Anal. Chem. 2010, 82, 3856–3861
Modification and Implications of Changes in Electrochemical Responses Encountered When Undertaking Deoxygenation in Ionic Liquids Chuan Zhao,† Alan M. Bond,*,† Richard G. Compton,‡ Aoife M. O’Mahony,‡ and Emma I. Rogers‡ School of Chemistry and ARC Special Research Center for Green Chemistry, Monash University, Clayton, Victoria 3800, Australia, and Department of Chemistry, Physical and Theoretical Chemistry Laboratory, Oxford University, South Parks Road, Oxford, OX1 3QZ, United Kingdom Physicochemical changes and substantially modified electrochemical behavior have been reported when ionic liquids are degassed with nitrogen. In conventional experiments in aqueous and organic media, degassing with N2 is commonly used to remove the electroactive dissolved oxygen. However, in hydrophilic ionic liquid media, degassing with N2 removes not only the dissolved oxygen but also a significant amount of the adventitious water present. Given the low viscosity of water, this in turn leads to a dramatic change of the viscosity of the degassed ionic liquid and hence mass transport properties that influence voltammetric responses. In the widely used and relatively viscous room temperature ionic liquid, 1-n-butyl-3-methylimidazolium tetrafluoroborate (BMIMBF4) containing the redox probe tetracyanoquinodimethane (TCNQ) and 9% (v/v) deliberately added water, 1 h degassing with very dry N2 under benchtop conditions results in a dramatic decrease of the TCNQ reduction current obtained under steady-state conditions at a 1 µm diameter microdisc electrode. This is reflected by a change of diffusion coefficient of TCNQ (DTCNQ) from 2.6 × 10-7 to 4.6 × 10-8 cm2 s-1. Karl Fischer titration measurements show that almost complete removal of the deliberately added 9% water is achieved by degassing under benchtop conditions. However, displacement of oxygen by nitrogen in the ionic liquid solution results in the decrease of electrochemical reduction current by 6%, implying that dissolved gases need not be inert with respect to solvent properties. Oxygen removal by placing the BMIMBF4 ionic liquid in a nitrogen-filled glovebox or in a vacuum cell also simultaneously leads to removal of water and alteration of voltammetric data. This study highlights that (i) important physicochemical differences may arise upon addition or removal of a solute from viscous ionic liquids; (ii) degassing with dry nitrogen removes water as well as oxygen from ionic liquids, which may have implications on the viscosity and structure of the * To whom correspondence should be addressed. E-mail: alan.bond@ sci.monash.edu.au. Fax: +61 3 99054597. Phone: +61 3 99051338. † Monash University. ‡ Oxford University.
3856
Analytical Chemistry, Vol. 82, No. 9, May 1, 2010
medium; (iii) particular caution must be exercised when deoxygenation is applied in ionic liquid media as part of the protocol used in electrochemical experiments to remove oxygen; (iv) gases such as oxygen, argon, and nitrogen dissolved in ionic liquids need not be innocent with respect to the properties of an ionic liquid. The use of vacuum based techniques to eliminate all volatile solutes, including water and oxygen, is advocated. In many branches of chemistry, the need to remove oxygen prior to undertaking a measurement often is of paramount importance.1-4 A major problem is that dissolved oxygen often introduces adverse effects in analytical measurements. For example, the presence of oxygen is deleterious in spectroscopic measurements related to luminescence analysis since fluorescent and phosphorescent compounds are highly susceptible to quenching by molecular oxygen.5,6 In high-performance liquid chromatography (HPLC), deaeration of the mobile phase is often necessary because dissolved gases in the system can cause a variety of problems.7,8 In the case of voltammetry in aqueous media, a major problem caused by dissolved oxygen arises from its reduction to hydrogen peroxide and water (eqs 1 and 2) during the application of negative potentials. These reduction processes generates high background currents and reactive species which may interfere with the reductive processes of interest.2-4 O2 + 2H+ + 2e- f H2O2
(1)
H2O2 + 2H+ + 2e- f 2H2O
(2)
