10. Field Demonstration of Advanced CDRMax Solvent at the USDOE’s National Carbon Capture Centre and the CO₂ Technology Centre Mongstad DA, Norway (2016)

Prateek Bumba*, Dr Avinash Patkar. P.E, a, Richard Mathera, Ramesh Kumara, Dr. James Halla Frank Mortonb, and Justin Anthonyb

^a Carbon Clean Solutions Limited, 47, Castle Street, Reading, Berkshire, RG1 7SR, United Kingdom ^b Southern Company, National Carbon Capture Center (NCCC), Wilsonville, AL, USA

1876-6102 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of GHGT-13.
doi: 10.1016/j.egypro.2017.03.1261

Original paper available here (doi.org)

Abstract

Carbon capture solvents should have a low Specific Reboiler Duty, resistance to oxidative degradation, should not be corrosive and should have low emissions to atmosphere. Carbon Clean Solutions Ltd have developed the CDRMax solvent for use on power plant flue gases. The solvent has been tested at both the National Carbon Capture Centre, USA and the Technology Centre Mongstad, Norway. At both centers the SRD, emissions to atmosphere and concentrations of metals in the solvent were measured. The results show that CDRMax was less corrosive and had a lower SRD than 30wt% MEA, had low emissions and was resistant to oxidative degradation.

1. Introduction

Carbon Clean Solutions Ltd is a company based in Reading, United Kingdom. The company offers its range of innovative solvents for gas treatment. This paper describes testing of the CDRMax solvent on a flue gas containing ~ 4 mole% CO₂. This is the concentration of CO₂ which occurs in flue gas from a gas-fired power plant. The solvent was tested during two campaigns. The first campaign was on the Pilot Scale Test Unit at the National Carbon Capture Centre, Alabama, USA from March 2014 to April 2014. The second test campaign was on the amine plant at the Technology Centre Mongstad, Norway from November 2015 until January 2016. During both campaigns the Specific Reboiler Duty (SRD) emissions to atmosphere and corrosion were measured.

2. NCCC Campaign

2.1 Overview of the NCCC facility

The Pilot Scale Test Unit (PSTU) at the NCCC facility is a post combustion capture test facility. It can operate on coal flue gas or a simulated natural gas flue gas. The flue gas is taken as a slipstream from an 800 MW coal fired boiler. Before entering the test unit, the flue gas is treated by SCR, ESP and FGD to reduce the concentration of NO_x, fly ash and SO_x. If required, the flue gas is diluted with air to achieve gas fired flue gas conditions. A final polishing NaOH wash reduces the concentration of SO_x still further and saturates the flue gas.

The flue gas and lean solvent flow rates can be adjusted so that the liquid to gas ratio can be varied. The absorber can be configured so that it can be operated with 12m or 18m height of packing. The dimensions of the equipment at the PSTU are shown in Table 1 below.

Column	Height, m	Outer diameter m	Number of Beds	Bed height m	Packing type
Pre-Scrubber	14	0.76	1	6.1	Mellapakplus M202Y
Cooler/Condenser	9.1	0.61	1	3.05	Mellapakplus M252Y
Absorber	32.9	0.66	3	6.1	Mellapakplus M252Y
Wash Tower	9.1	0.61	1	3.05	Mellapakplus M252Y
Regenerator	23	0.61	2	6.1	Mellapakplus M252Y

Table 1. The dimensions of equipment in the PSTU.

2.2 Overview of the NCCC test campaign

A series of parametric trials were conducted on the PSTU plant with the CDRMax solvent. These allowed the SRD to be measured under a range of conditions. The solvent was tested on both a coal flue gas and a simulated natural gas flue gas. However, only the simulated natural gas flue gas results are provided here. The ranges of the test conditions are provided in Table 2 below. During each test phase only one parameter was changed at a time to ensure that the impact of each change could be isolated. Laboratory analysis was carried out on solvent samples to determine water content, lean and rich loading as well as the concentration of metals.

Table 2. Operating conditions for the NCCC campaign.

Table 3 below shows the composition of the flue gas used in the trials at the NCCC.

