International Journal of Greenhouse Gas Control 5 (2011) 396–400

A systematic investigation of carbamate stability constants by 1H NMR

Nichola McCann *, Duong Phan, Debra Fernandes, Marcel Maeder

Department of Chemistry, University of Newcastle, Callaghan, NSW 2308, Australia

A R T I C L E I N F O

Article history:

Received 15 October 2009

Received in revised form 20 January 2010

Accepted 23 January 2010

Available online 19 February 2010

Keywords:

CO2 capture

Carbamate equilibrium constants

Amine solvents

A B S T R A C T

A systematic, experimental investigation into carbamate stability constants using 1H NMR spectroscopy

reveals that both steric hindrance and the acid dissociation constant of the parent amine have a

significant effect on the stability of the resulting carbamates. Increasing steric hindrance was found to

decrease carbamate stability, while increasing pK of the parent amine was found to increase carbamate

stability. The carbamate stability constants of monoethanolamine, propylamine, isobutylamine, 2-

amino-1-propanol, aminomethylpropanol and methyldiethanolamine were determined as were the

carbamate protonation constants of the first four of these amines.

� 2010 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

International Journal of Greenhouse Gas Control

journal homepage: www.e lsev ier .com/ locate / i jggc

1. Introduction

There is a general consensus that our greenhouse gas emissionsneed to be drastically reduced in the coming years if we are to avertthe potentially disastrous effects of anthropogenically inducedclimate change (Metz et al., 2005; Stern, 2006). Given the massivescale of this problem (recent estimates put global man-made CO2

emissions at 24 gigatons, Mikkelsen et al., 2010), the best solutionwill clearly involve a wide range of different technologies. It isbecoming increasingly apparent that CO2 capture and sequestra-tion could make a significant contribution to reducing the output ofCO2 from fossil fuelled power plants — currently one of the largestproducers of CO2 (Davison, 2007; Wise and Dooley, 2009). Aqueousamine solutions are regarded as amongst the most advancedtechnologies available for post-combustion CO2 capture (PCC) anda number of different amines and amine systems have beeninvestigated for this purpose. Many of the best performing amines,however, have significant drawbacks associated with their use,including high heats of regeneration (resulting in significantlyincreased energy requirements), slow kinetics (meaning that largeabsorption towers are required), high toxicity, and low stability.Not surprisingly therefore, there is substantial interest in thedevelopment of potential new, improved solvents and solventsystems.

A crucial step in developing new solvents is a good under-standing of the chemistry associated with CO2 and aqueous

* Corresponding author. Current address: Laboratory of Engineering Thermody-

namics (LTD), Department of Mechanical and Process Engineering, University of

Kaiserslautern, 67663 Kaiserslautern, Germany.

E-mail address: Nichola.McCann@mv.uni-kl.de (N. McCann).

1750-5836/$ – see front matter � 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.ijggc.2010.01.008

amines. CO2 can potentially react with an aqueous amine solutionvia two different pathways: in a simple acid–base reaction, or viacarbamate formation. In the acid–base reaction, the amine behavesas a base, with which the acidic CO2 can react. The chemicalequations representing the associated reactions are shown inEqs. (1)–(4):

H2Oþ CO2 !K1

HCO�3 þHþ (1)

HCO�3 !K2

CO2�3 þHþ (2)

AHþ !KAAþHþ (3)

H2O !KWOH� þHþ (4)

These reactions have been extensively investigated in a varietyof different contexts and the associated constants are well known.These constants are presented in Table A1 in the supportinginformation.

