Liquid–liquid electro-organo-synthetic processes in a carbon nanofibre membrane microreactor: Triple phase boundary effects in the absence of intentionally added electrolyte

Electrochimica Acta 56 (2011) 6764–6770

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Liquid–liquid electro-organo-synthetic processes in a carbon nanofibre membrane microreactor: Triple phase boundary effects in the absence of intentionally added electrolyte John D. Watkins a , Sunyhik D. Ahn a , James E. Taylor a , Steven D. Bull a , Philip C. Bulman-Page b , Frank Marken a,∗ a b

Department of Chemistry, University of Bath, Claverton Down, Bath BA2 7AY, UK School of Chemistry, University of East Anglia, Norwich, Norfolk NR4 7TJ, UK

a r t i c l e

i n f o

Article history: Received 3 December 2010 Received in revised form 17 May 2011 Accepted 19 May 2011 Available online 27 May 2011 Keywords: Carbon nanofibre membrane Microreactor Deuteration Electrolysis Electrosynthesis Supporting electrolyte Voltammetry Ferrocene Olefin Isotope

a b s t r a c t An amphiphilic carbon nanofibre membrane electrode (ca. 50 nm fibre diameter, 50–100 ␮m membrane thickness) is employed as an active working electrode and separator between an aqueous electrolyte phase (with reference and counter electrode) and an immiscible organic acetonitrile phase (containing only the redox active material). Potential control is achieved with a reference and counter electrode located in the aqueous electrolyte phase, but the electrolysis is conducted in the organic acetonitrile phase in the absence of intentionally added supporting electrolyte. For the one-electron oxidation of n-butylferrocene coupled to perchlorate anion transfer from aqueous to organic phase effective electrolysis is demonstrated with an apparent mass transfer coefficient of m = 4 × 10−5 m s−1 and electrolysis of typically 1 mg n-butylferrocene in a 100 ␮L volume. For the two-electron reduction of tetraethylethylenetetracarboxylate the apparent mass transfer coefficient m = 4 × 10−6 m s−1 is lower due to a less extended triple phase boundary reaction zone in the carbon nanofibre membrane. Nevertheless, effective electrolysis of up to 6 mg tetraethyl-ethylenetetracarboxylate in a 100 ␮L volume is demonstrated. Deuterated products are formed in the presence of D2 O electrolyte media. The triple phase boundary dominated mechanism and future microreactor design improvements are discussed. © 2011 Elsevier Ltd. All rights reserved.

1. Introduction Electro-organo-synthetic reactions offer versatile and clean laboratory protocols for example in hydrogenation/deuteration [1], decarboxylation [2], halogenation [3], and carboxylation [4] processes. New microreactor systems have been proposed for continuous flow-electrolysis processes [5,6]. In order to avoid the use of undesirable supporting electrolyte in organic media, novel particulate electrolyte systems [7], self-supported micro-gap reactions [8], and biphasic processes [9] have been proposed. Although dry reaction conditions are sometimes desirable, there are many processes which occur readily in the presence of moisture or in the presence of a separate aqueous phase [10]. Therefore, flow [11], ultrasoundassisted [12], and ultra-turrax-assisted [13] processes for “biphasic” aqueous-organic electrolysis systems have been developed. These systems have in common that electron transfer is localised in an aqueous electrolyte|organic|electrode triple phase boundary reac-

∗ Corresponding author. Tel.: +44 1225 383694; fax: +44 1225 386231. E-mail address: [email protected] (F. Marken). 0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2011.05.075

tion zone [14]. The use of the liquid|liquid interface as an reaction environment opens up a wide range of new chemistries where (i) ionic species (reactive or passive) are supplied from the aqueous phase into the organic phase, (ii) “external” pH control can be achieved by buffering the aqueous phase, and (iii) precipitation reactions can be avoided. The use of a novel “biphasic” microreactor with a carbon nanofibre membrane electrode separating the aqueous and the organic liquid phases is described here. Fig. 1A shows a schematic drawing explaining the experimental design. The liquid|liquid interface, here typically 2 mol dm−3 NaCl aqueous electrolyte and an immiscible acetonitrile phase, are in contact at a carbon nanofibre membrane of ca. 50–100 ␮m thickness. The carbon membrane acts as the working electrode allowing anion and cation transport into the organic phase. Fig. 1B shows a schematic reaction involving oxidation of molecule “M” in the organic phase coupled to the transfer of the anion “X− ” from the aqueous into the organic phase. A closer consideration of the carbon membrane (see the schematic drawing in Fig. 1C) suggests the presence of a complex interface with an extended triple phase boundary reaction zone. In this report it is shown that after optimisation of all

