September 2024

Carbon Capture/CO2 Mitigation

Retrofitting CO2 capture in a petroleum refinery

This article examines the application of solvent absorption-based post combustion capture for refinery/site-specific factors and quantifies the impact this will have on CCS CAPEX/OPEX. This analysis helps to determine the optimum approach for any individual refinery site.

IP: 18.191.240.118

Published in Hydrocarbon Processing’s August 2022 issue, the article “Evaluating options for decarbonizing petroleum refineries” estimated the carbon dioxide (CO2) emissions from refineries for 11 cases covering variations in crude quality, refinery configuration, cogeneration options, electricity sources, etc. The technology options for reducing CO2 emissions from each type of emissions source were presented and a qualitative evaluation was carried out. 

Carbon capture and storage (CCS) was one of the options identified as technically feasible for reducing CO2 emissions. The application of CCS at a particular refinery will compete with other options for CO2 reduction. The optimum mix of technologies will depend on site-specific factors and regulatory incentives. Capital and operational expenditures (CAPEX/OPEX) for retrofitting CCS at any existing refinery will vary significantly due to CO2 emissions profile (CO2 concentration in flue gas, flue gas flowrates from individual stacks), opportunities for heat integration, the marginal cost of steam generation, availability, and the cost of water and plot plan constraints. This article examines the application of solvent absorption-based post combustion capture for refinery/site-specific factors and quantifies the impact this will have on CCS CAPEX/OPEX. This analysis helps to determine the optimum approach for any individual refinery site.   

Methodology and model setup. A rate-based simulation model for CO2 capture from low partial pressure streams was developed and optimized. Monoethanolamine (MEA) is considered as a solvent for low CO2 partial pressure in flue gas streams. The model was optimized for the chemical process (temperatures, pressures, amine loading, amine circulation) and configured for estimating preliminary equipment sizes. FIG. 1 shows the typical schematic for an amine-based CCS plant.  

FIG. 1. Schematic of CO2 capture by using an MEA solvent. 

The model was built with an MEA solvent system using proprietary softwarea as a modeling tool. The thermodynamic package selected was amine sweetening using the Soave-Redlich-Kwong (SRK) equation of state. The softwarea was used to facilitate the computation of mass transfer characteristics of the system for sizing calculations. The following models were considered for the simulation: 

1. Thermodynamic method absorber: Mass + heat transfer (reactive + non-reactive) (proprietary thermodynamic method of the softwarea) 
2. Thermodynamic method stripper: Mass + heat transfer (reactive + non-reactive) (proprietary thermodynamic method of the softwarea). 

Additional model setup details included:  

1. A flue gas blower: The blower increases the pressure to overcome the pressure drop in downstream components.  
2. A direct contact cooler: Flue gas is cooled using a direct contact cooler.  
3. An absorber column: The column was run with structured packing and height estimated using in-built correlations in the softwarea. An absorber pump around the cooler was included to remove heat from the upper portion of the column. 
4. A CO2 stripper: The stripper stages were estimated using the mass + heat transfer (reactive + non-reactive) option available in the softwarea. 
5. Exchangers: All heat exchangers were configured to estimate the preliminary sizing information for comparative analysis.  
6. Cooling water: Supply and return conditions were fixed to estimate the overall cooling water circulation requirement and pump power. 

The following requirements must be considered when considering the deployment of CCS in a petroleum refinery:  

• Each refinery has a unique CO2 emissions profile depending on parameters such as crude quality, refinery configuration, source of electricity and source of fuel. 

• For a mid-sized refinery, CO2 emissions are spread all over the facility in approximately 15–20 stacks. 

• The flue gas flowrate and CO2 concentration in flue gases vary widely.  

• Steam, power and water costs vary significantly between sites and must be accounted for in the CCS design. 
• Plot plan constraints differ from each site, thus requiring the application of unique equipment arrangements.   

Inherent complexity and many CO2 capture options provide an opportunity to find optimum solutions through the application of innovative analytical approaches. 

Selected petroleum refinery configurations and associated CO2 emissions. Selected cases for crude oil, refinery configuration and associated CO2 emissions are taken from the earlier referred article and are summarized in TABLE 1. The source of electricity in each case is natural gas combined-cycle (NGCC)-based cogeneration. Refinery throughput is assumed to be 100,000 bpd. 

Two major sources of CO2 emissions are process heating and hydrogen (H2) production. Estimating the emissions from these sources requires the development of an overall refinery fuel balance. Refineries utilize fuel for process fired heaters and as supplemental fuel in SMRs for H2 production. The fuel fired can be gaseous fuel such as refinery offgas (internally produced fuel gas) and imported natural gas or liquid fuel (internally produced fuel oil). For this study, it was assumed that all process heating was with gaseous fuel. In Cases A–D, the internally produced offgas was insufficient to satisfy the gaseous fuel demand of the plant. Therefore, natural gas was imported as a supplementary fuel. The offgas and the natural gas were assumed to be mixed in a centralized refinery fuel gas system, and then distributed to all users of the gaseous fuel. The relative weight of the natural gas vs. the offgas was dependent on the refinery’s configuration and it is, therefore, different in the four cases. The compositions of flue gases from the various fired heaters of the refinery were calculated depending on the mixed fuel gas composition. 

