Anal. Chem. 2009, 81, 10262–10267
Technical Notes In Situ NMR Spectroelectrochemistry of Higher Sensitivity by Large Scale Electrodes Sabrina Klod, Frank Ziegs, and Lothar Dunsch* Department of Electrochemistry and Conducting Polymers, Leibniz-Institute of Solid State and Materials Research IFW Dresden, Helmholtzstrasse 20, D-01069 Dresden, Germany The combination of NMR spectroscopy and electrochemistry provides an in situ method to measure structural changes of the redox components in an electrochemical reaction by proton NMR experiments. As the use of metal thin film radio frequency (RF) transparent electrodes in NMR spectroelectrochemical studies is limited by layer thickness and electrodes size, we present a new spectroelectrochemical NMR cell design consisting of nearly metal free symmetrically arranged large scale carbon fiber electrodes. Due to the advantages of modern NMR spectroscopy, a cell rotation is not necessary for high resolution measurements. This makes the presented cell for in situ spectroelectrochemical NMR measurements easy to prepare. The cell design is universal for a large variety of NMR spectrometers and frequencies used for detection of different nuclei. The feasibility of this new in situ NMR spectroelectrochemical cell is demonstrated in a detailed study of the electrochemical behavior of p-benzoquinone in different aqueous solutions. The combination of electrochemical methods with optical, vibrational, or electron spin resonance (ESR) spectroscopy results in highly sophisticated methods used for getting detailed information on the structural changes of redox systems during their reactions.1 Especially, in situ spectroelectrochemical techniques, applying spectroscopic methods during an electrochemical experiment, are powerful tools for the determination of structural changes of intermediates and final products of electrochemical reactions. Within the structure elucidation methods, NMR spectroscopy is the preferred method for determination of rather small structural or electronic changes.2 The combination of NMR spectroscopy and electrochemistry has rarely been used, although it is of high interest. Because of changes in magnetic probe geometry by the introduction of the electrodes into the sample tube and the action of metals in a strong magnetic field, such * Corresponding author. E-mail:
[email protected]. (1) Gale, E. J., Ed. Spectroelectrochemistry. Theory and Practice; Plenum: New York and London, 1988. (2) ClaridgeT. D. W. High-Resolution NMR Techniques in Organic Chemistry; Elsevier: Amsterdam, 1999; See also Kleinpeter, E.; Klod, S.; Rudorf, W. D. J. Org. Chem. 2004, 69, 4317.
10262
Analytical Chemistry, Vol. 81, No. 24, December 15, 2009
measurements are very difficult to conduct. A simple construction of two platinum wire electrodes as cathode and anode in an NMR cell has a strong influence on the homogeneity of the magnetic field in the probe head, making simultaneous electrolysis and NMR measurements impossible.3 Thin films of semiconducting SnO2 electrodes in a symmetric arrangement inside a spectroelectrochemical cell provide better conditions for the 1H NMR studies.4 In proportion to the electrolyte volume, the electrode surface area is small and the resistance of such films is rather high. Therefore, the electrode potential is difficult to control, and furthermore, a long-running electrolysis is needed to obtain enough reaction products for the 13C NMR spectroscopy. Thin gold films have lower resistance and provide a wider electrochemical potential window, but the low surface area is still problematic as well as the restrictions in the cathodic potential range.5,6 Thus, 13 C NMR experiments in spectroelectrochemistry are not possible as in situ measurements. The flow cell techniques provide better conditions for the NMR measurements because the electrochemical reaction takes place outside of the radio frequency (RF) coils of the probe, but we do not have a real in situ method. Furthermore, these flow techniques require large solution volumes (and redox components) and are limited to one direction of an electrochemical reaction.3,7 Here, we present a new design of an in situ NMR electrochemical cell, which makes sensitive in situ NMR measurements during electrochemical experiments possible. This cell is very simple to prepare and universally applicable for a wide variety of spectrometers. The working electrode is located within the region of the RF coils, and therefore, any electrode reaction (including electrode potential steps or cycling) can be studied. As an example, we follow the reduction of p-benzoquinone to hydroquinone by an electrochemical reaction to demonstrate the power of this method combination. (3) Albert, K.; Dreher, E.-L.; Straub, H.; Rieker, A. Magn. Reson. Chem. 1987, 25, 919. (4) Mincey, D. W.; Popovich, M. J.; Faustino, P. J.; Hurst, M. M.; Caruso, J. A. Anal. Chem. 1990, 62, 1197. (5) Webster, R. D. Anal. Chem. 2004, 76, 1603. (6) Prenzler, P. D.; Bramley, R.; Downing, S. R.; Heath, G. A. Electrochem. Commun. 2000, 2, 516. (7) Richards, J. A.; Evans, D. H. Anal. Chem. 1975, 47, 964. 10.1021/ac901641m CCC: $40.75 2009 American Chemical Society Published on Web 11/13/2009
Figure 1. Electrochemical cell design for in situ NMR spectroelectrochemistry.