J.D. Watkins et al. / Electrochimica Acta 56 (2011) 6764–6770

6765

Fig. 1. (A) Schematic cartoon drawing of an electrochemical cell with working, counter, and reference electrode (W, C, R) in the aqueous phase and an immiscible organic phase separated by a carbon nanofibre membrane. (B) Schematic drawing of an oxidation process at the carbon membrane working electrode, which is accompanied by the transfer of an anion from the aqueous into the organic phase. (C) Expanded view of the triple phase boundary reaction zone within the carbon membrane. This cartoon drawing is not realistic and the true distribution of organic and aqueous phase within the membrane is currently not well understood.

other parts of the system, the process at the triple phase boundary is limiting the rate of electrolysis. 2. Experimental details 2.1. Chemical reagents Sodium perchlorate monohydrate (98%, Sigma Aldrich), sodium sulphate decahydrate (≥99%, ACS reagent, Sigma Aldrich), nbutylferrocene (98%, Alfa Aesar), phosphoric acid (Sigma Aldrich ACS reagent, 85 wt%), sodium chloride (≥95%, SigmaUltra reagent, Sigma Aldrich), sodium hydroxide (≥97%, ACS reagent, Sigma Aldrich), hydroquinone, and tetraethyl-ethylenetetracarboxylate (98+%, Lancaster Synthesis) were used without further purification. Demineralised and filtered water was taken from a Millipore water purification system with not less than 18 M cm resistivity. 2.2. Instrumentation

electrode could be checked for possible leaks). 100 ␮L of tetraethylethylenetetracarboxylate solution in acetonitrile was added into the interior of the glass microreactor tube and allowed to equilibrate with the aqueous electrolyte for 10 min. A micropropeller stirrer (1 mm diameter stainless steel with flattened end) was placed into the glass tube to agitate the acetonitrile solution with a rotation rate of ca. 300 Hz (Dremel tool). Constant potential electrolyses were conducted in chronoamperometry mode. After completion of the electrolysis, the organic phase was removed and the microreactor was rinsed with one aliquot of clean acetonitrile. This was followed by removal of acetonitrile in vacuo, and dissolution of product into deuterated chloroform for NMR and MS analysis. Tetraethyl ethane-1,1,2,2-tetracarboxylate: 1 H NMR (300 MHz; CDCl3 ): ıH = 4.22 (8H, q, J = 7.2 Hz, CH2 CH3 ), 4.13 (2H, s, CH(CO2 Et)2 ), 1.27 (12H, t, J = 7.2 Hz, CH2 CH3 ); 13 C{1 H} NMR (75 MHz, CDCl3 ): ıC = 167.2, 62.2, 51.6, 14.1; HRMS: m/z (ES) 341.1211, C14 H22 NaO8 [M+Na]+ requires 341.1212. In the deuterated product 1 H NMR lines at ıH = 4.13 were absent.

Voltammetric experiments were performed with a microAutolab III system (Ecochemie, Netherlands) in staircase voltammetry mode. The step potential was maintained at approximately 1 mV. The counter and reference electrodes were platinum gauze and KCl-saturated calomel (SCE, Radiometer), respectively. The working electrode was a 4 mm diameter carbon nanofibre membrane disc (see Fig. 2, “Bucky paper”, Nanolabs US, with low resistivity (∼0.1  cm) and relatively low impurity levels (Fe 0.36, Si 0.31, Al 0.23, Na 0.32, S 0.23 at%) and several recent applications e.g. in Hall measurements with Pt nanoparticles, for catalytic hydrogenation with Rh nanoparticles, and for processes in ionic liquids [15–17] mounted with Ambersil silicone (Silicoset 151) on a glass capillary of 3.5 mm inner diameter and 5 mm outer diameter. The electrical contact was made with a 1 mm stripe of pyrolytic graphite film (Goodfellow, UK) inside of the glass capillary. Solutions were deaerated with argon (Pureshield, BOC). The pH was measured with a glass electrode (3505 pH meter, Jenway). All experiments were conducted at a temperature of 22 ± 2 ◦ C.

