Skip to main content

Development in oxyfuel combustion of coal

Published by
World Coal,


Toby Lockwood discusses developments in oxyfuel combustion of coal in the latest report from the IEA Clean Coal Centre (CCC).

Alongside pre- and post-combustion capture, oxyfuel combustion is regarded as one of the three principal strategies for the capture of CO2 emissions from fossil fuel combustion. In oxyfuel capture, a more concentrated stream of CO2 is produced by replacing combustion air with a mixture of oxygen and recycled flue gases, effectively transferring most of the energetic penalty of the process to the production of oxygen rather than the CO2 capture itself.

Although less-established than the other two strategies, oxyfuel has made significant progress in the last decade, with a number of large pilots commissioned and three demonstration-scale projects in various stages of planning. Research challenges for the oxyfuel process include the need for a thorough understanding of how coal combustion and corrosion chemistry respond to the altered boiler atmosphere, as well as the optimisation of two major additions to a conventional coal plant: an air separation unit (ASU) for the production of oxygen and a compression and purification unit (CPU) for the processing of flue gases into dense phase CO2 ready for transportation and storage.

Challenges of oxyfuel combustion of coal

The oxyfuel gas mixture has an inherently destabilising effect on combustion, largely due to the increased heat capacity of CO2 relative to nitrogen. However, extensive study of oxyfuel combustion of pulverised coal up to the scale of large pilot plants has demonstrated that stable, flexible oxyfuel flames can be achieved using oxygen concentrations greater than that of air and specially designed burners. Several oxyfuel-tailored burners are now available, broadly based on the principle of promoting ignition close to the burner, either by the injection of pure oxygen or strong internal recirculation. Crucially for retrofit applications, appropriate adjustment of the proportion of recycled flue gas can also be used to match the heat transfer characteristics of an oxyfuel boiler to those of an air-fired boiler.

Elevated levels of several potentially corrosive species in the oxyfuel boiler environment have led to concerns over an increased risk of high temperature corrosion. In particular, the recycling of flue gas before it has undergone desulphurisation or drying leads to concentrated levels of SOx and water vapour, potentially limiting oxyfuel to less efficient recycle paths or lower sulfur coals. However, current testing suggests high temperature corrosion rates and mechanisms are within the bounds of experience from air-fired plant. On the other hand, acidic corrosion can be extremely aggressive, requiring careful monitoring of low temperature sections of oxyfuel plant.

While the well-established status of cryogenic air separation technology constitutes one of the strengths of oxyfuel capture, the considerable energetic and economic cost of the ASU means there is a strong incentive for its optimisation. Fortunately, with only relatively impure oxygen (~97%) required for oxyfuel combustion, there is increased scope for the development of new, highly efficient ASU designs.

In the last fifteen years, ASU manufacturers have achieved around a 20% reduction in energy consumption, primarily through improvements to the distillation process cycle. Further energy savings are possible when the ASU is thermally integrated with the plant steam cycle by using the waste heat of air compression for feedwater heating. New technologies have also been developed for the energy efficient storage of liquid oxygen, which allows the unit to be operated more flexibly and cost effectively. Air separation by oxygen-conducting ceramic membranes could offer a more efficient alternative to the cryogenic process, and the last few years have seen promising scale-up based on compact, modular designs.

Oxyfuel flue gases can contain over 90% CO2 once water has been condensed out, but are still likely to require substantial processing in a CPU before the CO2 can be efficiently and safely transported. Residual oxygen and nitrogen from air ingress must be removed by the compression and partial condensation of the flue gases, using sensitive compressor and heat exchange equipment, which impose their own stringent purity requirements. Sorbents are employed for further dehydration, whilst a number of technologies have been developed for the removal of the acid gases. The high pressure, concentrated flue gas in the CPU could actually facilitate the removal of conventional pollutants, rendering possible the removal of SOx, NOx, and mercury as condensates without the use of chemical reagents.

Current experience with oxyfuel combustion

The commissioning of Vattenfall’s 30 MWth oxyfuel pilot at Schwarze Pumpe in 2009 represents a landmark for oxyfuel technology, as the first plant to combine the boiler, ASU and CPU processes. Despite closure of the unit in 2014, it has contributed hugely to the understanding of the oxyfuel plant operation and acted as a catalyst for technology development.

Scale-up of the oxyfuel process has nonetheless continued with the opening of a 100 MWth unit at Callide, Australia, also constituting the first oxyfuel retrofit and first to generate electricity to the grid. This plant has so far provided important insight into the challenges in optimising and controlling a large, multi-burner unit, and has operated successfully since 2012. Elsewhere, a 30 MWth unit at the Ciuden Research Institute in Spain is significant for its use of circulating fluidised bed combustion, a boiler technology which may be particularly suited to oxyfuel.

However, further scale-up to a demonstration plant of over 100 MW, integrated with CO2 transport and storage, is required before oxyfuel combustion can be considered technologically ready for commercialisation. Currently, the 168 MW FutureGen 2.0 project in the US is the closest to realisation, with a final investment decision expected in late 2014. In the UK, the 426 MW White Rose oxyfuel project is undergoing a FEED study but has already secured strong economic backing from both the EU and UK governments. Interest in oxyfuel capture is also growing in China, where the development of a 200 MW plant is led by Shenhua Group and Dongfang Boiler Works. As for other carbon capture routes, the principal barriers to such oxyfuel demonstrations are political and financial rather than technological.

A second generation of oxyfuel technology is emerging using less conventional boiler designs, often based on pressurised combustion, which allows the latent heat of flue gas water vapour to be usefully recovered, as well as reducing air ingress and the demand on flue gas compression. Already developed to the 5 MWth scale in Italy, a range of pressurised technologies have attracted interest from the US Department of Energy. Other concepts pursued for the improvement of oxyfuel include the reduction or elimination of flue gas recycle.

Conclusion

Estimates of the potential economic and energetic performance of an optimised commercial-scale oxyfuel coal plant suggest that the technology can compete favourably with other capture technologies, with efficiency penalties as low as 6% suggested by some studies. Other strengths of the process include its basis in well-established gas processing technologies and potential for smaller or more integrated conventional plant equipment. Following the successful operation of several large, full-chain pilot plants, we are likely to see at least one oxyfuel demonstration plant in the next few years if a favourable political and economic climate for large carbon capture and storage projects prevails.

The full report “Developments in oxyfuel combustion of coal” by Toby Lockwood is available from the IEA CCC Bookshop.

Written by Toby Lockwood. Edited by

Read the article online at: https://www.worldcoal.com/power/11092014/world-coal-development-in-oxyfuel-combustion-of-coal1311/


 

Embed article link: (copy the HTML code below):