Commercial-Scale Front-End Engineering Design Study for MTR’s Membrane CO₂ Capture Process        

The US Department of Energy (DOE), through its Office of Fossil Energy and Carbon Management (FECM), has funded several Front-End Engineering and Design (FEED) studies for retrofitting existing fossil fuel power plants with carbon capture technology. These comprehensive studies are publicly available on the US DOE Office of Scientific and Technical Information (OSTI) website as part of the knowledge sharing requirements associated with the funding. 

This study is our effort to help stakeholders quickly understand the key findings in lengthy reports highlighting methodologies, and unique design aspects that could shape future carbon capture plant designs.  Our objective is to highlight the most relevant sections of each study. 

Overview

Conducted by Membrane Technology Research Inc. (MTR) in collaboration with Basin Electric Power Cooperative, Sargent & Lundy, Trimeric Corporation, and the Electric Power Research Institute, this FEED study examines the addition of a commercial-scale carbon capture facility to an existing coal-fired power plant. The host facility is Unit #1 of Basin Electric Power Cooperative’s Dry Fork Station, located in Gillette, Wyoming. The objective of this study was to design a system capable of capturing and compressing approximately 70% of the carbon dioxide (CO₂) using MTR’s membrane-based post-combustion carbon capture technology at a 90% capacity factor. 

FEED study link: Commercial-Scale Front-End Engineering Design (Feed) Study for MTR’s Membrane CO₂ Capture Process (Technical Report)

Total CO₂ captured: 2.394 MM tonnes/yr

Total As-Spent CAPEX ($MM USD in 2022): $1,558.36

Objective

This summary highlights the unique design and planning aspects from this FEED study which include: 

• Use of MTR’s membrane technology
• Adaptation to existing infrastructure
• Use of flue gas pretreatment – SO₂ polisher and direct contact cooler
• Evaluation of permeate compression equipment
• Selection of a Hybrid Wet Surface Air Cooler System
• Investigation of a comprehensive CO₂ dehydration system
• Use of variable frequency drives
• Use of equipment sparing and redundancy

Use of MTR’s Membrane Technology

What is it:

With a Technology Readiness Level (TRL) of 8, membrane-based technology has been successfully demonstrated at a commercial scale for carbon capture (to learn more about TRL, click here). The technology uses a semi-permeable membrane to selectively separate CO₂ from other gases present in the facility’s flue gas. The membrane is designed to allow CO₂ to pass through easily, while other gases are blocked or slowed down, much like a filter or strainer. This selective process makes it easier to capture and remove CO₂ from the host facility’s flue gas emissions. 

In this study:

This study utilizes MTR’s membrane-based post-combustion carbon capture technology, using two-stage Polaris™ membranes to selectively separate CO₂ from the produced flue gas. The Dry Fork Station Unit #1 currently operates with two separate flue gas trains, and the new carbon capture plant will maintain this same two-train configuration.

Each flue gas train is first pre-treated by the sulphur dioxide (SO₂) polisher and the direct contact cooler (DCC). The trains are then pressurized by a booster fan before entering the first stage membrane (Membrane A). Membrane A is designed to allow a CO₂-rich gas stream (permeate) to pass through the membrane, increasing the CO₂ concentration within the stream. The remaining CO₂-depleted flue gas stream (retentate) returns to the existing power plant stack and is released to the atmosphere. 

Using ten turbofans and two compressors per flue gas train, the permeate stream from Membrane A is drawn out by a vacuum created by the permeate compression system. From here, the permeate stream is directed to the second-stage membrane (Membrane B), where the CO₂ concentration is further increased within the stream. The remaining retentate stream from Membrane B is recirculated back to the inlet of Membrane A. Meanwhile, the permeate stream from Membrane B is drawn out by a vacuum using a second permeate compression system, which includes two turbofans for each flue gas train that combine into a shared six-stage compressor. 

After the second permeate compression system, the permeate stream is sent to the dehydration system to be chilled using a propylene refrigeration system and then superheated before entering the dehydration beds (i.e., molecular sieves). In the dehydration beds, the stream’s water content is reduced to less than 1 part per million by volume basis (ppmv). Once dry, the CO₂ rich stream is cooled again and goes through a distillation process to meet end use specifications which follow the recommended limits for Enhanced Oil Recovery provided by the U.S. DOE National Energy Technology Laboratory. Finally, the permeate stream goes through CO₂ product pumps and heaters to increase the pressure and overall temperature to meet CO₂ pipeline transport specifications.