3. Results and discussion

2.3. Procedure for electro-organo-syntheses

The carbon nanofibre membrane is amphiphilic as shown in Fig. 2D (both acetonitrile and aqueous electrolyte readily absorb into the membrane and in a biphasic mixture the membrane binds into the liquid|liquid interface; similar phenomena have been reported for example for carbon nanoparticles [18,19] and for carbon nanotubes [20]) and interacting with the aqueous phase to

Prior to all measurements aqueous solutions were thoroughly degassed using argon. After the carbon membrane electrode was fabricated, the glass capillary reactor was submerged ca. 5 mm deep into the aqueous electrolyte solution (so that the

3.1. Carbon nanofibre membrane oxidation of hydroquinone dissolved in the aqueous phase Initially, the carbon nanofibre membrane is investigated in contact with the aqueous electrolyte phase. Hydroquinone (2 × 10−3 mol dm−3 ) in 0.1 mol dm−3 phosphate buffer is employed as an aqueous redox system in contact with the high surface area carbon nanofibre electrode. The 2-electron oxidation of hydroquinone produces benzoquinone (see Eq. (1)) in a chemically reversible process, which at carbon nanofibre electrodes results in a well-defined voltammetric response (see Fig. 3A).

(1)

6766

J.D. Watkins et al. / Electrochimica Acta 56 (2011) 6764–6770

Fig. 2. (A–C) Scanning electron microscopy (SEM) images for the carbon nanofibre membrane with ca. 50–100 ␮m thickness and composed of carbon nanofibres of ca. 50 nm diameter. (D) Photographs of (i) the carbon membrane disc sinking to the bottom in liquid acetonitrile and in aqueous 1 mol dm−3 Na2 SO4 /0.1 mol dm−3 NaClO4 and (ii) the carbon membrane disc binding into the liquid|liquid interface in the two-phase system acetonitrile|aqueous 1 mol dm−3 Na2 SO4 /0.1 mol dm−3 NaClO4 (0.1 × 10−3 mol dm−3 n-butylferrocene added to colorise the organic phase yellow). (E) Typical EDS spectrum obtained at 10 kV acceleration. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

give a considerable capacitive background current as well as a Faradaic hydroquinone oxidation current is observed. The immersion depth of the carbon membrane did not significantly affect the voltammetric response consistent with a carbon membrane “filled” predominantly with aqueous phase. The capacitive current Ic = 50 ␮A (at a scan rate of 10 mV s−1 , see Fig. 3A) suggests a capacitance of ca. 5 mF (or a specific capacitance for the 0.6 mg carbon membrane disc of 8 F g−1 ) consistent with a highly porous carbon nanofibre electrode of this type [21]. The oxidation of hydroquinone occurs with a midpoint potential of ca. 0.38 V vs. SCE and with well-defined concentration dependence (see Fig. 3A and B). Uncompensated resistance (ca. 200 ) is causing a distortion of the voltammetric responses, but analysis of the scan rate dependency clearly shows a lack of diffusion contribution (see Fig. 3C and D). The voltammetric response is dominated by hydroquinone “trapped” in the porous carbon membrane (“thin film”) and diffusion of hydroquinone from solution towards the electrode

makes an insignificant contribution in this range of scan rates. The charge 1 mC (see Fig. 3D) is only slightly higher than that expected for physically trapped solution in the carbon membrane disc (ca. 0.25 mC for 2 × 10−3 mol dm−3 hydroquinone) and therefore only insignificant accumulation of hydroquinone occurs. In Fig. 3E the effect of adding the organic acetonitrile phase is demonstrated. The capacitive current response is not significantly affected consistent with aqueous electrolyte remaining in contact to the relatively hydrophilic carbon nanofibre membrane. However, the Faradaic current for the oxidation of hydroquinone is lower (by approximately a factor of 2) presumably due to the acetonitrile phase somewhat interacting with the carbon nanofibres and some displacement of hydroquinone solution in the carbon membrane. The voltammetric responses in this system did not show sensitivity towards the immersion depth of the carbon nanofibre membrane or to the fill height with organic phase (typically 50–100 ␮L) presumably due to the capillarity effect of the membrane.