Alternatively, CO2 can react with several amines to form acarbamic acid as represented in Eq. (5). This carbamic acid can thenalso be deprotonated to give the carbamate, as in Eq. (6). Under theconditions most commonly used for PCC, the product will existprimarily as carbamate:

RNH2 þ CO2 !Kcarb

RNHCO2H (5)

RNHCO2H !KcarbHRNHCO�2 þHþ (6)

Alternative mechanisms for carbamate formation have beensuggested and used previously (Caplow, 1968; Crooks and

N. McCann et al. / International Journal of Greenhouse Gas Control 5 (2011) 396–400 397

Donnellan, 1989; Arstad et al., 2007; Li et al., 2007; Ismael et al.,2009), however, the above equations most accurately representthe chemistry occurring within the system (McCann et al., 2009). Itshould be noted that the acid–base equilibria and carbamateformation occur simultaneously and interactively. That is, theproperties of the carbamate will influence absorption occurring viathe acid–base pathway, through changes in pH, concentrations,etc., and vice versa (McCann et al., 2008).

Carbamate formation has been linked to fast kinetics, butreduced CO2 capacity and increased regeneration energy. As such, athorough understanding of the factors that influence the formationof carbamate is essential for the development of any new aminesystem.

This topic has been investigated by a number of differentresearch groups: Sartori and Savage (1983) investigated thecarbamate stability constants of monoethanolamine (MEA), 2-amino-2-methyl-1-propanol (AMP) and diethanolamine (DEA) by13C NMR (nuclear magnetic resonance). They found that for MEAand AMP, the more sterically hindered amine (AMP) had a lowercarbamate stability (log Kcarb = 1.10 and <�1, respectively). Thestructure of DEA is significantly different to the other two, so it isdifficult to draw conclusions as to what contributes to the differentcarbamate stability (log Kcarb = 0.30).

Several additional amines (including MDEA (Bottinger et al.,2008), piperazine (Ermatchkov et al., 2003; Bottinger et al., 2008),ammonia (Mani et al., 2006), diglycolamine, diisopropylamine(Barth et al., 1984), 2-[(2-aminoethyl)amino]ethanol) (Jakobsenet al., 2008) and some synthetically prepared, more complexamines (Mikkelsen et al., 2009)) have also been investigated, boththrough NMR speciation studies and/or vapour liquid equilibrium(VLE) measurements. The list of amines for which carbamateequilibrium constants have been measured is, however, relativelyshort and diverse, making it extremely difficult to definitivelyidentify factors that influence the stability of the carbamate.

Carbamate stability constants have also been investigatedthrough molecular modelling studies (Chakraborty et al., 1988; daSilva and Svendsen, 2006). Earlier studies attributed decreasedcarbamate formation to the electronic effects of methyl substitu-tion at the a-carbon, rather than to a steric effect. A later studyconcluded that ‘trends in carbamate stability apparently cannot beexplained, in terms of any single molecular characteristic’ (da Silvaand Svendsen, 2006), clearly stating the need for a systematicinvestigation into those molecular characteristics that do influencecarbamate stability.

This work describes the first systematic experimental investi-gation into two factors that influence the stability of carbamates inaqueous solution. Six amines were investigated, including three forwhich carbamate stability constants have not previously beenmeasured. This allowed the composition of a series with fewstructural differences, and thus the correlation of trends in stabilityconstant with molecular structure. The hypothesis that sterichindrance reduces stability was shown to be valid. In addition, asecond factor affecting stability was found, and is nominallyattributed to the pK of the parent amine. With the addition of threenew carbamates, the pool of knowledge of carbamate stabilityconstants has also been significantly increased.

1.1. Methodology

Pure amines, 2-amino-1-propanol (AP) hydrochloride andKHCO3/K2CO3 were obtained from Sigma–Aldrich, HCl from AjaxFinechem. N-methyldiethanolamine (MDEA) was distilled undervacuum and stored under nitrogen, 2-amino-2-methyl-1-propanol(AMP), propylamine (PA), and isobutylamine (IBA) were crystal-lised from HCl and n-propanol to give the amines as hydrochloridesalts. All amine solutions were freshly made prior to each

experiment. HCl concentrations and amine protonation constantswere determined by titration with standardized NaOH on anautomated titration apparatus under a nitrogen atmosphere. 1HNMR spectra were recorded on a Bruker Avance DPX-300 at afrequency of 300.13 MHz at 30.0 8C. All samples were measured inH2O in a 5 mm tube containing an insert with either TSP (3-(trimethylsilyl)propionic acid-d4, sodium salt) or 1,4-dioxane inD2O as a reference.

