- Email: [email protected]
Multiple headspace extraction for gas detection in ionic liquids
Journal of Chromatography A, 1371 (2014) 15–19
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Multiple headspace extraction for gas detection in ionic liquids D. Müller, M. Fühl, K. Pinkwart, N. Baltes ∗ Fraunhofer Institut für Chemische Technologie, Joseph-von-Fraunhofer-Str. 7, D-76327 Pfinztal (Berghausen), Germany
a r t i c l e
i n f o
Article history: Received 29 July 2014 Received in revised form 6 October 2014 Accepted 7 October 2014 Available online 16 October 2014 Keywords: Ionic liquids Multiple headspace extraction Gas solubility Oxygen Carbon monoxide Ethylmethylimidazolium tetracyanoborate
a b s t r a c t In this study multiple headspace extraction was used for the first time to measure the saturation concentration of carbon monoxide and oxygen in various ionic liquids (ILs). Many processes in ILs involve the reaction of gases so that the reactant solubility is not a mere characteristical parameter, but understanding the solubility of gases in ILs is required for assessing the feasibility of possible applications. Multiple headspace extraction has proofed to be a powerful tool to obtain solubilities in good accordance with literature data. The measured saturation concentration for carbon monoxide and oxygen in ILs based on rarely researched tetracyanoborates and other anions was in the range of 1.5–6.5 mmol/L. The great advantage of multiple headspace extraction is that it is a nonexpensive method that can be realised in most analytical laboratories by combination of a simple gas chromatograph and an eligible headspace injector. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Gas analysis is an important analytical field, with emphasis placed on advanced materials and new methodologies for realising a necessary and reasonable sensitivity at a low price. According to the great number of publications over the recent years, ionic liquids show a promising potential in sensor and methodology development. Ionic liquids (ILs) are in general described as solvents composed entirely of (mainly organic) ions and with melting points under 373.15 K. Many of them show good chemical stability, a low volatility, an almost non existing inflammability and furthermore a thermal robustness up to 473.15 K or even more (for a review see Refs. [1–3]). Moreover, they normally have a very low coefficient of thermal expansion – a very useful property that might prevent leakage problems in high temperature applications. The gas solubility as well as the mass transport of the gas within the IL differ strongly with the nature of its ions so that the efficiency of gas absorbing processes (e.g. in separation and extraction [4–10]) or gas involving, catalytic reactions (e.g. hydroformulation [11,12], carbonylation [13–18], etc.) depends on the choice of suitable combinations of cations and anions. Electrochemical IL based gas sensing methodologies [19–29] also profit of a preferably large uptake of the gas in question by the IL-electrolyte. Hence, several papers were published within the last decade, presenting different methods for the evaluation of gas solubilities in numerous ILs. Ohlin et al. [30] used
∗ Corresponding author. Tel.: +49 721 4640 868; fax: +49 721 4640 318. E-mail address: [email protected] (N. Baltes). http://dx.doi.org/10.1016/j.chroma.2014.10.017 0021-9673/© 2014 Elsevier B.V. All rights reserved.
NMR-spectroscopy to investigate the solubility of carbon monoxide (CO) in various ILs. This method is apparently quite sensitive and only restricted by the fact that the gas of interest needs to contain particular isotopes. Furthermore, NMR-spectroscopy is a rather sophisticated and expensive method which is not established in many common analytic laboratories. An alternative method was proposed by Anthony et al. who determined the solubility of several gases in ILs using a gravimetric microbalance [31,32]. The main drawback of this method is that low-molecular weight gases like hydrogen or carbon monoxide usually have solubilities below the detection limit of the microbalance. Another disadvantage is a certain error rate because of buoyancy effects [33]. Latter provoke that the IL weighs generally less in a gas atmosphere than in a vacuum – independent from any gas solvation behaviour. Kumelan et al. investigated the solubility of oxygen and CO by means of an optical method, working at high pressure [34–38]. Another pressure depending method is the isochoric saturation technique which was used by Jaquemin et al. in order to obtain further knowledge about gas solubilities and the corresponding thermodynamical properties of several sorts of gas in dependence on temperatures between 283 and 343 K [39–41]. Blath et al. [42] used the same method to show that the solubility behaviour of nitrogen, methane and carbon dioxide is mainly influenced by the ion size of the IL in question and not the ion type. Shifflet and co-workers, on the other hand, presented solubility diagrams of ammonia, sulphur dioxide, etc. (in dependency of the temperature) and compared experimental results with fits of a theoretical model based on Van der Waals equation [43] (a number of predictive thermodynamic models and experimental methods for measuring the solubility of gases in ILs as well as the
16
D. Müller et al. / J. Chromatogr. A 1371 (2014) 15–19
corresponding results is given in an extensive review by Chen and co-workers) [44]. In this paper we present gas detection in ILs on basis of multiple headspace extraction (MHE). The great advantage of MHE is that this technique can be easily applied by using merely a gas chromatograph and a suitable injector. Von Wald et al. [45] used MHE for trace detection of solved, organic impurities in ILs. The groups of Laus [46] and Zhang [47], too, provided evidence of slightly volatile solvents in drugs or other toxins in soil by means of MHE and ILs. But to our knowledge MHE has not been used to quantitatively determine any gas solubility in ILs. Here, the solubility of oxygen and CO was investigated in several gas saturated ILs at 298 K by detecting the gas concentrations above the IL in the presence of different testing gas concentrations. An addition plot finally revealed the unknown gas concentration. 2. Experimental 2.1. Reagents The ionic liquids 1-methyl-3-octylimidazolium bis(trifluoromethylsulphonyl)imide (OMIM NTf2 ), 1-methyl-1-propylpiperidinium bis(trifluoromethylsulphonyl)imide (PP13 NTf2 ) and trihexyl bis(trifluoromethylsulphonyl)imide (tetradecyl)phosphonium (P6,6,6,14 NTf2 ) were purchased from Iolitec, Germany. 1-Hexyl-3tris(pentafluoroethyl)trifluorophosphate methylimidazolium (HMIM FAP), 1-butyl-3-methylimidazolium tetracyanoborate (BMIM TCB), 1-ethyl-3-methylimidazolium tetracyanoborate (EMIM TCB) and 1-hexyl-3-methylimidazolium tetracyanoborate (HMIM TCB) were produced by the Merck Company, Germany. All chemicals were used without further purification (specification of the supplier: water (KF) < 100 ppm, halides < 100 ppm). The supplier of the testing gas (8000, 5000, 3000 ppm CO and 5000, 3000, 1000 ppm oxygen ±2% rel. in argon) was Air Liquide, Germany. 2.2. Sample preparation A volume of 17 ml of each liquid was sparged with argon (purity: 99.9999%) at a temperature of 413 K over night before CO or oxygen was purged through for at least 2 h at ambient temperature. Each gas was additionally dried by means of moisture traps (Molecular Sieve 0.3 nm by Merck) before being introduced via needles through the septum seal of the liquid containing vials (another needle worked as gas outlet in order to prevent any high pressure). Hereafter, a determined quantity (about 1 ml) of the gas purged IL was added to a sealed 20 ml crimp neck vial (La-Pha-Pack GmbH (Germany); silicone/aluminium-septa) containing pure argon (offset detection). Similarly, CO or oxygen saturated liquids were transferred into vials containing test gas (of CO or oxygen) of different concentrations by means of 2 ml syringes with extra big needles. Latter have been necessary in order to avoid outgassing because the pressure within the syringes might become too low otherwise. The stitching holes were sealed immediately by means of cyanoacrylate (Cyanolit by Eleco, France) as minute leakages result in random errors. It is therefore recommended to execute the following proceedings on schedule, too, so that the offset detection is reproducible. 2.3. Chromatographic conditions The analyses of the solutes were performed on a Perkin Elmer (Germany) Clarus 600 gas chromatograph system equipped with a Perkin Elmer Clarus 500 flame ionisation detector and a headspace autosampler Turbo Matrix 16 (Perkin Elmer). For the detection of carbon monoxide a methaniser was connected upstream operating
