Elsevier

Chemical Engineering Science

Volume 170, 12 October 2017, Pages 508-517
Chemical Engineering Science

Mass transfer studies on catalyst-aided CO2 desorption from CO2-loaded amine solution in a post-combustion CO2 capture plant

https://doi.org/10.1016/j.ces.2017.02.004Get rights and content

Highlights

  • Solid acid catalyst applied in CO2 desorption increase the KLav.

  • Desorption experiments were conducted in packed column in full cycle operating mode.

  • Lewis acid catalyst replaces the role of HCO3 ions in CO2 lean regions.

  • Brφnsted acid catalyst donates protons which aid in carbamate breakdown.

  • Catalyst acid strength and Brφnsted/Lewis ratio are the most significant properties.

Abstract

Two solid acid catalysts of contrasting characteristics were studied on their effects on the liquid-side mass transfer performance for the desorption of CO2 from CO2-loaded 5 M monoethanolamine (MEA) solvent. The characteristics of the catalysts were both physical (surface area of catalyst which increases the interfacial area for mass transfer) and chemical (the catalyst acid sites) to catalyze CO2 desorption as illustrated by the use of protonated zeolite socony mobile-5 (HZSM-5) representing a higher Brφnsted/Lewis acid site ratio in a catalyst and γ-Al2O3 representing lower Brφnsted/Lewis acid site ratio. The experiments were conducted in a 1.07 m × 0.047 m2 packed column in full cycle operation. The packing structure consisted of a 0.51 m bed height of 6 mm inert marble randomly mixed with a specific mass of catalyst and a 50.8 mm Sulzer LDX structured packing acting as top and bottom support beds. The effects of amine flow rate and catalyst bed temperature on desorption performance were also analyzed. The catalyst physical activity in terms of the added interfacial area showed higher count for HZSM-5 as compared to γ-Al2O3. The mass transfer coefficient increased by an average of 16% and 22% using γ-Al2O3 and HZSM-5 catalysts, respectively at a CO2 desorption temperature of 85 °C. These results are consistent with the greater ability of HZSM-5 to act as a proton donor to facilitate MEACOO- breakdown throughout the loading range of desorption as compared with γ-Al2O3, an amphoteric oxide, which plays the role of HCO3- to release CO2 with lower energy penalties at leaner CO2 loadings. The high percentage increase implies a corresponding reduction in the size of the desorber column which further translates to lower capital costs.

Introduction

Climate Change reports issued recently cautions of calamitous consequences if carbon emissions are not reduced (Pachauri et al., 2014). However, spiralling energy demands and consumption rate demand the utilisation of a relatively cheap and abundant resource, hence the reliance on fossil fuels. Coal combustion produces nearly 40% of the world’s electricity (Miller, 2015). The proven and economically recoverable world reserves of coal are quoted to last for about 150 years which shows coal is going to remain a staple feed to power plants. Even though coal fired power plants contribute more to climate change issues than any other fossil fuel, however, not using coal at all is not a practical solution. It has been clearly stated by experts around the world that the single most important action to mitigate climate change is to reduce carbon dioxide (CO2) emissions from coal. Several technological fixes are under continual developments to increase their mitigation techniques of CO2 (Yang et al., 2008). Post Combustion CO2 Capture (PCCC) using reactive solvents is among the preferred abatement strategies to capture CO2 from large point sources of CO2 such as flue gases of coal fired power plants (Wang et al., 2011). Even though there exists high operational confidence and capture efficiency with the use of PCCC, its high cost greatly impedes industrial deployments because this removal technique is via chemical bonding of amines with CO2 (Liang et al., 2015a). Issues surrounding optimization of solvent and minimization of heat input for regeneration (Aroonwilas and Veawab, 2004, Liang et al., 2015a) are among the challenges that are being addressed.

The high cost of heat, which is essential for regeneration, continues to constitute the major drawback to industrial implementation. With the emerging market of trading CO2 for further utilisation, and the rational of recycling potent solvent to effectively clean flue gas from coal fired power plants, it becomes essential to tackle the expensive desorption aspect of the capture process. Based on the literature, it is clear that the CO2 desorption process (also known as solvent regeneration or CO2 stripping) has not attracted as much attention as the CO2 absorption process. Publications in CO2 desorption focuses on flow sheet modifications and optimizations to identify energy saving areas in an integrated carbon capture and coal fired plants. Oyenekan and Rochelle (2006) and Oyenekan and Rochelle (2007) have used innovative stripper configurations (vacuum and multipressure) to attain energy savings advantage. Earlier experimental studies in a packed regenerator column were only performed to validate process models. (Weiland et al. (1982) used fundamental physico-chemical data to design and analyse a packed column, and compared model predictions to pilot plant data with good correlation. (Tobiesen et al. (2005) also developed computational models and found energy savings in optimizing the amine loading levels and reboiler pressure. Their only comparison to experimental data was in validating solubility data. Other publications emerged discussing desorption reaction rates, and to a limited extent, on mass transfer for mainly solvent screening purposes (Eimer, 2014). Their contribution was to identify solvents characteristics that contribute to reducing the energy penalty in CO2 desorption from CO2-loaded amines.

