Pilot plant study of two new solvents for post combustion carbon dioxide capture by reactive absorption and comparison to monoethanolamine

Application of new solvents will substantially contribute to the reduction of the energy demand for the post combustion capture of CO₂ from power plant flue gases. The present work describes tests of such new solvents in a gas-fired pilot plant, which comprises the complete absorption/desorption process (column diameters 0.125 m, absorber/desorber packing height 4.25/2.55 m, packing type: Sulzer BX 500, flue gas flow 30–100 kg/h, CO₂ partial pressure 35–135 mbar). Two new solvents CESAR1 (0.28 g/g 2-amino-2-methyl-1-propanol+0.17 g/g piperazine+0.55 g/g H₂O) and CESAR2 (0.32 g/g 1, 2-ethanediamine+0.68 g/g H₂O), which were developed in an EU-project, were systematically studied and compared to MEA (0.3 g/g monoethanolamine+0.7 g/g H₂O). The two new solvents and MEA were studied in the same way in the pilot plant and detailed results are reported for all solvents. In the present study the structured packing Sulzer BX 500 is used. The measurements are carried out at a constant CO₂ removal rate of 90% by an adjustment of the regeneration energy in the desorber for systematically varied solvent flow rates. An optimal solvent flow rate leading to a minimum energy requirement is found from these studies. Direct comparisons of such results can be misleading if there are differences in the kinetics of the different solvent systems. The influence of kinetic effects is experimentally studied by varying the flue gas flow rate at a constant ratio of solvent mass flow to flue gas mass flow and constant CO₂ removal rate. Results from these studies indicate similar kinetics for CESAR1, CESAR2 and MEA. The direct comparison of the pilot plant results for these solvents is therefore justified. Both CESAR1 and CESAR2 show improvements compared to MEA. The most promising is CESAR1 with a reduction of about 20% in the regeneration energy and 45% in the solvent flow rate.     

Comparison of solvent performance for CO2 capture from coal-derived flue gas: A pilot scale study

The performance of a proprietary solvent (CAER-B2), an amine-carbonate blend, for the absorption of CO₂ from coal-derived flue gas is evaluated and compared with state-of-the-art 30 wt% monoethanolamine (MEA) under similar experimental conditions in a 0.1 MWth pilot plant. The evaluation was done by comparing the carbon capture efficiency, the overall mass transfer rates, and the energy of regeneration of the solvents. For similar carbon loadings of the solvents in the scrubber, comparable mass transfer rates were obtained. The rich loading obtained for the blend was 0.50 mol CO₂/mol amine compared to 0.44 mol CO₂/mol amine for MEA. The energy of regeneration for the blend was about 10% lower than that of 30 wt% MEA. At optimum conditions, the blend shows promise in reducing the energy penalty associated with using industry standard, MEA, as a solvent for CO₂ capture.     

Results from trialling aqueous NH3 based post-combustion capture in a pilot plant at Munmorah power station: Absorption

Australia's Commonwealth Scientific and Industrial Research Organization (CSIRO) and Delta Electricity have developed, commissioned and operated an A$7 million aqueous NH₃ based post-combustion capture (PCC) pilot plant at the Munmorah black coal fired power station in Australia. The results from the pilot plant trials will be used to address the gap in know-how on application of aqueous NH₃ for post-combustion capture of CO₂ and other pollutants in the flue gas and explore the potential of the NH₃ process for application in the Australia power sector. This paper is one of a series of publications to report and discuss the experimental results obtained from the pilot plant trials and primarily focuses on the absorption section.The pilot plant trials have confirmed the technical feasibility of the NH₃ based capture process. CO₂ removal efficiency of more than 85% can be achieved even with low NH₃ content of up to 6 wt%. The NH₃ process is effective for SO₂ but not for NO in the flue gas. More than 95% of SO₂ in the flue gas is removed in the pre-treatment column using NH₃. The mass transfer coefficients for CO₂ in the absorber as functions of CO₂ loading and NH₃ concentration have been obtained based on pilot plant data.     