J.D. Watkins et al. / Electrochimica Acta 56 (2011) 6764–6770

6767

Fig. 3. (A) Cyclic voltammograms (scan rate 10 mV s−1 ) for the oxidation of hydroquinone with a concentration of (i) 1, (ii) 2, (iii) 5, and (iv) 10 × 10−3 mol dm−3 in aqueous 0.1 mol dm−3 PBS pH 7 without organic phase and left to equilibrate for 5 min between individual experiments. (B) Plot of the charge under the voltammetric oxidation response (obtained after approximate baseline correction) vs. hydroquinone concentration. (C) Cyclic voltammograms (scan rate (i) 5, (ii) 10, (iii) 20, and (iv) 50 mV s−1 ) for the oxidation of 2 × 10−3 mol dm−3 hydroquinone in aqueous 0.1 mol dm−3 PBS pH 7. (D) Plot of the charge under oxidation (i) and reduction (ii) responses as a function of scan rate. (E) Cyclic voltammograms (scan rate 10 mV s−1 ) for the oxidation of 2 × 10−3 mol dm−3 hydroquinone in aqueous 0.1 mol dm−3 PBS pH 7 + 1 mol dm−3 Na2 SO4 with (i) and without (ii) organic acetonitrile phase (50 ␮L).

3.2. Carbon nanofibre membrane oxidation of n-butylferrocene dissolved in the organic phase Next, the electrolysis of a purely organic redox system dissolved in the acetonitrile phase is investigated. The model redox system n-butylferrocene is chosen due to the well-defined one electron oxidation–reduction response (see Eq. (2)).

(2) The oxidation of n-butylferrocene is accompanied by the transfer of an anion from the aqueous into the organic phase and therefore an aqueous electrolyte system has been chosen based on 2 mol dm−3 NaCl (to stabilize the aqueous – organic phase boundary) and 0.1 mol dm−3 NaClO4 (to provide a hydrophobic ClO4 − anion for preferential transfer into the organic phase). Fig. 4A shows a typical set of voltammetric responses consistent with the oxidation and back reduction of n-butylferrocene with a midpoint potential of ca. 0.28 V vs. SCE. Voltammetric responses are superimposed on a considerable capacitive current response and “drawn out” presumably due to transport effects caused by the absence of intentionally added supporting electrolyte in the organic phase (vide infra). Approximate peak currents are not dissimilar to those observed for the hydroquinone oxidation suggesting efficient transport of n-butylferrocene into the carbon nanofibre membrane material. The effect of the n-butylferrocene concentration on the voltammetric response also confirms the presence of a Faradaic current with

transport of n-butylferrocene into the carbon nanofibre membrane. The electrolysis process was investigated under controlled potential conditions and with a “micro-propeller” (see Section 2) causing agitation in the organic phase. For the oxidation of n-butylferrocene a switch in colour is observed when the green n-butylferricenium cation is formed in the acetonitrile phase. Photographic images for the conversion are show in Fig. 4C. Both the forward (oxidation at 0.9 V vs. SCE) and backward (reduction at −0.1 V vs. SCE) processes (see Eq. (2)) are clearly observed. Next, a bulk electrolysis experiment conducted with a 20 × 10−3 mol dm−3 n-butylferrocene solution in 100 ␮L acetonitrile. The plot in Fig. 4D shows the change in current and total electrolysis charge with time. The electrolysis charge is approximately 300 mC which is slightly higher compared to the anticipated 193 mC, presumably due to some degradation (and further electron transfer) of the chemically unstable n-butylferricenium product. The electrolysis process is complete in ca. 10 min. Experiments with faster agitation in the acetonitrile phase resulted in no change in electrolysis rate, which suggests a rate limiting process within the carbon membrane. In order to further quantify this chronoamperometry experiments were carried out. The electrolysis current in Fig. 4D shows a typical first order decay during the first 5 min consistent with a typical bulk electrolysis process following I(t) = Io × exp(− m(A/V)t) [22] where I(t) is the electrolysis current, Io is the initial electrolysis current, m is the apparent mass transfer coefficient, A is the geometric area 10−5 m2 , V is the volume of acetonitrile 10−7 m3 , and t is the time. The apparent mass transfer coefficient here is m = 4 × 10−5 m s−1 . The apparent mass transport coefficient, when dominated by diffusion, is given by m = D/ı where D is the diffusion coefficient (here assumed to be 1.56 × 10−9 m2 s−1 for n-butylferrocene in

