CO2 transportation research: Liquefaction of gas mixtures and recovering liquefaction cost (Flexible Funding 2020)

Prof. Kumar Patchigolla at Cranfield University was awarded funding in the UKCCSRC’s Flexible Funding 2020 call to look at the “CO₂ transportation research: Liquefaction of gas mixtures and recovering liquefaction cost”.

Increase in CO₂ emissions has led to global warming driving humanity to adopt Carbon Capture Utilisation and Storage (CCUS) processes more extensively. In this research project, work is carried out to lower the cost of CCUS deployment, in particularly for the marine sector by using a cold energy utilisation technology. Captured CO₂ is liquified and transported in a ship container at optimal conditions of -28.5°C and 15 bar. After the ship container has unloaded the liquid CO₂, the empty container tends to warm up due to heat transfer from ambient temperature. A Cold Thermal Energy Storage (CTES) unit is designed to utilise the cold energy of liquefied CO₂ to precool the ship container before loading, saving operational cost.

Fig. 1. Schematic of the (a) cold thermal energy storage 3D design and (b) cold energy utilisation technology

The designed CTES system consisted of heat exchanger, control centre, inlet and outlet valves as shown in the Fig. 1a. Heat exchanger thermodynamic calculations were done for the required flow of cold energy using liquefied CO₂. The material of the ship container was identified, and a process flowchart was created accordingly using Aspen HYSYS providing the flow rate of the liquefied CO₂ into the CTES system. It is concluded that the CTES unit of volume 15m³ utilising phase change material (PCM) can store 309 kWh of cold energy required to cool down a typical 11,500m³ ship container. Commercial PCM E-26 were chosen as the material costing £1.95/Kg and Carbon steel grade A517 was selected for manufacturing storage tank costing £0.97/kg. The cost estimates were excluding the packaging and transport costs. It has also been estimated that a typical ship taking around 200 trips per year, saving 309 kWh of energy per trip and with current UK electricity cost of 1 MWh being £1000, the overall cost saving by the proposed CTES system is £200,000 per year. This concept has the potential to its return within several years of implementation.

Additional research has been secured to explore its potential implementation with Oman shipping company.

Furthermore, as part of this project, we were able to arrange a site visit to Uniper Grian Power Plant in Kent, UK, where they are having significant long-term regasification capacity at the Grain LNG terminal, to convert LNG back to natural gas. We would like to extend our thanks to Uniper colleagues for their support in showing our future “Heat Engineers” around the power plant.

Fig.2. Uniper power plant site visit for Prof. Patchigolla’s group along with MSc Advanced Heat Engineering students at Cranfield University

Read more on Kumar’s flexible funding page 2020 project page

An open-technology and open-access post-combustion capture initiative – Professor Jon Gibbins, Director, UK CCS Research Centre and Professor of CCS, University of Sheffield; and Dr Stavros Michailos, Researcher Co-Investigator, University of Sheffield (now Lecturer, University of Hull)

Post-combustion CO₂ capture (PCC) plants are the basis for the majority of projects in the current wave of UK CCS deployment.  Applications actively being considered include gas-fired and biomass power plants, energy-from-waste plants, refineries and steam-methane reformer retrofits.  PCC plants are also widely applicable globally, particularly for retrofit applications.

This proposal addressed two issues:

1. Knowledge-sharing for cost reduction
2. Residual CO₂ emissions from PCC plants in the context of net-zero

The delivery of the objectives of this project was enhanced by a number of subsequent open-access projects that it facilitated.  The following non-experimental projects are completed:

• UPCC: Ultra-high Post-Combustion CO₂ Capture – a University of Sheffield QR SPF project (Dec 2020 to March 2021) – £18k funding for additional modelling work
• AECOM for BEIS, Start-up and Shut-down times of Power CCUS Facilities – showing solutions for 95% CO₂ capture at all times (Feb-July 2020), Gibbins acted as an academic reviewer. https://www.gov.uk/government/publications/start-up-and-shut-down-times-of-power-carbon-capture-usage-and-storage-ccus-facilities
• BAT Review for New-Build and Retrofit Post-Combustion Carbon Dioxide Capture Using Amine-Based Technologies for Power and CHP Plants Fuelled by Gas and Biomass as an Emerging Technology under the IED for the UK – UKCCSRC review by Gibbins and Lucquiaud https://ukccsrc.ac.uk/best-available-techniques-bat-information-for-ccs/ – 95% capture including start and stop are recommended in the associated guidance https://www.gov.uk/guidance/post-combustion-carbon-dioxide-capture-best-available-techniques-bat

