implemented. Despite these savings, the CAC is still high, mainly due to the CAPEX associated with the PCC plant.

As can be seen, there is a lot of information in the literature focused on reducing the energy penalty of the PCC plant, as well as its capital cost; nevertheless, very few papers paying attention to the effect of the W r e g e n on the η 2 n d have been published. Also, to the authors’ knowledge, no work has been reported on the combination of the η 2 n d with CAC applied to NGCC plants. The relationship between these two indicators is key because they allow estimating costs for NOAK plants based upon scientific limits. This is mainly important for Mexican decision makers, since knowing information about cost scenarios for NOAK plants would allow them to design appropriate public policies on energy and climate change.

This study aims to analyze the cost projection of FOAK and NOAK type plants for two promising NGCC configurations, namely: conventional NGCC and EGR. A thermo-economic analysis was performed considering the η 2 n d and CAC as main indicators. Several essential variables influencing the overall cost of the technology were considered, such as W r e g e n , technology maturity, learning rate, carbon tax credit, and carbon capture level. Most of the related work published in the literature has considered these variables separately. Our contribution is the first attempt to establish and carry out an integral effort to consider the different inputs reported by other authors to project the cost of carbon mitigation in NGCCs taking into account the thermodynamic constraints of the PCC technology.

The work is organized as follows. First, the methodology is described. Then, the results are presented and discussed. Finally, a conclusion is reached.

Table 3 shows the η 2 n d values for NGCC and EGR configuration at different capture levels. For each case, η 2 n d increases as a function of the carbon capture level. The η 2 n d ranges from 14.2% to 15.2% and 13.3%–14.2% for NGCC and EGR cases, respectively. This result is in good agreement with those published in the literature (Bolland and Undrum, 2003; Lindqvist et al., 2014; Liu et al., 2017). The increments on η 2 n d are because the W r e a l required for the CO₂ separation is reduced as a function of the capture level. Additionally, at a specific capture rate, we can see that EGR shows a lower W f a n values than NGCC. This is because of the lower amount of flue gases to be treated in the PCC plant. Meanwhile, W r e g e n and W p u m p presents similar values for both cases, which is due to optimization work done for sizing the absorber and stripper columns (Díaz-Herrera et al., 2020), which results in similar values in the reboiler duty and power consumption for solvent pumping, respectively. For example, the reboiler duty for NGCC and EGR cases is 3.76 and 3.74 MJ per kg of CO₂ desorbed, respectively. Since W r e g e n does not significantly change for both cases and contributes to most of the energy consumption required for the CO₂ separation process (∼83%–88% of W r e a l ), the η 2 n d does not show significant variations.

Figures 3, 4 show the CAPEX for conventional NGCC and EGR power plants with PCC technology classified by type of plant at different carbon capture levels, respectively. As expected, the PCC package is the technology with the highest cost reduction, with a total CAPEX reduction of around 40% for a NOAK compared to the FOAK plant. While, for the power plant and CO₂ compression packages, the CAPEX reduction is marginal. This is because the PCC’s learning technology rate is higher compared to the power plant and the CO₂ compression system (see Table 2). For FOAK plants, PCC shows, in general terms, a higher cost than the power plant package. However, it is estimated that the cost of the PCC plant could decrease in the future (NOAK plants), even making it potentially cheaper than the power plant package (see more details in Supplementary Table S4).

The CAPEX reduction in the PCC technology has a significant effect on LCOE. From Figure 5, we can see that CAPEX is one of the most crucial component cost. The estimated CAPEX reduction for a NOAK plant (mainly due to PCC technology) could potentially decrease the LCOE by 10%–11% in comparison with the FOAK plant type. For example, a NOAK NGCC plant with a 90% carbon capture level (NN-90%) has an LCOE equal to $92.5/MWh, which is 10.4% lower than a FOAK plant type (NF-90% = $103.1/MWh). This percentage of the reduction in the LCOE is in good agreement with the results published by Rubin et al. (Rubin et al., 2007).

On the other hand, Figures 6, 7 show the effect of the W r e g e n on the overall efficiency of the CO₂ capture process ( η 2 n d ) and CAC for conventional NGCC and EGR cases, respectively. For each curve, the initial point represents the BAU value for solvent regeneration (100% typical or traditional energy consumption), and consequent points represent steps of 10% decrement in the reboiler energy requirements, until reaching 0% (theoretical value). Considering that η 2 n d typically ranges from 5% to 40% for real-world separation processes (House et al., 2011), the block in yellow represents the probable future cost scenario for each case. In general, it can be seen that lower values of the W r e g e n represent lower CAC values and a higher percentage of the efficiency of the CO₂ capture process ( η 2 n d ). For example, in the NF-90% case, the effect of reducing the percentage of the steam consumption in the reboiler from 100% to 0%, reduces the CAC from $169.0 to $132.9/tCO₂ (21.4% cost reduction), while the η 2 n d increases from 14.7% to 88.0% (see Supplementary Table S5 in the Supplementary Information section). Nevertheless, 0% of steam consumption is thermodynamically impossible; thus, this value represents the theoretical CAC value for a FOAK NGCC power plant at a 90% capture level. For EGR cases, note that η 2 n d reaches values higher than 100%. This is because the W min has a higher value than the sum of W f a n and W p u m p (see Table 3). Unfortunately, this is not possible, as the theoretical value is limited to a η 2 n d equal to 100%, resulting in CAC values between $128.9 and $134.0/tCO₂ for FOAK EGR plants.