The typical temperature levels of flue gases to the stacks depended on the presence of heat recovery coils in the convective section (e.g., for steam generation and/or superheating) and the presence of air preheating facilities (APH). For this article, all heaters are assumed to be with APH. Cases A and D were selected for detailed analysis described in the following sections. 

CO2 capture for Case A refinery. A summary of process heating services, duty, flue gas flowrate and estimated equipment sizes for capturing 90% of CO2 emissions from Case A’s refinery configuration is detailed in TABLE 2.  

The following are additional details for Case A CCS: 

• CO2 capture = 90% 

• Rich amine flowrate = 1,283 t/hr 

• Common regenerator size (diameter x height) = 5.8 m x 28 m 

• Regenerator reboiler duty = 82 MMKcal/hr 

• Regenerator condenser duty = 14 MMKcal/hr 

• Low-pressure (LP) steam requirement = 160 t/hr 

• Total cooling duty = 128 MMKcal/hr. 

TABLE 3 summarizes the process heating services, duty, flue gas flowrate and estimated equipment sizes for capturing 90% of the CO2 emissions from the refinery configuration in Case D.  

The following are additional details for Case D CCS: 

• CO2 capture = 90% 

• Rich amine flowrate = 2,878 t/hr 

• Common regenerator size (diameter x height) = 8.6 m x 28 m 

• Regenerator reboiler duty = 180 MMKcal/hr 

• Regenerator condenser duty = 32 MMKcal/hr 

• LP steam requirement = 360 t/hr 

• Total cooling duty = 290 MMKcal/hr. 

For both scenarios, it was considered more cost efficient to locate only the absorption section of an amine-based capture process near the CO2 emissions point and transport the CO2 absorbed in the rich solvent to the regeneration and CO2 compression section located further away. Key findings included:  

1. Emissions originate from many sources, with most sources contributing < 400 t/d of CO2. This requires many blowers, flue gas coolers and absorbers to be near individual stacks. Current state-of-the-art CCS plants design single absorbers to cater to 3,000 tpd–4,000 tpd of CO2 capture. Installing multiple small absorbers and associated equipment will not be cost effective. 
2. There was a significant requirement of LP steam—160 t/hr for Case A and 360 t/hr for Case D. Typically, refineries have extensive heat integration, and generating such a quantity of additional steam will not be possible through heat integration with waste heat alone. Therefore, it will require setting up a dedicated additional steam generation facility.  
3. There was a significant requirement for cooling duty—128 MMKcal/hr for Case A and 290 MMKcal/hr for Case D. This adds significant capital and operating costs and requires optimization. 

Each of these three key findings is discussed in subsequent sections to discover opportunities for optimization.    

Overcoming spatial constraints and achieving economies of scale. Space restrictions could severely impact the technical and economic feasibility of installing CO2 capture equipment and their supporting utilities in refineries. Existing (brownfield) facilities were not designed to accommodate spacious capture equipment, thus possibly making retrofit applications of CCS more challenging and costly. Spatial constraints can be solved by placing the CO2 capture unit (or part of it) farther away from the emissions point source. This implies flue gas transport over longer distances, requiring large-diameter and expensive stainless-steel ducting and additional energy for blowers. Three examples of layout alternatives are presented in FIG. 2 that can be considered depending on potential spatial constraints.   

FIG. 2. Layout options for CCS. 

Examples (a) and (b) depict a dedicated blower, cooler, absorber, stripper and compressor train for a single flue gas stack. This will be a practical option only for a few large CO2 sources (e.g., an SMR H2 unit, an FCCU) in larger crude capacity refineries to give economies of scale for the carbon capture equipment. Even in these cases, it is possible to have a common stripper/compressor for multiple sources. Conceptually, the configurations similar to option (c) provide an opportunity to maximize equipment sizes as well as overcome layout constraints. Depending on site-specific plot constraints, an optimum solution will need to be worked out. 

Meeting the steam requirement for CO2 capture. Additional steam can be produced via the following alternatives, either singly or in combination. 

        Waste heat integration. Refineries generally have an LP steam generation of approximately 4 kg/cm2 (g). This corresponds to steam pressures in the steam cycle and typically demand a temperature of approximately 150°C. This implies that waste heat below approximately 160°C (assuming a 10°C approach) is not utilized. The steam used for the stripper reboiler in the model was 2.5 kg/cm2 (g) and at 138°C. There may be opportunities at some sites for steam generation at these temperature levels.  

