Carbon Capture and CO2 Compression KPIs: Capture Rate, Reboiler Duty and Energy Benchmarks
Capture rate, reboiler duty, parasitic power, compression energy, availability and solvent makeup: how to measure each KPI, what moves it, and which figures have a cited source.
Track a capture and compression plant on six KPIs: capture rate, specific reboiler duty, parasitic power, compression energy, plant availability and consumable intensity. Measure each one per tonne of CO2 captured so plants of different sizes compare cleanly, and use the ranges below as reference points for spotting a gap, not as targets.
Every figure here shows its basis. A superscript number links to a public source in the Sources list at the end, and the sentence states that source's year and scope, including whether it is a modeled design case or a measured plant. A figure marked no cited source is kept exactly as originally published because no public source stating it has been found, so treat it as illustrative.
At a glance
Carbon capture and CO2 compression KPI benchmarks at a glance
GJ/t and kWh/t are per tonne of CO2. Ranges are as originally published unless a source is cited, and each section states the cited source's year and scope. Where the original used a word other than typical or strong, that word is kept in the cell. Not stated means no figure was given for that column.
Method: How the figures in this guide are sourced
A figure with a superscript number is stated by the cited source, listed under Sources with its publisher, date and access date. Several sources are modeled reference designs or single demonstration plants rather than surveys of operating plants, and the sentence says so. A figure marked no cited source is kept exactly as originally published because no public source stating it has been found. It is illustrative, not measured industry data. Worked examples are arithmetic, not benchmarks.
Capture rate
Capture rate is the headline KPI: the share of CO2 in the flue gas that the plant removes. In the U.S. Department of Energy's 2022 NETL baseline study, 90 percent is the reference capture rate, and solvent-based post-combustion capture at coal plants has shown the ability to capture 90 percent of the CO2 in the flue gas1.
U.S. regulation no longer requires that level. In its September 2026 partial repeal of the Carbon Pollution Standards, EPA determined that 90 percent carbon capture and storage is not the best system of emission reduction for coal-fired units12. The 45Q tax credit asks only that capture equipment on a power unit have a design capacity of at least 75 percent of the unit's baseline carbon oxide production13.
World-class amine plants sustain 92 to 95 percent no cited source, while poorly tuned units drift to 82 to 88 percent as solvent degrades no cited source. For a measured reference point, Petra Nova, the first U.S. commercial-scale post-combustion unit on coal flue gas, recorded 93 percent capture efficiency at its commercial operation date and a 90.2 percent three-year average at full load14.
A plant that watches only capture rate usually meets its permit and quietly loses margin every shift. Chase rate and energy together, never rate alone, or you buy percentage points at a terrible heat price.
How to measure
Measure capture rate continuously from inlet and outlet CO2 analyzers.
Put both analyzers on the same wet or dry basis.
Report it daily next to reboiler duty, so rate and energy are read together.
What moves it
Liquid to gas (L/G) ratio: raising L/G from 2.5 to 3.2 can recover 3 to 5 points no cited source, at a cost in pumping and reboiler energy.
Specific reboiler duty is the most important energy KPI, measured in gigajoules (GJ) of regeneration heat per tonne of CO2. World-class advanced solvents hit 2.4 to 2.6 GJ per tonne3, 4.
NETL's 2015 capture handbook reports 2.4 to 2.6 GJ per tonne for a proprietary amine blend, against 3.55 for monoethanolamine (MEA)3. NETL's 2022 baseline puts the CANSOLV process at about 2.5 MJ per kg of CO2, the same as GJ per tonne, for its coal plant cases4.
Conventional MEA sits at 3.5 to 4.0 GJ per tonne2. That low end assumes good heat integration: the same handbook says MEA needs 3.6 to more than 7 GJ per tonne depending on how much heat integration the plant has2. A fouled or poorly integrated plant exceeds 4.5 no cited source.
Tightening the lean or rich exchanger approach from 12 to 5 degrees C alone can cut duty by 0.2 to 0.3 GJ per tonne no cited source. NETL's 2022 reference design specifies plate-and-frame rich/lean exchangers with a 5 degree C approach15.
Trend duty against heat exchanger fouling, since fouling silently adds 0.3 to 0.5 GJ per tonne over a run no cited source.
How to measure
Divide reboiler steam heat by tonnes of CO2 captured over the same period.
Trend it daily, next to capture rate.
Log exchanger approach temperatures so fouling shows up before duty climbs.
What moves it
Lean and rich solvent loading optimization.
Cross exchanger approach temperature.
Heat integration with the compression intercoolers.
Exchanger cleaning before fouling builds up over a run.
Parasitic power is the share of host plant output consumed by capture and compression, a make or break number for power stations. In NETL's 2022 reference cases for new natural gas combined cycle (NGCC) plants with 90 percent capture, adding capture costs 6 percentage points of efficiency on a higher heating value basis, 11 to 12 percent relative to the same plant without capture5.
Those are modeled design cases for new F- and H-class plants, not a survey of operating units. World-class integration holds the penalty near 7 percent no cited source.
Per tonne, NETL's 2015 handbook puts the total parasitic load of 30 wt% MEA capture, steam and electricity including compression, at about 385 kWh per tonne of CO216. It says technologies in development reach less than 200 to 250 kWh per tonne16.
