Carbon monoxide separation by reactive absorption in chlorocuprate(I)-based ionic liquid

Gabriel Zarca, Inmaculada Ortiz, Ane Urtiaga

Dep. Chemical and Biomolecular Engineering, Universidad de Cantabria, Santander
39006, Spain
zarcag@unican.es
1. Introduction
High amounts of CO are often release in the waste gas streams of industrial processes when a partial combustion of hydrocarbons is performed, e.g. carbon black manufacturing, steel-making and silicon carbide processes. The selective recovery and valorisation of this CO could be sought as a means to obtain a valuable by-product such as hydrogen or chemicals, e.g. methanol, formaldehyde, etc. Available technologies at commercial scale to address the CO separation are based on cryogenic distillation, liquid absorption, and adsorption processes. Nevertheless, if nitrogen is present in the gas mixture, the use of an absorption process may be considered as the most attractive technology at medium scale given that cryogenic distillation becomes less efficient with increasing N2 content and pressure swing adsorption process has a lower yield [1]. Nowadays, the CO absorption process is based on the complexation/decomplexation reaction between CO and copper(I) aluminum chloride salt (CuAlCl4) dissolved in an aromatic hydrocarbon, usually toluene, and has several concerns regarding the solubility of the complexing salt and its long-term activity. In this work, a novel approach to the separation and purification of CO is explored taking advantage of the outstanding properties offered by ionic liquids, namely, their negligible vapour pressure, high solvent capacity, wide liquid range and thermal and chemical stability.
2. Thermodynamic study
Taking that into account, we proposed employing the ionic liquid 1-hexyl-3- methylimidazolium chlorocuprate(I) as an efficient medium to provide an improved CO solubility over N2. In the Lewis basic regime, that is when there is a chloride excess, chlorocuprate(I) anionic species are formed, which react with CO in a 1:1 ratio leading to a copper(I)-CO adduct [2]. In this way, the CO solubility in this medium was greatly
enhanced, whereas the N2 physical solubility was very low; therefore, this chemical system provides high CO/N2 solubility-selectivity, up to 90 at 303 K with a 2.7 mol copper(I) L-1, and contrary to the COSORB process, it employs a non-volatile and non- toxic solvent in which copper(I) can be solubilised in high amounts without the need of
adding another salt such us AlCl3 [3].
3. Kinetic study
Afterwards, the reaction kinetics between CO and the chlorocuprate(I) species have been studied in a batch reactor with flat gas-liquid interface. First, the mass transfer coefficients were assessed in the absence of chemical reaction employing carbon dioxide as an inert model gas with almost identical diffusion properties to CO. The influence of several process variables on the mass transfer coefficients were analysed, namely, temperature and stirring speed. In addition, the effect of copper(I) concentration was studied since it significantly affects the physical properties of the ionic liquid and thus, variable mass transfer coefficients have to be considered. Eventually, a potential proportionality between the mass transfer coefficients and the ionic liquid viscosity to the power of -0.5 was found and a dimensionless correlation to estimate the mass transfer coefficients in the studied range of operating conditions was proposed [4].
Then, in order to characterize the reaction kinetics, reactive absorption experiments were performed with CO in the pseudo-first order kinetic regime in which the absorption rate is not diffusion-controlled. The experimental enhancement factors that stand for the way in which the mass transfer is enhanced by the chemical reaction were determined as function of temperature and copper(I) concentration and the Hatta number was obtained with two different approaches: first, assuming that the reaction took place in the fast pseudo-first order regime; and secondly, employing the relationship between the enhancement factors and Hatta number proposed by DeCoursey. The results showed that the reaction apparently was placed in the pseudo- first order regime and therefore, the overall kinetic constants were directly obtained from the absorption rates [5].
In summary, the information on the thermodynamics and reaction kinetics derived in this study, together with the characterization of the mass transport, allow the design and optimization of a novel ionic liquid-based separation process that represents a promising alternative to selectively recover CO from industrial flue gases.
References
1. J.A. Hogendoorn, W.P.M. van Swaaij, G.F. Versteeg, The absorption of carbon monoxide in COSORB solutions: absorption rate and capacity, Chem. Eng. J. 59 (1995) 243-252.
2. D.R. Smith, J.A. Quinn, Facilitated transport of carbon monoxide through cuprous chloride solutions, AIChE J. 26 (1980) 112-120.
3. O.C. David, G. Zarca, D. Gorri, A. Urtiaga, I. Ortiz, On the improved absorption
of carbon monoxide in the ionic liquid 1-hexyl-3-methylimidazolium chlorocuprate, Sep. Purif. Technol. 97 (2012) 65-72.
4. G. Zarca, I. Ortiz, A. Urtiaga, Carbon monoxide recovery from flue gases by reactive absorption in imidazolium chlorocuprate(I) ionic liquid: Mass transfer coefficients, Chinese J. Chem. Eng. Accepted manuscript (2014).
5. G. Zarca, I. Ortiz, A. Urtiaga, Kinetics of the carbon monoxide reactive uptake by an imidazolium chlorocuprate(I) ionic liquid, Chem. Eng. J. Accepted manuscript (2014).