Biological growth and fouling in amine and other organic solvent CO2 capture systems
After recent experiences with some stored amine samples in our laboratory it seemed that biological growth in amine solutions was a possibility, and one that I admit I had not thought much about until then, although I was aware of the possibility of using biological treatments on amine capture plant waste streams (e.g. Henry, 2016).
Following up with a literature search I found that, unfortunately, biological fouling has already been reported from experience at Boundary Dam 3, one of the few existing large amine post-combustion capture plants and the only one that has been run for anywhere near the 10 years that this unit will soon have achieved. This length of service is important because the biological fouling problems are stated to have become particularly serious only after nearly 7 years. I will review this experience in more depth but, before doing so, it is important to note that an absence of reporting does not necessarily mean an absence of biological growth in other systems. For example, as a purely theoretical suggestion, accelerated solvent degradation has often been observed in amine systems operated without proper reclaiming after an number of months of ‘incubation’ and apparently the shift towards enhanced degradation cannot be reversed except by a thorough clean – characteristics that appear, at least superficially, to be consistent with biological contamination attached to the internal surfaces of the plant.
The documented experience at Boundary Dam 3 is, however, a matter of record, although the full interpretation of the stated observations and the wider issues raised still give lots of room for conjecture. The only public domain report of which I am aware (now, I failed to note it at the time it was presented!) on the biological fouling experience at Boundary Dam is a free-to-access paper from GHGT-16, held in Lyon in October 2022, ‘Improving the Operating Availability of the Boundary Dam Unit 3 Carbon Capture Facility’ by a joint CCS Knowledge Center/SaskPower team. This paper started with a summary note that “the packing in the absorbers and the strippers have also experienced accumulations of fly ash materials and organic deposits that limit the gas flow rate and reduce operational capacity as a result” and then went on to explain that “organic buildup in the CO2 absorber wash water and the SO2 absorber caustic polisher sections have recently [i.e. relative to October 2022] developed into a significant fouling issue. These living organisms are believed to feed on trace constituents present in the system. Research is being performed in this area to understand and mitigate the biological buildup issue. The current mitigation plan is to perform chemical shocking to remove this particular buildup and has to be done while the plant is off-line due to the potential to contaminate the amine solutions. On-line cleaning procedures are also being developed and tested. It is not believed that this organic fouling is related to fly ash. Figures 10 and 11 show the buildup of pink slime, the organic fouling, in CO2 wash water and SO2 caustic polisher sections.” Images included a close-up of packing in the CO2 absorber water wash with a layer of slime over part of the top, a gloved hand holding a sample of the slime, slime buildup on the flow distributor troughs in the CO2 absorber wash water section and growth built up on the walls above the SO2 caustic polisher.
Apparently in connection with both fly ash and slime buildup in the packing it was stated that “differential pressure measurement in critical areas such as flue gas demisters, absorber beds, and heat exchangers are key to identifying the plugging issues. Higher pressure drops than the designed value suggests plugging of the equipment that require maintenance and cleaning. Depending on the severity of the pressure drop, the maintenance and cleaning may be performed at a regular scheduled outage or, in the worst case, may require an unscheduled shutdown.”
The paper noted that “examples of scheduled outage tasks include….. SO2 caustic polisher packing cleaning and CO2 wash water section packing cleaning. Differential pressures across the beds are continually monitored.” A full packing replacement, apparently for both beds, around June 2021 was also indicated in a figure showing the differential pressure across the SO2 caustic polisher and the CO2 wash water sections.
A summary of the latest maintenance planning includes cleaning various units at intervals of 6 months to 2 years plus “SO2 and CO2 absorber amine packings replacement” every 6-8 years. The paper concludes “The BD3 CCS Facility is a first-of-its-kind commercial-scale implementation of carbon capture technology. Early operations encountered difficulties in achieving desired operating conditions. The main factors contributing to reduced capture performance and availability stem from unplanned interactions between the flue gas particulate and the capture plant systems. Further, organic growth in some of the packed beds, unrelated to particulate, has also reduced plant output. Efforts have been made in many areas to mitigate these operational issues and to improve reliability. Adaptations to allow on-line maintenance combined with proactive and coordinated maintenance planning have contributed to a significant increase in capture plant reliability, but unanticipated events related to the CO2 compressor have recently hindered the ability for the overall plant to achieve availability targets. Maintenance outage plans will continue to be refined to address both on-going and arising maintenance needs and to achieve increasing reliability. The lessons learned from the BD3 CCS Facility experience should be invaluable to improving the performance and reliability of future carbon capture systems.”
Recommended by LinkedIn
It is worth noting that biological fouling is a normal aspect of wet cooling tower operation, with the ubiquity of airborne organisms making contamination and adaptive mutation to suit almost any conditions just a matter of time. It is also well-known that biofilms and suspended solids in the liquid flowing over the packing in a cooling tower make a bad combination. As one reference notes, “Another factor that influences fouling is the suspended solids’ concentration in the water. Any microbiological colonies that do become established in fill [packing] or heat exchangers typically secrete a slimy film for protection. This slime naturally captures silt.” The same reference also illustrates the potential attraction of amine and other organic solvent capture plants for suitable bugs, by analogy with contaminated water sources, “One increasingly popular [water] source, either by choice or mandate, is effluent from a publicly owned treatment works, i.e., municipal wastewater treatment plant discharge. These streams often have significant concentrations of ammonia, nitrite/nitrate, phosphate and organics, or in the industry vernacular, “bug food.” Untreated, such makeup water can cause explosive growth of microorganisms in cooling systems.”
And now some personal conjecture. Biological fouling has not been widely observed in flue gas amine capture plants to date because all pilot testing and virtually all the larger plants built so far have run for shorter, often much shorter, periods than the 7 years achieved at Boundary Dam before problems apparently became significant. Pilot plants are also thoroughly cleaned, e.g. with hot caustic solution, at frequent intervals to remove traces of one solvent before testing another. Anecdotally, it appears that similar cleaning may also be used fairly frequently in many smaller amine units capturing CO2 for utilisation. And how often has biological growth occurred and not been recognised? But with a new wave of commercial amine plants being planned, and with the example of Boundary Dam, a more relevant question is probably “what can be done to control the problem when it does occur?” It might be suggested that the solution is to prevent biological infection in the first place but this would require stopping all air ingress at all times or ensuring internal conditions everywhere in the capture plant that are so extreme that nothing at all can live there, neither of which seem easy to guarantee. Management, rather than avoidance, seems clearly relevant.
(As an aside here, note that the many amine plants used for gas sweetening run with the absorber at high pressure and no oxygen present and the stripper at moderate pressure, so quite different condition from flue gas capture. Conversely, PCC plants on gas turbine flue gas will see much higher oxygen contents than at Boundary Dam and direct air capture (DAC) plants even higher oxygen levels, plus inherently extreme scope for organism exposure.)
Management of biological fouling generally implies making conditions unappealing, so the problem doesn’t grow to intolerable proportions. This should be particularly feasible if the scope for such countermeasures has been incorporated in plant design and operations planning from the start. Possibilities might include:
Our team is also considering setting up a SMART mini-pilot unit on a flue gas slipstream, deliberately infecting it with the suspect bottle of biologically-contaminated amine that we think we have in our lab and then seeing what occurs. We don’t, however, expect to be invited to try this out on sites that have, or even may in the future have, a PCC plant, even if we route the SMART flue gas vent line back to the combustor and promise to be COVID-careful with biosecurity. As with COVID, though, once adequate mitigating measures are in place biological ‘infections’ of solvent PCC plants will probably become, largely, an inconvenience rather than a showstopper.