Additional information:

Discover how membrane technology stacks up against other carbon capture technologies by visiting our CCUS Technology Comparison Guide.

What is it:

Designing and implementing new equipment in a way that can easily integrate or leverage pre-existing components, operational conditions, and the physical layout of the host facility.

In this study:

The Dry Fork Station Unit #1 operates with two separate flue gas trains, each supported by its own induced draft fan and leading back to a single stack for release to atmosphere. To avoid significant changes to the flue gas path, the new carbon capture plant design integrates into the existing power station’s layout.  This is accomplished by maintaining the two-train configuration with new ductwork, routing the flue gas from the existing induced draft fans to initial capture components. 

To overcome the increased flue gas resistance introduced by new capture system components such as the DCC and ductwork, an assessment was conducted to determine if the existing induced draft fans required modifications. By slightly tipping the fan blades, the design was able to increase the fan’s static pressure by approximately 15%, which was enough to move the flue gas from the DCC to the new booster fan at atmospheric pressure. This modification avoided the cost of installing entirely new fans, as the existing fan motors and supporting electrical infrastructure were found capable of accommodating the increased power demand.

The study also maintained Dry Fork Station’s goal of ensuring the carbon capture plant also operates as a zero-liquid discharge (ZLD) facility. To do so, a high efficiency reverse osmosis system was selected to treat the cooling tower and wet surface air cooler blowdown streams. The waste from these systems is managed using a lime fixation system, chemical precipitation and microfiltration to generate solid waste. 

Additional information:

In regions facing water scarcity or where minimizing liquid waste disposal is a priority, recycling and reusing waste streams becomes an important strategy. To eliminate or reduce liquid waste generated by the plant, a ZLD facility uses a variety of treatment and management strategies. In ZLD plants, most of the liquid waste is recovered for reuse, and the remainder is transformed into a solid for disposal. This approach may help a facility avoid the complexities of water and waste disposal regulations, as no liquid effluent is discharged from the system.

What is it? 

Flue gas pretreatment involves conditioning flue gas streams to achieve optimal capture performance and to protect equipment downstream of the unit. This step typically involves cooling and removing impurities, such as acid gases, particulates and moisture from the flue gas before entering the main capture system. 

In this study:

The new pretreatment system consists of two DCC vessels (one for each flue gas train) aligning with the host facility’s existing flue gas path. The purpose of the DCC vessel is to lower the flue gas temperature and to reduce acid gases and moisture content before entering the capture membranes. Each DCC vessel features an integrated design with the lower section containing a SO₂ polishing unit and the upper section housing the DCC.

First, the flue gas enters the SO₂ polishing section where SO₂ is removed to prevent the contamination of condensed water in the following upper cooling stage. This enables the recovery and reuse of high-quality makeup water for cooling towers or other plant systems. This section uses a 25% sodium hydroxide solution which is recirculated to maintain a pH level of 7–8, preventing unwanted CO₂ absorption while ensuring effective SO₂ removal.  In this study, the system was designed to reduce SO₂ levels from 27 ppmv to less than 5 ppmv at the DCC vessel outlet. The MTR membrane technology removes SO₂ from the flue gas more efficiently than CO₂, therefore a lower SO₂ value at the membrane inlet is required.

The upper section of the vessel containing the DCC lowers the flue gas temperature from 106°C to 32°C. Water that condenses out of the flue gas during cooling is recycled back to the wet surface air cooler.

Additional information:

Depending on the type of technology used, flue gas pretreatment configurations can vary.  Some common configurations include separate DCC and SO₂ polisher units, or as in this study, a combined design within a single vessel. In combined configurations, the DCC unit can be placed in either in the top or the bottom section depending on operational requirements, and the flue gas may contact the DCC or the SO_2​ polisher unit first. Capture technologies with low SO₂ tolerance may require different configurations or even a dedicated SO₂ treatment unit to ensure effective SO₂ removal, overall maximizing a facility’s CO₂ capture efficiency.

What is it? 

Permeate compression equipment selection studies are conducted to compare the installation cost of different equipment setups (i.e., fans and compressors) to generate a vacuum to draw out the permeate stream from Membrane A, and then increase its pressure to enter Membrane B. This vacuum serves as the driving force that pulls the CO₂ molecules across the membrane material to facilitate separation.