Related experimental projects are still ongoing:

• PCC-CARER – UKCCSRC/industry funding https://ukccsrc.ac.uk/research/flexible-funding/flexible-funding-2021/prof-jon-gibbins-university-of-sheffield/
• FOCUSS – BEIS/industry funding https://www.ssethermal.com/news-and-views/2022/06/carbon-capture-rates-in-focuss-as-sse-thermal-secures-grant-from-beis/

Delivery of the project objectives was started by early workshops, as envisaged in the proposal, and then continued through stakeholder engagement on the BAT review, a number of conference presentations, an open-access journal peer-reviewed paper and updates on the project web page.  A list is given at the end of this blog.  This amounted to more material and engagement than anticipated in the proposal, due to assistance by the unforeseen related projects, and activity was extended over a considerably longer period than originally envisaged.  The project also helped to get a new detailed engineering study for an amine capture plant made available on the UKCCSRC web site https://ukccsrc.ac.uk/open-access-sherman-feed/ and publicised to potential users.

In addition to high levels of CO₂ capture from combustion flue gases, the research outcomes also suggested a way to reduce the regeneration energy penalty for CO₂ capture from air using amines.  This has resulted in a new UKCCSRC project with co-investigator Stavros Michailos as PI, Co-DAC: Low-energy Direct Air Capture potential when combined with a Post Combustion Capture plant, with the concept described in a GHGT-16 presentation/paper https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4283821.  Stavros Michailos has also progressed to a new position as Lecturer at the University of Hull.

As a result of computer modelling in this project, a clear theoretical understanding has been gained of how PCC plant operating parameters can be adjusted to achieve high levels of CO₂ capture and the factors that determine the amount of heat required when doing so.  In general, previous studies have not undertaken an holistic analysis and have therefore concluded that there are greater barriers to achieving high capture levels than, in fact, exist – the problems they encountered being due to pre-selection of operating parameters such as stripper pressure or absorber liquid/gas ratio.

An outline illustration of a PCC plant is shown below.  This project modelled the performance of such a plant using a non-proprietary capture solvent, a solution of 35% w/w monoethanolamine (MEA) in water.

One key finding was that lean solvent loadings of ~0.1 mol CO₂/mol MEA are required to achieve high capture levels (95-99%) in the absorber.  This is not a surprising result as the leaner a solvent is, the more aggressively it will remove CO₂ from a gas mixture.  The problem is how to produce this low lean loading without requiring excessive energy.  Previous studies, using a fixed stripper pressure, have concluded that a lot of energy would be required but we were able to show that, provided the pressure in the stripper/reboiler was high enough, the amount of heat required per tonne of CO₂ captured would not go up.  The high pressure is important to suppress wasteful water vapour production in the reboiler; some water vapour is essential to ‘strip’ the CO₂ from the solvent when it is heated, but a lot of heat is required to produce it.

It was also found that a high rich loading in the solvent leaving the absorber was essential to minimise heat demand in the stripper/reboiler; not doing so means that ‘unused’ solvent is being circulated and heated and also involves more water vapour being emitted from the stripper with the CO₂.  High solvent rich loadings require that enough packing is available in the absorber to transfer enough CO₂ from the flue gas to the solvent.  Up to 24 m of packing height was found necessary to capture the 99% of the CO₂ in the flue gas from a gas turbine that comes from the burnt fuel, leaving only the 1% that came from the air going into it.  But this study emphasised that the solvent flow through the absorber is also critical: too low and it is insufficient to capture the CO₂, too high and the CO₂ is captured but the rich loading will be lower than the optimum value.

Work is now ongoing to test these theoretical findings on the amine capture pilot plant at the University of Sheffield’s Translational Energy Research Centre (www.terc.ac.uk), shown below, at a scale of around 1 tonne of CO₂ capture per day.