The monoethanolamine (MEA) titration data were taken fromour previously reported results (McCann et al., 2009). For theremaining amines, solutions of amine hydrochloride (AP, PA andIBA) and K2CO3 were titrated with varying concentrations of HCl.Initial concentrations of amine, (bi-)carbonate and HCl are given inBates and Pinching, 1951; Cox et al., 1968; Hamborg et al., 2007;Harned and Davis, 1943; Harned and Scholes, 1941; Maeda et al.,1987; Soli and Byrne, 2002; Table A2 of the supportinginformation. In the case of MDEA and AMP, no carbamateformation was observed in concentrated solutions of amine and(bi-)carbonate (1.7 M MDEA, 2 M KHCO3 and 2.3 M AMP�HCl,2.7 M K2CO3), so full titrations were not conducted. Instead,maximum carbamate stability constants were estimated assumingthat a maximum of 5% of the amine could be present as thecarbamate without being detectable by NMR. In reality, referencepeaks with integrations of approximately 2% of the amine peakswere measured, so this estimate is rather conservative.

1H NMR and data analyses were conducted as describedpreviously (McCann et al., 2009). All measurements were analysedtogether in a single global analysis for each amine. Activitycoefficient corrections were applied to all charged species based onthe extended Debye–Huckel equation (Eq. (7)):

logg i ¼�Az2

i

ffiffiffiffi

mp

1þ Briffiffiffiffi

mp (7)

In this equation, the activity coefficient gi is a function of theionic strength, m, the charge zi of the ith component and of theparameters A and B that are defined by the dielectric constant ofthe solvent and the temperature. In water at 30 8C, A = 0.51,B = 0.33. The radii ri are not known for several of the ionic species;the values were estimated based on published values for similarcompounds (Stern and Amis, 1959; Barner and Scheuerman, 1978).

2. Results

Acid dissociation constants for the majority of the aminesinvestigated are well known. However, values at 30 8C are notavailable for PA or IBA. These values were determined potentio-metrically as part of this work. The value for IBA (10.33 � 0.02) issimilar to the published value at 25 8C (10.48) (Christensen et al.,1969). To our knowledge, no value for the acid dissociation constantof PA has previously been published.

Fig. 1 shows the 1H NMR spectra associated with the titration ofan MEA/KHCO3 solution with HCl at 30 8C. In the initial spectra,where the pH is relatively high, the peaks at approximately 2.6 and3.4 ppm correspond to the two methylene groups of thedeprotonated MEA (the NCH2 and OCH2 groups respectively). AsHCl is added to the solution, the pH drops and the amine group isincreasingly protonated. This is accompanied by a downfield shiftin both peaks to approximately 3 and 3.6 ppm.

The carbamate peaks in the earlier measurements occur atapproximately 3 and 3.4 ppm (NCH2 and OCH2 groups, respective-ly); as the pH decreases, the N-methylene peak moves slightly up-field, although this shift is difficult to see on the scale in Fig. 1. It ispossible that the O-methylene peak also moves with pH, althoughoverlap with the amine peak makes this impossible to see. Theaddition of HCl also results in dissociation of the carbamate, to give

Fig. 1. Titration of an MEA=HCO�3 solution with HCl (see also Fig. 3 for the

concentration profiles for this reaction).

Fig. 2. Chemical shift of the carbamate peak as HCl is added to a MEA/KHCO3

solution. Note that below pH 7, pH values are approximate only.

Fig. 3. Equilibrium concentration profiles for the titration of MEA/KHCO3 with HCl.