Reduction of amine mist emissions from a pilot-scale CO2 capture process using charged colloidal gas aphrons

In the CO₂ capture process from coal-derived flue gas where amine solvents are used, the flue gas can entrain small liquid droplets into the gas stream leading to emission of the amine solvent. The entrained drops, or mist, will lead to high solvent losses and cause decreased CO₂ capture performance. In order to reduce the emissions of the fine amine droplets from CO₂ absorber, a novel method using charged colloidal gas aphron (CGA) generated by an anionic surfactant was developed. The CGA absorption process for MEA emission reduction was optimized by investigating the surfactant concentration, stirring speed of the CGA generator, and capture temperature. The results show a significant reduction of MEA emissions of over 50% in the flue gas stream exiting the absorber column of a pilot scale CO₂ capture unit.     

CO2 capture solvent selection by combined absorption-desorption analysis

A method was developed for selection of promising solvents based on CO₂ absorption experiments at 40 °C and 9.5 kPa CO₂ partial pressure followed by desorption of the same solvents at 80 °C down to 1.0 kPa CO₂ partial pressure. Experiments conducted on 13 solvent systems under atmospheric conditions revealed the solvents absorption and desorption characteristics and these were compared with 1.0 M, 2.5 M, 5.0 M and 10.0 M MEA. Results showed that absorption or stripping data alone were not sufficient in making robust solvent selection decisions, and that combined data analysis was necessary. 1.0 M tetraethylenepentamine (TEPA) and 5.0 M MEA showed the best performance in terms of absorption rate. 1.5 M Bis-(3-dimethylaminopropyl) amine (TMBPA) was easy to desorb, has high absorption capacity; and when promoted it showed the best performance in terms of CO₂ carrying capacity. At the test conditions, 1.5 M TMBPA promoted with 1.0 M PZ showed the best potential for efficient CO₂ removal at reduced cost of all systems tested. Its cyclic capacity in mol CO₂/mol amine was found to be 70% higher than that of 5 M MEA.     

Aqueous piperazine derivatives for CO2 capture: Accurate screening by a wetted wall column

Concentrated aqueous piperazine (PZ) has been identified as a better solvent for CO₂ capture than monoethanolamine (MEA), because it has a greater rate of CO₂ absorption and greater CO₂ capacity. This work evaluates the effect of substitute groups on PZ performance. Many previous screening studies measured absorption/desorption with CO₂/N₂ sparging, which lacks accuracy and cannot be used to estimate actual absorber performance. In this work a wetted wall column was used to accurately measure absorption/desorption rate at typical rich and lean CO₂ loading (α) and simulate performance of real packing. The method also provides accurate measurement of CO₂ solubility at 40-100 °C. This study provided rate and solubility data at 40-100 °C and practical ranges of CO₂ loading for 8 m 1-methylpiperazine (1-MPZ), 8 m 2-methylpiperazine (2-MPZ), 4 m 2-MPZ/4 m PZ, 7.7 m N-(2-hydroxyethyl)piperazine (HEP), 6 m 1-(2-aminoethyl)piperazine (AEP), 8 m 2-piperidine ethanol (2-PE), and 2 m trans-2,5-dimethylpiperazine (2,5-DMPZ). With the measurements of CO₂ flux (NCO₂) and equilibrium driving force, liquid film mass transfer coefficients (k_g′) are calculated. The rate decreases as of 1-MPZ = PZ > 2-MPZ/PZ > 2-MPZ > HEP > MEA > AEP = 2-PE. This method also allows bracketing and determination of equilibrium CO₂ partial pressure (^*PCO₂) at each condition. Semi-empirical solubility models of CO₂ for each amine were regressed from experimental solubility data to find the lean and rich CO₂ loading corresponding to 0.500 kPa and 5 kPa CO₂ partial pressure respectively. Based on the solubility model, the actual operating capacity of the solvents without overstripping decreases in the sequence of 2-PE > 2-MPZ > 2-MPZ/PZ > 1-MPZ > PZ > HEP > AEP > MEA. The enthalpy of CO₂ absorption (ΔH_abs) of all the piperazine derivatives is around 70 kJ/mol CO₂.     

