Special Focus: Hydrogen Infrastructure Development

 

K. BRUN, Elliott Group, Jeannette, Pennsylvania

Because of H₂’s potential in the context of developing a decarbonized energy infrastructure, technologies for efficiently producing, transporting, storing and utilizing it have attracted significant investment. 

Of the three H₂ production processes commonly used on an industrial scale (steam reforming, partial oxidation gasification and electrolysis), two require hydrocarbons for feedstock and emit significant amounts of carbon dioxide (CO₂) as a byproduct—the outlier is electrolysis, which requires only water and an electric power source and can be powered by renewable energy. Steam reforming and partial oxidation gasification, by contrast, are complex chemical processes that convert a fossil fuel (natural gas or coal, respectively) into H₂, carbon monoxide (CO) and other compounds. 

Due to the low cost of fossil fuels and corresponding plant process equipment, 98% or more of H₂ now produced is derived using one of these two methods. Converting natural gas to H₂ is a proven, relatively inexpensive process, particularly in North America where natural gas is abundant; even if gas prices were to rise, the conversion of coal to H₂ would still be both commercially viable and less expensive than electrolysis. Therefore, the most realistic path to decarbonization involves the sequestration of CO₂ emitted during the H₂ production process, rather than its elimination. As such, three important gas streams must be managed, even in a decarbonized economy: 

1. The production of natural gas, followed by its transportation to a site where it can be converted into H₂
2. The transport of H₂ to its end-use site, either an industrial facility or a power plant
3. The transport of CO₂ to an appropriate geological sequestration and end-storage site.

These gas streams will require substantial compression. As the complexities of both natural gas and H₂ compression have been extensively covered elsewhere, this article is chiefly concerned with solutions for the compression and transport of CO₂. While CO₂ compression has been successfully undertaken for many years as part of acid/sour gas injection and enhanced oil recovery projects, the scale-up required for carbon separation and sequestration in a decarbonized H₂ scenario would challenge even technologies now considered state-of-the-art. Because of this, a need exists for new CO₂ compression applications for separation, transport and storage injection. 

CO₂ compression: An overview. The pressure of CO₂ gas separated from the newly produced H₂ is strongly dependent on the type of separation process utilized: as such, it can vary from only slightly above atmospheric pressure to several hundred psi. Additionally, significant uncertainty surrounds geological formation injection pressure, since it depends strongly on the type of formation and its drilled depth of injection. The generally accepted rule, however, is that for each km of depth of injection, ~1,150 psi of gas pressure is required. Since many of the geological formations presently under consideration for CO₂ storage are relatively shallow, injection pressures below 2,000 psi should be expected to occur frequently. A typical carbon separation and storage pressure application requires CO₂ to be compressed from below 50 psia to above 2,100 psia, as shown in FIG. 1.

FIG. 1. Compression start and endpoint on CO₂ pressure-enthalpy diagram.

Many viable thermodynamic path options, including refrigeration and liquid pumping, near-isothermal, and high-pressure ratio compression, exist to move this compression process from its start to its endpoint. Put differently, the available options are to compress the CO₂ and remain in the gas state on the right side of the vapor dome, refrigerate the CO₂ and pump it in the liquid state on the left side of the dome or utilize some combination of these methods.

CO₂ sequestration and storage. For most power plant carbon capture and sequestration applications, the following new compression duties are required: 

1. Pipeline header injection and recompression transport
2. Injection into geological storage reservoirs for sequestration
3. Separation processes (membrane, thermal or chemical)
4. Power plant cycle compression.

According to industry convention, CO₂ as a supercritical (dense phase) fluid above 2,100 psi should be transported in pipelines. At 2,100 psi, CO₂ is well above its critical point and will be supercritical at almost all ambient temperatures. Fluids in a dense phase share some physical properties of liquids, such as a very low compressibility; they also share some physical properties of gases and will expand in space to fill voids. The advantage of transporting CO₂ at supercritical pressures, therefore, is that its density does not change much with pressure: from a thermodynamic perspective it is essentially pumped rather than compressed. This significantly reduces the power demand for the pumping stations along a CO₂ pipeline. 

However, there are two disadvantages: the added injection compression ratio required at the pipeline header station and the significantly higher material costs when building a pipeline designed for a maximum allowable operating pressure above 2,100 psi. Since the CO₂ available from separation is at low, near-atmospheric pressures (< 100 psi), the pipeline header station must always use a compressor; for a 2,100-psi CO₂ pipeline, a high-pressure ratio header compressor with many intercooled stages will be needed to handle the significant volume reduction. In such a case, however, beyond the header station the gas is simply pumped.