N. McCann et al. / International Journal of Greenhouse Gas Control 5 (2011) 396–400398

a greater proportion of the amine as the free (albeit protonated)amine.

The chemical shift of the carbamate peak as HCl was added isshown in Fig. 2. The results shown in this figure includemeasurements where the calculated aqueous CO2 concentrationwas above 30 mM, and so cannot be used in the determination ofconstants. In addition, the pH values below 7 are approximate only,as some gaseous CO2 would have escaped from the solution.Nevertheless, a clear change in chemical shift is visible, and theinflection point (corresponding to the pK of the protonation) of the

Table 1Determined equilibrium and protonation constants at 30 8C for the six amines investigat

constant. Log Kcarb is defined as in Eq. (8).

Amine Log Kcarb pKcarbH Lit log Kcarb

MEA 1.54 (2) 7.49 (5) 1.10 at 40 8C1.43 at 40 8C

PA 1.67 (3) 7.2 (3)

IBA 1.69 (4) 7.3 (6)

AP 0.72 (3) 8.0 (2)

AMP <�0.7 N/A <�1 at 40 8C�0.56 at 40 8C

MDEA <�1.3 N/A Carbamate does

curve is at approximately 7.5. This is in excellent agreement withthe pKcarbH value determined in this work (7.49).

The change in chemical shift for the carbamate is much smallerthan that observed for the amine. This observation is unsurprising,given that for the amine, the N-methylene group is very close to thesite of protonation (the N of the N-methylene group). In contrast, inthe case of the carbamate, the N-methylene group is much furtherremoved from the site of protonation (the carboxylate group). Theincreased separation between the NMR-measured group and thesite of protonation decreases the amplitude of change in chemicalshift. These observations lend support to the suggestion that it isthe COO� group of the carbamate that is protonated (as shown inEq. (6)), rather than the nitrogen group, as has been previouslysuggested (Caplow, 1968) with the zwitterion mechanism.

Fig. 3 shows the concentration profiles associated with themeasurement shown in Fig. 1. The decreasing (red) concentrationsare the amine species (protonated and deprotonated) and theincreasing (blue) concentrations correspond to the carbamatespecies. The markers represent experimental values, the linescalculated values. The data markers represent only the resolvedpeaks: O and D represent the concentrations for resolved peaks forthe NCH2 group (first and repeat measurements), the mar-kers � and + represent the OCH2 group. As seen in Fig. 1, themajor species at high pH is the carbamate, this species converts tothe amine as HCl is added and the pH drops.

Titrations with the remaining amines showed similar results,although the spectra were more complex, due to the increasedcomplexity of molecular structure. Nevertheless, individual peaksattributable to both the amine and the carbamate could beidentified in all measurements, and peak positions over the courseof the titration showed similar trends to those discussed above.

The determined log KCarb constants as well as the associatedpKcarbH values are given in Table 1. Here for clarity and ease ofcomparison, the log Kcarb constants are defined as shown in Eq. (8),

ed in this study and literature values, where available for the carbamate equilibrium

References

Sartori and Savage (1983)

Park et al. (2002)

Sartori and Savage (1983)

Park et al. (2002)

not form Poplsteinova Jakobsen et al. (2005), Bottinger et al. (2008)

Fig. 5. Effect of steric hindrance and pKa on carbamate stability constant.

N. McCann et al. / International Journal of Greenhouse Gas Control 5 (2011) 396–400 399

the pKcarbH are defined as acid dissociation constants.

RNH2 þHCO�3 !Kcarb

RNHCO�2 (8)

As no carbamate formation was detected for MDEA and AMP, nocarbamate protonation constant for these amines could bedetermined. Where available, log Kcarb values reported in theliterature are also presented in Table 1 and are in good agreementwith the determined constants. There are many carbamatestability constants reported in the literature for MEA and AMP.Similarly, there are many reports that MDEA does not form acarbamate. The purpose of this work is not a review of thisliterature so only a select few of the literature reports arementioned here. No carbamate protonation constants (other thanMEA, reported previously by us) have been reported for theremaining amines, but the protonation constant for ammoniumcarbamate is similar (6.76 at 25 8C (Christensson et al., 1978)).