at 623 K with a constant flow rate of air (450 ml/min) and hydrogen (45 ml/min). The detection of oxygen occurred by means of a thermal conductivity detector working at 476 K. Furthermore, a ValcoPLOT capillary column (Valco Vici (USA); stainless steel; molecular sieve; length 30 m, inner diameter 0.53 mm; film thickness 20 m) was used. Helium (purity: 99.9999%) was used as carrier gas. At the headspace autosampler the incubation temperature was set to 313 K for 45 min (30 min for oxygen detection) with a head pressure of 30 psi. The autosampler was connected to the GC via a heated transfer line with a temperature of 393 K. The headspace syringe was set to 488 K. The pressurising time was 5 min and the injection time was 0.04 min. The injector at the GC was held at 333 K by working in splitless injection mode. The oven was operated with the following temperature programme: 313 K for 5 min, an increase in temperature to 473 K, at a rate of 15 K/min. The final temperature was hold for 2 min. 3. Results and discussion 3.1. Method and validation In order to determine the solubility of a certain gas in an IL, latter was dried, eventually saturated with the gas to be investigated and transferred into a headspace vial containing argon or testing gas. Assuming that partition and activity coefficient are constant and independent of analyte concentration one gets a linear relationship between the concentration of the analyte in the sample and its concentration in the headspace. The GC response therefore depends linearly on the solvated amount of gas. As MHE is carried out stepwise and the gas concentration decreasing continuously (following a first order mechanism) the measured peak area A can be described by [48]: Ai = A1 · e−q(i−1)
(1)
where q is a rate constant of the extraction process. The total peak area of consecutive analyses from the same sample vial, which is proportional to the total amount of gas originally solvated in the IL, can be described by the following equation:
i→∞
Atotal =
i=1
Ai =
A1 A1 = 1 − e−q 1−Q
(2)
Q is the area ratio of the consecutive peaks: Q =
A(i+1) A2 A3 = = = e−q A1 A2 Ai
(3)
Eq. (1) can be linearised by logarithmic calculation. The corresponding plot of ln (Ai ) vs. (i−1) (Fig. 1) shows a slope of −q and a y-intercept of ln (A1 ). Considering the initial weight of the IL-sample the weight-related total peak area (rel. Atotal ) can be calculated from such determined values by means of Eqs. (2) and (3). Fig. 2 shows a standard addition plot of total peak areas obtained from a series of repeating CO extractions performed in the presence of differently concentrated testing gas atmospheres above OMIM NTf2 . The saturation concentration was determined by the negative intersection of the fit with the abscissa (here 1.3 mmol/L). Yet, even if the squared regression coefficient of the linear fit is about 0.996, we found deviations between subsequent measurement series themselves. Such deviations might originate from too strong a pressure of the crimp pincers, which can result in uparching of the cap and the covering membrane and therefore in very small leakages. Another source of error might have been temperature effects on the syringe which was used to transfer the gas saturated IL into the crimp neck vial. Because of these uncertainties the solubility and the corresponding deviation were not determined by (weighted) linear
D. Müller et al. / J. Chromatogr. A 1371 (2014) 15–19
17
with t representing the t-value for n numbers of measurement series, a probability p and f = n − 1 degrees of freedom. The mean average value of 5 saturation concentrations (each single concentration was estimated as shown in Fig. 2) and the corresponding uncertainty are finally presented in Table 1 for a number of ILs. In order to calculate Henry’s law constants it was necessary to know the mole fraction and the density of the corresponding ILs. Latter was measured between 298 and 333 K in a dry glove box. Like Hayan et al. [50] we could not find any significant change within this temperature range meaning that the expansion coefficient is negligible small. The density values shown in Table 1 represent the average of five experimental results and usually show a standard deviation of 2%. 3.2. Solubility of carbon monoxide and oxygen
Fig. 1. MHE measurement above CO saturated OMIM NTf2 . The total peak area Atotal is 378.56 with Q = 0.2866 and A1 = 270.086. Concentration of the CO testing gas: 1.068 mmol/L.
Fig. 2. Single MHE measurement series above CO saturated OMIM NTf2 . Weightrelated total peak areas (symbols) in dependency on CO concentration in the testing gas. The standard addition plot (black line) shows a CO concentration of 1.3 mmol/L solvated in gas saturated IL.
regression of a collection of subsequent measurement series since the results of such an analysis can be unreliable and unpredictedly affected. Instead, the experimental uncertainty k for the gas solubility was estimated from the standard deviation of the average for a confidence interval of 75% [49]: s k = t(p;f ) √ n
(4)
The experimentally determined solubilities of CO in various ILs are shown in Table 1. Similar to the values measured by Blath et al. the first three values were dependent on the solvent in the same following order: P6,6,6,14 NTf2 > HMIM FAP > BMIM TCB [42]. It is noticeable, moreover, that all three tetracyanoborate based ILs seem to absorb less CO than the other ILs being investigated. A reason for such a behaviour might be the different macromolecular ionic structure and the presence of a certain conformational equilibrium in the ionic domains. The small differences of the solubility of the three TCB ILs suggest on the other hand that the solubility is rather insensitive to the cations chain length – a phenomenon also well known from carbon dioxide [33,51,9]. A proper investigation of the structures of ILs and their interactions with CO, however, is not a goal of this paper so that the effects on the solubility should be discussed elsewhere. For CO absorbed by OMIM NTf2 we found a Henry’s law constant of 990 bar at room temperature. According to Ohlin et al. [30], at 295 K, Henry’s constant of CO in OMIM NTf2 amounts to 670 bar, which is about 33% smaller than the number reported here. Only the solubility of HMIM FAP (540 bar) does not agree within the limit of error with the value measured by other groups (1670 bar) [42]. It is noteworthy, however, that all solubilities are in more or less the same range – a result that is confirmed by numerous reference data recorded in various ILs [30,35–37,39,41]. Hence, there is a high probability of overlap especially after consideration of the errors in measurement. This is a well known problem in trace analysis of gases in ILs, especially if the investigated gas shows a low solubility. Gravimetric microbalance measurements of oxygen dissolved in BMIM PF6 can show errors of more than 50% [31]. The relative high discrepancies in case of OMIM NTf2 and BMIM TCB in particular are supposed to be a consequence of minor differences of the experimental conditions. If minimal amounts of gas get lost through the headspace septum or due to mere experimental inobservance the differences of GC-signals will be easily affected. On the other hand,
Table 1 Physical solubility and properties of CO and O2 in a variation of ionic liquids at room temperature under normal conditions. Cation 1 2 3 5 6 7 8 9 a b c d e
HMIM EMIM P6,6,6,14 HMIM OMIM BMIM PP13 EMIM
Anion FAP TCB NTf2 TCB NTf2 TCB NTf2 TCB
Gas CO CO CO CO CO CO O2 O2
k = t(0.75;4) s/sqrt(n). Measured at room temperature. Henry’s constant, KH = PCO /CO , calculated for a partial pressure of 1 bar. From Ref. [41]. From Ref. [30].