Optimization of CO2 capture process with aqueous amines using response surface methodology

Amine is one of candidate solvents that can be used for CO₂ recovery from the flue gas by conventional chemical absorption/desorption process. In this work, we analyzed the impact of different amine absorbents and their concentrations, the absorber and stripper column heights and the operating conditions on the cost of CO₂ recovery plant for post-combustion CO₂ removal. For each amine solvent, the optimum number of stages for the absorber and stripper columns, and the optimum absorbent concentration, i.e., the ones that give the minimum cost for CO₂ removed, is determined by response surface optimization. Our results suggest that CO₂ recovery with 48 wt% DGA requires the lowest CO₂ removal cost of $43.06/ton of CO₂ with the following design and operating conditions: a 20-stage absorber column and a 7-stage stripper column, 26 m³/h of solvent circulation rate, 1903 kW of reboiler duty, and 99°C as the regenerator-inlet temperature.     

Modelling of a post-combustion carbon dioxide capture absorber using potassium carbonate solvent in Aspen Custom Modeller

The process models for an equilibrium CO₂ absorber and a rate based CO₂ absorber using potassium carbonate (K₂CO₃) solvents were developed in Aspen Custom Modeller (ACM) to remove CO₂ from a flue gas. The process model utilised the Electrolyte Non-Random Two Liquid (ENRTL) thermodynamic model and various packing correlations. The results from the ACM equilibrium model shows good agreement with an inbuilt Aspen Plus® model when using the same input conditions. By further introducing a Murphree efficiency which is related to mass transfer and packing hydraulics, the equilibrium model can validate the experimental results from a pilot plant within a deviation of 10%. A more rigorous rate based model included mass and energy flux across the interface and the enhancement effect resulting from chemical reactions. The rate based model was validated using experimental data from pilot plants and was shown to predict the results to within 10%. A parametric sensitivity analysis showed that inlet flue gas flowrate and K₂CO₃ concentration in the lean solvent has significant impact on CO₂ recovery. Although both models can provide reasonable predictions based on pilot plant results, the rate based model is more advanced as it can explain mass and heat transfer, transport phenomena and chemical reactions occurring inside the absorber without introducing an empirical Murphree efficiency.     

CO2 capture using oxygen enhanced combustion strategies for natural gas power plants

This paper describes a series of experiments conducted with natural gas in air and in mixtures of oxygen and recycled flue gas, termed O₂/CO₂ recycle combustion. The objective is to enrich the flue gas with CO₂ to facilitate its capture and sequestration. Detailed measurements of gas composition, flame temperature and heat flux profiles were taken inside CANMET's 0.3 MWth down-fired vertical combustor fitted with a proprietary pilot scale burner. Flue gas composition was continuously monitored. The effects of burner operation, including swirling of secondary stream and air staging, on flame characteristics and NO_x emissions were also studied. The results of this work indicate that oxy-gas combustion techniques based on O₂/CO₂ combustion with flue gas recycle offer excellent potential for retrofit to conventional boilers for CO₂ emission abatement. Other benefits of the technology include considerable reduction and even elimination of NO_x emissions, improved plant efficiency due to lower gas volume and better operational flexibility.     

Landfill gas (LFG) processing via adsorption and alkanolamine absorption

Landfill gas (LFG) was upgraded to pure methane using the adsorption and absorption processes. Different toxic compounds like aromatics and chlorinated compounds were removed using granular activated carbon. The activated carbon adsorbed toxic trace components in the following order: carbon tetrachloride > toluene > chloroform > xylene > ethylbenzene > benzene > trichloroethylene ≈ tetrachloroethylene. After removing all trace components, the gas was fed to absorption apparatus for the removal of carbon dioxide (CO₂). Two alkanolamines, monoethanol amine (MEA) and diethanol amine (DEA) were used for the removal of CO₂ from LFG. The maximum CO₂ loading is obtained for 30 wt.% MEA which is around 2.9 mol L^− 1 of absorbent solution whereas for same concentration of DEA it is around 1.66 mol L ^− 1 of solution. 30 wt% MEA displayed a higher absorption rate of around 6.64 × 10^− 5 mol L^− 1 min^− 1. DEA displayed a higher desorption rate and a better cyclic capacity as compared to MEA. Methane obtained from this process can be further used in the natural gas network for city.     