Project outputs

Read More on Jon’s Flexible funding 2020 project page

We’re excited to share a blog report from one of our Flexible Funding 2020 projects, Salman Masoudi Soltani’s Biomass combustion ash in carbon capture. Here’s what the team have to say about their research:

“Bioenergy with carbon capture and storage (BECCS) is a net-negative technology that is recognised by many as a prominent tool in the uphill battle against climate change. However, combustion of biomass is associated with production of large quantities of problematic waste residue, Biomass Combustion Ash (BCA). Coal ash, for example, can find secondary applications within industry as an additive to concrete, whereas BCA has a higher amount of alkali/alkali-earth metals, which render such use inapplicable. However, in our research group we endeavour to produce solid adsorbents from this abundant waste source to be used as the CO₂ capturing media, thus, leading to a potential route for in-situ decarbonisation of the power sector via waste valorisation.

Industrial-grade BCA was first collected from a UK power plant and then characterised to identify the most appropriate route and/or material for CO₂ capture. The fly ash, due to the high content of aluminium and silica has been proposed as a precursor for zeolite production. Indeed, the zeolitic sorbent obtained via a fusion-hydrothermal synthesis route has was developed.

The bottom ash, however, is a material rich in unburnt carbon, that we have extracted via a simple and industrially mature technique. The produced carbon has a promising CO₂ uptake, that can be further enhanced via physical and chemical activation methods.

Figure 1: The produced BCA-derived sorbents

The results obtained from this investigation have been published as part of the IEEE 2021 International Conference on Nanotechnology for which our research group has received the best paper/presentation award. Furthermore, posters created by the RAs have been featured at the Doctoral Research Poster Conference at Brunel University London, where they have also been recognised with the main prize of the fair.

Additionally, as part of this funding call, we were able to arrange a site visit to Drax Power Plant in Selby, UK – the largest biomass combustion power plant in the world providing 6% of the country’s electricity supply. There we had an extensive tour around the facilities gathering insight on both, their biomass combustion ash waste management techniques as well as their carbon capture pilot units. We would like to extend our thanks to Dr Jeni Reeve and Dr Ben Dooley for their continued support throughout this research project as well as other members of Drax Group UK for their help along the way.”

Figure 2: Drax power plant site visit. From left to right: Mike Gorbounov (RA), Ben Petrovic (RA), Dr Salman Masoudi Soltani (PI).

Thanks to Salman and team for the report! See more on the project here.

Today we’re sharing a diary-style blog report from Dr Andrew Cavanagh (University of Edinburgh), on ‘Storage Capacity and Reserve Provision – The gigatonne challenge embedded in European net zero’, as part of Prof Stuart Haszeldine’s outputs for his Flexible Funding 2020 project, SCARP North Sea: Storage Challenge and Reserve Provision, a fundamental gigatonne CO₂ challenge for European net zero.

“February 2021: I just found out this is a requirement of the grant agreement. My first blog in years. Project has been reshaped by the pandemic as a desktop study. It’s one of those rare opportunities to dive deep into North American continental geology and look for synergies and parallels between two emergent 21^st century industries: shale gas fracking and CO₂ storage. Enjoying the background reading a lot.

March 2021: Working through the Ellsworth summary of earthquakes in Oklahoma. It’s a brilliant summary, covering the background of induced seismicity studies in the USA and this unintended experiment in triggered earthquakes. Initial impression is that the scale of land use and fluid disposal holds up as an analogue. Maybe. Impressive that the largest American modern earthquake east of the Rockies was an induced event in Oklahoma! Magnitude 6 no less.

March 2021: Interesting similarity in the background seismicity for the North Sea gigatonne storage areas and Oklahoma before the fracking. Both in terms of magnitude and frequency, both regions appear to have a similar profile of occasional earthquakes that barely register as ‘events’ in the public consciousness but fit on a Gutenberg-Richter trend as the infrequent upper-bound expression of more frequent smaller earthquakes that suggest a low seismicity area creaking and occasionally crackling in a field of intracratonic stresses.

May 2021: I’ve been busy writing up the gap work relating to net zero demand and immature CO₂ storage resources. It’s increasingly relevant to gigatonne storage as the European policy is indicative of a global supply squeeze that will likely focus 80% of storage delivery into the 2030s and 2040s. This effectively doubles demand on top of the doubling entailed in a shift from 2ºC to 1.5ºC as the acceptable limit. So, this will push storage rates to their technical limit. Reasonable to ask if that limit will be public perception and government regulation of induced seismicity.

August 2021: Mind the gap since last entry. Finished gap work and drafted for submission. The scale and rate of required storage globally and in the North Sea is truly extraordinary. To a first approximation, the only waste disposal industry that has ever approached this amount of subsurface injection is the fracking-associated produced water disposal in North America. This makes a proper understanding of the Oklahoma induced seismicity vital, as it is sure to draw lazy comparisons that play on the anxiety associated with all earthquakes. Question is, does the analogue comparison hold?

