Evaluation of the protic ionic liquid, N,N-dimethyl-aminoethylammonium formate for CO2 capture

https://doi.org/10.1016/j.ijggc.2014.11.011Get rights and content

Highlights

  • Protic ionic liquid vapour–liquid equilibria with CO2 compared to MEA.

  • Protic ionic liquid reaction rates with CO2 tested in wetted wall column apparatus.

  • Protic ionic liquid tested in gas–solvent membrane contactor.

Abstract

A protic ionic ethylenediamine liquid N,N-dimethyl-aminoethylammonium formate (DMEDAH formate) was evaluated for the removal of carbon dioxide from carbon dioxide and nitrogen mixed gas streams. This evaluation was conducted via a series of vapour–liquid equilibrium (VLE) experiments, wetted wall measurements and solvent contactor trials with the results compared to the performance of the industrial standard 30 wt% monoethanolamine (MEA). The VLE of the ionic liquid indicated that the sorption process included a chemical reaction component and is comparable to MEA. The mass transfer coefficient, calculated via the wetted wall column apparatus is an order of magnitude lower than MEA, which is likely due to the higher viscosity of the ionic liquid and its impact on carbon dioxide diffusivity. The solvent contactor trials indicated the solvent may be used in such an arrangement with good performance at low CO2 loadings, but would benefit from the use of alternative membrane materials.

Introduction

Carbon dioxide capture processes are most commonly conducted using fast aqueous amine solutions such as 30 wt% monoethanolamine (MEA) in water. Although these systems exhibit positive characteristics such as fast kinetics and favourable vapour–liquid equilibria; high regeneration energies due to high water content and reaction enthalpies, in addition to impurity formation, makes it an expensive process. Room temperature ionic liquids (RTILs) are a new type of solvent that could be an improvement on this current system. To date much work has been conducted on the development of RTILs as physical solvents, however there has also been renewed focus on the development of RTILs as chemical solvents (Bates et al., 2002, Brennecke and Gurkan, 2010, Sanchez et al., 2007, Wang et al., 2010, Yang et al., 2011) as discussed in a number of recent reviews(MacFarlane et al., 2014).

Although novel solvent development is a priority, so is the development of process designs that can adequately contact these RTILs with the CO2 rich gas stream and promote CO2 mass transfer. A key limitation to this is the generally high viscosity of RTILs. Of these works, Zhang et al., 2011, Zhang et al., 2013 developed a rotating packed bed reactor to improve the mass transfer performance for a viscous amino-functionalized RTIL. Iliuta et al. (2014) took a different approach and mixed an amine, diethanolamine (DEA) with a RTIL ([hmim][NTf2]) in a stirred cell reactor, which upon loading with carbon dioxide formed a carbamate precipitate. The precipitate could then be separated from the solution and heated to release the CO2 thereby reducing the regeneration energy penalty. In this process the ionic liquid did not undergo a chemical reaction with the CO2. Yang et al. (2014) conducted a similar study although the combination was 30 wt% MEA, 40 wt% [bmim][BF4] and 30 wt% water and the solvent was contacted with the CO2 rich stream via a polypropylene membrane contactor. This process design was found to have a number of benefits over the industry standard 30 wt% MEA including: (1) the energy consumption was found to be 37.2% lower, (2) the MEA loss was reduced, and (3) mixing the ionic liquid with water reduced the solution viscosity, thereby improving CO2 diffusivity. Other methods to incorporate RTILs into CO2 capture processes have been investigated, including their immobilisation onto porous microspheres (Wang et al., 2013). However, in these systems, although reasonable capacities and kinetics have been observed, it has been found that the ionic liquid can leak into the bulk gas phases thereby reducing working capacity.

In this work, potential use of the ionic liquid, N,N-dimethyl-aminoethylammonium formate (DMEDAH formate) was evaluated for carbon dioxide capture via a number of testing methods including evaluation of:

  • (1)

    physical properties, to understand how easily the solvent may be utilised in an gas liquid solvent contactor;

  • (2)

    vapour–liquid equilibrium (VLE), to obtain an appreciation of the volume of solvent that may be required;

  • (3)

    mass transfer kinetics, to gauge the size of equipment and recirculation rates required; and

  • (4)

    operation with a polypropylene membrane contactor, to assess system practicality.

DMEDAH formate was selected for this trial as it was previously reported to show similar CO2 uptake characteristics as MEA (Vijayraghavan et al., 2013), and appreciable CO2 desorption was shown to occur at low temperature (above 25 °C). Therefore, this ionic liquid could provide an energy efficient reversible CO2 capture medium, unlike MEA which can require heating up to 140 °C before CO2 desorption. In addition, the low temperature CO2 desorption characteristic of the ionic liquid would also result in a significant reduction in thermal and oxidative degradation processes that result in solvent loss and formation of toxic compounds.