(1) Rollie, M. E.; Patonay, G.; Warner, I. M. Ind. Eng. Chem. Res. 1987, 26, 1–6. (2) Wallace, G. G. TrAC, Trends Anal. Chem. 1985, 4, 145–148. (3) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; Wiley: New York, 2001. (4) Bond, A. M.; Grabaric, B. S. Anal. Chem. 1979, 51, 337–341. (5) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Plenum: New York, 1983. (6) Warner, I. M. In Instrumental Analysis, 2nd ed.; Christian and O’Reilly: New York, 1986. (7) Meyer, V. R. Practical High-Performance Liquid Chromatography, 4th ed.; Wiley: New York, 2004. (8) Snyder, L. R.; Kirkland, J. J. Introduction to Modern Liquid Chromatography; Wiley: New York, 1974. 10.1021/ac100378g 2010 American Chemical Society Published on Web 04/14/2010
In nonaqueous solvents, reduction leads to the formation of the superoxide anion.9,10 Many approaches have been developed for solution deoxygenation, including vacuum degassing,1,8 nitrogen purging,11 argon purging,12 and chemical methods using strongly reducing agents as oxygen scavengers.13 The aim of all these methods is to remove or substantially reduce (by 95% to >99%) the oxygen dissolved in the solution of interest. However, the most widely used method of oxygen removal is to purge the sample with dry nitrogen (prepurified of any trace oxygen), which is assumed to be an inert gas. The method is quite effective and also simple to implement. Purging a solution with an inert gas reduces the partial pressure of oxygen above the solution, and consequently the solubility of dissolved oxygen in the solution decreases according to Henry’s law1 (eq 3) P ) kS
(3)
where P is the pressure of oxygen above the solution, k is the Henry’s law constant, and S is the solubility of oxygen in the solution. In aqueous or organic solvents, purging usually is undertaken for 5-30 min, depending on bubbling rates and also rates of oxygen uptake, although an even longer time may be required in systems where the solubility of oxygen is very high or a very low oxygen concentration is required. Over the past few years, the use of ionic liquids (ILs) as an alternative to conventional aqueous or organic media has expanded rapidly in almost every area of analytical chemistry, e.g., electrochemistry,14-18 chromatography,19,20 spectroscopy,21 and even mass spectrometry.22 In most cases, principles and approaches developed over many decades for aqueous or molecular solvent media have been directly extended to the IL case. Thus, deoxygenation in ionic liquids also has been commonly achieved by purging with inert gases. Recent work23-25 has shown that, partly because of the very high solubility levels, dissolved gases very significantly alter the transport properties of ionic liquids. Moreover, ionic liquids can behave differently to conventional molecular liquids as a result of their structures and properties (9) Zhang, D.; Okajima, T.; Matsumoto, F.; Ohsaka, T. J. Electrochem. Soc. 2004, 151, D31–D37. (10) Ortiz, M. E.; Nunez-Vergara, L. J.; Squella, J. A. J. Electroanal. Chem. 2003, 549, 157–160. (11) Fox, M. A.; Staley, S. W. Anal. Chem. 1976, 48, 992–998. (12) Lund, W.; Opheim, L. N. Anal. Chim. Acta 1975, 79, 35–45. (13) Arthur, P. Anal. Chem. 1964, 36, 701. (14) Buzzeo, M. C.; Evans, R. G.; Compton, R. G. ChemPhysChem 2004, 5, 1106– 1120. (15) Hapiot, P.; Lagrost, C. Chem. Rev. 2008, 108, 2238–2264. (16) Zhao, C.; Burrell, G.; Torriero, A. A. J.; Separovic, F.; Dunlop, N. F.; MacFarlane, D. R.; Bond, A. M. J. Phys. Chem. B 2008, 112, 6923–6936. (17) Wang, H.; Zhao, C.; Bhatt, A. I.; MacFarlane, D. R.; Lu, J. X.; Bond, A. M. ChemPhysChem 2009, 10, 455–461. (18) Endres, F.; MacFarlane, D.; Abbott, A. Electrodeposition from Ionic Liquids; John Wiley & Sons: Hoboken, NJ, 2008. (19) Anderson, J. L.; Ding, J.; Welton, T.; Armstrong, D. W. J. Am. Chem. Soc. 2002, 124, 14247–14254. (20) Armstrong, D. W.; He, L. F.; Liu, Y. S. Anal. Chem. 1999, 71, 3873–3876. (21) Tran, C. D.; Lacerda, S. H. D. Anal. Chem. 2002, 74, 5337–5341. (22) Li, Y. L.; Gross, M. L. J. Am. Soc. Mass Spectrom. 2004, 15, 1833–1837. (23) Barrosse-Antle, L. E.; Aldous, L.; Hardacre, C.; Bond, A. M.; Compton, R. G. J. Phys. Chem. C 2009, 113, 7750–7754. (24) Barrosse-Antle, L. E.; Hardacre, C.; Compton, R. G. J. Phys. Chem. B 2009, 113, 2805–2809. (25) Barrosse-Antlle, L. E.; Hardacre, C.; Compton, R. G. J. Phys. Chem. B 2009, 113, 1007–1011.