6768

J.D. Watkins et al. / Electrochimica Acta 56 (2011) 6764–6770

3.3. Carbon nanofibre membrane reduction of tetraethyl-ethylenetetracarboxylate dissolved in the organic phase The reduction of olefins is a well known 2-electron 2-proton process and in particular the reduction of activated olefins with ester functionality is readily performed under electrolysis conditions [24]. Here tetraethyl-ethylenetetracarboxylate (see Eq. (3)) is employed as a model olefin for the study and optimisation of the electrolysis conditions.

(3)

Fig. 4. (A) Cyclic voltammograms (scan rate 10 mV s−1 ) for the oxidation of nbutylferrocene with a concentration of (i) 1, (ii) 2, (iii) 5, and (iv) 10 × 10−3 mol dm−3 in acetonitrile in contact to aqueous in 0.1 mol dm−3 NaClO4 + 2 mol dm−3 NaCl. (B) Plot of the peak current for the n-butylferrocene oxidation vs. the concentration of nbutylferrocene in the organic phase. (C) Photographic images of 20 × 10−3 mol dm−3 n-butylferrocene in acetonitrile during the course of forward and backward electrolysis: (i) initial solution, (ii) electrolysis at 0.9 V vs. SCE with stirring for 1 min, (iii) and for 15 min, (iv) back-electrolysis at −0.1 V vs. SCE for 15 min, and (v) for 30 min. (D) Plots of the logarithm of electrolysis current (i) and charge (ii) for oxidation of a 20 × 10−3 mol dm−3 n-butylferrocene solution (100 ␮L, at 0.7 V vs. SCE). The dashed line indicates the initial first order current decay line (see text).

acetonitrile [23]) and ı is the apparent diffusion layer thickness. Here the parameter ı can be compared for example to the membrane thickness. Based on the slope in Fig. 4D an apparent diffusion layer thickness of ı = 40 ␮m is obtained. Diffusion of n-butylferrocene into the carbon membrane against a potential gradient is likely to explain this result at least in part. Therefore, faster electrolysis experiments will require for example a thinner carbon membrane. However, the extent of the triple phase boundary reaction zone causes additional rate limitations (vide infra). In fact, it is probably oversimplified in the case investigated here to use the apparent mass transport coefficient in the conventional sense.