August 2021: Short answer is no.

September 2021: The more I dig into Oklahoma, the more of an anomaly it becomes. It is neither representative of seismic response in most States that have seen similar levels of fracking, nor does the analogue comparison with CO₂ storage in the North Sea hold up. The premise is beginning to unravel, but good to get there in case the comparison is made by others in the future. Questions arising regarding the pressure model.

October 2021: Now convinced that while the observed correlation (injection activity and earthquakes) is strong and clear, the inferred correlation (pressure change and fault failure) is unproven. While Oklahoma earthquakes were clearly caused by the injected wastewater, the highly cited papers include an inferred correlation for pressure as the ‘obvious’ causative mechanism. The pressure models are terrible. Appear to be poorly constructed, over-simplified and weak representations of the subsurface geology, based on poor data control. The visually convincing pressure blush underplays the extraordinarily small magnitude of the inferred (i.e. simulated) pressure changes. These are not measured or tested in any way. It’s just another superficially convincing pressure model that does not add up.

October 2021: So, the Oklahoma pressure models are pretty much unbelievable. Pressure changes of approximately 10 kPa are difficult to measure at a depth of a few kms without sensitive in-situ monitoring. There are no measurements, none, to back up the models. For context, the models infer pressure changes of less than 10 kPa/month, or 1/10^th of an atmosphere. It is hard to believe that the faults associated with the seismic swarm are being triggered by sub-10 kPa changes on a monthly timescale.

November 2021: The narrative of the write-up has shifted towards explaining the non-representative behaviour of Oklahoma with respect to fracked shale plays in North America and its unlikely suitability as an analogue for the North Sea and CO₂ storage with respect to seismic hazard, though a number of lessons can be learnt. Firstly, the media and the public respond strongly to the emotive nature of ‘earthquakes’. Magnitude, frequency and impact all play a role, though any earthquake will do in a pinch on a slow news day. Public understanding and media responsibility remain key to stakeholder perception of the significance of induced seismicity associated with subsurface industries.

December 2021: Added a brief section on the commonplace poor practise of forward modelling pressure to support edge-case scenarios that confirm bias despite a lack of supporting observations. There has been a long history of this in CO₂ storage. The relevant papers are almost always poorly referenced and thinly veiled attacks on the feasibility of storage. Calling this the ‘pressure Gruffalo’.

December 2021: Report finished. Paper drafting and poster to follow. Last blog entry.”

The team have also shared several photos of a deserted Edinburgh as part of their fieldwork during the lockdowns:

Read More on Stuart’s flexible funding 2020 project page

We’re pleased to share a blog report below from Stavros Michailos, Mohammed S. Ismail and Lin Ma of Energy 2050 at the University of Sheffield, on Lin’s Flexible Funding 2020 project, Evaluation of different CCUS systems based on the MCFC technology for decarbonising the power generation sector:

Carbon capture utilisation and storage (CCUS) is pivotal for delivering net-zero GHG emissions and strategically significant to the UK economy. The study details the performance of three CCUS systems based on MCFC via multiscale modelling, i.e. multiphysics modelling of the MCFC unit and process modelling. A distinct advantage of the MCFC over other CO₂ capture technologies is that, instead of absorbing energy (electricity), it generates it. Mass and energy balances have been established in COMSOL Multiphysics® and Aspen Plus software. Based on the simulations, a levelised cost analysis has been undertaken to appraise the capture cost of the investigated technologies. The MCFC concentrates CO₂ that can be stored or further utilised. Figure 1 is a visual representation of the scenarios examined in the study by means of simplified block flow diagrams (BFD).

Figure 1. Block flow diagrams of the investigated CCUS scenarios

Multiphysics model of MCFC

A two-dimensional MCFC multiphysics model has been created to inform the process model of the entire system. Figures 2 displays the profiles at a typical cell potential (0.7 V) for the temperature, CO₂ concentration along the cathode channel and H₂ concentration along the anode channel. The temperature is almost uniform across the various components of the fuel cell with slightly increased temperatures at the anode side of the fuel cell (Figure 1a). Figure 1b shows that the concentration of CO₂ (mol/m³) expectedly decreases from the inlet to the outlet, with very low amount of CO₂ at the cathode electrodes where the cathodic half reaction takes place. This observation also applies to H₂ concentration in the anode side of the fuel cell (Figure 1c).