Impact of liquid absorption process development on the costs of post-combustion capture in Australian coal-fired power stations

Australian power generators produce approximately 170 TWh per annum of electricity using black and brown coals that accounts for 170 Mtonne of CO₂ emissions per annum or over 40% of anthropogenic CO₂ emissions in Australia. This paper describes the results of a techno-economic evaluation of liquid absorption based post-combustion capture (PCC) processes for both existing and new pulverised coal-fired power stations in Australia. The overall process designs incorporate both the case with continuous capture and the case with the flexibility to switch a CO₂ capture plant on or off depending upon the demand and market price for electricity, and addresses the impact of the presently limited emission controls on the process cost. The techno-economic evaluation includes both air and water cooled power and CO₂ capture plants, resulting in cost of power generation for the situations without and with PCC. Whilst existing power plants in Australia are all water cooled sub-critical designs, the new power plants are deemed to range from supercritical single reheat to ultra-supercritical double reheat designs, with a preference for air-cooling. The process evaluation also includes a detailed sensitivity analysis of the thermodynamic properties of liquid absorbent for CO₂ on the overall costs. The results show that for a meaningful decrease in the efficiency and cost penalties associated with the post combustion CO₂ capture, a novel liquid sorbent will need to have heat of absorption/desorption, sensible heat and heat of vaporisation around 50% less in comparison with 30% (w/w) aqueous MEA solvent. It also shows that the impact of the capital costs of PCC processes is quite large on the added cost of generation. The results can be used to prioritise PCC research in an Australian context.     

Kinetics of carbon dioxide absorption and desorption in aqueous alkanolamine solutions using a novel hemispherical contactor—II: Experimental results and parameter estimation

In Part 1 of this paper, detailed design of the hemispherical apparatus and a rigorous mathematical model applied to CO₂ absorption and desorption in and from aqueous alkanolamine solutions was presented with some preliminary results. This part of the paper provides detailed results on CO₂-amine kinetics under absorption and desorption conditions and present new estimates of the kinetic parameter for aqueous solutions of monoethanolamine (MEA), diethanolamine (DEA), methyl-diethanolamine (MDEA) and 2-amino-2-methyl-1-propanol (AMP). The absorption experiments were conducted at near atmospheric pressure with pure humidified CO₂ at 293-323 K using initially unloaded solutions. The desorption experiments were performed at 333-383 K for CO₂ loadings between 0.02 to 0.7 mol of CO₂ per mole of amine using humidified nitrogen gas as a stripping medium at total system pressure ranging from 110 to 205 kPa.The new rigorous mathematical model discussed in Part 1 was used in conjunction with a non-linear regression technique to estimate the kinetic parameters. In all cases, the new model predicts the experimental results well. Also, the new results clearly demonstrate that the theory of absorption with reversible chemical reaction could be used to predict desorption rates. The zwitterion mechanism adequately describes the reactions between CO₂ and carbamate forming amines such as MEA, DEA and AMP. The reactions between CO₂ and aqueous MDEA solutions are best described by a base-catalyzed hydration reaction mechanism. The kinetic data obtained show that desorption experiments could be used to determine both forward and backward rate constants accurately. The absorption experiments, on the other hand, could only be used to determine forward rate constants. It was found that at all operating conditions used in this study, the kinetic parameters for MEA, DEA and AMP obtained using absorption data could not be extrapolated to predict desorption rates. However, for MDEA, these data could be used successfully to obtain reasonably good predictions of desorption rates.     

Experimental investigation of liquid holdup in structured packings

Liquid holdup is an important hydrodynamic parameter for characterizing the gas/liquid flow pattern in packed beds. In this paper, a study of liquid holdup in 3 different structured packings: Mellapak 2X from Sulzer, Koch-Glitsch Flexipac 2Y HC, and Montz-Pak B1-250M is presented, using air/water, air/water/sugar solutions with liquid viscosity up to 12 cP and air/30 wt% MEA in a 0.5 m ID absorption column with a packing height of 5 m. As expected, at a given liquid load, the liquid holdup was close to constant as a function of gas flow, with an increase at high gas velocities. In general, the Sulzer packing had a higher liquid holdup than observed in the two other packings. A possible explanation for this could be the lack of enhanced draining of liquid as seen with the modifications of the end-section of the Koch-Glitsch and Montz packings. Liquid holdup was found to increase with increasing liquid viscosity. The influence was higher at high liquid load than at low liquid load. Our results indicate a higher dependency at high liquid load and a lower at low liquid load. There was a reasonable agreement between our results and the data found in the literature.     