It is clear that the overall performance of any column depends on the solvent kinetics, which forms a strong basis of the mass transfer performance and energy penalty of the system. Several solvents have emerged in application to carbon capture. These solvents include but are not limited to organic solvents, ionic liquids, amines, switchable solvents and others (Orhan et al., 2016, Ozkutlu et al., 2016, Yüksel et al., 2015). Aqueous solutions of amines have been employed and proven to exhibit tremendous absorption characteristics. Monoethanolamine (MEA), the most widely used amine has faster kinetics implying shorter columns for absorption. However, the CO2 desorption characteristics of MEA are poor as it exhibits high energy requirement for solvent regeneration with associated poor thermal stability at high desorption temperature (Aroonwilas and Veawab, 2004). MEA also exhibits high degradation and corrosion rates when additives such as catalysts and inhibitors are added to the system. Other emerging industrial amines with better desorption properties have also been applied in the capture process. These are found in the class of tertiary amines and sterically hindered amines. The structure and activity relationships studies to rank amines for behavioural traits has also helped catalogue several amines with their distinguishing effect in the capture process (Singh et al., 2007, Singh et al., 2009). With the advent of these potential amines, there are still numerous MEA absorption plants operating throughout the world. This makes it an appropriate baseline solvent of choice for CO2 desorption studies.

Evidently, a solvent process alone may not possess all the desired features. Consequently, other innovative technological approaches to reach the energy penalty reduction goal for solvent regeneration are required. A recent innovative strategy introduced in desorption was the application of a solid acid catalyst (Idem et al., 2011). This represents an excellent example of a gas-liquid-solid reacting system. Based on their study of desorption chemistry analysis, a predominantly Lewis acid catalyst (example γ-Al2O3) and a predominantly Brφnsted acid catalyst (Protonated zeolite socony mobil-5 (H-ZSM-5)) were selected to catalyze desorption of CO2 from the bicarbonate (HCO3-) and carbamate ions (MEACOO-) respectively. A heat duty reduction using H-ZSM-5 catalyst of 1.56 GJ/tonne CO2 at a temperature reduced from 110–120 °C to 90–95 °C was realised. To corroborate these findings, Shi et al. (2014b) developed computational potential energy surface (PES) diagram for the catalytic desorption systems and further studied the desorption process with single and blended amines. Their results on desorption performance with MEA with no catalyst as baseline was recorded as follows: MEA–DEAB with H-ZSM-5 (38%) > MEA–DEAB with γ-Al2O3 (40%) > MEA–DEAB with no catalyst (51%) > MEA with H-ZSM-5 (65%) > MEA with γ-Al2O3 (73%) > MEA–MDEA with γ-Al2O3/no catalyst (74%), all relative to MEA with no catalyst (100%). They concluded on the significant effect of HCO3 role in catalytic desorption present in substantial amounts in the blended amines. Liang et al. (2015b) researched using hybrid catalyst (consisting of a physical mixture of γ-Al2O3 and H-ZSM-5) with MEA solvent at different loading regions. The reason was to rank catalyst based on operating loading regions that can be seen in industrial power plants. At low/lean loading regions, the catalyst activity followed a γ-Al2O3/H-ZSM-5 = /1 > γ-Al2O3 > H-ZSM-5 > H-Y (protonated zeolite Y) rankings and at high/rich loading regions a ranking of H-ZSM-5 > γ-Al2O3/H-ZSM-5 = 2/1 > γ-Al2O3 > H-Y.