3. Discussion

The purpose of this investigation was a systematic investigationinto two factors that influence carbamate stability and the aminesinvestigated were selected to represent this. As the benchmarkamine for CO2 capture, MEA was of significant interest; theremaining amines have similar structures. The chemical structuresof the amines used are given in Fig. 4. The amines investigated canbe divided into 2 main groups (PA/IBA and MEA/AP/AMP), asshown by the circles within the figure. Within each group, themajor difference between the amines is the level of sterichindrance around the amine; steric hindrance is lowest in theamines at the top of the figure and highest in the amines at thebottom. MDEA belongs to neither group, but is included as anexample of very high steric hindrance.

Alternatively, the amines could be divided into groups movingfrom left to right across the figure. In each horizontal group, thelevel of steric hindrance is relatively constant, but the number ofhydroxyl groups on the molecule is increased. Concurrently, the pKof the amine increases as the number of hydroxyl groups increases.We make no effort here to differentiate between these two factors,or ascribe trends to only one or the other — such investigations arebeyond the scope of this work. From here on, however, the factorwill be called a pK effect, for simplicity and clarity.

The log Kcarb constants given in the table are all well defined.Their values, as a function of pK and steric hindrance are presentedgraphically in Fig. 5. In this figure, the level of steric hindrance is an

Fig. 4. Amines investigated in this study.

approximate, relative estimation, based on the structure of thecompound, rather than a measured or calculated value. Forexample, the carbon b to the amine group is relatively far removedfrom the site of carbamate formation, and addition of a methylgroup at this site would increase steric hindrance around theamine only very slightly. In contrast, the addition of two methylgroups at the carbon a to the amine group would be expected toincrease steric hindrance significantly. As previously hypothesised,the carbamate stability decreases with increasing steric hindrance.

Fig. 5 also shows that the amines with lower pK values typicallyhad lower carbamate equilibrium constants. It is possible that thistrend is associated with electronic effects, potentially arising fromthe additional hydroxyl-group, which also give rise to a decreasedpK of the parent amine. However, further work is required to moreprecisely identify the cause of the decreased carbamate stabilityconstant and is outside the scope of this work.

In contrast to the carbamate formation constants, the proton-ation constants, as seen in Table 1, are not as well defined. In orderto prevent gaseous CO2 from being released from the solution, onlymeasurements with aqueous CO2 concentrations below 30 mMwere used. Consequently, the pH of the solutions typicallyremained reasonably high (above �7), with the result that thecarbamate pK values around 7 were not well defined. Nevertheless,the change in chemical shift of the N-methylene peak of thecarbamate over the course of the titration supports the suggestionof a protonation. Previous reports of a protonated ammoniumcarbamate with a similar pK value (Christensson et al., 1978) alsosupport the assignment of carbamate protonation values.

Within the reported errors, the carbamate acid dissociationconstants are very similar for all four amines. That is, changes inboth steric hindrance and pK of the parent amine have little effecton the carbamate acid dissociation constant. This is not surprisingas the site of protonation (the COO� group) is far removed, bothspatially and electronically from where these effects occur.

4. Conclusions

A thorough understanding of the factors affecting carbamatestability is essential for the efficient and effective development ofnew amine-based solvent systems for CO2 capture. This workrepresents the first systematic experimental investigation into thistopic. We found conclusive experimental evidence to support thehypothesis that increasing steric hindrance of the parent aminedecreases the stability of the carbamate. We also found that

N. McCann et al. / International Journal of Greenhouse Gas Control 5 (2011) 396–400400

increasing pK of the parent amine leads to increased stability of thecarbamate. We have determined carbamate stability constants andthe associated protonation constants of three new amines. Suchstudies are of fundamental importance in the directed design ofnew solvents and solvent systems for CO2 capture in the future.