In addition, we can observe significant changes in the CAC value between FOAK and NOAK plants. For example, for a conventional NGCC plant with a 90% carbon capture rate (NF-90%), the BAU value for a FOAK plant is $169.0/tCO₂ compared to the $132.4/tCO₂ for a NOAK plant (NN-90%), which represents a cost reduction equal of 21.6% (see Supplementary Table S5 in the Supplementary Information section). A similar percentage in the cost reductions shows the EGR cases; for example, in an EGR plant with a 90% carbon capture level, the BAU value for a FOAK plant (EF-90%) is $164.5/tCO₂ compared to the $127.3/tCO₂ for a NOAK plant (EN-90%), representing a cost reduction of 22.6%.

Regarding NOAK-type plants, Figure 8 shows the expected carbon mitigation cost for future NGCC and EGR power plants as a function of the capture rate at η 2 n d = 40%. Using the assumed value, it is observed that EGR shows a better economic performance than NGCC, with a minimum value of $102.5/tCO₂. This is equivalent to a CAC reduction of 37.3% compared to the BAU value for a similar FOAK plant (EF_85% = $163.5/tCO₂). At 85% capture, the CAC for the conventional NOAK NGCC plant is $104.1/tCO₂. Maintaining this same CAC value, the carbon capture rate can increase from 85% to 90.8% if the EGR configuration is implemented.

One of the thermodynamic implications of assuming a value of η 2 n d equal to 40% is much higher energy efficiency in the reboiler duty, estimated at a value between 1.1 and 0.8 MJ/kg of CO₂ desorbed. This is equivalent to a reduction in the energy penalty ranging from 72% to 78% of the BAU value of W r e g e n . If we place this number in perspective, most of the current innovative solvent-based PCC configuration plants reach reboiler duties between 2.3 and 2.9 MJ/kg CO₂ (Liu, 2020; Muhammad et al., 2020; Vega et al., 2020).

Considering a PCC’s learning rate equal to 11% and an η 2 n d = 40%, the results show that the estimated CAC for NOAK plants could be around $100–110/tCO₂. Despite the fact that NOAK plants considerably reduce the CAC compared to the FOAKs, it is noted that the cost of carbon mitigation is higher than the 45Q tax credit for geological purposes ($85/tCO₂). One possible way that could help to reduce PCC plant costs, without increasing the carbon tax credit, is through a higher learning rate. It is valid to assume that PCC technology could achieve learning rates similar to those of other energy technologies in the short term. The historical learning rate for wind power deployment in the United States (1985–1994 period) is estimated to be 32%, and that for solar PV in the EU region (period of 1985–1995) is 35% (McDonald and Schrattenholzer, 2001). For this purpose, the NF-90% case was considered and the value of the PCC technology learning rate varied from 0% to 35% (see Figure 9). It can be seen from this graph that the NOAK NGCC plant could reach the value of 45Q carbon tax credit from a learning rate close to 25%. This could have positive impact on the deployment of CCS at an early stage of the technology. Additionally, it can be observed that the NOAK plant could reduce its CAC to a value, in the best scenario, as low as $69/tCO₂ (59.2% lower than the BAU value for the NF-90% case = $169/tCO₂). This indicates that the NOAK plants, even with the technological and energy improvements assumed in this work, would have to be supported with an equivalent carbon tax to guarantee their operation during their useful life.

This study performed a scenario analysis for estimating the cost of FOAK and NOAK combined cycle plants equipped with PCC. Cost projections for conventional NGCC and EGR configurations were evaluated considering the η 2 n d and the CAC as main indicators. Several critical variables influencing the overall cost of the technology were considered, such as W r e g e n , technology maturity, learning rate, carbon tax credit, and carbon capture level. From the results obtained, the following remarks can be made:

• Among the major technology sub-sections involved in a CCS project, PCC shows the highest cost reduction, with a total CAPEX reduction of about 40% for a NOAK compared to a FOAK plant. While, for the power plant and CO₂ compression technology sub-sections, the CAPEX reduction is marginal. This CAPEX reduction for NOAK plants potentially represents a decrease in the LCOE by 10%–11%, and the CAC by 21%–23% compared with the BAU values for FOAK plants in similar conditions.