CGas ± k/mM (n = 5) 4.91 1.85 6.46 2.01 2.94 1.55 5.67 4.34
± ± ± ± ± ± ± ±
a
1.05 0.67a 0.96a 0.71a 1.40a 0.88a 0.29a 0.31a
/g cm−3 b
KH /barc
1.61 1.07 1.12 1.01 1.37 1.02 1.44 1.07
540 2560 230 1790 990 2600 600 1090
± ± ± ± ± ± ± ±
130 (1670)d 980 40 (220)d 660 490 (670)e 1530 (1990)d 40 100
18
D. Müller et al. / J. Chromatogr. A 1371 (2014) 15–19
in case of a better gas solubility these minimal traces of escaped gas probably will not have such grave consequences so that the error of measurement e.g. in case of carbon dioxide is supposed to be smaller. In order to avoid e.g. outgas reactions in the syringe and to diminish experimental incertainties in general it is therefore advisable for future studies to automate as many processes as possible so that the experimental conditions (pressure, time, temperature) are more reproducible. Apart from experimental difficulties and the accuracy of the measurement technique, an important source of error can be generally found in the uncertainty of the IL-density and possible contaminations which results in direct effects on the solubility measurements and is related to the solubility of the tested gas [33]. As the density is easily influenced by contaminations like water it is recommended to use only ILs of a good quality which have been effectively dried. In case of oxygen trace analysis a possible perturbation might occur due to atmospheric oxygen which easily diffuses through tiny leaks into the oxygen free headspace vial. We therefore (a) sealed the headspace vials as good as possible, (b) used aluminium covered caps instead of the common teflon covered ones, (c) always recorded a blind measurement (filling the head space vials with pure nitrogen) and substracted the average of five total peak areas from the total peak areas obtained in the oxygen testing measurements and (d) followed a strict timetable in order to improve reproducibility. Following this procedure, a Henry’s law constant of 1.09 kbar was found for oxygen absorbed by EMIM TCB at 298 K. Again, this result corresponds quite well with values already found for the rather poor solubility of oxygen in other imidazolium based ILs [20,32]. 4. Conclusions and outlook In this work the absorbing capacity of carbon monoxide and oxygen in ionic liquids was quantitatively investigated for the first time by multiple headspace extraction. Both gases showed a solubility of 1–7 mmol/L in various ionic liquids including rarely researched tetracyanoborates. The obtained values were in good agreement with common literature data. Multiple headspace extraction has shown to be a powerful tool to estimate gas solubilities in ionic liquids. Thus, it is possible to obtain a great number of results in a nonexpensive and easy to realise way which might be especially interesting for statistical assessments, practical applications of ionic liquids as well as for calculating corresponding thermodynamical functions. Acknowledgements The authors would like to thank the Fraunhofer organisation for funding within the Fraunhofer project Ionic Liquids for Electrochemical Applications. We would also like to thank Mrs. Joy Bell for assistance in manuscript revising. References [1] Physical chemistry of ionic liquids, Phys. Chem. Chem. Phys. 12 (8) (2010). [2] H. Ohno, Electrochemical Aspects of Ionic Liquids, John Wiley and Sons, Inc., Hoboken, New Jersey, 2005. [3] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis, vols. 1, 2, Wiley-VCH Verlag GmbH, Weinheim, 2008. [4] A. Blahut, V. Dohnal, Interactions of volatile organic compounds with the ionic liquids 1-butyl-1-methylpyrrolidinium tetracyanoborate and 1-butyl1-methylpyrrolidinium bis(oxalato)borate, J. Chem. Thermodyn. 57 (2013) 344–354. [5] M.B. Shiflett, A. Yokozeki, Separation of CO2 and H2 S using room-temperature ionic liquid [bmim][PF6], Fluid Phase Equilib. 294 (1–2) (2010) 105–113. [6] T.W. Stephens, E. William, P. Twu, J.L. Anderson, G.A. Baker, M.H. Abraham, Correlation of the solubilizing abilities of 1-butyl-1-methylpiperidinium bis(trifluoromethylsulfonyl)imide and 1-butyl-1-methylpyrrolidinium tetracyanoborate, J. Solut. Chem. 41 (7) (2012) 1165–1184.
[7] H. Vatani, A.S. Yazdi, Ionic-liquid-mediated poly(dimethylsiloxane)-grafted carbon nanotube fiber prepared by the sol–gel technique for the head space solid-phase microextraction of methyl tert-butyl ether using GC, J. Sep. Sci. 37 (1–2) (2014) 127–134. [8] A. Yokozeki, M.B. Shiflett, Separation of carbon dioxide and sulfur dioxide gases using room-temperature ionic liquid [hmim][Tf2N], Energy Fuels 23 (2009) 4701–4708. [9] M.T. Mota-Martinez, M. Althuluth, M.C. Kroon, C.J. Peters, Solubility of carbon dioxide in the low-viscosity ionic liquid 1-hexyl-3-methylimidazolium tetracyanoborate, Fluid Phase Equilib. 332 (2012) 35–39. [10] M.B. Shiflett, A. Yokozeki, Separation of carbon dioxide and sulfur dioxide using room-temperature ionic liquid [bmim][MeSO4 ], Energy Fuels 24 (2010) 1001–1008. [11] A. Sharma, C. Julcour, A.A. Kelkar, R.M. Deshpande, H. Delmas, Mass transfer and solubility of CO and H-2 in ionic liquid. Case of [Bmim][PF6 ] with gas-inducing stirrer reactor, Ind. Eng. Chem. Res. 48 (8) (2009) 4075–4082. [12] M. Sobota, M. Schmid, M. Happel, M. Amende, F. Maier, H.P. Steinruck, N. Paape, P. Wasserscheid, M. Laurin, J.M. Gottfried, J. Libuda, Ionic liquid based model catalysis: interaction of [BMIM][Tf2 N] with Pd nanoparticles supported on an ordered alumina film, Phys. Chem. Chem. Phys. 12 (35) (2010) 10610–10611. [13] J.X. Li, S.R. Yang, W.Q. Wu, H.F. Jiang, Novel palladium-catalyzed cascade carboxylative annulation to construct functionalized gamma-lactones in ionic liquids, Chem. Commun. 