Performance modeling of Allam cycle integrated with a cryogenic air separation process

https://doi.org/10.1016/j.compchemeng.2021.107263Get rights and content

Highlights

  • Natural gas fired Allam cycle integrated with a cryogenic air separation is modeled.

  • Performance modeling accounts for operational codependence of the cycle and ASU.

  • Integration of booster compressor heat significantly improves overall efficiency.

  • ASU compressors and design constraints can significantly impact optimum cycle design.

  • Multi-variable optimization led to an efficiency comparable with that of Allam et al.

Abstract

The performance of an integrated Allam cycle and a cryogenic air separation process is numerically simulated. In contrast to previous research which has primarily focused on the design of the cycle alone, this work accounts for the operational co-dependence of the cycle and a co-located air separation unit (ASU). Heat integration between the cycle and ASU includes adiabatic heat of the main air compressor (MAC), and booster compressor of the ASU overlooked in past studies. The discharge pressure of the recycled CO2 compressor is determined based on its saturation pressure at cycle minimum temperature. The cycle efficiency exhibits a maximum at a MAC discharge pressure of 4.2 bar whereas it consistently decreases with discharge pressure of the booster. A genetic algorithm optimization is applied to estimate an optimum operational point. A maximum efficiency of 59.7% is obtained at turbine inlet temperature, inlet pressure, and outlet pressure of 1500 K, 305.5 bar, and 28.1 bar, respectively.

Introduction

Air combustion of hydrocarbon fuels for energy production contributes significantly to CO2 emissions. In 2018, the CO2 emissions from burning petroleum fuels, natural gas, and coal accounted for 45%, 31%, and 24%, respectively, in the United States. In the same year, the electric power generation sector produced about 33% of the total U.S energy-related carbon dioxide emissions where 98% of that came from combustion of coal and natural gas (Energy Information Administration (EIA), 2019). Several technologies have been developed or under development to avoid the emissions of, mainly, CO2 formed during air-fuel combustion. While post-combustion capture methods are viable solutions for reduction of CO2 in the existing fossil fuel-fired power plants (Merkel et al., 2010; Kanniche et al., 2010), implementation of the concept of oxyfuel combustion in future power cycles appears to be a promising technique for clean power generation from fossil fuels (Sifat and Haseli, 2019; Stanger et al., 2015).

A novel power cycle that employs the concept of oxyfuel combustion has been developed by a British engineer, Rodney Allam. The construction of a 25 MWe natural gas fired demonstration plant has recently been completed in La Porte, Texas. The Allam cycle (named after its inventor) operating with natural gas includes an air separation unit (ASU), a fuel compressor, an oxidant plant, a combustor, a turbine, a multi-stream economizer, a water separator, a two-stage CO2 compressor with intercooling, and a recycled CO2 pump with a thermal efficiency as high as 59% (Allam et al., 2013, 2017). According to the findings of Barba et al. (2016), the Allam cycle is the best design among several oxy-combustion power cycles based on a political, environmental, social, technological, legislative, and economic risk analysis.

Detailed performance modeling of Allam cycle, often using a commercial software, has been documented in recent studies (Zaryab et al., 2020; Scaccabarozzi et al., 2016; Mancuso et al., 2015; Rogalev et al., 2019; Mitchell et al., 2019; Fernandes et al., 2019; Wimmer and Sanz, 2020). The layout of the cycle studied and operating parameters in these studies are slightly different from the original design (Allam et al., 2017). The cycle examined in Refs. Zaryab et al. (2020), Scaccabarozzi et al. (2016), Mancuso et al. (2015), Rogalev et al. (2019), Wimmer and Sanz (2020) employs a four-stage compressor with intercooling which raises the pressure of the recycled CO2 stream leaving the water separator to 80 bar. Located downstream of the recycled CO2 compressor is a pump which further increases the pressure to that of the oxygen stream coming from the ASU. The original design (Allam et al., 2017) uses a two-stage compressor with an outlet pressure that is equal to the pressure of the oxygen stream (100 bar) produced by the ASU so the need for an extra pump to match the pressure of the recycled CO2 and O2 streams is eliminated.