EXPERIMENTAL SECTION The electrode construction was designed to fit into a 5 or 10 mm NMR tube as the electrochemical cell (see Figure 1). It consists of a carbon fiber filament as the working and the counter electrode. Additionally, a thin chlorinated silver wire was located close to the working electrode as a quasi-reference electrode. To study the electrolysis products at the working electrode, only this part of the cell system was located in the range of the RF coils in the NMR probe. The electrodes were connected to the potentiostat (details see below) by means of copper wires. To house the carbon fiber filament working and counter electrodes, glass capillaries, din ) 1.05, dout ) 1.4 +/- 0.05 mm and din ) 1.5, dout ) 2.0 +/- 0.05 mm (Hilgenberg GmbH), were sealed at one end by epoxy resin and wrapped at the other with a poly(tetrafluoroethylene) (PTFE) tape. With carbon fibers in NMR measurements, electron paramagnetic resonance (EPR) silent ones are required. The carbon fiber filament used was 2 mm × 8 µm, chosen for its lack of an EPR signal, and came from the stock used in an earlier study.8 It was made from poly(acrylonitrile) (PAN) fibers (former Chemiefaserwerk Premnitz) and carbonized in a pilot plant at Monokrystaly Turnov (now Czech Republic). The contact between the carbon fiber and the copper wire was made by conductive epoxy resin (8) Weber, J.; Dunsch, L.; Neudeck, A. Electroanalysis 1995, 7, 255–259. See also Dunsch, L. In Dresden Polymer Discussion, 3rd ed.; Dresden University of Technology, 1991; p 164-178.
ECOLIT 342, (Panacol Elasco GmbH). For highly flexible shielded cables from the cell to the potentiostat, a microphone cable type 1804 AJ5C, (Belden Electronics Division) were used. Due to this conductive epoxy resin, an overall resistance of about 35 Ω in the 2 m cable from the potentiostat to the electrodes was achieved through an antiinterference choke 220 µH at the output of the NMR magnet. A PTFE covered silver wire with an AgCl tip as the pseudoreference electrode was soldered to a copper wire, arranged together with the other connecting wires of the working and counter electrode. The electrode system of all three electrodes was inserted in the 5 mm NMR tube and sealed by a PTFE tape. For all electrochemical experiments, a PG 390USB (HEKA Elektronik) or a PAR 173 (Princeton Applied Research) in combination with an ITC 16 interface (HEKA Instrutech) and the software package Potpulse/Potmaster (HEKA Elektronik) were used. The potential was either changed in steps or kept constant at a certain value. The electrolysis time is strongly correlated to the electrolyte volume of the NMR tube due to the limited diffusion inside the tube. Approximately 6 h are necessary for the complete electrolysis of 0.5 mL of filling volume. The operation of this electrode system was checked by the oxidation of N,N,N′,N′-tetramethyl-p-phenylendiamin, the “Wurster’s Blue”. Inserted in the 5 mm NMR tube, the characteristic blue coloring of the solution at the potential of the first electron transfer and the vanish of the color at the potential of the second electron transfer could be observed visually. A test of the redox system by a platinum working electrode (with Pt as a counter electrode and Ag/Ag+ as pseudo reference) (see Figure 2) gave the same reduction and oxidation peak potentials. Due to the large electrode surface, the voltammetric peaks are much broader and the current is higher. The p-benzoquinone and p-hydroquinone were dissolved in deuterated water with Li2SO4, as the supporting electrolyte, and H2SO4 for the adjustment of the pH in all NMR experiments. The solutions were degassed carefully by nitrogen before mounting the electrodes into the cell. These electrochemical experiments were performed with a PG 284 potentiostat (HEKA Elektronik) and the software package Potpulse using a three electrode system with platinum wires as working and counter electrodes and a silver wire as the pseudoreference electrode at a scan rate 25 mV/s. The NMR measurements were performed on an Avance II 500 MHz spectrometer (Bruker Biospin GmbH) equipped with a standard 5 mm BBO probe head; a heating and cooling unit included using the software package TopSpin 2.1(Bruker Biospin GmbH). Due to the high quality of the modern shim systems, only a slight decrease in the S/N ratio was observed between the spinning and nonspinning mode. The cell was shimmed with only a little higher effort compared to standard tubes. Using the nonspinning mode, we avoid the difficulties of slighting electrical contacts in a rotating cell. A standard pulse sequence was used to obtain 1H and 13C spectra. The spectra were calibrated by the signal of the lock solvent. Temperature control via the heating and cooling unit attached to the probe leads to a temperature accuracy of 0.1 K. The NMR signal of p-benzoquinone is reduced to about one-quarter of the S/N ratio, and Analytical Chemistry, Vol. 81, No. 24, December 15, 2009
10263
Figure 2. Cyclic voltammograms of p-benzoquinone at a platinum electrode system (left side) and carbon fiber electrode system (right side).