In this study: 

This study compared three potential equipment configurations to identify the most cost-effective and reliable solution to increase the permeate pressure exiting from Membrane A (1.4psia) to meet the inlet pressure required to enter Membrane B. The configurations were:

• Four compressors
• Sixty fans
• A hybrid arrangement combining twenty fans and four compressors

The four compressor configuration requires two compressors per flue gas train. This design was deemed high-risk because it posed a single point of failure that threatened the entire operation. The risk was further compounded by the fact that the chosen compressor vendor lacked prior experience in building the specific size and type of compressor needed for CO₂ applications. The highly integrated nature of this design and its reliance on unique first of a kind vendor services rendered this option too risky to pursue.

The sixty fan configuration was comprised of thirty fans per flue gas train and was immediately disqualified from further consideration due to its large capital cost. This option was approximately 36% more expensive than the four compressor configuration, representing an additional ~$49 million in capital expenditure. 

The hybrid configuration, comprised of twenty fans and four compressors (ten fans and two compressors per flue gas train), was selected as the optimal design for the facility. Compared to the higher risk four compressor scenario, this option increased costs by only 2% (~$3 million). The key advantage of this configuration was its ability to continue partial operations in the event of equipment failure, offering a crucial advantage over the four compressor design that could shut down a portion or the entire capture facility. Incorporating the twenty fans was a key design decision as it routed the initial compression load to proven turbofans, reducing the pressure ratio on the first compressor stage and keeping it within its design limit. In addition, the fans help stabilize flow conditions at the compressor inlet, which is critical because fluctuations can lead to surge or stall and compromise system reliability.

What is it? 

A hybrid Wet Surface Air Cooler (WSAC) is a closed tube cooling system that utilizes both dry (air-cooled) and wet (evaporative) cooling sections to drop the process fluid’s temperature. Compared to conventional cooling systems (e.g., dry air coolers and evaporative air coolers), WSACs generally have a higher capital cost and can operate in several different cooling modes:

• Dry mode: Air cooling only using fans
• Wet mode: Spraying water externally over the heat exchanger tubes enables indirect evaporative cooling
• Hybrid mode: A combination of both approaches 

Compared to fully air-cooled systems, the advantage of WSAC coolers is that they can achieve a final cooling temperature that is closer to the lowest possible temperature set by the ambient humidity. By integrating the strengths of both cooling systems, hybrid WSACs ensure the cooling demands of the capture plant are met.

In this study: 

In this study, two WSACs are used to cool the circulating water in the upper cooling section of each DCC, which cools the flue gas stream before it enters the main capture system. Since there are two flue gas trains, this results in a total of four WSAC units.  A hybrid WSAC design was required due to site constraints on both water supply and disposal. This design minimizes well-water consumption by relying on DCC condensate collected from the flue gas as cooling-water makeup.

During summer months, the circulating water is cooled before returning to the DCC by first flowing through the dry cooling section, followed by the wet cooling section. By removing heat from the water stream in the dry stage, the system reduces the evaporative demand in the wet stage. This leads to more efficient cooling and optimized water utilization.

During the colder winter months, the system can bypass the dry cooling section and route the circulating water through the wet cooling section only. This prevents overcooling the circulating water and avoids freezing the wet cooling water in the WSAC basin by maintaining continuous water circulation. Additionally, the capture process produces more water in the winter because the incoming flue gas is colder, increasing the amount of moisture that must be removed. By directing all the circulating water through the wet cooling section, the system maximizes evaporation, helping to eliminate the surplus water. Operating exclusively with wet cooling is especially useful when the overall carbon capture process produces more water than needed.  

Wastewater in the form of blowdown is generated by the WSAC cooling system. To achieve the goal of operating Dry Fork Station as a zero-liquid discharge facility, the WSAC blowdown is sent to the high-efficiency reverse osmosis system for treatment, recycling the high-quality permeate back to the system.

Additional information:

During high efficiency reverse osmosis, mineral constituents are chemically precipitated while also removed by microfiltration. The process generates a solid waste stream which requires proper disposal.

What is it?

The dehydration system is an integral part of the capture process. To ensure the product is properly conditioned for transportation, storage and/or utilization, the system is designed to remove water from the CO₂​-rich stream. To explore different dehydration technologies, visit CO₂ Compression and Dehydration Technologies.