Figure 3 shows that the cell voltage and power density increase with increasing current density after 3500 A/m² (Figure 3(a-b)); this is attributed to the availability of amount of CO₂ that is higher than that what is needed for the half reaction at the anode electrode. However, before 3500 A/m², the fuel cell performance appears to slightly improve with decreasing CO₂ concentration; this seems to be due to availability of amount of oxygen that is substantially higher than what is needed to complete the half reaction at the cathode electrode. Evidently, the carbon capture factor, which is the ratio between the flow rate of CO₂ concentrated at the anode and the flow rate of CO₂ available in the flue gas, substantially decreases with increasing CO₂ concentration in the flue gas.

Process modelling

Aspen Plus has been utilised to model the electrolyser and the Fischer-Tropsch (FT) plant. In particular, for the FT plant, CO₂ and H₂ are fed to a reverse water gas shift reactor to convert CO₂ to CO. The produced syngas is sent to the FT reactor, which generates a mixture of light gases, syncrude oil and water. After proper separation a part of the unreacted gases and light hydrocarbons are sent to a CHP unit modelled as combined cycle turbine and the rest is recycled to the FT reactor. The syncrude oil is fractionated to naphtha, kerosene and diesel cuts. The CO₂ that is not utilised is sent for compression and permanent storage. Figure 4 presents the basic mass balances of the Scenarios 2 and 3.

Figure 4. Mass balances of the investigated scenarios

Conclusions and future work

The study revealed that MCFC is an interesting CO₂ capture technology with prospects to play a role in the mid-term for decarbonising scenarios. The stand-alone case, i.e. scenario 1, with certain technology improvements can raise the competitiveness of the MCFC. The multiphysics model of the MCFC has been built and used to inform the process model on the amount of CO₂ concentrated at the anode of the MCFC at a typical operating cell potential. The parametric study shows that the MCFC performance in terms of power density improves with decreasing temperature, increasing CO₂ concentration in the flue gas and increasing H₂ concentration in the fuel mixture. Conversely, the CO₂ capture factor was found to increase with increasing operating temperature, decreasing CO₂ concentration in the flue gas and decreasing H₂ concentration in the fuel mixture. In the future, experimental testing of the MCFC in the Translational Energy Research Centre (TERC) will enable the improvement of the models developed in the current project.

Read more on Lin’s flexible funding 2020 project page

Related to Prof Yong Yan’s recent blog report on his Flexible Funded 2020 project, we’re pleased to share more about the paper that has been published as a result.

Professor Yong Yan from the University of Kent and his collaborators from North China Electric Power University have just published a research paper entitled “A novel heterogeneous ensemble approach to variable selection for gas-liquid two-phase CO₂ flow metering”. It’s been published in International Journal of Greenhouse Gas Control (vol.110, pp.103418, 2021).

Variable selection is an important pre-processing step in the development of effective data-driven models for CO₂ flow measurement in CCS systems. In order to effectively quantify the importance of potential input variables to the desired output, ensemble learning is proposed and incorporated into variable selection methodology. This paper presents a tree-based heterogeneous ensemble approach to variable selection and its application to gas-liquid two-phase CO₂ flow measurement using a Coriolis mass flowmeter and other low-cost sensors. The importance of each variable is determined through combining the importance scores from four tree-based algorithms, including decision tree regression, bootstrap aggregating of regression trees, gradient boosting decision tree and gradient boosting random forest. Then the backward elimination algorithm is applied to remove the relatively less important variables and hence a small set of input variables for data-driven models. The selection results demonstrate that the significant variables for CO₂ mass flow measurement include apparent mass flow rate, time shift, differential pressure and pressure drop while observed density, density drop, observed flow velocity and outlet temperature for prediction of gas volume fraction. To assess the validity of the selected variables, data-driven models based on gradient boosting random forest are developed. Results suggest that the relative error of the model output is mostly within 1% for CO₂ mass flowrate measurement and 5% for gas volume fraction prediction by taking the selected variables as model inputs.

A copy of the full paper is available on https://kar.kent.ac.uk/89452/.