How to measure
Compare gross and net output over the same period, with the capture island's loads metered.
Express the difference as a share of gross output.
Split it into steam extraction, compression, and blowers and pumps.
What moves it
Steam extraction, which follows specific reboiler duty.
Compression energy is the cleanest cross plant KPI. Compressing CO2 to pipeline pressure takes 90 to 110 kWh per tonne6, 7.
The two sources bracket that range. The IPCC's 2005 special report puts compression to 110 bar at around 0.4 GJ per tonne, about 111 kWh6. NETL's 2015 handbook models a six-stage intercooled train to 2,215 psia at about 288 kJ per kg, or about 80 kWh per tonne of gas power before mechanical losses7.
Best in class compressor trains with optimized staging reach 88 kWh per tonne no cited source; worn or badly staged machines pass 125 no cited source.
Blowers and pumps
Blowers and pumps add their own load on top of compression. The IPCC's 2005 report gives typical electricity needs for post-combustion capture, excluding compression, of 0.06 to 0.11 GJ per tonne at coal plants, about 17 to 31 kWh, and 0.21 to 0.33 GJ per tonne at gas combined cycles, about 58 to 92 kWh8. Gas plants sit higher because their flue gas is more dilute.
How to measure
Meter compressor train power and divide it by tonnes of CO2 delivered.
Meter blowers and pumps separately from the compressor.
State the discharge pressure with the figure, since it sets the energy.
What moves it
Staging and intercooler performance.
Machine wear, which drives badly staged or worn trains well above design.
Availability and throughput decide whether the plant earns. World-class capture islands run 90 to 95 percent availability no cited source; new or troubled units sit at 75 to 85 percent while they chase foaming, corrosion, and compressor trips no cited source.
Because capital charges are fixed, every point of availability below target adds directly to cost per tonne. A realistic capacity factor target for a mature bolt-on unit is 85 percent9, 10. NETL's 2022 baseline assumes 85 percent for coal and gas plants with capture, and Petra Nova's reliability analysis expected an 85 percent mean capacity factor with a well-planned maintenance program and strategic spare parts9, 10.
Reaching it is hard. Petra Nova's final report tabulates 41, 34 and 29 full-day equivalents of outage attributed to the capture facility and its balance of plant in 2017, 2018 and 201910. Treat sustained operation above 90 percent as genuinely strong performance worth protecting with tight maintenance planning no cited source.
How to measure
Measure skid throughput against nameplate every hour.
Log the limiting constraint each hour: absorber flooding, compressor surge, or steam supply.
Separate capture island outages from host unit and offtake outages.
What moves it
Foaming control.
Corrosion management.
Fewer compressor trips.
Tight maintenance planning with strategic spare parts.
Consumable intensity separates a tuned plant from a leaky one. Solvent makeup at 1.2 to 1.8 kg per tonne is normal for MEA11. The IPCC's 2005 report puts MEA consumption around 1.6 kg per tonne of CO2 separated when flue gas sulfur oxides are held near 10 ppm11.
Watch absorber pressure drop as a health KPI. A 20 to 30 percent rise above clean baseline flags fouling, foaming, or maldistribution before capture rate even falls no cited source.
CO2 leak and vent loss should stay under 1 percent of captured mass no cited source. Anything past 2 to 3 percent means you pay to capture gas you never deliver no cited source.
How to measure
Reconcile solvent makeup in kg per tonne of CO2 captured every week.
Trend absorber pressure drop against its clean baseline.
Reconcile captured mass against delivered mass to find leak and vent loss.
What moves it
Control of solvent oxidation and thermal degradation.
Entrainment control in the absorber.
Flue gas pretreatment that keeps sulfur oxides low.
Turn the KPIs into an improvement program by ranking gaps in dollars per tonne, not percentages. Closing a 10 point availability gap often beats squeezing another capture point.
For example, a plant at 3.9 GJ per tonne targeting 3.4 saves 0.5 GJ per tonne. At 6 dollars per GJ of steam, that is 0.5 x 6 = 3 dollars per tonne.
Control limits to review weekly
Reboiler duty within 0.2 GJ per tonne of target no cited source.
Compression within 5 kWh per tonne of target no cited source.
Solvent makeup within 0.3 kg per tonne of target no cited source.
How to build your own baseline
Measure every KPI per tonne of CO2 captured so plants of different sizes compare cleanly.
Write down each definition, including analyzer basis and discharge pressure, and keep it fixed.
Generate the raw numbers with the capture efficiency, regeneration heat load and compressor power calculators, then trend them daily.
Benchmark against your own best sustained month before chasing published figures.
Recheck targets when flue gas composition or steam price changes, since both shift every target you set.
DOE/NETL Carbon Capture Program: Carbon Dioxide Capture Handbook. U.S. DOE National Energy Technology Laboratory. published August 2015. Chapter 4, Table 4-2 (six-stage compression, 23.5-2,215 psia, 86% polytropic efficiency), pp. 91-92. Accessed September 30, 2026.
DOE/NETL Carbon Capture Program: Carbon Dioxide Capture Handbook. U.S. DOE National Energy Technology Laboratory. published August 2015. Chapter 1, Section 1.3.3.1 (chilled ammonia), p. 27; Section 1.4.1 Summary of Advances, p. 32. Accessed September 30, 2026.