Mitchell et al. (2019) have analyzed an Allam cycle where a cryogenic ASU process developed by Allam (2018) produces oxygen at about combustion pressure (~300 bar) with 99.5% purity. The supercritical oxygen is then heated up to near ambient temperature before it is sent to the combustor so there is no need to an oxidant pump nor mixing oxygen stream with a fraction of the recycled CO2 as in the original design (Allam et al., 2017). The cycle analyzed by Mitchell et al. (2019) features a bypass stream compressor proposed by the cycle developers (Allam et al., 2018) which is taken from the turbine exhaust flow, compressed up to 80 bar, and returned back to the economizer. This bypass stream exiting the economizer is mixed with the recycled CO2 flow pressurized by a two-stage compressor to 80 bar. Because the bypass flow contains water, the cycle requires an additional water separator after the mixing point. The cycle efficiency is found to be 58% quite close to 58.5% reported in Ref. Allam et al. (2018).

Table 1 compares the operating parameters of the Allam cycle reported in the above studies. The net cycle efficiency reported in these studies is moderately lower than that reported by the cycle developers. Mitchell et al. (2019) attributed this discrepancy, in part, to the exclusion of the adiabatic heat of the bypass flow compressor. They reported a thermal efficiency of 56.6% for the cycle that would use heat of ASU only. On the contrary, Scaccabarozzi et al. (2016) and Mancuso et al. (2015) have found a net efficiency of around 55% under identical operating conditions; 1.6 percentage points lower than the above figure. The cycle efficiency reported by Rogalev et al. (2019) for a cycle similar to that examined by Scaccabarozzi et al. (2016) and Mancuso et al.,(2015) is nearly the same as that calculated by Mitchell et al. (2019). However, the ASU specific work assumed by Rogalev et al. (2019) is notably less than that employed in Refs. Scaccabarozzi et al. (2016), Mancuso et al. (2015) due to using oxygen with 91-92% purity. The lowest oxygen purity recommended by the cycle designers is 98% (Allam et al., 2013; Allam, 2018).

An important aspect of the Allam cycle is a co-located ASU which, in addition to supplying the required oxygen of the process, thermodynamically interacts with the cycle. Previous studies have primarily focused on performance assessment of the Allam cycle whereas the operational impact of ASU on the overall cycle efficiency has not been adequately studied yet. An accurate process model should account for the co-dependence of the cycle and ASU. Air separation process is commonly modeled by mere assigning a specific separation work usually in units of kJ/kgO2 as well as a transference of the adiabatic heat of the main air compression (MAC) of ASU. Cryogenic air separation processes producing oxygen at elevated pressures typically include an additional compressor, referred to as booster (Allam, 2009; Beysel, 2009). Even the air separation process developed by Allam, (2018) employs two booster compressors. Inclusion of the adiabatic heat generated by booster compressor(s) in the cycle heat integration has been overlooked in past studies. It is therefore necessary to account for detailed features of ASU in the performance modeling of the Allam cycle.

Unlike past studies which have primarily focused on the design of the Allam cycle itself, the present article develops a detailed process model to account for the co-dependence of the operation of the cycle and a co-located cryogenic air separation process. The model will be used to assess the impact of various heat integration scenarios and the operation of ASU compressors on the overall performance of the integrated system. More specifically, the goal is to clarify discrepancies in the reported thermal efficiency of the natural gas fired Allam cycle as a wide range of values can be found in the literature contradicting the 59% figure of the developers (Allam et al., 2013, 2017). A further task of this work is to investigate the impact of design constraints on the cycle performance and optimum operational design. The objectives and tasks laid out above have not been pursued adequately in the literature.