50 (11) (2014) 1381–1383. [14] Y.Z. Lei, R. Zhang, L.J. Wu, Q. Wu, H. Mei, G.X. Li, Palladium-catalyzed carbonylation of quaternary ammonium halides to tertiary amides, Appl. Organomet. Chem. 28 (4) (2014) 310–314. [15] H.Y. Song, F.X. Jing, R.H. Jin, Z. Li, J. Chen, Novel functional ionic liquids as metalfree, efficient and recyclable catalysts for the carbonylation of formaldehyde, Catal. Lett. 144 (4) (2014) 711–716. [16] Q.H. Zhang, F. Shi, Y.L. Gu, J. Yang, Y.Q. Deng, Efficient and eco-friendly process for the synthesis of N-substituted 4-methylene-2-oxazolidinones in ionic liquids, Tetrahedron Lett. 46 (35) (2005) 5907–5911. [17] T. Fukuyama, T. Inouye, I. Ryu, Atom transfer carbonylation using ionic liquids as reaction media, J. Organomet. Chem. 692 (1–3) (2007) 685–690. [18] Y.B. Ma, F. Shi, Y.Q. Deng, Oxidative carbonylation of aniline with a mesoporous silica gel immobilised Se-functionalised ionic liquid catalyst, J. Chem. Res. 6 (2010) 344–347. [19] M.C. Buzzeo, R.G. Evans, R.G. Compton, Non-haloaluminate room-temperature ionic liquids in electrochemistry – a review, ChemPhysChem 5 (8) (2004) 1106–1120. [20] M.C. Buzzeo, C. Hardacre, R.G. Compton, Use of room temperature ionic liquids in gas sensor design, Anal. Chem. 76 (15) (2004) 4583–4588. [21] N. Dossi, R. Toniolo, A. Pizzariello, E. Carrilho, E. Piccin, S. Battiston, G. Bontempelli, An electrochemical gas sensor based on paper supported room temperature ionic liquids, Lab Chip 12 (1) (2012) 153–158. [22] X.J. Huang, L. Aldous, A.M. O’Mahony, F.J. del Campo, R.G. Compton, Toward membrane-free amperometric gas sensors: a microelectrode array approach, Anal. Chem. 82 (12) (2010) 5238–5245. [23] X.X. Jin, L. Yu, D. Garcia, R.X. Ren, X.Q. Zeng, Ionic liquid high-temperature gas sensor array, Anal. Chem. 78 (19) (2006) 6980–6989. [24] R. Wang, T. Okajima, F. Kitamura, T. Ohsaka, A novel amperometric O-2 gas sensor based on supported room-temperature ionic liquid porous polyethylene membrane-coated electrodes, Electroanalysis 16 (1–2) (2004) 66–72. [25] R. Wang, S. Hoyano, T. Ohsaka, O2 gas sensor using supported hydrophobic room-temperature ionic liquid membrane-coated electrode, Chem. Lett. 33 (1) (2004) 6–7. [26] Z. Wang, P. Lin, G.A. Baker, J. Stetter, X. Zeng, Ionic liquids as electrolytes for the development of a robust amperometric oxygen sensor, Anal. Chem. 83 (18) (2011) 7066–7073. [27] S.Q. Xiong, Y. Wei, Z. Guo, X. Chen, J. Wang, J.H. Liu, X.J. Huang, Toward membrane-free amperometric gas sensors: an ionic liquid-nanoparticle composite approach, J. Phys. Chem. C 115 (35) (2011) 17471–17481. [28] N. Baltes, F. Beyle, S. Freiner, F. Geier, M. Joos, K. Pinkwart, P. Rabenecker, Trace detection of oxygen – ionic liquids in gas sensor design, Talanta 116 (2013) 474–481. [29] J. Ghilane, C. Lagrost, P. Hapiot, Scanning electrochemical microscopy in nonusual solvents: inequality of diffusion coefficients problem, Anal. Chem. 79 (19) (2007) 7383–7391. [30] C.A. Ohlin, P.J. Dyson, G. Laurenczy, Carbon monoxide solubility in ionic liquids: determination, prediction and relevance to hydroformylation, Chem. Commun. 9 (2004) 1070–1071. [31] J.L. Anthony, E.J. Maginn, J.F. Brennecke, Solubilities and thermodynamic properties of gases in the ionic liquid 1-n-butyl-3-methylimidazolium hexafluorophosphate, J. Phys. Chem. B 106 (29) (2002) 7315–7320. [32] J.L. Anthony, J.L. Anderson, E.J. Maginn, J.F. Brennecke, Anion effects on gas solubility in ionic liquids, J. Phys. Chem. B 109 (13) (2005) 6366–6374. [33] Y.F. Hu, Z.C. Liu, C.M. Xu, X.M. Zhang, The molecular characteristics dominating the solubility of gases in ionic liquids, Chem. Soc. Rev. 40 (7) (2011) 3802–3823. [34] J. Kumelan, A.P.S. Kamps, I. Urukova, D. Tuma, G. Maurer, Solubility of oxygen in the ionic liquid [bmim][PF6]: experimental and molecular simulation results, J. Chem. Thermodyn. 37 (6) (2005) 595–602. [35] J. Kumelan, A.P.S. Kamps, D. Tuma, G. Maurer, Solubility of CO in the ionic liquid [bmim][PF6 ], Fluid Phase Equilib. 228 (2005) 207–211. [36] J. Kumelan, A.P.S. Kamps, D. Tuma, G. Maurer, Solubility of the single gases H-2 and CO in the ionic liquid [bmim] [CH3 SO4 ], Fluid Phase Equilib. 260 (1) (2007) 3–8.
D. Müller et al. / J. Chromatogr. A 1371 (2014) 15–19 [37] J. Kumelan, A.P.S. Kamps, D. Tuma, G. Maurer, Solubility of the single gases carbon monoxide and oxygen in the ionic liquid [hmim][Tf2 N], J. Chem. Eng. Data 54 (3) (2009) 966–971. [38] J. Kumelan, D. Tuma, G. Maurer, Simultaneous solubility of carbon dioxide and hydrogen in the ionic liquid [hmim][Tf2 N]: experimental results and correlation, Fluid Phase Equilib. 311 (2011) 9–16. [39] J. Jacquemin, P. Husson, V. Majer, M.F.C. Gomes, Low-pressure solubilities and thermodynamics of solvation of eight gases in 1-butyl-3-methylimidazolium hexafluorophosphate, Fluid Phase Equilib. 240 (1) (2006) 87–95. [40] J. Jacquemin, P. Husson, A.A.H. Padua, V. Majer, Density and viscosity of several pure and water-saturated ionic liquids, Green Chem. 8 (2) (2006) 172–180. [41] J. Jacquemin, M.F.C. Gomes, P. Husson, V. Majer, Solubility of carbon dioxide, ethane, methane, oxygen, nitrogen, hydrogen, argon, and carbon monoxide in 1-butyl-3-methylimidazolium tetrafluoroborate between temperatures 283 K and 343 K and at pressures close to atmospheric, J. Chem. Thermodyn. 38 (4) (2006) 490–502. [42] J. Blath, M. Christ, N. Deubler, T. Hirth, T. Schiestel, Gas solubilities in room temperature ionic liquids – correlation between RTiL-molar mass and Henry’s law constant, Chem. Eng. J. 172 (1) (2011) 167–176. [43] A. Yokozeki, M.B. Shiflett, Gas solubilities in ionic liquids using a generic van der Waals equation of state, J. Supercrit. Fluids 55 (2) (2010) 846–851. [44] Z.G. Lei, C.N. Dai, B.H. Chen, Gas solubility in ionic liquids, Chem. Rev. 114 (2) (2014) 1289–1326.