Behaviour of absorption/stripping columns for the CO2-MEA system; Modelling and experiments

A model is presented for the steady-state simulation of a CO₂ recovery pilot plant with aqueous monoethanolamine (MEA) solutions. CO₂ absorption is performed in a column packed with 2.54 cm ceramic Pall rings. CO₂ recovery is achieved in a 20 sieve tray steam stripping column. The packed column absorption model was fitted to the experimental data using the specific interfacial area of the irrigated packing as an adjustable parameter. The equivalent average bubble diameter was used as the adjusting parameter in the sieve tray stripping column. Modelling of both towers reproduces within 3% average error concentrations measured in a pilot plant. Measured temperatures were also well correlated.     

Mass transfer investigation and operational sensitivity analysis of amine-based industrial CO2 capture plant

In this article, the industrial process of CO₂ capture using monoethanolamine as an aqueous solvent was probed carefully from the mass transfer viewpoint. The simulation of this process was done using Rate-Base model, based on two-film theory. The results were validated against real plant data. Compared to the operational unit, the error of calculating absorption percentage and CO₂ loading was estimated around 2%. The liquid temperature profiles calculated by the model agree well with the real temperature along the absorption tower, emphasizing the accuracy of this model. Operational sensitivity analysis of absorption tower was also done with the aim of determining sensitive parameters for the optimized design of absorption tower and optimized operational conditions. Hence, the sensitivity analysis was done for the flow rate of gas, the flow rate of solvent, flue gas temperature, inlet solvent temperature, CO₂ concentration in the flue gas, loading of inlet solvent, and MEA concentration in the solvent. CO₂ absorption percentage, the profile of loading, liquid temperature profile and finally profile of CO₂ mole fraction in gas phase along the absorption tower were studied. To elaborate mass transfer phenomena, enhancement factor, interfacial area, molar flux and liquid hold up were probed. The results show that regarding the CO₂ absorption, the most important parameter was the gas flow rate. Comparing liquid temperature profiles showed that the most important parameter affecting the temperature of the rich solvent was MEA concentration.     

Dynamic modelling and analysis of post-combustion CO2 chemical absorption process for coal-fired power plants

Post-combustion capture by chemical absorption using MEA solvent remains the only commercial technology for large scale CO₂ capture for coal-fired power plants. This paper presents a study of the dynamic responses of a post-combustion CO₂ capture plant by modelling and simulation. Such a plant consists mainly of the absorber (where CO₂ is chemically absorbed) and the regenerator (where the chemical solvent is regenerated). Model development and validation are described followed by dynamic analysis of the absorber and regenerator columns linked together with recycle. The gPROMS (Process Systems Enterprise Ltd.) advanced process modelling environment has been used to implement the proposed work. The study gives insights into the operation of the absorber-regenerator combination with possible disturbances arising from integrated operation with a power generation plant. It is shown that the performance of the absorber is more sensitive to the molar L/G ratio than the actual flow rates of the liquid solvent and flue gas. In addition, the importance of appropriate water balance in the absorber column is shown. A step change of the reboiler duty indicates a slow response. A case involving the combination of two fundamental CO₂ capture technologies (the partial oxyfuel mode in the furnace and the post-combustion solvent scrubbing) is studied. The flue gas composition was altered to mimic that observed with the combination. There was an initial sharp decrease in CO₂ absorption level which may not be observed in steady-state simulations.     

Preliminary analysis of process flow sheet modifications for energy efficient CO2 capture from flue gases using chemical absorption

The energy penalty associated with solvent based capture of CO₂ from power station flue gases can be reduced by incorporating process flow sheet modifications into the standard process. A review of modifications suggested in the open and patent literature identified several options, primarily intended for use in the gas processing industry. It was not immediately clear whether these options would have the same benefits when applied to CO₂ capture from near atmospheric pressure combustion flue gases. Process flow sheet modifications, including split flow, rich split, vapour recompression, and inter-stage cooling, were therefore modelled using a commercial rate-based simulation package. The models were completed for a Queensland (Australia) based pilot plant running on 30% MEA as the solvent. The preliminary modelling results showed considerable benefits in reducing the energy penalty of capturing CO₂ from combustion flue gases. Further work will focus on optimising and validating the most relevant process flow sheet modifications in a pilot plant.     