Acknowledgements

This research has been supported by the Australian Govern-ment through the CSIRO Energy Transformed Flagship and the AsiaPacific Partnership on Clean Development and Climate initiative(APP).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in

the online version, at doi:10.1016/j.ijggc.2010.01.008.

References

Arstad, B., Blom, R., et al., 2007. CO2 absorption in aqueous solutions of alkano-lamines: mechanistic insight from quantum chemical calculations. J. Phys.Chem. A 111 (7), 1222–1228.

Barner, H.E., Scheuerman, R.V., 1978. Handbook of Thermochemical Data forCompounds and Aqueous Species. John Wiley and Sons, New York.

Barth, D., Rubini, P., et al., 1984. Determination of the thermodynamic equilibriumparameters for the formation of amino-alcohol carbamates in aqueous solutionby carbon-13 nuclear magnetic resonance. Bull. Soc. Chim. Fr. 7–8, 227–230.

Bates, R.G., Pinching, G.D., 1951. Acidic dissociation constant and related thermo-dynamic quantities for monoethanolammonium ion in water from 0 8C to 50 8C.J. Res. Natl. Bur. Stand. (U.S.) 46, 349–352 Research Paper No 2205.

Bottinger, W., Maiwald, M., et al., 2008. Online NMR spectroscopic study of speciesdistribution in MDEA–H2O–CO2 and MDEA–PIP–H2O–CO2. Ind. Eng. Chem. Res.47 (20), 7917–7926.

Caplow, M., 1968. Kinetics of carbamate formation and breakdown. J. Am. Chem.Soc. 90 (24), 6795–6803.

Chakraborty, A.K., Bischoff, K.B., et al., 1988. Molecular orbital approach tosubstituent effects in amine–CO2 interactions. J. Am. Chem. Soc. 110 (21),6947–6954.

Christensen, J.J., Izatt, R.M., et al., 1969. Thermodynamics of proton ionization indilute aqueous solution. XI. pK. DH. deg., and DS. deg. Values for protonionization from protonated amines at 25.deg. J. Chem. Soc. A 8, 1212–1223.

Christensson, F., Koefoed, H.C.S., et al., 1978. Equilibrium constants in the ammo-nium carbonate–carbaminate system. The acid dissociation constant of carba-mic acid. Acta Chem. Scand. A A32 (1), 15–17.

Cox, M.C., Everett, D.H., et al., 1968. Thermodynamics of the acid dissociation ofsome alkylammonium ions in water. J. Chem. Soc. B 11, 1373–1379.

Crooks, J.E., Donnellan, J.P., 1989. Kinetics and mechanism of the reaction betweencarbon dioxide and amines in aqueous solution. J. Chem. Soc., Perkin Trans. 2(4), 331–333.

da Silva, E.F., Svendsen, H.F., 2006. Study of the carbamate stability of amines usingab initio methods and free-energy perturbations. Ind. Eng. Chem. Res. 45 (8),2497–2504.

Davison, J., 2007. Performance and costs of power plants with capture and storage ofCO2. Energy (Oxford, UK) 32 (7), 1163–1176.

Ermatchkov, V., Perez-Salado Kamps, A., et al., 2003. Chemical equilibrium con-stants for the formation of carbamates in (carbon dioxide + piperazine + water)from 1H-NMR-spectroscopy. J. Chem. Thermodyn. 35 (8), 1277–1289.

Hamborg, E.S., Niederer, J.P.M., et al., 2007. Dissociation constants and thermody-namic properties of amino acids used in CO2 absorption from (293 to 353) K. J.Chem. Eng. Data 52 (6), 2491–2502.

Harned, H.S., Davis Jr., R., 1943. The ionization constant of H2CO3 in water and thesolubility of CO2 in water and aqueous solutions from 0 to 50 8C. J. Am. Chem.Soc. 65, 2030–2037.