In this study:

The design goal in this study was to reduce the water content of the CO₂-rich gas stream to less than 1 ppmv, where the maximum water content within this stream could be up to 10 ppmv. To meet this, an adsorption-based dehydration system using molecular sieves to adsorb and remove water from the gas stream was selected.  Achieving this level of dryness was needed to prepare the stream for liquefaction and purification. 

Before the gas reaches the units dehydration beds, it is first chilled with an intermediate pressure refrigerant and passed through a knockout drum or feed separator to reduce its water content. This pre-treatment step reduces the size, capital costs and overall operating cost of the dehydration system.

The dehydration system consists of two identical parallel trains. Each train includes two sets of dehydration beds, totaling four beds per train (eight beds overall). At any time, one dehydration bed in each set facilitates drying (i.e., molecular sieves adsorb and remove water from the flue gas stream), while the other works to ensure regeneration (i.e., the dehydration bed is slowly heated to desorb the water from the solid adsorbent). The dehydration unit is fully automated, switching dehydration beds between drying and regeneration modes without operator intervention. Automated switching valves and the regeneration heater are controlled by the control system to maintain temperature set points. This configuration ensures efficient and reliable water removal, providing flexibility between drying and regeneration modes. 

Additional Information:

Using molecular sieves for adsorption-based dehydration systems is common for processes where CO₂ must be liquified through refrigeration.  When CO₂ is cooled into a liquid, any leftover water in the gas would freeze and clog the equipment. For safe and efficient refrigeration, molecular sieves can help achieve the extremely low water content required.  

What is it?

Variable Frequency Drive (VFD) is a technology that controls the frequency of electrical power sent to an alternating current (AC) motor. By adjusting the frequency, the VFD adjusts the motor’s speed, allowing it to operate at various speeds instead of a constant rate. VFDs provide several key benefits in electrical systems:

• Smooth startup: Motors start gradually and smoothly and then run at the optimal speed for the job, saving energy and reducing wear.
• Surge protection: Prevents sudden electrical surges during startup, protecting equipment and lowering electricity costs.
• Power regulation: Regulates power delivered to the motor, allowing the correct cable size to be selected, safely carrying the load.

In this study:

VFDs were used extensively on major rotating equipment throughout the process to ensure efficient operation and flexible control. Key applications included:

• Flue gas booster fan: Equipped with a VFD to automatically adjust the fan motor speed to overcome pressure drops across ducting and Membrane A.
• Membrane A and B permeate fans, hybrid WSAC system fans, circulating water pumps and product pumps: All fitted with VFDs.
• Membrane A and B permeate compressors and refrigeration system compressors: VFDs are used specifically for starting purposes, reducing inrush currents during compressor startup.

Additional Information:

VFDs present several drawbacks that must be considered when used for industrial applications. Their high initial cost can be prohibitive, especially for smaller operations. Energy savings may also take time to offset the equipment’s initial capital investment. Installation and maintenance require specialized expertise, and improper handling can lead to inefficiencies or equipment failure. VFDs also generate electrical noise and harmonics, which may interfere with nearby sensitive devices unless mitigated with additional components (e.g., line reactors, DC link chokes, harmonic filters etc.).

What is it?

The equipment sparing and redundancy philosophy dictates how duplicate components or backup systems are incorporated into a plant design to ensure reliability, maximize operational availability, and protect against single-point failures in the process.

In this study: 

In this study, the equipment sparing and redundancy philosophy focused on achieving high reliability for critical functions, particularly in the electrical and auxiliary systems, while minimizing unnecessary duplication of costly mechanical equipment. 

Although pumps represent a significant part of a project’s capital investment, they were identified as an exception to the design’s cost reduction efforts due to their critical role in maintaining process uptime and operational reliability. All pumps had a backup, primarily arranged in either a 2 x 100% configuration (i.e., one operating, and one standby) or a 3 x 50% configuration (i.e., two operating, and one standby). The need for sufficient compressed air to support process instrumentation also provided a critical demand to design the air system with a 2 x 100% arrangement. 

Additional Information:

Below is a comprehensive list of where redundancy and sparing were applied in the capture plant and the reasons for their implementation:

Redundancy for the auxiliary power system, instrumentation, and control systems can be found in section 7.2 of Appendix B in the FEED study.