Kinetics of absorption of carbon dioxide into aqueous solution of 2-(1-piperazinyl)-ethylamine

The absorption of CO₂ into aqueous solution of 2-(1-piperazinyl)-ethylamine (PZEA) were studied at 303, 313, and 323 K within the amine concentration range of 0.083-1.226 kmol m⁻³ using a wetted wall column absorber. The experimental results were used to interpret the kinetics of the reaction of CO₂ with PZEA within the amine concentration range of 0.150-1.226 kmol m⁻³ for the above mentioned temperature range. Based on the pseudo-first-order condition for the CO₂ absorption, the overall second order reaction rate constants were determined from the kinetic measurements. The reaction order was found to be in between 0.99 and 1.03 with respect to amine for the later mentioned concentration range. The kinetic rate parameters were calculated and presented at each experimental condition. The second-order rate constants k₂, were obtained as 31867.6, 56354.2, and 100946 m³ kmol^-1 s^-1 at 303, 313, and 323 K, respectively, with activation energy of 47.3 kJ mol⁻¹. This new amine in the field of acid gas removal can be used as an activator by mixing with other alkanolamine solvents due to its very high rate of reaction with CO₂.     

Simulations of chemical absorption in pilot-scale and industrial-scale packed columns by computational mass transfer

A complex computational mass transfer model (CMT) is proposed for modeling the chemical absorption process with heat effect in packed columns. The feature of the proposed model is able to predict the concentration and temperature as well as the velocity distributions at once along the column without assuming the turbulent Schmidt number, or using the experimentally measured turbulent mass transfer diffusivity. The present model consists of the differential mass transfer equation with its auxiliary closing equations and the accompanied formulations of computational fluid dynamics (CFD) and computational heat transfer (CHT). In the mathematical expression for the accompanied CFD and CHT, the conventional methods of k-ε and are used for closing the momentum and heat transfer equations. While for the mass transfer equation, the recently developed concentration variance and its dissipation rate ε_c equations (Liu, 2003) are adopted for its closure. To test the validity of the present model, simulations were made for a pilot-scale randomly packed chemical absorption column of 0.1 m ID and 7 m high, packed with 1/2^″ ceramic Berl saddles for CO₂ removal from gas mixture by aqueous monoethanolamine (MEA) solutions (Tontiwachwuthikul et al., 1992 ) and an industrial-scale randomly packed chemical absorption column of 1.9 m ID and 26.6 m high, packed with 2^″ stainless steel Pall rings for CO₂ removal from natural gas by aqueous MEA solutions (Pintola et al., 1993). The simulated results were compared with the published experimental data and satisfactory agreement was found between them in both concentration and temperature distributions. Furthermore, the result of computation also reveals that the turbulent mass transfer diffusivity D_tvaries along axial and radial directions. Thus the common viewpoint of assuming constant D_t throughout the whole column is questionable, even for the small size packed column. Finally, the analogy between mass transfer and heat transfer in chemical absorption is demonstrated by the similarity of their diffusivity profiles.     

Absorption of carbon dioxide by the absorbent composed of piperazine and 2-amino-2-methyl-1-propanol in PVDF membrane contactor

This paper tests the performance of microporous polyvinylidinefluoride (PVDF) hollow fiber in a gas absorption membrane process (GAM) using the aqueous solutions of piperazine (PZ) and 2-amino-2-methyl-1-propanol (AMP). Experiments were conducted at various gas flow rates, liquid flow rates and absorbent concentrations. Experimental results showed that wetting ratio was about 0.036% when used with the aqueous alkanolamine solutions, while that was 0.39% with aqueous piperazine solutions. The CO₂ absorption rates increased with increasing both liquid and gas flow rates at N_Re < 20. The increase of the PZ concentration showed an increase of absorption rate of CO₂. The CO₂ absorption rate was much enhanced by the addition of PZ promoter. The resistance of membrane was predominated as using a low reactivity absorbent and can be neglected as using absorbent of AMP aqueous solution. The resistance of gas-film diffusion was dominated as using the mixed absorbents of AMP and PZ. An increase of PZ concentration, the resistance of liquid-film diffusion decreased but resistance of gas-film increased. Overall, GAM systems were shown to be an effective technology for absorbing CO₂ from simulated flue gas streams, but the viscosity and solvent-membrane relationship were critical factors that can significantly affect system performance.