19
[45] G. Von Wald, D. Albers, H. Cortes, T. McCabe, Background vapor from six ionic liquids and the partition coefficients and limits of detection for 10 different analytes in those ionic liquids measured using headspace gas chromatography, J. Chromatogr. A 1201 (1) (2008) 15–20. [46] G. Laus, M. Andre, G. Bentivoglio, H. Schottenberger, Ionic liquids as superior solvents for headspace gas chromatography of residual solvents with very low vapor pressure, relevant for pharmaceutical final dosage forms, J. Chromatogr. A 1216 (32) (2009) 6020–6023. [47] J. Zhang, H.K. Lee, Headspace ionic liquid-based microdrop liquid-phase microextraction followed by microdrop thermal desorption-gas chromatographic analysis, Talanta 81 (1–2) (2010) 537–542. [48] B. Kolb, L.S. Ettre, Static Headspace-Gas Chromatography, John Wiley & Sons, Inc., 2006. [49] Lothar Papula Mathematik für Ingenieure und Naturwissenschaftler, Band 3, Friedr. Vieweg & Sohn Verlagsgesellschaft mbH, Braunschweig/Wiesbaden, 1997. [50] M. Hayyan, F.S. Mjalli, M.A. Hashim, I.M. AlNashef, X.M. Tan, Electrochemical reduction of dioxygen in bis (trifluoromethylsulfonyl) imide based ionic liquids, J. Electroanal. Chem. 657 (1–2) (2011) 150–157. [51] R. Babarao, S. Dai, D.E. Jiang, Understanding the high solubility of CO2 in an ionic liquid with the tetracyanoborate anion, J. Phys. Chem. B 115 (32) (2011) 9789–9794.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Multiple headspace extraction for gas detection in ionic liquids D. Müller, M. Fühl, K. Pinkwart, N. Baltes ∗ Fraunhofer Institut für Chemische Technologie, Joseph-von-Fraunhofer-Str. 7, D-76327 Pfinztal (Berghausen), Germany
a r t i c l e
i n f o
Article history: Received 29 July 2014 Received in revised form 6 October 2014 Accepted 7 October 2014 Available online 16 October 2014 Keywords: Ionic liquids Multiple headspace extraction Gas solubility Oxygen Carbon monoxide Ethylmethylimidazolium tetracyanoborate
a b s t r a c t In this study multiple headspace extraction was used for the first time to measure the saturation concentration of carbon monoxide and oxygen in various ionic liquids (ILs). Many processes in ILs involve the reaction of gases so that the reactant solubility is not a mere characteristical parameter, but understanding the solubility of gases in ILs is required for assessing the feasibility of possible applications. Multiple headspace extraction has proofed to be a powerful tool to obtain solubilities in good accordance with literature data. The measured saturation concentration for carbon monoxide and oxygen in ILs based on rarely researched tetracyanoborates and other anions was in the range of 1.5–6.5 mmol/L. The great advantage of multiple headspace extraction is that it is a nonexpensive method that can be realised in most analytical laboratories by combination of a simple gas chromatograph and an eligible headspace injector. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Gas analysis is an important analytical field, with emphasis placed on advanced materials and new methodologies for realising a necessary and reasonable sensitivity at a low price. According to the great number of publications over the recent years, ionic liquids show a promising potential in sensor and methodology development. Ionic liquids (ILs) are in general described as solvents composed entirely of (mainly organic) ions and with melting points under 373.15 K. Many of them show good chemical stability, a low volatility, an almost non existing inflammability and furthermore a thermal robustness up to 473.15 K or even more (for a review see Refs. [1–3]). Moreover, they normally have a very low coefficient of thermal expansion – a very useful property that might prevent leakage problems in high temperature applications. The gas solubility as well as the mass transport of the gas within the IL differ strongly with the nature of its ions so that the efficiency of gas absorbing processes (e.g. in separation and extraction [4–10]) or gas involving, catalytic reactions (e.g. hydroformulation [11,12], carbonylation [13–18], etc.) depends on the choice of suitable combinations of cations and anions. Electrochemical IL based gas sensing methodologies [19–29] also profit of a preferably large uptake of the gas in question by the IL-electrolyte. Hence, several papers were published within the last decade, presenting different methods for the evaluation of gas solubilities in numerous ILs. Ohlin et al. [30] used
∗ Corresponding author. Tel.: +49 721 4640 868; fax: +49 721 4640 318. E-mail address: [email protected] (N. Baltes). http://dx.doi.org/10.1016/j.chroma.2014.10.017 0021-9673/© 2014 Elsevier B.V. All rights reserved.
NMR-spectroscopy to investigate the solubility of carbon monoxide (CO) in various ILs. This method is apparently quite sensitive and only restricted by the fact that the gas of interest needs to contain particular isotopes. Furthermore, NMR-spectroscopy is a rather sophisticated and expensive method which is not established in many common analytic laboratories. An alternative method was proposed by Anthony et al. who determined the solubility of several gases in ILs using a gravimetric microbalance [31,32]. The main drawback of this method is that low-molecular weight gases like hydrogen or carbon monoxide usually have solubilities below the detection limit of the microbalance. Another disadvantage is a certain error rate because of buoyancy effects [33]. Latter provoke that the IL weighs generally less in a gas atmosphere than in a vacuum – independent from any gas solvation behaviour. Kumelan et al. investigated the solubility of oxygen and CO by means of an optical method, working at high pressure [34–38]. Another pressure depending method is the isochoric saturation technique which was used by Jaquemin et al. in order to obtain further knowledge about gas solubilities and the corresponding thermodynamical properties of several sorts of gas in dependence on temperatures between 283 and 343 K [39–41]. Blath et al. [42] used the same method to show that the solubility behaviour of nitrogen, methane and carbon dioxide is mainly influenced by the ion size of the IL in question and not the ion type. Shifflet and co-workers, on the other hand, presented solubility diagrams of ammonia, sulphur dioxide, etc. (in dependency of the temperature) and compared experimental results with fits of a theoretical model based on Van der Waals equation [43] (a number of predictive thermodynamic models and experimental methods for measuring the solubility of gases in ILs as well as the
16
D. Müller et al. / J. Chromatogr. A 1371 (2014) 15–19
corresponding results is given in an extensive review by Chen and co-workers) [44]. In this paper we present gas detection in ILs on basis of multiple headspace extraction (MHE). The great advantage of MHE is that this technique can be easily applied by using merely a gas chromatograph and a suitable injector. Von Wald et al. [45] used MHE for trace detection of solved, organic impurities in ILs. The groups of Laus [46] and Zhang [47], too, provided evidence of slightly volatile solvents in drugs or other toxins in soil by means of MHE and ILs. But to our knowledge MHE has not been used to quantitatively determine any gas solubility in ILs. Here, the solubility of oxygen and CO was investigated in several gas saturated ILs at 298 K by detecting the gas concentrations above the IL in the presence of different testing gas concentrations. An addition plot finally revealed the unknown gas concentration. 2. Experimental 2.1. Reagents The ionic liquids 1-methyl-3-octylimidazolium bis(trifluoromethylsulphonyl)imide (OMIM NTf2 ), 1-methyl-1-propylpiperidinium bis(trifluoromethylsulphonyl)imide (PP13 NTf2 ) and trihexyl bis(trifluoromethylsulphonyl)imide (tetradecyl)phosphonium (P6,6,6,14 NTf2 ) were purchased from Iolitec, Germany. 1-Hexyl-3tris(pentafluoroethyl)trifluorophosphate methylimidazolium (HMIM FAP), 1-butyl-3-methylimidazolium tetracyanoborate (BMIM TCB), 1-ethyl-3-methylimidazolium tetracyanoborate (EMIM TCB) and 1-hexyl-3-methylimidazolium tetracyanoborate (HMIM TCB) were produced by the Merck Company, Germany. All chemicals were used without further purification (specification of the supplier: water (KF) < 100 ppm, halides < 100 ppm). The supplier of the testing gas (8000, 5000, 3000 ppm CO and 5000, 3000, 1000 ppm oxygen ±2% rel. in argon) was Air Liquide, Germany. 2.2. Sample preparation A volume of 17 ml of each liquid was sparged with argon (purity: 99.9999%) at a temperature of 413 K over night before CO or oxygen was purged through for at least 2 h at ambient temperature. Each gas was additionally dried by means of moisture traps (Molecular Sieve 0.3 nm by Merck) before being introduced via needles through the septum seal of the liquid containing vials (another needle worked as gas outlet in order to prevent any high pressure). Hereafter, a determined quantity (about 1 ml) of the gas purged IL was added to a sealed 20 ml crimp neck vial (La-Pha-Pack GmbH (Germany); silicone/aluminium-septa) containing pure argon (offset detection). Similarly, CO or oxygen saturated liquids were transferred into vials containing test gas (of CO or oxygen) of different concentrations by means of 2 ml syringes with extra big needles. Latter have been necessary in order to avoid outgassing because the pressure within the syringes might become too low otherwise. The stitching holes were sealed immediately by means of cyanoacrylate (Cyanolit by Eleco, France) as minute leakages result in random errors. It is therefore recommended to execute the following proceedings on schedule, too, so that the offset detection is reproducible. 2.3. Chromatographic conditions The analyses of the solutes were performed on a Perkin Elmer (Germany) Clarus 600 gas chromatograph system equipped with a Perkin Elmer Clarus 500 flame ionisation detector and a headspace autosampler Turbo Matrix 16 (Perkin Elmer). For the detection of carbon monoxide a methaniser was connected upstream operating