Section snippets

Integrated cycle-ASU process

Fig. 1 depicts a schematic of an Allam cycle integrated with a cryogenic ASU. The power cycle consists of a fuel compressor, a combustor, a cooled turbine, an economizer heat exchanger, a water separator, a two-stage intercooled CO2 compressor, a cooler, and a pump. The cycle is powered by methane that is supplied at 70 bar. The combustor receives a fuel stream, a recycled CO2 stream preheated in the economizer by recovering the thermal energy of the hot turbine exhaust and the ASU air, and an

Model description

Previous studies (Allam et al., 2017; Scaccabarozzi et al., 2016; Mancuso et al., 2015; Rogalev et al., 2019; Mitchell et al., 2019) have used commercial software for numerical simulation of the performance of Allam cycle. A detailed thermodynamic model describing the operation of the cycle integrated with a cryogenic air separation unit has not been reported yet. To fill this gap, a process model is developed for the integrated Allam cycle and the ASU shown in Fig. 1. The details of the model

Overall cycle efficiency

The calculated net cycle efficiency of the natural gas fired Allam cycle reported in past studies is in the range 53.9-58% whereas the efficiency reported by Allam et al. (2013, 2017) is about 59%. The reason for the discrepancies in cycle efficiency, aside from potential differences in the modeling methodologies employed, is partially due to the differences in the operating parameters assumed in previous studies – see Table 1 and compare the process parameters reported in Refs.

Impact of ASU

From the results and discussion of the preceding section, it is evident that the adiabatic heat of air compression substantially impacts the overall performance of the integrated Allam cycle and ASU. Due to the importance of the ASU heat integration, the process model developed in this study is used to assess the effect of the ASU compressors discharge pressure.

Fig. 5 displays the overall efficiency and oxygen temperature at the exit of MHE varying with the discharge pressure of the booster.

Impact of design constraints

The results presented so far are obtained subject to certain design constraints including a temperature difference of 5 K at the cold-end side of economizer and a minimum cycle temperature of 290 K. As discussed in Section 4, the underlying reason for discrepancy in the efficiency of Allam cycle reported in past works is mainly due to the differences in the assumptions and operational parameters employed in those studies. Here, we present illustrative examples to show the impact of design

Genetic algorithm optimization

The last section is devoted to find optimum operational conditions subject to the design constraints associated with the integrated Allam cycle and ASU system shown in Fig. 1. The optimization objective is to maximize the net efficiency with respect to four independent variables and subject to the design constraints as listed in Table 4. The optimum design point is searched using a genetic algorithm embedded (ESS, n.d.) in EES software. The number of individuals and generations was set to 16

Discussion and future work

Over the past 5 years, numerous research groups (Zaryab et al., 2020; Scaccabarozzi et al., 2016; Mancuso et al., 2015; Rogalev et al., 2019; Mitchell et al., 2019; Fernandes et al., 2019; Wimmer and Sanz, 2020) have studied the thermodynamic performance of the natural gas fired Allam cycle often by means of a commercial process simulator, e.g., Aspen Plus. These studies have greatly contributed to the understanding of the basic operation of the Allam cycle so the current knowledge of the

Conclusions

There has been an apparent confusion surrounding the power generation efficiency of the Allam cycle as one may find a range of values in different studies that are lower (notably in some cases) than the 59% figure enunciated by the cycle designers. The primary objective of this article was to investigate the cycle efficiency through a newly developed process model which accounts for the co-dependence of the cycle integrated with a cryogenic air separation unit. A net cycle efficiency of 58.2%

CRediT authorship contribution statement

Y. Haseli: Conceptualization, Methodology, Software, Investigation, Writing - original draft, Writing - review & editing. N.S. Sifat: Methodology, Investigation, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

The research fund provided by Central Michigan University is gratefully acknowledged.

References (35)

Cited by (25)

  • Allam cycle: Review of research and development

    2023, Energy Conversion and Management
View all citing articles on Scopus
View full text