        Steam extraction from steam turbine-based power plant. Several refineries (including the examples considered above) have dedicated combined-cycle cogeneration power plants. Steam extraction depends on the configuration of the plant’s steam cycle. The objective is to determine the most optimum extraction with a minimum impact on the power plant. This requires expertise and modeling software that are typically used for modeling gas turbine combined cycles. The purpose of this study was limited to identifying methodology and streams for achieving integration. For this purpose, the simulation was performed using proprietary simulation softwareb for a base case steam cycle without steam extraction for CCS, and a modified cycle for a combined-cycle gas turbine (CCGT) with CCS. Both comprise two gas turbine generators, two heat recovery steam generators and one steam turbine generator. A three-pressure reheat system was employed.  

For the base case without carbon capture, the steam turbine gross power output was 169 MW. For CCS, the LP steam for the reboilers was taken from the crossover pipe between the medium-pressure and LP sections. The condensate return from the carbon capture plant was around 120°C. Because it represented a significant proportion of the boiler feedwater, the temperature profile in the economizer section changed, and the heat recovery steam generator (HRSG) exit gas temperature increased by 15°C, compared to the case without carbon capture. The steam turbine gross power output was reduced to 133 MW, and additional LP steam was generated for use in the stripper reboiler. If the refinery can source the power deficit from alternate renewable sources, the LP steam can be sourced by modifying the steam cycle. This was a very simplified evaluation to determine feasibility  based on purely thermodynamic considerations and must be evaluated in detail by rigorous modeling software and the feasibility of equipment modifications. 

       Steam generation from additional boilers. Multiple options are possible:  

• Natural gas/coal-fired boiler with CO2 capture 

• Electrical boiler. 

For sites with an abundant availability of cheap/renewable electricity, electrical boilers present a good opportunity for generating LP steam. Fuel-fired boilers will need CO2 capture, otherwise it offsets a significant portion of captured CO2. 

Cooling options study. Amine-based carbon capture units are highly energy intensive and have a very high cooling requirement. This section will focus on alternative methods of achieving the required cooling duty. The following options were considered: 

• Case 1: All cooling duty was by cooling water shell-and-tube exchangers. The electricity cost was $0.10/kWh, and the water cost was $3.75/1,000 gal. 

• Case 2: All cooling, other than the absorber pump around the cooler, was done with air-cooled heat exchangers. The electricity cost was $0.10/kWh, and the water cost was $3.75/1,000 gal. 

• Case 3: All cooling duty was performed by cooling water shell-and-tube exchangers. The electricity cost was $0.10/kWh, and the water cost was $10/1,000 gal.  

• Case 4: All cooling, other than the absorber pump around the cooler, was done with air-cooled heat exchangers. The electricity cost was $0.10/kWh, and the water cost was $10/1,000 gal. 

• Case 5: All cooling, other than the absorber pump around the cooler, was done with air-cooled heat exchangers. The electricity cost was $0.15/kWh, and the water cost was $3.75/1,000 gal. 

Since exchangers entail a significant capital investment and operating cost, one useful approach to compare alternatives is annualizing all costs. The capital cost must be annualized by assuming the interest rate and lifetime of operation of the exchanger. For this study, the interest rate was 10%, and the project life was 20 yr.  

For this study, the exchanger equipment costs have been estimated utilizing methodology suggested in literature.¹ An industry plant cost indexc was used to arrive at current costs. Since the objective of this study was comparative analysis, these preliminary costs served the purpose. This analysis provided good insights and options for selecting the cooling system when considering the overall lifecycle cost. TABLE 4 summarizes the findings. As shown, annualized costs are lower when utilizing cooling water exchangers.  

Takeaway. Post-combustion carbon capture is an option being suggested to reduce carbon footprint. Of the various technologies available for CO2 capture, capturing CO2 by chemical absorption is the technology that is closest to commercialization. The addition of carbon capture entails significant capital and operating costs. The application of this technology on a large scale for LP flue gases is still very limited, and design practices to reduce costs continue to evolve. This article investigated the retrofit of post-combustion carbon capture in petroleum refineries. Factors specific to refinery configuration that play a significant role in determining the capital and operating costs for CCS are plot plan constraints, the availability and cost of water, and the marginal cost of steam generation. Design options were evaluated and approached to systematically evaluate these options for site-specific applications.   

NOTES  

^a Bryan Research & Engineering’s ProMax® version 5.0.2 software 

^b Aspen HYSYS 

^c Chemical Engineering’s Plant Cost Index (August 2023) 

LITERATURE CITED  

1 Peters, M., K. Timmerhaus and R. West, Plant Design and Economics for Chemical Engineers, 5th Ed., McGraw Hill, December 9, 2022. 

 

The Author

Related Articles

From the Archive

Comments