Prof Yong Yan, University of Kent, shares his report blog as an output from his project report for Monitoring of CO₂ flow under CCS conditions through multi-modal sensing and machine learning, which was funded under our Flexible Funding 2020 call:

“Following the approval of my application for the UKCCSRC Flexible Funding 2020, I conducted the proposed work in collaboration with North China Electric Power University (NCEPU) and the industrial partners from 1 May 2020 to 31 August 2021. The following is a report on the activities that were undertaken during the project. Accurate measurement and real-time monitoring of CO₂ flow across the entire CCS chain are essential to gain an in-depth understanding of the physical and chemical characteristics of CO₂ flow in pipelines. Advances in these areas will also ensure the success of large-scale injection projects and enable the long-term monitoring of captured CO₂ in storage sites. However, due to the significant changes in CO₂ properties and impurities in composition, single-sensor technologies such as those based on Coriolis, ultrasonic, differential pressure (DP), acoustic emission (AE) or any other principle alone are not sufficient to provide accurate and robust monitoring across the range of CCS conditions particularly under two-phase gas-liquid flow and transient flow conditions. This research project aims to implement and assess the performance and operability of the multi-modal sensing and machine learning techniques for the monitoring of CO₂ under CCS conditions on a dedicated CO₂ test facility.

Experimental Setup
Experimental work was conducted on a CO₂ two-phase flow test rig at NCEPU. The first part of the experimental work was to measure the mass flow rate of two-phase CO₂. Several sensors and flow instruments, including a Coriolis mass flowmeter (CMF, model Optimass 6400 supplied by Krohne Ltd), a DP transducer that was installed across the Coriolis flowmeter and a capacitive sensor were installed on the test section of the rig, as shown in Figure 1. Various two-phase flow conditions were created under both steady-state and transient flow conditions. For experiments under steady-state flow conditions, the mass flowrate of liquid CO₂ is set from 300 kg/h to 3050 kg/h, resulting in the reference gas volume fraction (GVF) from 0% to 87%. Experiments under variable load, start-up and shutdown conditions were conducted to investigate the real-time performance of the instruments and established models. The mass flowrate of gas (liquid) phase CO₂ was fixed at 70 kg/h (1500 kg/h) while that of liquid (gas) phase CO₂ increased and decreased, resulting in variations in GVF.

Figure 1. Experimental setup for the mass flowrate measurement of two-phase CO₂ flow

The second part of the experimental programme was the detection and monitoring of the leakage from CO₂ transportation pipelines under CCS conditions. The experimental setup is shown in Figure 2. Pure CO₂ and CO₂-N₂ mixture in the pipeline were set to leak into the atmosphere by passing the medium through a vent valve. Several tests including different initial conditions were conducted to investigate the release behaviours of CO₂ and CO₂-N₂ mixtures under CCS conditions. AE sensors and temperature transducers are used to acquire signals for the leakage detection and leak rate measurement. Several researchers from NCEPU supported the experimental programme.

Figure 2. Experimental setup for the leakage monitoring of CO₂

Performance of Data Driven Models for Mass Flow Metering of Two-Phase CO₂
Mass flow rate, density, capacitance, pressure and temperature data of two-phase CO₂ flow were recorded in real time during the tests. Several data driven models based on recurrent neural networks (RNNs) and least squares support vector machines (LS-SVM) were developed to measure the mass flowrate using the data recorded. The relative errors in the predicated mass flow rate are found mostly within ±1.5% under steady-state flow conditions. Figure 3 shows the performance of the data driven model based on RNN under transient flow conditions. The RNN model is capable of tracking the change in CO₂ mass flowrate under dynamic CCS conditions with a relative error within ±4%.

(a) Comparison of mass flowrates

(b) Comparison of relative errors

Figure 3. Comparison of mass flowrates of CO₂ and their relative errors under transient flow conditions

In addition to the data analysis, a study of variable selection for the development of data driven models to improve the measurement accuracy has been undertaken. A gradient boosting random forest (GBRF) model is applied to variable selection for gas-liquid CO₂ two-phase flow measurement. The relative errors are within ±1% with the strategically selected variables for the data driven models.

Performance of Data Driven Models for the Leakage Detection of CO₂
LS-SVM models coupled with AE sensors and temperature transducers have been developed to measure the leakage rate of CO₂ flow from a pipeline and its volume fraction of N₂ in the leakage. In a practical leakage, highly pressurized CO₂ is released rapidly through the leak hole, which generates AE signals. In order to quantify the leakage, several features are extracted from the raw AE signals such as peak value, peak frequency and AE energy. The temperature transducers measure the temperature drop signals at the leak hole, which is highly correlated with the volume fraction of impurity. The features of the AE signals and the temperature drop are inputs to the LS-LVM model.