Table 3. Flue gas composition for the NCCC campaign on a dry basis.

2.4 Optimization of SRD by varying the liquid to gas ratio

A series of tests were carried out to determine the liquid to gas ratio that would provide the lowest SRD. For these tests a constant flue gas rate of 3,630 kg/hr was maintained, the flue gas temperature was 40 OC and the stripper pressure was 0.7 bar(g). The solvent flow was set to 2,000 kg/hr for the first test then increased to 2,722 kg/hr for the second test before finally being increased to 3,085 kg/hr for the third test. This gave liquid to gas ratios of 0.65, 0.75 and 0.85 kg of solvent per kg of flue gas respectively. For each test condition the steam to the reboiler was adjusted until the CO₂ capture rate was 90%. As can be seen in Figure 1 below the optimal L/G was found to be close to 0.75 kg/kg.

Figure 1. Optimisation of SRD for CDRMax at NCCC by varying the liquid to gas ratio while maintaining a constant stripper pressure of 0.7 bar(g).

2.5 Optimization of SRD by varying the stripper pressure

To determine the optimal stripper pressure, a series of energy measurements were made while the liquid to gas ratio was held at the optimum of 0.75 kg/kg. The stripper pressure was varied from 0.8 to 1.54 bar(g) and at each condition the steam flow to the reboiler was adjusted until the CO₂ capture rate was 90%. As shown in Figure 2 below there is an optimal stripper pressure close to 1.0 bar(g) where the SRD was 3.1 MJ/kg CO₂ . For the 1.0 bar(g) condition the actual CO₂ capture rate was above 90%, therefore, the SRD would be lower for a capture rate of 90%. In addition, it was noted that the cross-over lean rich heat exchanger was not optimal for the CDRMax solvent. CCS Ltd simulated the SRD for the same operating conditions but with an optimized cross-over heat exchanger and an advanced stripper configuration. The SRD was projected to be reduced to 2.8 MJ/kg CO₂.

Figure 2. Optimisation of SRD for CDRMax at NCCC by varying the stripper pressure at a constant liquid to gas ratio of 0.75 kg/kg.

The NCCC have reported an optimal SRD for 30 wt% MEA of 3.5 MJ/kg CO₂ at a CO₂ capture rate of 86% [1]. Therefore the optimal CDRMax value of 3.1 MJ/kg CO₂ at a capture rate of 90% provides an 11% reduction in the SRD. With design modifications, CCS Ltd estimate that the energy savings could be 20%.

Table 4. Comparison between 30 wt% MEA, CDRMax and projected CDRMax values if optimal cross over heat exchanger and stripper design is implemented.

2.5 Effect of absorber height

To determine the effect of the absorber packing height on SRD for 90% CO2 capture efficiency, two tests were carried out with a packing height of 18m and 12m. The flue gas rate, L/G and the flue gas temperature and stripper pressure were maintained constant for both conditions while the steam flow to the reboiler was adjusted to achieve a target CO2 capture rate of 90%. There was a reduction in the SRD of 5.4% when the packing height was increased from 12m to 18m.

2.6 Emissions of ammonia

Ammonia emissions are an indicator of solvent stability against the oxidative degradation. The oxygen content in the flue gas was 15.9 mole% on dry basis. During the NCCC campaign three tests of ammonia emissions through the flue gas outlet from the water wash were made and average concentration was found to be 3.2 ppmv.

2.7 Corrosion

During a campaign using 30 wt% MEA, the NCCC had reported chromium concentrations in the solvent [2]. After 300 hours of operation the chromium concentration was 45,090 µg/L. For the CDRMax solvent, after 500 hours of operation the chromium concentration was 2,120 µg/L. Even though the CDRMax campaign was longer, the concentration of chromium was 20 times lower than for the MEA campaign.

Table 5. Concentration of cations in 30wt% MEA and CDRMax reported in µg/L.