Figure 3. 1H NMR spectra (a) and
13
C NMR spectra (b) of p-benzoquinone (black) and phydroquinone (red).
a signal broadening of 4 times is observable by incorporation of the electrode system into the NMR tube.
Scheme 1. Reduction of p-Benzoquinone in the Presence of a Proton Donor
RESULTS AND DISCUSSION As a system to study the feasibility of our new in situ NMR cell, the well-known redox system of p-benzoquinone and phydroquinone was used.9,10 Deduced to the high symmetry, the 1 H NMR spectra of these two compounds consist of signals at 6.80 and 6.72 ppm, in aqueous solution, and are very similar (see Figure 3). Both structures can be distinguished in a mixture by their proton signals. In water, the hydroxyl group of hydroquinone is not detectable due to fast proton exchange phenomena. The 13C NMR spectra aquired without the electrode system within 500 scans (c.f. Figure 3) consist of two lines for each compound and differ markedly. Due to aromatic character and 10264
Analytical Chemistry, Vol. 81, No. 24, December 15, 2009
the electron pushing effect of the hydroxyl group, the carbon signals of hydroquinone are shifted to higher field. On the other hand, the electron withdrawing effect of the quinoidic carbonyl group leads to a strong low field shift of these two carbon signals (9) Beck, F. Elektroorganische Chemie; Akademie-Verlag: Berlin, 1974. (10) Tomilov, A. P.; Mairanovskii, S. G.; Floshin, M. Y.; Smirnov, V. A. The Electrochemistry of organic compounds, 1st ed.; John Wiley and Sons: New York, 1972.
Figure 4. Cyclic voltammograms of p-benzoquinone in aqueous solution at pH 1 (red) and pH 7 (green).
Figure 5. In situ NMR spectra of the reduction of p-benzoquinone during the electrolysis in aqueous solutions of pH 1 (a) at a potential of -350mV and (b) at a potential of +350mV.
in benzoquinone. Thus, these two structures are easy to distinguish by 13C NMR spectroscopy. The reduction of p-benzoquinone to hydroquinone is a wellknown electrochemical reaction9,10 with a two electron transfer in the presence of proton donors such as sulphuric acid (see
Scheme 1). The first reduction potential is -180 mV (Ag/Ag+), while the reoxidation peak arises at -120 mV (c.f. Figure 4) pointing to a reversible reaction. In the absence of proton donors, the cyclic voltammogram changes remarkably. The reduction peak shifts to -400 mV, and the reoxidation peak shifts to -320 mV Analytical Chemistry, Vol. 81, No. 24, December 15, 2009
10265
Figure 6. In situ NMR spectra of the reduction of p-benzoquinone during the electrolysis in aqueous solution of pH 7 (a) at a potential of -500mV and (b) at a potential of +500mV.