at 623 K with a constant flow rate of air (450 ml/min) and hydrogen (45 ml/min). The detection of oxygen occurred by means of a thermal conductivity detector working at 476 K. Furthermore, a ValcoPLOT capillary column (Valco Vici (USA); stainless steel; molecular sieve; length 30 m, inner diameter 0.53 mm; film thickness 20 m) was used. Helium (purity: 99.9999%) was used as carrier gas. At the headspace autosampler the incubation temperature was set to 313 K for 45 min (30 min for oxygen detection) with a head pressure of 30 psi. The autosampler was connected to the GC via a heated transfer line with a temperature of 393 K. The headspace syringe was set to 488 K. The pressurising time was 5 min and the injection time was 0.04 min. The injector at the GC was held at 333 K by working in splitless injection mode. The oven was operated with the following temperature programme: 313 K for 5 min, an increase in temperature to 473 K, at a rate of 15 K/min. The final temperature was hold for 2 min. 3. Results and discussion 3.1. Method and validation In order to determine the solubility of a certain gas in an IL, latter was dried, eventually saturated with the gas to be investigated and transferred into a headspace vial containing argon or testing gas. Assuming that partition and activity coefficient are constant and independent of analyte concentration one gets a linear relationship between the concentration of the analyte in the sample and its concentration in the headspace. The GC response therefore depends linearly on the solvated amount of gas. As MHE is carried out stepwise and the gas concentration decreasing continuously (following a first order mechanism) the measured peak area A can be described by [48]: Ai = A1 · e−q(i−1)
(1)
where q is a rate constant of the extraction process. The total peak area of consecutive analyses from the same sample vial, which is proportional to the total amount of gas originally solvated in the IL, can be described by the following equation:
i→∞
Atotal =
i=1
Ai =
A1 A1 = 1 − e−q 1−Q
(2)
Q is the area ratio of the consecutive peaks: Q =
A(i+1) A2 A3 = = = e−q A1 A2 Ai
(3)
Eq. (1) can be linearised by logarithmic calculation. The corresponding plot of ln (Ai ) vs. (i−1) (Fig. 1) shows a slope of −q and a y-intercept of ln (A1 ). Considering the initial weight of the IL-sample the weight-related total peak area (rel. Atotal ) can be calculated from such determined values by means of Eqs. (2) and (3). Fig. 2 shows a standard addition plot of total peak areas obtained from a series of repeating CO extractions performed in the presence of differently concentrated testing gas atmospheres above OMIM NTf2 . The saturation concentration was determined by the negative intersection of the fit with the abscissa (here 1.3 mmol/L). Yet, even if the squared regression coefficient of the linear fit is about 0.996, we found deviations between subsequent measurement series themselves. Such deviations might originate from too strong a pressure of the crimp pincers, which can result in uparching of the cap and the covering membrane and therefore in very small leakages. Another source of error might have been temperature effects on the syringe which was used to transfer the gas saturated IL into the crimp neck vial. Because of these uncertainties the solubility and the corresponding deviation were not determined by (weighted) linear
D. Müller et al. / J. Chromatogr. A 1371 (2014) 15–19
17
with t representing the t-value for n numbers of measurement series, a probability p and f = n − 1 degrees of freedom. The mean average value of 5 saturation concentrations (each single concentration was estimated as shown in Fig. 2) and the corresponding uncertainty are finally presented in Table 1 for a number of ILs. In order to calculate Henry’s law constants it was necessary to know the mole fraction and the density of the corresponding ILs. Latter was measured between 298 and 333 K in a dry glove box. Like Hayan et al. [50] we could not find any significant change within this temperature range meaning that the expansion coefficient is negligible small. The density values shown in Table 1 represent the average of five experimental results and usually show a standard deviation of 2%. 3.2. Solubility of carbon monoxide and oxygen
Fig. 1. MHE measurement above CO saturated OMIM NTf2 . The total peak area Atotal is 378.56 with Q = 0.2866 and A1 = 270.086. Concentration of the CO testing gas: 1.068 mmol/L.
Fig. 2. Single MHE measurement series above CO saturated OMIM NTf2 . Weightrelated total peak areas (symbols) in dependency on CO concentration in the testing gas. The standard addition plot (black line) shows a CO concentration of 1.3 mmol/L solvated in gas saturated IL.
regression of a collection of subsequent measurement series since the results of such an analysis can be unreliable and unpredictedly affected. Instead, the experimental uncertainty k for the gas solubility was estimated from the standard deviation of the average for a confidence interval of 75% [49]: s k = t(p;f ) √ n
(4)
The experimentally determined solubilities of CO in various ILs are shown in Table 1. Similar to the values measured by Blath et al. the first three values were dependent on the solvent in the same following order: P6,6,6,14 NTf2 > HMIM FAP > BMIM TCB [42]. It is noticeable, moreover, that all three tetracyanoborate based ILs seem to absorb less CO than the other ILs being investigated. A reason for such a behaviour might be the different macromolecular ionic structure and the presence of a certain conformational equilibrium in the ionic domains. The small differences of the solubility of the three TCB ILs suggest on the other hand that the solubility is rather insensitive to the cations chain length – a phenomenon also well known from carbon dioxide [33,51,9]. A proper investigation of the structures of ILs and their interactions with CO, however, is not a goal of this paper so that the effects on the solubility should be discussed elsewhere. For CO absorbed by OMIM NTf2 we found a Henry’s law constant of 990 bar at room temperature. According to Ohlin et al. [30], at 295 K, Henry’s constant of CO in OMIM NTf2 amounts to 670 bar, which is about 33% smaller than the number reported here. Only the solubility of HMIM FAP (540 bar) does not agree within the limit of error with the value measured by other groups (1670 bar) [42]. It is noteworthy, however, that all solubilities are in more or less the same range – a result that is confirmed by numerous reference data recorded in various ILs [30,35–37,39,41]. Hence, there is a high probability of overlap especially after consideration of the errors in measurement. This is a well known problem in trace analysis of gases in ILs, especially if the investigated gas shows a low solubility. Gravimetric microbalance measurements of oxygen dissolved in BMIM PF6 can show errors of more than 50% [31]. The relative high discrepancies in case of OMIM NTf2 and BMIM TCB in particular are supposed to be a consequence of minor differences of the experimental conditions. If minimal amounts of gas get lost through the headspace septum or due to mere experimental inobservance the differences of GC-signals will be easily affected. On the other hand,
Table 1 Physical solubility and properties of CO and O2 in a variation of ionic liquids at room temperature under normal conditions. Cation 1 2 3 5 6 7 8 9 a b c d e
HMIM EMIM P6,6,6,14 HMIM OMIM BMIM PP13 EMIM
Anion FAP TCB NTf2 TCB NTf2 TCB NTf2 TCB
Gas CO CO CO CO CO CO O2 O2
k = t(0.75;4) s/sqrt(n). Measured at room temperature. Henry’s constant, KH = PCO /CO , calculated for a partial pressure of 1 bar. From Ref. [41]. From Ref. [30].