FIG. 2. Intercooled centrifugal barrel compressor feeding a dense phase pump. 

Fortunately, transport at 2,100 psi is not required for all applications: the actual transport pressure of CO₂ depends on the separation process outlet starting pressure, the distance the CO₂ must be transported and the geological sequestration injection pressure (which is often well below 2,100 psi). If a lower-pressure CO₂ pipeline is utilized, conventional compressors are preferred for the header station and for recompression along the line. The transport pressure is selectable depending on the carbon sequestration application; it is not always advantageous to go with supercritical CO₂. 

Purely from a compression stage thermodynamic perspective, CO₂ is a heavy gas but relatively easy to compress. That ease notwithstanding, CO₂ presents several technical challenges that must be addressed to make its compression or pumping process efficient and reliable. These include:

• Most equations of state for CO₂ are still inaccurate at high pressures and temperatures.
• CO₂ is a heavy gas, resulting in amplified rotor dynamic and impeller-dynamic forces.
• CO₂ has a strong thermodynamic path dependence and multi-phase behavior.
• CO₂ forms carbonic acid in the presence of water, which then drives corrosion.
• CO₂ is soluble in elastomeric materials, which can lead to rapid decompression failures.
• When rapidly expanded, CO₂ quickly forms liquids and dry ice, which can be a problem at the shaft seals.
• CO₂ has a low sonic speed, which results in higher shock losses and a reduced operating range.
• CO₂ selectively leeches certain elements from common metals and has a very low viscosity at high pressures.

All of these represent manageable, if complex, engineering and design challenges. 

Compression and pumping options. CO₂ has a high-pressure ratio per compressor impeller stage. Because of this, it also has a significant specific volume decrease with pressure along with a very high heat of compression. This means that CO₂ heats up when compressed and requires stage intercooling to maintain the gas temperature at reasonable levels so as not to damage the compressor seals and bearings. Furthermore, because of its rapid density change with pressure, a significant flow volume reduction requires a wide range of aerodynamic high-to-low flow compression stages.

FIG. 3. Integrally geared compressor with interstage cooling.

The following types of compressors are typically considered for high-pressure-ratio carbon sequestration applications: 

• Reciprocating
• Screw
• Centrifugal barrel
• Centrifugal horizontally split
• Integrally geared
• Hybrid centrifugal with dense phase pump.

Since both reciprocating and screw compressors are severely flow limited, in practice they cannot be used for large-scale carbon sequestration applications; the other options all rely on proven centrifugal compressor or pump impellers and differ primarily in their layouts and stage arrangements. The best configuration for the application type under discussion is either an intercooled, barrel, straight-through centrifugal compressor with a dense phase pump (FIG. 2) or an integrally geared intercooled compressor (FIG. 3). Both arrangements require 7–8 compression or pumping stages, with intercoolers between them, to meet the specified compression ratio (for simplicity’s sake, the schematics shown do not include all stages). Industry opinions vary as to which is the better selection: both arrangements have pros and cons depending on operating conditions and range, application-specific standards, plant type, maintenance expectations and service cycle.

FIG. 4. An 8-impeller-stage, barrel-type, centrifugal compressor with multiple nozzles for intercooling or side streams.

Compressor designs. The two designs discussed represent the most promising options, not the only ones. Several other machinery solutions for compression and pumping of CO₂ are commercially available, such as the multi-stage barrel centrifugal compressor shown in FIG. 4 or the horizontally split centrifugal compressor in FIG. 5. In both cases, two casing section intercoolers and one discharge cooler are required to avoid overheating and achieve efficient CO₂ compression; typical operating conditions for these machines range from < 50 psia (for suction pressures) to > 1,100 psia (for discharge pressures). At 1,100 psia, the compressor discharge gas will be in the supercritical state, meaning that after cooling the CO₂ it can be fed directly into the dense phase pump for higher-pressure pipeline transport or storage injection. Although the technical challenges of CO₂ compression are often application-specific and must be individually addressed for new carbon sequestration technologies, a wide toolkit is available for further development as necessary.

FIG. 5. A multi-stage, horizontally split compressor with nozzles for two intercoolers driven by an electric motor through a gearbox.

KLAUS BRUN is the Director of Research and Development at Elliott Group, leading a group of 60+ professionals in the development of turbomachinery and related systems for the energy industry. He holds nine patents, has authored more than 350 papers, and is the editor of three textbooks on energy systems and turbomachinery. In the past, Dr. Brun has held positions in product development and management, engineering and executive management at Southwest Research Institute, Solar Turbines, General Electric and Alstom.