(cf. Figure 4). The reduction changes from a two electron transfer to a single electron transfer without a second electron transfer up to -1000 mV. Here, a semiquinone radical is formed instead of the hydroquinone which is not detectable by standard NMR techniques due to its paramagnetic state. The goal of this spectroelectrochemical study was the influence of different pH values on the reaction mechanism in aqueous solutions, we started the study at very low pH (Figure 5). The pH value was set by sulphuric acid as the proton donor in the electrode reaction. A single sharp line at 6.95 ppm for p-benzoquinone was observed in the 1H NMR spectrum at the beginning. During the 13C experiment with up to 5000 scans, only the nuclear Overhauser effect (NOE) enhanced signal of the proton attached carbons at 137 ppm is observable, while the second line of the carbonyl carbon was not detectable. The rather small solubility of p-benzoquinone, causing very low concentrations, might be the reason for the absence of these NMR lines. Immediately after applying an electrolysis potential of -350 mV, a second proton signal at 6.85 ppm arises, which increases with electrolysis time. Simultaneously, the proton signal of p-benzoquinone at 6.95 ppm decreases. Reaching an equal concentration of both structures after 3 h of electrolysis, we tried to measure the 13 C spectrum of the solution but got no NMR signal even after 5000 scans. After 6 h of electrolysis, the first proton signal at 6.95 ppm was completely vanished and only the second proton signal at 6.85 ppm was observed. At this point, the carbon signal of the proton attached carbons of hydro10266
Analytical Chemistry, Vol. 81, No. 24, December 15, 2009
quinone at 115 ppm was detected in the 13C NMR experiment, while the second carbon signal at 149 ppm was not observed. To test the reversibility of the reduction, we applied a potential of +350 mV to the same solution and obtained the 1H NMR spectra at several time steps. A decrease of the second proton signal at 6.85 ppm was clearly detected while the first signal at 6.95 ppm appeared and increased during electrolysis time. After this long time experiment, a slight shift of the two signals to higher field and a signal broadening are observable, caused by a change of the pH by the reduction of protons and a molecular hydrogen formation followed by gas bubble movement. With the conclusion of this part, the complete reduction of p-benzoquinone, to the corresponding hydroquinone, and the reoxidation can be detected by an in situ NMR spectroelectrochemical experiment. At pH 7 (aqueous solution without sulphuric acid, see Figure 6) the electrochemical reaction results in a different behavior in NMR spectroscopy. A single NMR line at 6.80 ppm addressed to p-benzoquinone is observed in the 1H NMR experiment. In 13 C NMR spectroscopy, no signal is observable, even for a long accumulated acquisition. At this electrode potential, it is not possible to follow the electrochemical reaction by 13C NMR spectroscopy. After applying an electrode potential of -500 mV, the intensity of the proton signal of p-benzoquinone at 6.80 ppm decreases but no additional NMR line appears. Further electrolysis gives a remarkable decrease of the signal at 6.80 ppm. A new NMR line is not detectable because of the
Scheme 2. Reduction Mechanism of p-Benzoquinone in the Absence of a Proton Donor
production of the semiquinone radical. After an extended electrolysis time of about 1 h, a second NMR line arises with very low intensity. The signal at 6.72 ppm increases slower than the other NMR signal at 6.80 ppm decreases. The electrolysis was finished when both NMR signals showed very low intensity, indicating a low concentration of both structures in solution. When the potential to +500 mV for reoxidation was changed, the NMR spectrum does not change even after several hours of electrolysis. Due to the formation of quinhydrone being a slightly soluble complex, the back reaction at the electrode is impossible as the quinhydrone precipitates. The semiquinone radical not detectable in NMR experiments undergoes a disproportionation to benzoquinone and hydroquinone at the same ratio. Both leads to the formation of the nearly insoluble quinhydrone (see scheme 2) that is not available for electrochemical reactions any more. The precipitation of the quinhydrone complex is confirmed by cell inspection after electrolysis.
In varying the pH of the solution, the change in the reaction mechanism in the electrochemical reduction of p-benzoquinone can be clarified by in situ NMR spectroscopy. On the basis of this easy access to the NMR data of reaction products of electrode reactions, further studies of electrode reactions even in the anodic potential range are under way. CONCLUSION A new cell design for in situ NMR spectroelectrochemical studies is presented which makes sensitive studies of electrochemical reactions at solid electrodes possible. The new NMR spectroelectrochemical cell is a universal piece of equipment applicable for most NMR spectrometers and suitable for a large potential window. An in situ NMR spectroelectrochemical study with this new spectroelectrochemical cell shows the remarkable difference in the electrochemical behavior of p-benzoquinone in aqueous solution at different pH values. The results of these experiments proved the feasibility of the new cell design for in situ spectroelectrochemical NMR measurements. ACKNOWLEDGMENT The authors thank Dr. E. Dmitrieva (IFW Dresden) for her support in electrochemistry and Dr. M. Zalibera (STU Bratislava) for his work in NMR spectroscopy. Received for review July 23, 2009. Accepted October 30, 2009. AC901641M
Analytical Chemistry, Vol. 81, No. 24, December 15, 2009
10267
