Developing inexpensive capture of CO2 from power stations and other sources is a high priority. So far, progress has been slow at pushing down the cost and increasing the percentage of CO2 that is captured.

A recent US study offered a more optimistic future for CCS.[1] Researchers at Pacific Northwest National Laboratory developed a new technique and then did detailed modelling of how much a large scale plant attached to a power station would cost per tonne captured.. Their estimate – around $40 a tonne of CO2 for a coal-fired power station – is probably the lowest figure offered by scientists working in this critical field. Other forecasts have generally been more than $70 a tonne for gas-fired plants and slightly lower for coal.[2]

This note looks at the possible implications of the proposed approach for the future of carbon capture. It suggests that even at these very low carbon capture prices it may still be cheaper to use hydrogen rather than natural gas for making electricity.

This is a complicated piece of analysis, meaning that there are probably mistakes in the workings but, very roughly, hydrogen costs of $1.55 per kg in the US will compete with natural gas and this form of cheap CCS at 2021 prices. In the UK, the higher gas price implies that hydrogen is competitive at around $2.60 per kg. These prices are attainable within a few years and US support may mean that hydrogen is already competitive with natural gas if CCS and carbon pricing is included.

Low cost CCS

PNNL has been working on solvent development for more than a decade and proposes what appears to be a new class of CO2 absorbing chemicals. These solvents, CO2 binding organic liquids or CO2BOLs, require much smaller amounts of water to capture the gas.

The benefit of this is that far smaller amounts of energy are required to ‘boil off’ the CO2 after it has been captured. It can then be permanently stored, perhaps in a geologic formation. Other techniques may require as much as a third of the energy of a power plant to be devoted to separating out the carbon dioxide from the solvent once it has been captured.

The approach which PNNL has modelled in detail (but which has not yet been built at commercial scale), calls for the creation of a very large metal box. The CO2BOL is dropped in streams from the roof, at a rate of up to 4 million litres an hour or about 4,000 cubic metres. The flue gas enters the chamber at the bottom and rises, encountering the falling liquid which absorbs a very large fraction of the CO2. The liquid is then extracted and heated to remove the carbon dioxide.

The PNNL researchers say that their work suggests a need for a plant costing about $750m to capture the emissions of a substantial coal-fired station, which might be several million tonnes of CO2 a year. They claim potential capture rates of 90%,

In addition, the carbon dioxide must be transferred by pipeline to a point at which the gas is then piped into permanent storage. US government research estimates that the storage cost across much of the country is as low as $10 a tonne. (This seems to exclude the pipeline transmission cost, which will not be large in most circumstances).

The total costs of the full CCS process will be about $39 for the capture, $10 for the final storage and perhaps $3 for the transport to the well per tonne of CO2. This adds up to $52 a tonne. The capture cost is for a coal-fired power station; gas would be higher because the CO2 concentrations are much lower in the exhaust stream. Let’s assume the total price for capture of 90% of flue gas from a gas CCGT plant of about $60 a tonne.

What do these figures mean?

The purpose of this short section is to estimate how much carbon capture will add to the price of a megawatt hour of gas-produced electricity. It includes estimated figures for the cost of the greenhouse gases not successfully captured by the proposed new process, assuming that some form of carbon taxation applies. I copy figures from the IPCC for some of these costs.[3]

I use the following process to estimate the impact of fully accounting for the CO2 and methane emissions.

· First I use the IPCC figures to estimate what the emissions are from the drilling and transportation of natural gas to the typical power station (‘fugitive emissions’). The numbers include methane losses at the well but they are much lower than are currently being seen in many gas producing areas, such as the shale formations of the US. The IPPC says that typically the upstream emissions of methane from natural gas production and transportation process are about 120 kg per megawatt hour of electricity produced at a CCGT.

· Next, I use an estimate for the average CO2 output from the gas-fired power station. This will vary substantially depending on the age of the plant and the quality of its turbines and will vary slightly depending on how many hours a week the plant operates. The figure is 370 kg per megawatt hour.

· I assume the new CCS plant will collect and store 90% of all CO2 produced by the combustion of natural gas. The PNNL researchers say it is possible the percentage may be higher but experience suggests that getting over 90% capture is extremely difficult.

· The price of carbon emissions is forecast at $100 per megawatt hour in my analysis. The EU ETS price is currently about €88, equivalent to about $95.

· I have guessed that the extra gas required for the carbon capture processes is approximately 15%. (I could not find the projected figure in the paper itself). This figure is important because it means that the total CO2 produced to make electricity is assumed to be 15% higher than an unabated plant. This increases the fugitive emissions, the uncaptured CO2 and the tonnage of CO2 that is captured in the process of making a megawatt hour of electricity.

