Compressing CO2: The Pressure Is Set by the Phase, Not by the Distance

A CO₂ compressor's discharge pressure is not chosen from a catalogue. It is set by where the CO₂ has to arrive and in what phase. Above 7.38 MPa and 31 °C, CO₂ is a supercritical fluid with liquid-like density and gas-like viscosity, and the reasoning that produces a pipeline pressure is the same reasoning that sizes the machine. Two Chinese machines now bracket that region: a 13 MPa-class single-shaft centrifugal unit installed for an oilfield capture duty, and a 15 MPa-class unit that completed a first trial run at a 4 Mt/y CCS demonstration in Shaanxi (1)(2).

Why CO₂ is an awkward gas to compress

Three properties drive the design, and none behaves as a compressor designer would expect from a permanent gas.

The critical point is low, and the process crosses it. CO₂ becomes supercritical above 7.38 MPa and 31 °C. Near that point its physical properties vary steeply with small changes in pressure and temperature, which is what makes the high-pressure end of a CO₂ machine a vibration problem rather than a straightforward aerodynamic one.

The density is high at the top of the range. Supercritical CO₂ occupies a small volume for its mass, so the later stages of a centrifugal machine are narrow and the impeller flow channels become tight. The Chinese illustration for the duty is a compression ratio that takes 140 bottles of gas at the same volume and puts them into one bottle (2).

Impurities move the critical point. Captured CO₂ is not pure CO₂. Depending on the capture route it carries nitrogen, argon, hydrogen, oxygen, sulphur dioxide, hydrogen sulphide and water. The mixture composition determines the critical pressure and temperature, and therefore the window in which the pipeline can operate in dense or supercritical phase — and the two are linked, because the compressor's work input changes with the mixture (3)(4). Water and sulphur compounds also make the fluid corrosive to pipe steel and welds (3), which is a materials question downstream of the machine and a limit on how much of each impurity is acceptable upstream of it.

Two destinations, two pressures

The pattern in that table is the technical content of this article. The machine class is not chosen by how far the CO₂ travels; it is chosen by the phase the receiving formation or the receiving pipeline requires. Saline-aquifer storage and oilfield injection in the same country can therefore sit a step apart in pressure, and both sit far above the pressure at which CO₂ leaves the capture unit.

The machine side

Shaangu's CO₂ centrifugal compressor technology is listed in the Ministry of Industry and Information Technology's recommended catalogue of energy-saving and carbon-reduction technology equipment. As the company describes it, the technology uses high-precision CO₂ physical-property calculation software and covers CCUS plants of 150,000 to 3,000,000 tonnes a year with discharge pressures from 2 to 20 MPa across grades; the design uses a short-and-thick dry gas seal, a shaft configuration the company calls a whip-resistant shaft, and vortex-suppression structures to reduce the effect of aerodynamic forces on the rotor, with a claimed polytropic efficiency above 80% under intelligent control (1).

A second product line covers 300,000 to 1,000,000 tonnes a year at discharge pressures up to 15 MPa, and the company states its machines are in service at a number of national demonstration projects, including a 500,000 t/y CCUS project for the National Energy Investment Group that the company says was selected among the National Energy Administration's ten major science and technology achievements, and a coal-capture integrated project with an expected reduction of 107,600 t of CO₂ a year (6).

For the largest single-train duties the integrally geared route is the other answer, and it is treated separately on this site: an eight-stage integrally geared machine compressing to 12.6 MPa at a coal-power capture plant, with per-stage impeller speeds chosen individually rather than shared on one shaft (Supercritical CO2 Compression: What an Eight-Stage Integrally Geared Machine Changes).

Why the oilfield duty pushes the pressure up, in the maker's account

Shaangu states that the oilfield storage scenario places an unusually high requirement on CO₂ compressor discharge pressure, and that because CO₂ physical properties change abruptly at high pressure the consequence can be severe unit vibration, seal failure and a sharp rise in operating loss. The company describes the machine as matched to the requirements of long-distance pipeline transport and deep reservoir injection (1).

The same account names the site conditions: ambient temperatures above 40 °C in a normal summer, ground-surface peaks above 50 °C, high wind and drought during construction (1).

The pipeline constraint

China's CO₂ pipeline stock is thin, and the reason is not the compressor. One industry account, citing the China Petroleum Pipeline Research Institute, reports that supercritical CO₂ pipeline mileage in China is under 1,000 km. Compared with natural gas pipelines, the service environment for the pipe and the girth welds is more severe, phase control during transport is more demanding, decompression during a leak can produce low-temperature brittle fracture, and an impurity-bearing supercritical CO₂ stream is corrosive to the pipe body and to welds. The account concludes that the pipeline technical system still needs to be completed and the transport standard system is yet to be established (7).

The same account gives the scale of the demand: 14 CCUS demonstration projects in operation in China, 39 under construction, and more than 4 million tonnes a year being stored. It also records one operator's cumulative position — 30 Mt-scale capture and injection capacity built up over nearly twenty years, with more than 500,000 tonnes of CO₂ stored to date (7).

The boundary of this set of numbers

Six limits belong alongside the achievement.

The 2–20 MPa and 150,000–3,000,000 t/y window is a product range, not a project record. Those figures describe what the supplier offers across a series (1). They do not mean a machine has been built and run at every point in that rectangle.

The efficiency figure is the supplier's own. The polytropic efficiency above 80% comes from the company's description of its technology (1), not from a third-party performance test.

The 13 MPa machine has completed installation, not commissioning. The release covers the end of installation work; no running data, discharge temperature or availability figure has been published for it (1).

The comparison to 140 bottles is an illustration, not a measurement. It is the company's own way of conveying a high pressure ratio on a dense fluid (2). It should not be read as a flow or a pressure value.

The 15 MPa machine's evidence is a successful first trial run. That is a commissioning milestone: the machine reached speed and pressure and its parameters met target (2). It is not an endurance record.

And the acceptable impurity level is not a fixed number today. The impurity question is documented in the literature and the mixture composition is known to shift the phase window (3)(4), but with the transport standard system still to be established (7), the limits that decide what a compressor has to tolerate are being set project by project rather than against a national figure.

What the trend looks like

The CO₂ pressure ladder is being set from the far end. Geology and phase behaviour decide the discharge pressure, and the machine range is then built backwards from it — which is why a single supplier now describes a 2–20 MPa window and why the two flagship Chinese machines sit at 13 and 15 MPa rather than at one design point. The direction of travel is toward higher pressure on the injection side and toward a standardised phase window on the transport side. The first is a machine problem and it is being solved; the second is a standards problem and it is still open.

What this article does not cover

Note that this article does not select a compressor for a capture project, does not calculate stage count, power or intercooling, does not model the phase behaviour of a specific stream, does not set or interpret an impurity limit, does not size a pipeline, and does not give prices. It describes why CO₂ compression is set by phase rather than distance, the machine classes Chinese projects are using, the two destinations that fix the pressure, and the limits of what has been demonstrated.

Related on this site

• Supercritical CO2 Compression: What an Eight-Stage Integrally Geared Machine Changes
• A 146 km Supercritical CO₂ Pipeline and a 900,000 t/y Purification and Boosting Station Break Ground in Shaanxi
• Compressed-Air Energy Storage: What the Compressor Train Has to Do

Sources