Measure CO2 flows accurately for CCUS

Joonaki, E, TÜV SÜD National Engineering Laboratory

Energy demand is predicted to double over the next two decades. While the balance of energy sources will change during this period, all scenarios envisaged by organizations such as the Intergovernmental Panel on Climate Change (IPCC) and the UK’s Committee on Climate Change recognize that fossil fuels will still be part of the energy mix until 2050.

With the legislative adoption of net-zero targets by the UK and other countries around the world, in addition to drastic emissions cuts, “negative emissions” technologies will likely be a necessary part of the transition to net-zero. The latest IPCC report, “Climate change 2022: Mitigation of climate change,”1 includes actively removing carbon from the atmosphere through carbon capture, utilization and storage (CCUS) as a key part of keeping global warming to within 1.5°C–2°C.

A recent report from the International Energy Forum (IEF), “Strategies to scale carbon capture, utilization and storage,”2 found that CCUS deployment must reach at least 5.6 gigatons (Gt) of carbon dioxide (CO2) by 2050 from just 40 metric MMtpy today to meet the Paris Agreement and UN Sustainable Development Goals.

In addition to offsetting the direct end use of fossil fuels, there is broad agreement that CCUS will be required for the large-scale production of blue hydrogen (H2), which is produced from fossil fuel sources through reforming and the carbon produced (both directly and indirectly) sequestered.

Despite these drivers, the IEF report found several obstacles to reaching economies of scale, including significant upfront costs and energy penalties, poor market signals and regulatory hurdles. The IPCC report concludes that “a political commitment to formal integration (of CCUS) into existing climate policy frameworks is required, including reliable measurement, reporting and verification…of carbon flows.”

This article examines some issues associated with the underpinning measurements of CO2 that will be required to address the IPCC requirement.

Implementation of large-scale CCUS will require accurate knowledge of how much CO2 has flowed through pipelines and then been used or sequestered, much the same as custody transfer metering in the oil and gas industry today. Both the IPCC and IEF reports mention the use of CCUS hubs to reduce transportation costs, with multiple sources of CO2 feeding into shared infrastructure and pipelines for delivery to utilization or storage facilities. The EU Emissions Trading System contains a requirement for uncertainties of ±1.5% on mass flow measurements of CO2. With available flow metering technology, single-phase flowmeters will be required for CCUS applications, especially for transportation through pipelines, where the CO2 will be in the liquid or dense supercritical phases.

Depending on the flow metering technology used, it is required to know the values of key thermophysical properties of the flowing fluid. For example, density is required for conversion from volumetric to mass flowrate, while both density and viscosity are required to convert the transit time measured by an ultrasonic flowmeter to volumetric flowrate.

CO2 streams are never pure—they have varying degrees of contaminants depending on the process from which the CO2 is being captured and the capture technology employed. Therefore, it is necessary to measure the composition of the process streams in real time, especially where CO2 from multiple sources is fed into shared infrastructure.

Phase diagrams of pure CO2 (FIG. 1) have several phase transition boundaries close to the temperature and pressure regions relevant for pipeline transport (e.g., the critical point is 31°C and 74 bar). Furthermore, even low levels of impurities significantly perturb the fluid properties and open two-phase region boundaries.These factors impact flow measurement in several ways. First, given the use of single-phase flowmeters, it is important to understand the fluid properties of the process stream in real time to ensure that the stream stays in the intended phase (i.e., gaseous, supercritical, liquid). If the prevailing phase changes from that for which the meter is intended, or if multi-phase flow develops, the accuracy of the flowmeter will deteriorate significantly to the point where fiscal measurements are impossible.

FIG. 1. Phase diagrams of CO2-rich streams in the presence of some common industrial impurities: nitrogen (N2), methane (CH4), carbon monoxide (CO) and H2 at various concentrations from 10 mol%–30 mol%. Solid lines are bubble lines and square dot lines are dew lines.

Secondly, unless a Coriolis flowmeter is used, a conversion from volume to mass flowrate is required—this relies upon detailed knowledge of fluid properties, and pressure and temperature conditions, to determine the instantaneous density of the process stream.