Harned, H.S., Scholes Jr., S.R., 1941. The ionization constant of HCO3� from 0 8C to

50 8C. J. Am. Chem. Soc. 63, 1706–1709.Ismael, M., Sahnoun, R., et al., 2009. A DFT study on the carbamates formation

through the absorption of CO2 by AMP. Int. J. Greenhouse Gas Control 3, 612–616.

Jakobsen, J.P., da Silva, E.F., et al., 2008. NMR study and quantum mechanicalcalculations on the 2-[(2-aminoethyl)amino]ethanol–H2O–CO2 system. J.Magn. Reson. 191 (2), 304–314.

Li, J., Henni, A., et al., 2007. Reaction kinetics of CO2 in aqueous ethylenediamine,ethyl ethanolamine, and diethyl monoethanolamine solutions in the tempera-ture range of 298–313 K, using the stopped-flow technique. Ind. Eng. Chem. Res.46 (13), 4426–4434.

Maeda, M., Hisada, O., et al., 1987. Estimation of salt and temperature effects on ionproduct of water in aqueous solution. Bull. Chem. Soc. Jpn. 60 (9), 3233–3239.

Mani, F., Peruzzini, M., et al., 2006. CO2 absorption by aqueous NH3 solutions:speciation of ammonium carbamate. Bicarbonate and carbonate by a 13C NMRstudy. Green Chem. 8 (11), 995–1000.

McCann, N., Maeder, M., et al., 2008. Simulation of enthalpy and capacity ofCO2 absorption by aqueous amine systems. Ind. Eng. Chem. Res. 47, 2002–2009.

McCann, N., Phan, D., et al., 2009. Kinetics and mechanism of carbamate formationfrom CO2(aq), carbonate species, and monoethanolamine in aqueous solution. J.Phys. Chem. A 113, 5022–5029.

Metz, B., Davidson, O., et al., 2005. IPCC Special Report on Carbon Dioxide Captureand Storage. Prepared by Working Group III of the Intergovernmental Panel onClimate Change. Cambridge University Press, Cambridge, UK.

Mikkelsen, M., Jørgensen, M., et al., 2010. The teraton challenge. A review of fixationand transformation of carbon dioxide. Energy Environ. Sci. 3, 43–81.

Mikkelsen, M., Jørgensen, M., et al., 2009. Synthesis and characterisation of zwit-terionic carbon dioxide fixing reagents. Int. J. Greenhouse Gas Control.

Park, S.H., Lee, K.B., et al., 2002. Correlation and prediction of the solubility of carbondioxide in aqueous alkanolamine and mixed alkanolamine solutions. Ind. Eng.Chem. Res. 41 (6), 1658–1665.

Poplsteinova Jakobsen, J., Krane, J., et al., 2005. Liquid-phase composition deter-mination in CO2–H2O–alkanolamine systems: an NMR study. Ind. Eng. Chem.Res. 44 (26), 9894–9903.

Sartori, G., Savage, D.W., 1983. Sterically hindered amines for CO2 removal fromgases. Ind. Eng. Chem. Fundam. 22, 239–249.

Soli, A.L., Byrne, R.H., 2002. CO2 system hydration and dehydration kinetics and theequilibrium CO2/H2CO3 ratio in aqueous NaCl solution. Mar. Chem. 78 (2–3),65–73.

Stern, K.H., Amis, E.S., 1959. Ionic size. Chem. Rev. 1–64.Stern, N., 2006. The Economics of Climate Change. www.hm-treasury.gov.uk.Timimi, B.A., Everett, D.H., 1968. Thermodynamics of the acid dissociation of some

aminoalcohols in water. J. Chem. Soc. B 11, 1380–1386.Wise, M.A., Dooley, J.J., 2009. The value of post-combustion carbon dioxide capture

and storage technologies in a world with uncertain greenhouse gas emissionsconstraints. Int. J. Greenhouse Gas Control 3 (1), 39–48.