What is the price of this pattern of emissions? Some of the emissions will result in carbon charges, which will (or should be) be imposed by future emissions trading schemes in Europe and elsewhere. The other cost is the $40 a tonne for the carbon capture.

Uncaptured emissions

Fugitive emissions – 120 kg per MWh – inflated by 15% = 138 kg per MWH

@ $100 a tonne = $13.8 per MWH

Uncaptured emissions – 37 kg per MWH – inflated by 15% = 42.6 kg per MWh

@ $100 a tonne = $4.26 per MWh

Captured emissions

Captured emissions – 333 kg per MWh – inflate by 15% = 383 kg per MWh

@$40 a tonne = $15.32 per MWh

Storage costs

383 kg per MWh of CO2 to be stored at $13 a tonne = $4.98 per MWh

If these estimates are accurate, the total costs of carbon capture and possible carbon taxation at a gas fired power station is about $38.36 per megawatt hour of electricity produced.

This is the burden imposed by using CCS. The figure is actually unlikely to be as low as this. Initial estimates tend to be overly optimistic but nevertheless I’ll use this number.

How do the costs of gas with CCS and hydrogen power stations compare?

In the US in 2021, the average natural gas price to power stations was $5.15 per million Btu.[4] This is equivalent to about $17.57 per megawatt hour. If a power plant is 60% efficient, it will require $29.28 of natural gas to make a MWh of electricity.

For comparison in the UK, which is usually a low cost location in Europe, the current (February 2023) wholesale price of natural gas is about £47 a megawatt hour, meaning a megawatt hour of electricity will take about £78 ($93.65) of natural gas to make.

A working natural gas power station in the US using CCS and paying for remaining emissions would face costs of

· About $29.28 for natural gas

· About $38.36 for CCS and possible carbon tax

· The total is $77.64 per megawatt hour of electricity.

How do we compare this to hydrogen? The direct comparison is between $77.64 and the cost of hydrogen needed to make a MWh in a combined cycle power plant. We don’t need to compare the running costs of the two plants because they will be essentially the same. And hydrogen will also offer about 60% efficiency in a power plant.

Let’s look at a range of green hydrogen costs

· $1 per kg

· $1.50 per kg

· $2.00 per kg

· $2.50 per kg

To make a megawatt hour of electricity in a CCGT using hydrogen will take about 50 kg of the gas. The full cost needed to compare with natural gas with CCS at $77.64 per megawatt hour will be as follows.

At $1 per kg = $50 per MWh

At $1.50 per kg = $75 per MWh

At $2.00 per kg = $100 per MWh

At $2.50 per kg =$125 per MWh

What this says is that hydrogen has to be below around $1.55 per kg to compete with natural gas with CCS in the US. This is a tough target, although electrolyser manufacturer NEL has often said it is attainable by 2025 in good locations. The official US Earthshot target is $1 per kg by 2031.[5] The support offered by the Inflation Reduction Act would certainly get the cost down to approximately this level already.

What about the UK, where natural gas prices are currently very much higher? Including CCS costs at US prices (for want of good figures in the UK) suggests that the full price of making electricity with gas in the UK is now around $132 ($93.65 gas costs plus $38.36 CCS costs) a MWh. At this cost, hydrogen below about $2.60 per kg would be competitive with natural gas. This is clearly also possible within a few years.

There are many contestable assumptions in this analysis but I wanted to show that, even with extremely aggressive cost assumptions for CCS, hydrogen is not obviously uncompetitive with natural gas for making electricity.

[1] I am very grateful to Thad Curtz for pointing out this research. https://www.cnbc.com/2023/01/24/new-technique-from-us-national-lab-to-remove-co2-at-record-low-cost.html covers the topic. The paper is at https://www.sciencedirect.com/science/article/abs/pii/S0959652622052702. The paper says $40 a tonne, the CNBC piece $39 a tonne.

[2] This doesn’t make coal ‘better’ because over twice as much CO2 is produced for each megawatt hour of electricity produced compared to gas.

[3] IPCC Annex III, Technology Specific Cost and Performance Parameters, S. Schlömer et al. https://www.ipcc.ch/site/assets/uploads/2018/02/ipcc_wg3_ar5_annex-iii.pdf

[4] https://www.eia.gov/todayinenergy/detail.php?id=55519

[5] https://www.energy.gov/eere/fuelcells/hydrogen-shot