3. TCM campaign

3.1 Overview of the TCM test facility

The Technology Centre Mongstad is one of the largest post combustion capture test facilities in the world. It is highly instrumented to give a detailed understanding of the process conditions which occur within each section of the plant. The facility can operate with flue gas rates of 60,000 Sm3/hr and with flue gases from different sources. Figure 3 below shows the Process Flow Diagram for the plant. Flue gas from a Combined Heat and Power plant can be operated on its own. The flue gas is derived from natural gas combustion so is equivalent for CCGT flue gas.

Product CO2 from the stripper can be recycled into the CHP flue gas to increase the CO2 content of the inlet flue gas to 13 mole%, this provides a synthesised coal or cracker flue gas but without aerosols. Controlled amounts of cracker gas (RFCC) can be added so the effect of aerosols on emissions can be observed. During the test campaign on CDRMax all three modes of operation were used. However, only the results obtained on CHP flue gas are reported here.

The incoming flue gas is cooled to a desired temperature and saturated in a direct contact cooler. In this vessel the gases are contacted with a recirculating stream of brackish water. The absorber can be operated with up-to 24 m height of packing. Above the absorber is a two stage water wash. Make-up water is added to the higher bed which then cascades into the lower bed before being purged into the absorber. The stripper has 8 m of packing and is heated by a plate reboiler. The details of the equipment are provided in Table 6 below.

Table 6. Dimensions of equipment at the TCM.

3.2 Overview of the TCM campaign

The objective of the test campaign was to demonstrate the CDRMax solvent to determine the SRD, emissions to atmosphere, solvent degradation and corrosion of the plant. The test campaign started on the 17th November 2015 and concluded on the 6^th January 2016. The CDRMax solvent was operated on a flue gas containing 3.7 mole% CO₂ (dry basis). The SRD was measured at different process conditions which are shown in Table 7 below. The composition of the CHP flue gas on a dry basis is provided in Table 8 below.

Throughout the campaign, the emissions of solvent and ammonia were measured in the depleted flue gas using a Proton Transfer Time of Flight Mass Spectrometer (PTR-TOF-MS) and periodically by iso-kinetic sampling.

Further details of the emissions monitoring equipment can be found in [3].

Table 7. Operating conditions for the TCM campaign.

Table 8. Flue gas composition during TCM campaign on a dry.

3.3 Energy required for the regeneration of solvent

3.3.1 Comparison between 30wt% MEA and CDRMax

The TCM have previously reported test results for 30 wt% MEA [4]. The paper describes that streaming was observed in the stripper column. This was corrected by the addition of anti-foam. It was found that after the addition of anti-foam the SRD was reduced. During this test, the flue gas rate was 47,000 Sm³/hr, the flue gas temperature was 25 ^OC, lean solvent temperature was 25 ^OC, the CO₂ capture rate was 85 %, stripper pressure 0.9 bar(g) and 24 m of absorber packing were used. The liquid to gas ratio was optimized to provide the lowest SRD value. CCS Ltd replicated this test with the CDRMax solvent to provide a comparison.

Figure 4 below shows results of the testing. As can be seen, the 30 wt% MEA with no anti-foam the SRD was 4.0 MJ/kg CO₂. Following the addition of anti-foam, for 30 wt% MEA the SRD was reduced to 3.7 MJ/kg CO₂. For CDRMax, the SRD was 3.3 MJ/kg CO2. The CDRMax solvent showed a 16.7 % energy reduction compared to 30 wt% MEA without anti-foam and a 10% reduction from 30 wt% with anti-foam. If the CCS Ltd patented stripper configuration were incorporated into the design then the SRD would be reduced further still.

Figure 4. Comparison of SRD between 30 wt% MEA (with and without anti-foam) and CDRMax solvents.