Clearly, the study in a batch scale apparatus on catalytic contribution was to prove savings in reboiler heat duty by reducing the desorption temperature and increasing the rate of desorbed CO2. However, the practicality of the catalytic capture operations is masked in the batch desorption process, and as such, industrial deployment problems could be missed. The catalytic capture operation were devoted to determining its influence on the heat duty of the system. Influences on other areas, especially the mass transfer, was not addressed. Mass transfer influence based on heterogeneous catalysis is scarce in the literature. Mass transfer analysis in the desorber is also an area that has received little attention. Thus, it can be concluded that reliable experimental estimations of the mass transfer enhancements in the catalytic system has not yet been reported in the literature. The novelty of exploiting the combined advantage of reducing both the size of the regeneration/desorber column and its auxiliary equipment as well as operational heat duty simultaneously remains unverified.

This paper, experiments in a full cycle bench-scale CO2 capture plant using a single amine: namely 5 M MEA, was conducted for the purpose of analyzing and developing the mass transfer characteristics of a catalyst-aided CO2 desorption from a CO2 loaded amine solution. Analysis used the two film theory, and developed mass transfer theory for the catalyst-aided process was used to elucidate the mass transfer improvements that are derived from using solid acid catalysts such as γ-Al2O3 and H-ZSM-5 to replace the conventional packing in the desorber as a function of the characteristics of the catalysts and operating variables.

Section snippets

Determination of mass transfer coefficient

According to Eimer (2014), the theory of mass transfer in desorption has been established to be a little more complex than absorption. However, Astarita and Savage (1980) delineated situations where absorption can be likened to desorption as follows:

  • 1.

    Mass transfer must lie within or between the diffusional and fast reaction regimes.

  • 2.

    There is only a single overall reaction.

  • 3.

    The total capacity of the liquid for dissolved gas greatly exceeds its purely physical capacity (Henry’s Law).

  • 4.

    The gas phase

Chemicals and material

MEA solvent was obtained from Sigma Aldrich, Canada with purity >99%. 100% CO2 and N2 were supplied by Praxair Inc., Canada. Gas analyser calibrations used a 15% premixed CO2 with N2 also purchased by Praxair Inc., Canada. γ-Al2O3 was acquired from Zeochem, US while H-ZSM-5 catalyst was purchased from Zibo Yinghe Chemical Company Ltd. Standard 1 N hydrochloric acid was supplied by Sigma Aldrich, Canada and used for titrations.

Mass transfer experiment: apparatus and method

The experiment was conducted in a bench-scale full cycle unit shown in

Results and discussion

The results are presented as the overall mass transfer coefficient based on the liquid side in the desorber. Data collected for analysis included the solution concentration, the rich loading, lean loading, the inlet CO2 concentration, off gas concentration, the inlet and off gas mass flow, the liquid flow rate, product CO2 flow rate and the average desorber temperature. The data was analyzed with Aspen Plus version 8.4. Liquid side analysis control volume included unit 14 and 15 on Fig. 2.

Conclusions

The overall liquid side mass transfer coefficient (KLav) was analyzed from the full cycle carbon capture plant with a randomly packed desorption column comprising of two solid acid catalysts, namely, H-ZSM-5 and γ-Al2O3. The determined KLav ranged from 0.16 to 0.89 h−1 at temperatures ranging from 75–85 °C, flow rate from 50 to 80 ml/min and catalyst weight from 0 to 200 g.

There is an up to 97% increment in interfacial area created when catalyst is added to system. The interfacial area added by

Acknowledgements

The financial support provided by Natural Science and Engineering Research Council of Canada (NSERC) and the Canada Foundation for Innovation (CFI) as well as the Clean Energy Technologies Research Institute (CETRI) and Faculty of Graduate Studies and Research (FGSR), University of Regina is gratefully acknowledged.

References (34)

  • M. Wang et al.

    Post-combustion CO2 capture with chemical absorption: a state-of-the-art review

    Chem. Eng. Res. Des.

    (2011)
  • H. Yang et al.

    Progress in carbon dioxide separation and capture: a review

    J. Environ. Sci.

    (2008)
  • A. Aroonwilas et al.

    Characterization and comparison of the CO2 absorption performance into single and blended alkanolamines in a packed column

    Indust. Eng. Chem. Res.

    (2004)
  • M. Caplow

    Kinetics of carbamate formation and breakdown

    J. Am. Chem. Soc.

    (1968)
  • P. Danckwerts

    Significance of liquid-film coefficients in gas absorption

    Indust. Eng. Chem.

    (1951)
  • D. Eimer

    Gas Treating: Absorption Theory and Practice

    (2014)
  • Geankoplis, C.J., 2003. Transport Processes and Separation Process Principles. Prentice Hall Professional Technical...
  • Cited by (39)

    View all citing articles on Scopus
    View full text