CGas ± k/mM (n = 5) 4.91 1.85 6.46 2.01 2.94 1.55 5.67 4.34
± ± ± ± ± ± ± ±
a
1.05 0.67a 0.96a 0.71a 1.40a 0.88a 0.29a 0.31a
/g cm−3 b
KH /barc
1.61 1.07 1.12 1.01 1.37 1.02 1.44 1.07
540 2560 230 1790 990 2600 600 1090
± ± ± ± ± ± ± ±
130 (1670)d 980 40 (220)d 660 490 (670)e 1530 (1990)d 40 100
18
D. Müller et al. / J. Chromatogr. A 1371 (2014) 15–19
in case of a better gas solubility these minimal traces of escaped gas probably will not have such grave consequences so that the error of measurement e.g. in case of carbon dioxide is supposed to be smaller. In order to avoid e.g. outgas reactions in the syringe and to diminish experimental incertainties in general it is therefore advisable for future studies to automate as many processes as possible so that the experimental conditions (pressure, time, temperature) are more reproducible. Apart from experimental difficulties and the accuracy of the measurement technique, an important source of error can be generally found in the uncertainty of the IL-density and possible contaminations which results in direct effects on the solubility measurements and is related to the solubility of the tested gas [33]. As the density is easily influenced by contaminations like water it is recommended to use only ILs of a good quality which have been effectively dried. In case of oxygen trace analysis a possible perturbation might occur due to atmospheric oxygen which easily diffuses through tiny leaks into the oxygen free headspace vial. We therefore (a) sealed the headspace vials as good as possible, (b) used aluminium covered caps instead of the common teflon covered ones, (c) always recorded a blind measurement (filling the head space vials with pure nitrogen) and substracted the average of five total peak areas from the total peak areas obtained in the oxygen testing measurements and (d) followed a strict timetable in order to improve reproducibility. Following this procedure, a Henry’s law constant of 1.09 kbar was found for oxygen absorbed by EMIM TCB at 298 K. Again, this result corresponds quite well with values already found for the rather poor solubility of oxygen in other imidazolium based ILs [20,32]. 4. Conclusions and outlook In this work the absorbing capacity of carbon monoxide and oxygen in ionic liquids was quantitatively investigated for the first time by multiple headspace extraction. Both gases showed a solubility of 1–7 mmol/L in various ionic liquids including rarely researched tetracyanoborates. The obtained values were in good agreement with common literature data. Multiple headspace extraction has shown to be a powerful tool to estimate gas solubilities in ionic liquids. Thus, it is possible to obtain a great number of results in a nonexpensive and easy to realise way which might be especially interesting for statistical assessments, practical applications of ionic liquids as well as for calculating corresponding thermodynamical functions. Acknowledgements The authors would like to thank the Fraunhofer organisation for funding within the Fraunhofer project Ionic Liquids for Electrochemical Applications. We would also like to thank Mrs. Joy Bell for assistance in manuscript revising. References [1] Physical chemistry of ionic liquids, Phys. Chem. Chem. Phys. 12 (8) (2010). [2] H. Ohno, Electrochemical Aspects of Ionic Liquids, John Wiley and Sons, Inc., Hoboken, New Jersey, 2005. [3] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis, vols. 1, 2, Wiley-VCH Verlag GmbH, Weinheim, 2008. [4] A. Blahut, V. Dohnal, Interactions of volatile organic compounds with the ionic liquids 1-butyl-1-methylpyrrolidinium tetracyanoborate and 1-butyl1-methylpyrrolidinium bis(oxalato)borate, J. Chem. Thermodyn. 57 (2013) 344–354. [5] M.B. Shiflett, A. Yokozeki, Separation of CO2 and H2 S using room-temperature ionic liquid [bmim][PF6], Fluid Phase Equilib. 294 (1–2) (2010) 105–113. [6] T.W. Stephens, E. William, P. Twu, J.L. Anderson, G.A. Baker, M.H. Abraham, Correlation of the solubilizing abilities of 1-butyl-1-methylpiperidinium bis(trifluoromethylsulfonyl)imide and 1-butyl-1-methylpyrrolidinium tetracyanoborate, J. Solut. Chem. 41 (7) (2012) 1165–1184.
[7] H. Vatani, A.S. Yazdi, Ionic-liquid-mediated poly(dimethylsiloxane)-grafted carbon nanotube fiber prepared by the sol–gel technique for the head space solid-phase microextraction of methyl tert-butyl ether using GC, J. Sep. Sci. 37 (1–2) (2014) 127–134. [8] A. Yokozeki, M.B. Shiflett, Separation of carbon dioxide and sulfur dioxide gases using room-temperature ionic liquid [hmim][Tf2N], Energy Fuels 23 (2009) 4701–4708. [9] M.T. Mota-Martinez, M. Althuluth, M.C. Kroon, C.J. Peters, Solubility of carbon dioxide in the low-viscosity ionic liquid 1-hexyl-3-methylimidazolium tetracyanoborate, Fluid Phase Equilib. 332 (2012) 35–39. [10] M.B. Shiflett, A. Yokozeki, Separation of carbon dioxide and sulfur dioxide using room-temperature ionic liquid [bmim][MeSO4 ], Energy Fuels 24 (2010) 1001–1008. [11] A. Sharma, C. Julcour, A.A. Kelkar, R.M. Deshpande, H. Delmas, Mass transfer and solubility of CO and H-2 in ionic liquid. Case of [Bmim][PF6 ] with gas-inducing stirrer reactor, Ind. Eng. Chem. Res. 48 (8) (2009) 4075–4082. [12] M. Sobota, M. Schmid, M. Happel, M. Amende, F. Maier, H.P. Steinruck, N. Paape, P. Wasserscheid, M. Laurin, J.M. Gottfried, J. Libuda, Ionic liquid based model catalysis: interaction of [BMIM][Tf2 N] with Pd nanoparticles supported on an ordered alumina film, Phys. Chem. Chem. Phys. 12 (35) (2010) 10610–10611. [13] J.X. Li, S.R. Yang, W.Q. Wu, H.F. Jiang, Novel palladium-catalyzed cascade carboxylative annulation to construct functionalized gamma-lactones in ionic liquids, Chem. Commun. 