3.3.2 Optimization of SRD

Parametric tests were carried out to optimize the specific reboiler duty for the CDRMax solvent. The inlet flue gas and lean solvent temperatures were held constant at 35 ^OC. The flue gas rate was 56,000 Sm³/hr and the CO₂ capture rate was 90 %. The CDRMax solvent was operated at three different stripper pressures of 0.8, 1.0 and 1.2 bar(g). At each pressure the liquid to gas ratio was varied and the steam rate to the reboiler adjusted until the target CO₂ capture rate was achieved. The lowest SRD found for each stripper pressure is plotted in Figure 5 below. As can be seen the lowest recorded SRD of 3.25 MJ/kg CO₂ was found at 1.0 bar(g). The temperature of gas inlet the absorber and lean solvent temperature are varied to find the best fit of operating conditions for the CDRMax keeping the stripper pressure at 1 bar(g) the corresponding liquid to gas ratio.

Figure 5. Optimisation of stripper pressure for CDRMax solvent.

3.4 Emissions

3.4.1 Emissions of solvent in the depleted flue gas

The emissions of solvent were monitored in the depleted flue gas throughout the campaign using a PTR-TOF-MS instrument. The results are shown in Figure 6 below and are reported on a 24 hour average basis. The permitted level of emissions was 6 ppmv which is shown by the dashed line. As can be seen, the emissions remained below 1 ppmv for the duration of the campaign.

Figure 6. Emissions of solvent in the depleted flue gas during the CDRMax campaign. Values are 24 hour average and measured by PTR-TOF-MS. The dashed line is the emissions permit limit of 6 ppmv.

To verify the PTR-TOF-MS instrument values, periodic iso-kinetic samples of emissions were also taken. The condensate from the samples were analysed by LC-MS. The concentrations in the condensate were then used to calculate the concentration of the solvent in the gas phase. The results of the iso-kinetic sampling are shown in Table 9 below. As can be seen, the values are all below 1 ppmv which gives good agreement with the PTR-TOF-MS measurements.

Table 9. Emissions of solvent during the TCM campaign. Values determined by iso-kinetic sampling. Condensate analysed by LC-MS.

3.4.2 Emissions of ammonia in the depleted flue gas

The emissions of ammonia in the depleted flue gas were measured by the PTR-TOF-MS throughout the campaign. The results are shown in Figure 7 below. The emissions limit for ammonia was 33 ppmv and this is shown on the figure by a dashed line. As can be seen, the emissions achieved during the campaign were generally less than 2 ppmv. Emissions of ammonia are indicative of oxidative degradation of the solvent. The ammonia emissions are stable and low, this indicates that the CDRMax solvent is resistant to oxidative degradation.

Figure 7. Emissions of ammonia in the depleted flue gas during the CDRMax campaign. Values measured by PTR-TOF-MS and reported on a 24 hour average basis. The dashed line represents the emissions permit limit of 33 ppmv on a 24 hour average basis.

3.5 Operating hours

The cumulative operating hours achieved during the campaign are shown in Figure 8 below. As can be seen there were very few interruptions in the progress of the campaign. The interruptions that did occur were minor plant issues and none were associated with the solvent operation. The solvent on-stream factor over the campaign was 100% while the overall on-stream factor was 97%.

Figure 8. Cumulative operating hours during the campaign. The solvent operation on-stream factor was 100%.

3.6 Corrosion

The TCM stripper and process pipework is constructed from stainless steel and 22 % chromium Duplex steel. Figure 9 below shows the concentration of molybdenum, chromium and nickel in the CDRMax solvent during the campaign. All of these elements are constituents of stainless steel so their change in concentration is related to corrosion. The concentration of molybdenum, nickel and chromium all remained below 0.5 mg/L throughout the campaign. The chromium concentration remained at or below 0.1 mg/L at the end of the 1,200 hour campaign which was the limit of quantification. The result from the NCCC campaign was that the concentration of chromium was

2.1 mg/L after 500 hours of operation. The higher value observed at the NCCC over a shorter campaign might be explained by the fact that the absorber at the NCCC is fabricated from stainless steel while at the TCM the absorber is a polypropylene lined concrete structure. Therefore, the surface area of stainless steel is relatively smaller at the TCM.

Figure 9 Concentration of molybdenum, chromium and nickel in the CDRMax solvent during the TCM campaign.