based on aggregation of ions. For example, the water content becomes an even more important issue in ionic liquids.16-18,26 Water is present in almost every ionic liquid as an adventitious impurity. The presence of a trace amount of water can drastically alter the physicochemical properties of the ionic liquids such as viscosity, conductivity, and diffusivity and consequently mass transport properties of electrochemical processes as has been appreciated since the early voltammetric studies using ionic liquids.27 Very recently, it was discovered that water, when present as an impurity in ionic liquid, presumably because of its modified structure, can be photooxidized to oxygen and thus serve as an electron donor for photochemical synthesis.26 In this study, we report some unexpected effects of purging an ionic liquid with nitrogen that arrive in the main because the presence of the gas reduces not only the dissolved oxygen concentration but also the water concentration and hence alters a range of physicochemical properties. In particular, the influence of deoxygenation with nitrogen on data obtained from the voltammetry of tetracyanoquinodimethane (TCNQ) at microdisc electrodes is reported in the commonly used ionic liquid 1-n-butyl3-methylimidazolium tetrafluoroborate (BMIMBF4). EXPERIMENTAL SECTION Reagents and Materials. High purity grade (g99.0%) 1-nbutyl-3-methylimidazolium tetrafluoroborate (BMIMBF4) was purchased from Merck and handled under the normal benchtop laboratory conditions or under other conditions as specified. Other chemicals used as supplied by the manufacturer were acetonitrile (CH3CN, Merck or Fischer Scientific); acetone (Merck); ferrocene (Fe(C5H5)2 or Fc, BDH) and tetracyanoquinodimethane (TCNQ, Aldrich). Tetrabutylammonium hexafluorophosphate (Bu4NPF6) was purchased from GFS and recrystallized twice from ethanol before use and tetrabutylammonium perchlorate (Bu4NClO4, Fluka, Puriss electrochemical grade, 99.99%) were used as received without further purification. Instrumentation and Procedures. Voltammetric experiments under benchtop laboratory conditions and in a homemade N2-filled glovebox were undertaken inside a Faraday cage with a BAS 100B/W electrochemical workstation and a low current amplifier (Bioanalytical System, West Lafayette, IN). TCNQ solutions of known concentration were prepared for these voltammetric studies by weighing the solid TCNQ with an ME36S microbalance (Sartorius, Australia). TCNQ only dissolves slowly in viscous BMIMBF4, and an ultrasonic bath (Unisonics Pty Ltd., Australia) was employed for 30 min to assist with the dissolution process. The solution was stored in the dark before voltammetric measurements were undertaken in a gastight single-compartment cell. Typically, 1 mL of ionic liquid containing a known water content was used for the electrochemical measurements. N2 gas containing 119 ppm.29
CONCLUSIONS Significantly modified voltammetric behavior for reduction of TCNQ is observed when the oxygen concentration present in BMIMBF4 is lowered by nitrogen purging, by placing in a nitrogen-filled glovebox, or by use of a vacuum cell. All approaches lead to removal of oxygen to a level below that this is voltammetrically detectable. However, the oxygen removal is accompanied by simultaneous removal of water. Nitrogen purging could remove water from very “wet” ionic liquids to a level of about 2000 ppm under fairly typical benchtop laboratory conditions. Leaving “wet’ ionic liquids in a nitrogen-filled glove box achieves slow drying of the sample, and if left long enough can give levels of about 400 ppm and approach the water level present inside the glove box. Application of vacuum for an extended period of time with heating could dry the hygroscopic ionic liquid BMIMBF4 to a water level of about 100 ppm. Thus, even under the most favorable conditions, the water and TCNQ concentration are approximately equal. After more than a decade of intensive electrochemical study in room temperature ionic liquids, a large number of diffusion coefficient data have been reported by different groups around the world. However, a large variation is evident in D values. In most studies, the principles and approaches developed over decades for aqueous or molecular solvent media have been directly applied to the IL case. However, it is now emerging that extrapolation to ILs is not straightforward.23-25 We have now established that replacement of oxygen by nitrogen also removes water and leaves water and nitrogen as solutes present in addition to the electroactive species. Considering the nonadditivity effect of electrochemical response when multisolutes are present in an ionic liquid, it is apparent that variable outcomes in studies with TCNQ dissolved in a ionic liquid would be obtained when more than one solute is present. We therefore recommend that the water level in the sample should be estimated and details of the oxygen removal level given in order to make sensible comparison of literature data. In this sense, the vacuum “T-cell” is advantageous as it removes oxygen and does not replace it with nitrogen and also leads to significant removal of water. Generally, measurement of residual water is always recommended before the electrochemical determination of D values, and the implication of degassing with N2 to remove oxygen needs careful analysis to avoid further uncertainties of reported D values. ACKNOWLEDGMENT This study was financed by the Australian Research Council Linkage Grant LP0668123 and Orica Ltd., Australia. Received for review February 9, 2010. Accepted March 30, 2010. AC100378G
Analytical Chemistry, Vol. 82, No. 9, May 1, 2010
3861