50 (11) (2014) 1381–1383. [14] Y.Z. Lei, R. Zhang, L.J. Wu, Q. Wu, H. Mei, G.X. Li, Palladium-catalyzed carbonylation of quaternary ammonium halides to tertiary amides, Appl. Organomet. Chem. 28 (4) (2014) 310–314. [15] H.Y. Song, F.X. Jing, R.H. Jin, Z. Li, J. Chen, Novel functional ionic liquids as metalfree, efficient and recyclable catalysts for the carbonylation of formaldehyde, Catal. Lett. 144 (4) (2014) 711–716. [16] Q.H. Zhang, F. Shi, Y.L. Gu, J. Yang, Y.Q. Deng, Efficient and eco-friendly process for the synthesis of N-substituted 4-methylene-2-oxazolidinones in ionic liquids, Tetrahedron Lett. 46 (35) (2005) 5907–5911. [17] T. Fukuyama, T. Inouye, I. Ryu, Atom transfer carbonylation using ionic liquids as reaction media, J. Organomet. Chem. 692 (1–3) (2007) 685–690. [18] Y.B. Ma, F. Shi, Y.Q. Deng, Oxidative carbonylation of aniline with a mesoporous silica gel immobilised Se-functionalised ionic liquid catalyst, J. Chem. Res. 6 (2010) 344–347. [19] M.C. Buzzeo, R.G. Evans, R.G. Compton, Non-haloaluminate room-temperature ionic liquids in electrochemistry – a review, ChemPhysChem 5 (8) (2004) 1106–1120. [20] M.C. Buzzeo, C. Hardacre, R.G. Compton, Use of room temperature ionic liquids in gas sensor design, Anal. Chem. 76 (15) (2004) 4583–4588. [21] N. Dossi, R. Toniolo, A. Pizzariello, E. Carrilho, E. Piccin, S. Battiston, G. Bontempelli, An electrochemical gas sensor based on paper supported room temperature ionic liquids, Lab Chip 12 (1) (2012) 153–158. [22] X.J. Huang, L. Aldous, A.M. O’Mahony, F.J. del Campo, R.G. Compton, Toward membrane-free amperometric gas sensors: a microelectrode array approach, Anal. Chem. 82 (12) (2010) 5238–5245. [23] X.X. Jin, L. Yu, D. Garcia, R.X. Ren, X.Q. Zeng, Ionic liquid high-temperature gas sensor array, Anal. Chem. 78 (19) (2006) 6980–6989. [24] R. Wang, T. Okajima, F. Kitamura, T. Ohsaka, A novel amperometric O-2 gas sensor based on supported room-temperature ionic liquid porous polyethylene membrane-coated electrodes, Electroanalysis 16 (1–2) (2004) 66–72. [25] R. Wang, S. Hoyano, T. Ohsaka, O2 gas sensor using supported hydrophobic room-temperature ionic liquid membrane-coated electrode, Chem. Lett. 33 (1) (2004) 6–7. [26] Z. Wang, P. Lin, G.A. Baker, J. Stetter, X. Zeng, Ionic liquids as electrolytes for the development of a robust amperometric oxygen sensor, Anal. Chem. 83 (18) (2011) 7066–7073. [27] S.Q. Xiong, Y. Wei, Z. Guo, X. Chen, J. Wang, J.H. Liu, X.J. Huang, Toward membrane-free amperometric gas sensors: an ionic liquid-nanoparticle composite approach, J. Phys. Chem. C 115 (35) (2011) 17471–17481. [28] N. Baltes, F. Beyle, S. Freiner, F. Geier, M. Joos, K. Pinkwart, P. Rabenecker, Trace detection of oxygen – ionic liquids in gas sensor design, Talanta 116 (2013) 474–481. [29] J. Ghilane, C. Lagrost, P. Hapiot, Scanning electrochemical microscopy in nonusual solvents: inequality of diffusion coefficients problem, Anal. Chem. 79 (19) (2007) 7383–7391. [30] C.A. Ohlin, P.J. Dyson, G. Laurenczy, Carbon monoxide solubility in ionic liquids: determination, prediction and relevance to hydroformylation, Chem. Commun. 9 (2004) 1070–1071. [31] J.L. Anthony, E.J. Maginn, J.F. Brennecke, Solubilities and thermodynamic properties of gases in the ionic liquid 1-n-butyl-3-methylimidazolium hexafluorophosphate, J. Phys. Chem. B 106 (29) (2002) 7315–7320. [32] J.L. Anthony, J.L. Anderson, E.J. Maginn, J.F. Brennecke, Anion effects on gas solubility in ionic liquids, J. Phys. Chem. B 109 (13) (2005) 6366–6374. [33] Y.F. Hu, Z.C. Liu, C.M. Xu, X.M. Zhang, The molecular characteristics dominating the solubility of gases in ionic liquids, Chem. Soc. Rev. 40 (7) (2011) 3802–3823. [34] J. Kumelan, A.P.S. Kamps, I. Urukova, D. Tuma, G. Maurer, Solubility of oxygen in the ionic liquid [bmim][PF6]: experimental and molecular simulation results, J. Chem. Thermodyn. 37 (6) (2005) 595–602. [35] J. Kumelan, A.P.S. Kamps, D. Tuma, G. Maurer, Solubility of CO in the ionic liquid [bmim][PF6 ], Fluid Phase Equilib. 228 (2005) 207–211. [36] J. Kumelan, A.P.S. Kamps, D. Tuma, G. Maurer, Solubility of the single gases H-2 and CO in the ionic liquid [bmim] [CH3 SO4 ], Fluid Phase Equilib. 260 (1) (2007) 3–8.
D. Müller et al. / J. Chromatogr. A 1371 (2014) 15–19 [37] J. Kumelan, A.P.S. Kamps, D. Tuma, G. Maurer, Solubility of the single gases carbon monoxide and oxygen in the ionic liquid [hmim][Tf2 N], J. Chem. Eng. Data 54 (3) (2009) 966–971. [38] J. Kumelan, D. Tuma, G. Maurer, Simultaneous solubility of carbon dioxide and hydrogen in the ionic liquid [hmim][Tf2 N]: experimental results and correlation, Fluid Phase Equilib. 311 (2011) 9–16. [39] J. Jacquemin, P. Husson, V. Majer, M.F.C. Gomes, Low-pressure solubilities and thermodynamics of solvation of eight gases in 1-butyl-3-methylimidazolium hexafluorophosphate, Fluid Phase Equilib. 240 (1) (2006) 87–95. [40] J. Jacquemin, P. Husson, A.A.H. Padua, V. Majer, Density and viscosity of several pure and water-saturated ionic liquids, Green Chem. 8 (2) (2006) 172–180. [41] J. Jacquemin, M.F.C. Gomes, P. Husson, V. Majer, Solubility of carbon dioxide, ethane, methane, oxygen, nitrogen, hydrogen, argon, and carbon monoxide in 1-butyl-3-methylimidazolium tetrafluoroborate between temperatures 283 K and 343 K and at pressures close to atmospheric, J. Chem. Thermodyn. 38 (4) (2006) 490–502. [42] J. Blath, M. Christ, N. Deubler, T. Hirth, T. Schiestel, Gas solubilities in room temperature ionic liquids – correlation between RTiL-molar mass and Henry’s law constant, Chem. Eng. J. 172 (1) (2011) 167–176. [43] A. Yokozeki, M.B. Shiflett, Gas solubilities in ionic liquids using a generic van der Waals equation of state, J. Supercrit. Fluids 55 (2) (2010) 846–851. [44] Z.G. Lei, C.N. Dai, B.H. Chen, Gas solubility in ionic liquids, Chem. Rev. 114 (2) (2014) 1289–1326.
19
[45] G. Von Wald, D. Albers, H. Cortes, T. McCabe, Background vapor from six ionic liquids and the partition coefficients and limits of detection for 10 different analytes in those ionic liquids measured using headspace gas chromatography, J. Chromatogr. A 1201 (1) (2008) 15–20. [46] G. Laus, M. Andre, G. Bentivoglio, H. Schottenberger, Ionic liquids as superior solvents for headspace gas chromatography of residual solvents with very low vapor pressure, relevant for pharmaceutical final dosage forms, J. Chromatogr. A 1216 (32) (2009) 6020–6023. [47] J. Zhang, H.K. Lee, Headspace ionic liquid-based microdrop liquid-phase microextraction followed by microdrop thermal desorption-gas chromatographic analysis, Talanta 81 (1–2) (2010) 537–542. [48] B. Kolb, L.S. Ettre, Static Headspace-Gas Chromatography, John Wiley & Sons, Inc., 2006. [49] Lothar Papula Mathematik für Ingenieure und Naturwissenschaftler, Band 3, Friedr. Vieweg & Sohn Verlagsgesellschaft mbH, Braunschweig/Wiesbaden, 1997. [50] M. Hayyan, F.S. Mjalli, M.A. Hashim, I.M. AlNashef, X.M. Tan, Electrochemical reduction of dioxygen in bis (trifluoromethylsulfonyl) imide based ionic liquids, J. Electroanal. Chem. 657 (1–2) (2011) 150–157. [51] R. Babarao, S. Dai, D.E. Jiang, Understanding the high solubility of CO2 in an ionic liquid with the tetracyanoborate anion, J. Phys. Chem. B 115 (32) (2011) 9789–9794.