3.7 Health, Safety and Environment

Before the campaign could be authorized it was necessary to be confident that the plant could operate within its environmental permit using the CDRMax solvent. CCS Ltd estimated what the emissions from the plant would be while operating on the CDRMax solvent. The effect of atmospheric conditions on solvent emissions was considered to estimate the degradation products which might form in the atmosphere. The dispersion and hence ground level concentrations of emissions and products of degradation were then estimated using simulation software which considered local climatic conditions. The analysis concluded that it would be possible to operate the CDRMax solvent and achieve the environmental emissions permit. Throughout the campaign there were no exceedances of the environmental permit so this conclusion was valid. A REACH format Safety Data Sheet for the CDRMax solvent was provided to inform the site of precautions to be taken when handling the solvent. There were no accidents or near misses during the test campaign.

4. Conclusions

At the NCCC and operating on flue gas containing 4.5 mole% CO₂, the lowest recorded SRD for CDRMax was 3.10 MJ/kg CO₂. For 30wt% MEA the equivalent value was 3.5 MJ/kg CO₂. This is an 11% reduction in the SRD. During the CDRMax trial the CO₂ capture rate was 92.5% while for 30 wt% MEA it was 86%. From simulation work, CCS Ltd projects that the SRD could be reduced to 2.8 MJ/kgCO2 for the CDRMax solvent if the process designed for CDRMax solvent.
At the TCM and operating on a CHP flue gas containing 3.7 mole% CO₂ test facility the CDRMax solvent was operated on equivalent conditions to 30wt% MEA. For 30wt% MEA the lowest SRD was 3.7 MJ/kg CO₂, for CDRMax the SRD was 3.3 MJ/kg CO₂. This is a 10% reduction in the SRD. An SRD of 3.1 MJ/kg CO2 was achieved for CDRMax at different operating conditions. The SRD can be further reduced by optimizing the cross-over exchanger and using the CDRMax solvent designed process configuration.
For the TCM campaign, ammonia emissions in the depleted flue gas remained significantly below the emission limit for the duration of the 1,200 hours of operation. Testing during the NCCC campaign showed ammonia emissions of 3.2 ppmv. This indicates very low levels of oxidative degradation.
For the TCM campaign, solvent emissions in the depleted flue gas remained below 1 ppmv which is significantly below the emission limit.
The concentration of chromium in the CDRMax solvent at the NCCC was 2.1 mg/L after 500 hours of operation. The NCCC had measured a chromium concentration of 45 mg/L for 30wt% MEA after 300 hours of operation. This is 20 times higher than observed for CDRMax. At the TCM the concentration of chromium in the CDRMax solvent after 1,200 hours of operation was 0.1 mg/L.
During the TCM campaign, the CDRMax solvent was operated for 1,200 hours and a solvent on-stream factor of 100% was achieved.
No breaches of the environmental permit occurred at the TCM test facility during testing of the CDRMax solvent.
No accidents or lost time incidents occurred at either test facility during testing of the CDRMax solvent.

5. Acknowledgements

Carbon Clean Solutions Ltd wishes to acknowledge the contribution of the following organizations to the test work in this paper. These are: the US Department of Energy and the National Carbon Capture Centre, the CO2 Technology Centre Mongstad, Norway and the United Kingdom’s Department of Industry, Energy and Industrial Strategy.

6. References

Morton, F. The National Carbon Capture Center: Cost-effective test bed for carbon capture R&D, GHGT 11, Energy Procedia, 2013
Wheeldon, J. 2013, NCCC Post Combustion CO₂ Capture Program, Presentation made at NETL CO₂ Capture Technology Meeting, July 8-11, 2013.
Morken, A.K, Emission results of amine plant operations for MEA testing at the CO₂ Technology Centre Mongstad, GHGT 12, Energy Procedia, 2014.
Brigman N, Results of amine plant operations from 30wt% and 40wt% aqueous MEA testing at the CO₂ Technology Centre Mongstad, GHGT 12, Energy Procedia, 2014.

10. Field Demonstration of Advanced CDRMax Solvent at the USDOE’s National Carbon Capture Centre and the CO2 Technology Centre Mongstad DA, Norway (2016)