In practice, a mixture of flow metering technologies is likely to be used within any individual CCUS hub. Coriolis flowmeters are likely to be the preferred technology for lower flowrate applications (e.g., captured streams from small-scale CO2 sources and for injection into geological storage facilities). However, higher flowrates in large-diameter pipelines will probably use full-bore ultrasonic flowmeters. Therefore, at a high-level, the composition of the process stream (one of the primary measurements) will feed into the determination of the fluid properties, which in turn will feed back into the primary flowrate measurement.

A further complication is that there are many different impurities in the flue gas, initial capture product, or in the final CO2 product ready for transportation, as explained earlier. Some impurities originate from the combustion air or the fuel, while others are combustion products. Also, some impurities in the fuel (e.g., sulfur) react in the combustion process, generating new compounds that were not present in the initial CCUS mixture. Typical impurities for CCUS are listed in TABLE 1.

The effects of the impurities alter the phase boundary and create a two-phase region, as well as change physical properties such as density and viscosity. Three categories can be identified for the impurities, based on their amount in CCUS fluids, as shown in TABLE 2: major impurities, minor impurities and trace impurities.

Although some of the major, minor and trace impurities are considered for their health, safety and environment (HSE) repercussions, they might not be essential to be factored into modeling. They may not be a vital term for fluid modeling conceptually, especially compared to much larger concentrations of components such as H2, N2 and O2. It is worth noting that the concentration magnitude does not indicate the level of importance. Some components [such as mercury (Hg), N2 and ammonia (NH3)] often exist in low concentrations in CCUS fluids, but they can strongly influence the overall fluid properties—this is one of the potential challenges of transportation of CCUS fluids.

Equally, monoethylene glycol (MEG) and triethylene glycol (TEG) are listed as minor impurities but are considered of greater importance due to their harmful effects on the CCUS system (e.g., damage to elastomers and liquid dropout in lines).

Therefore, balanced decision-making is needed to account for impurities based on their effects on thermodynamic properties and the modeling task load. Since the type and concentration of minor and trace impurities are process-specific, it would be ideal to consider different types and concentrations of impurities according to the source process.

Based on the current research on the physical characteristics of CO2 blends, a review/gap analysis was conducted at the TÜV SÜD National Engineering Laboratory. The assessment was performed to determine the lowest and highest pressure and temperature ranges at which fluid property measurements have been previously conducted

The study’s findings pointed out a general lack of experimental data and specific areas where there are few or no relevant research studies. The review's main goal was to collect information on physical properties, including density, and vapor-liquid equilibrium (VLE). Although data on ternary and other multi-component blends were collected wherever available, the focus was primarily on binary mixtures and blends with minor/trace contaminants. The results showed that there were insufficient data available for CO2 blends with minor or trace contaminants. Only solubility data for CO2/Hg, CO2/MEG, CO2/NH3 and CO2/TEG mixtures were discovered. Since minor/trace components are present in such low concentrations in CO2-rich streams, solubility data are readily available. Mercury, for example, is rarely present in quantities large enough to have a significant impact on total fluid density or phase equilibria. These components are considerably more significant in determining the solubility of the bulk phase to analyze dropout across the system at various pressures and temperatures.

Takeaways. The ability to accurately measure the amount of CO2 captured, transported and used or stored will be a fundamental foundation of large-scale CCUS. This presents some interesting technical challenges that require an integrated approach to real-time determination of process stream composition, bulk flowrate and fluid properties. The essential technologies exist, but the challenges of integrating these and making them economically viable should not be underestimated. GP


  1. Intergovernmental Panel on Climate Change (IPCC), “Climate change 2022: Mitigation of climate change,” 2022, online:
  2. International Energy Forum (IEF), “Strategies to scale carbon capture, utilization and storage,” September 2021, online:

Edris Joonaki is a Consultant and Technical Lead at TÜV SÜD National Engineering Laboratory, a provider of technical consultancy, research, testing and program management services. Part of the TÜV SÜD Group, the company is a global center of excellence for flow measurement and fluid flow systems and is the UK’s National Measurement Institute for Flow Measurement.



{{ error }}
{{ comment.comment.Name }} • {{ comment.timeAgo }}
{{ comment.comment.Text }}