Concluding Remarks
The complex flow conditions and significant changes in phase of CO₂ pose challenges in the measurement and monitoring of CO₂ flow in CCS chains. In this project, we have used a combination of a Coriolis mass flowmeter, a DP transducer and a capacitive sensor with data driven models to measure the CO₂ flow rate under both steady-state and transient flow conditions. Several AE sensors are utilized to detect and measure the leakage of CO₂ flow a transportation pipeline. The experimental data is still being processed at the time of writing. Variable selection results have recently been published in a leading journal (see a separate blog on UKCCSRC web). The results for CO₂ leakage detection and monitoring will be published in the near future.”

See more on Yong’s Flexible Funded 2020 project here, and find the published paper here.

CO₂-FROST: CO₂ frost formation during cryogenic carbon capture with tomography analysis is a research project funded as part of the UKCCSRC flexible funding 2020 programme. Here, David Cann (University of Chester), Yuan Chen (University of Edinburgh), Dr Carolina Font Palma (University of Hull) and Dr Jiabin Jia (University of Edinburgh) share their report on their research.   

Novel carbon capture technologies are increasingly needed to reach the target of net zero carbon emissions. Our project studies a cryogenic carbon capture (CCC) process based on physical separation by freezing out the CO₂ that avoids dealing with hazardous chemicals, high CO₂ removal levels and delivers high purity CO₂, which could be further used in the food industry or to produce chemicals.

To better understand cryogenic CO_s separation, we set the task of conducting an experimental programme at the University of Chester to determine whether tomography could be used to detect CO₂ frost inside a packed bed column.

The University of Chester’s cryogenic carbon capture rig uses cold bed material to sufficiently cool a gas mixture of CO₂ and nitrogen that is fed into the capture column. This unit was built to support fundamental research to develop the patented advanced cryogenic carbon capture (A3C) process from PMW Technology Ltd. The capture column was constructed using a PTFE pipe. Since it is not possible to view the inside of the capture column, it is difficult to determine exactly how and where frost is forming inside the capture column. Temperature profiles from thermocouples inside the capture column can be used to determine whether frost is present in the bed material but cannot provide any indication of frost thickness, regions of high frost coverage and low frost coverage etc.

For the first time, Electrical Capacitance Tomography (ECT) is applied to determine the formation of frost inside the capture column. The ECT sensor uses electrical capacitance to reconstruct images to represent high and low relative permittivity. Using an ECT sensor around the outside of the capture column provides a non-intrusive way of measuring frost formation, as the low relative permittivity of CO₂ frost in comparison to the ceramic bed material that the capture column uses should mean that the deposition of CO₂ frost onto the bed material surface would reduce the relative permittivity of the capture column where frost is forming.

Fig 1. Yuan Chen from the University of Edinburgh and David Cann from the University of Chester when working at Thornton Science Park.

First experimental campaign

The first experiments focused on the set up and calibration of the ECT sensor around the carbon capture column. Once our university allowed business travel, Yuan Chen arrived at the University of Chester with the ECT sensor, which was installed around the capture column and calibrated for empty column and ceramic bed filled conditions. The ECT sensor used high and low calibration values for relative permittivity in order to create reconstructed images.

In order to graphically represent the desublimation of CO₂ frost on ceramic bed material, we set the cold bed material as the high calibration and the frosted bed material as the low calibration. As CO₂ desublimes and covers the ceramic bed material with frost, the overall relative permittivity inside the capture column reduces.

The cryogenic experiments cause water frost to accumulate on the outside of the capture column when the capture column gets cold. The defrosting of the water frost overnight caused some issues with water interfering with the ECT sensor. Thus, the sensor was removed from the capture column to be dried and reattached.

Fig 2. Experimental setup of cryogenic column with ECT sensor

As with any experimental work, we faced several challenges. Due to deliveries of liquid nitrogen for experimental runs and the need for the capture column to be left to defrost overnight, there are a limited number of experimental runs that can be attempted each week.

The calibration points set for the ECT sensor to graphically reconstruct images to determine frost formation were very close to each other. We noticed that the difference in relative permittivity due to the desublimation of CO₂ was insufficient for the ECT sensor to reliably detect CO₂ frost.  The relative permittivity of the ceramic bed material was too large, estimated to be within the range of 25-30, whereas the relative permittivity of CO₂ frost is 1.6. As a result, we realised that we would need to redesign experiments.