To assist in the interpretation of the solvent performance, the results are compared to the industry standard 30 wt% MEA.

Section snippets

Physical properties

Contact angle measurements between 100, 30 and 0 wt% DMEDAH formate (balance Milli-Q water) with polypropylene and PTFE were conducted using the sessile drop technique. This was achieved with a contact angle goniometer interfaced to FTÅ200 software (First Ten Ångstroms Inc., Software Version 2.0 Build Number 187, UK) which used “drop shape” analysis to analyse drop shape and size and to calculate the contact angle. A Sterlitect laminated membrane filter of pore size 0.05 μm provided surfaces of

Physical properties

An important physical property for solvent absorption processes, especially those utilising ionic liquids, is solution viscosity (Bates et al., 2002). Viscosity of DMEDAH formate has been reported previously (Vijayraghavan et al., 2013) (105 mPa s compared to 30 wt% MEA at 2.3 mPa s). High viscosity impacts on the mass transfer coefficient due to its relationship with diffusivity. Diffusivity is generally described by the semi-empirical Wilke–Chang equation (Wilke and Chang, 1955), which suggests

Conclusions

CO2 separation through use of the RTIL, DMEDAH formate has been demonstrated and valuable information regarding the vapour–liquid equilibrium, mass transfer kinetics and use in membrane contactors gathered. DMEDAH formate was found to have a lower working capacity over the temperature range of 15–35 °C and slower kinetics than the industrial standard, 30 wt% MEA. However at low CO2 loadings and before the membrane pores became wetted, it exhibited similar performance to MEA in a gas–solvent

References (21)

There are more references available in the full text version of this article.

Cited by (19)

  • Effects of the structure on physicochemical properties and CO<inf>2</inf> absorption of hydroxypyridine anion-based protic ionic liquids

    2022, Journal of Molecular Liquids
    Citation Excerpt :

    As a branch of ILs, protic ionic liquids (PILs) could be easily prepared by the transfer of proton from a Brønsted acid to a Brønsted base, leading to form the strong hydrogen-bonding network of PILs and improve theirs gas absorption capacity [4–6]. Compared to aprotic ionic liquids (APILs), PILs attracted more and more attention due to simple preparation, low costs, and tunable physicochemical properties [7–9]. Accordingly, PILs have been widely applied in gas absorption [10–13] and gas conversion [14,15].

  • Protic ionic liquids with low viscosity for efficient and reversible capture of carbon dioxide

    2019, International Journal of Greenhouse Gas Control
    Citation Excerpt :

    Protic ionic liquids (PILs) can be easily obtained through a neutralization reaction of Brønsted acid and base. Recently, PILs have caught considerable attention in CO2 capture fields due to their simpler synthetic routes and lower costs compared to most of aprotic ionic liquids (AILs) (Mumford et al., 2015; Xu, 2017; Oncsik et al., 2018). For example, Alcantara et al. (2018a, 2018b) studied the CO2 absorption performance in four potential PILs which were synthetized through one-step neutralization of diethylamine/ethanolamines and butanoic acids.

  • Amino acid based poly(ionic liquid) materials for CO<inf>2</inf> capture: Effect of anion

    2019, Journal of Molecular Liquids
    Citation Excerpt :

    Since then, a number of other researchers [8,17–22] have also studied CO2 absorption in amine-functionalized cation and anion ILs. Some of the best performing materials are the diamino functional protic ILs, which is DEEDAH formate with 0.47 mol/mol absorption capacities [23,24]. Two types of amino-functionalised ILs have been employed for CO2 absorption.

  • A novel experimental apparatus for the study of low temperature regeneration CO<inf>2</inf> capture solvents using hollow fibre membrane contactors

    2018, International Journal of Greenhouse Gas Control
    Citation Excerpt :

    Pseudo steady-state was achieved approximately 12 min into the experiment, where the CO2 removal rate was approximately 5.15 × 10−3 g m−2 s−1. This type of method is commonly used in fundamental CO2 capture studies (Mumford et al., 2015; Albo and Irabien, 2012), and with some basic assumptions (Albo and Irabien, 2012) can be used to calculate a value for the overall CO2 mass transfer coefficient, Koverall. For DMPDAH acetate + 50% H2O, Koverall was determined to be (3.71 ± 0.2) × 10-5 m s−1, a value comparable to that determined by Albo and Irabien, (2012) for the aprotic IL 1-ethyl-3-methylimidazolium ethylsulfate (EMISE), (3.7 ± 0.18) × 10−5 m s−1.

View all citing articles on Scopus
View full text