Fig 3. Online progress meeting of CO₂-FROST team

Second experimental campaign

After the second lockdown, we managed to carry out more experiments. Instead of measuring the desublimation of CO₂ frost in the capture column, we decided that we would monitor water frost formation. Thus, the bed material was wetted with water before being added to the capture column. The relative permittivity of water is very high at approximately 80, but the change in relative permittivity from water to ice would be sufficient for the ECT sensor to be able to detect the freezing of water inside the column. The freezing of water as cold nitrogen gas is fed into the column would indicate the flow regime of the gas phase within the capture column, to determine whether there is uniform flow or if there are dead flow regions within the bed material close to the gas feed pipe.

The ECT sensor was able to give much clearer results from these water frost experiments. Overall, the flow of gas through the column appears to be fairly uniform as the colour change from red to blue for the ECT sensor reconstructed images is steady throughout the cross section of the column with the exception of some spots close to the column wall. It is most likely that these areas close to the column wall were spots of the column that had no water presence prior to the introduction of cooling gas, meaning that from the high calibration point of wetted bed material there was no phase change to be detected.

Fig 4. Reconstructed images for wetted ceramic double end calibration method. Each picture represents one-minute intervals

The project presented interesting practical challenges; including how to accurately take calibration readings for CO₂ frosted bed material, how to determine when the bed was frosted, as well as how to evenly wet the bed material without flooding the column. In the future, we recommend testing CO₂ frost formation using other bed materials. Our encouraging results will be presented at a conference shortly.

Congratulations to Salman Masoudi Soltani, whose research paper for his UKCCSRC 2020 flexible funding project has just been published!

Salman’s paper, Influence of surface modification on selective CO₂ adsorption: A technical review on mechanisms and methods, highlights some key issues and discussion point around adsorbents, and includes:

• a critical review of surface modification methods employed to enhance CO₂ adsorption
• assessment of CO₂ adsorption mechanisms in the context of surface functional groups
• evaluation of the methods used to introduce functional groups on adsorbent surfaces
• a critical discussion of the reagents used in surface modification of adsorbents.

To date, a number of separation technologies in carbon capture and storage have been explored, including physical absorption, chemical absorption, cryogenics, oxyfuel combustion, membranes and adsorption. Among these many available CCS technologies, capturing the CO₂ on solid sorbents has been gaining traction in recent years given its cost-effectiveness as a result of its ease of application, relatively small energy requirements and applicability in a wide range of processes.

However, even with the myriad materials such as zeolites, carbons, metal organic frameworks, mesoporous silicas and polymers, the challenge to identify a sorbent with optimal capacity, kinetics, selectivity, stability and ultimately, viability, still persists. By tailoring these solid materials through comprehensive campaigns of surface modification, the pitfalls of each can be mollified and the strengths enhanced. This highly specific tailoring must be well informed so as to understand the mechanisms by which the CO₂ is adsorbed, the surface chemistry that has influence on this process, and what methods exist to facilitate the improvement of this.

Salman’s research endeavours to identify the surface functional groups that interact with the CO₂ molecules during adsorption and the methods by which these functional groups can be introduced. It also provides a comprehensive review of the recent attempts and advancements made within the scientific community in the experimental applications of such methods to enhance CO₂ capture via adsorption processes.

• Read the paper, Influence of surface modification on selective CO₂ adsorption: A technical review on mechanisms and methods, here
• See Salman’s PhD student Mikhail Gorbounov present some of these results as a poster in our ECR Takeover on 19th November 2020 (with abstracts here) as part of our Autumn Web series

The UK Carbon Capture and Storage Research Community (UKCCSRC) is supported by the Engineering and Physical Sciences Research Council (EPSRC) as part of the UK Research and Innovation (UKRI) Energy Programme.

The mission of the UKCCSRC is to ensure that carbon capture and storage (CCS) plays an effective role in helping the UK achieve net-zero greenhouse gas emissions by 2050.

• Our People
• Our Partners
• Members Directory
• Jobs and Opportunities
• Systems, policy and public perception
• Disclaimer and Terms of Use
• Cookie Policy

Location

D03 Ella Armitage Building
University of Sheffield
40 Leavygreave Road
Sheffield
S3 7RD
UK

Social

• 

Copyright UKCCSRC - 2012 / 2026