In this process tetraethyl-ethylenetetracarboxylate remains in the organic phase during reduction and the transfer of protons from the aqueous phase (vide infra) ensures charge neutrality. Fig. 5A shows voltammetric responses for the reduction of (i) 0, (ii) 10, (iii) 20, and (iv) 40 × 10−3 mol dm−3 tetraethylethylenetetracarboxylate. The cathodic process commences at a potential of ca. −0.6 V vs. SCE and the process is accompanied by the transfer of two protons from the aqueous into the organic phase (see Eq. (3)). The current peaks are significant but somewhat drawn out probably due to some uncompensated resistance within the carbon nanofibre membrane and/or membrane pH gradient effects (vide infra). The effect of the tetraethyl-ethylenetetracarboxylate concentration in the organic phase on the reduction peak current suggests an approximately linear relationship up to ca. 0.2 mol dm−3 (see Fig. 5B). For even higher concentrations the current within the potential window does not increase any further probably due to resistance and charge transfer limitations within the carbon nanofibre membrane. However, high concentrations suitable for micro-scale bulk electrolysis of (up to 6 mg tetraethylethylenetetracarboxylate) are possible in this carbon nanofibre membrane microreactor. Bulk electrolysis data are summarised in Fig. 5C. During electrolysis halving or doubling the speed of agitation in the organic compartment did not significantly affect the cyclic voltammetry data or the electrolysis time. Effects from agitation become significant only at much lower stirring rate. Fast agitation (stirring with a micro-propeller with 300 Hz) was employed in all cases to provide a high rate of mass transport in electrolysis experiments and in this way to make membrane effects more obvious. During electrolysis the formation of product was monitored by 1 H NMR and GC/MS methods and an almost linear increase in product yield is apparent in the first hour of the electrolysis (see Fig. 5C). Current efficiencies in this period remain at only typically 20–40% presumably due to some oxygen entering the system through the organic phase. Beyond the 2 h electrolysis time, loss of product probably due to side reaction such as ester hydrolysis is likely and detrimental. Additional experiments conducted in electrolyte media containing 1 mol dm−3 phosphate buffer pH 7, in 1 mol dm−3 H3 PO4 , in 1 mol dm−3 Na2 SO4 , or in 1 mol dm−3 Li2 SO4 with 0.1 mol dm−3 phosphate buffer pH 7 did not improve yields or performance, which suggests that the process is limited primarily due to the process within the carbon membrane. Further insight into the electrolysis process can be obtained from chronoamperometry data (see Fig. 5D). The current–time dependence is similar to that observed for n-butylferrocene (see Fig. 4D), however, the electrolysis time is longer and the slope is 4.4 × 10−4 s−1 . The calculation of the apparent transfer coefficient m = 4 × 10−6 m s−1 and the apparent diffusion layer thickness in this case (using m = D/ı where D = 1.2 × 10−9 m2 s−1 is estimated with the Wilke–Chang expression [25]) gives ı = 260 ␮m, which is unrealistically high. Therefore the process

J.D. Watkins et al. / Electrochimica Acta 56 (2011) 6764–6770

6769

within the carbon nanofibre membrane is considerably slower for tetraethyl-ethylenetetracarboxylate compared to the process for n-butylferrocene and this effect cannot be explained by mass transport alone. A less “active” triple phase boundary zone within the membrane is likely. A key difference in these two processes is the presence of ionic species. The oxidation of n-butylferrocene results in the transport of n-butylferricenium cations and perchlorate anions back into the acetonitrile phase, which is improving ionic conductivity within the carbon nanofibre membrane. Similar cases where ion transfer into the organic phase increased the extent of the triple phase boundary reaction zone have been reported [26,27]. The extent of the triple phase boundary reaction zone in Fig. 1C is therefore sensitive to the type of electrolysis process and future improvements in the electrolysis efficiency and speed have to focus on the potential drop within the triple phase boundary reaction zone. Additional experiments were conducted in the presence of D2 O electrolyte solution in order to confirm the H+ /D+ transfer into the organic phase as suggested in Eq. (3). Under these conditions deuterium isotope labelling of the reduced tetraethylethylenetetracarboxylate is effective and the doubly deuterated product is the only product detected under these electrolysis conditions. 4. Conclusions A novel carbon nanofibre membrane reactor for “biphasic” electrosynthesis has been developed and tested. The oxidation of n-butylferrocene and the reduction of tetraethylethylenetetracarboxylate in acetonitrile in the absence of added electrolyte in the organic phase have been demonstrated. As a key factor in the electrolysis process the extent of the triple phase boundary reaction zone has been identified. In future, improved carbon membranes will be developed and in particular porous carbon tubes for flow-through electrolyses will be desirable. The new microreactor electrolysis system described here is versatile and potentially suitable for operation in organic synthesis laboratories. A wider range of electro-organo-synthetic processes is now under investigation. Acknowledgements J.D.W. thanks EPSRC for financial support through a DTA studentship. Help from Charles Y. Cummings with SEM experiments is gratefully acknowledged. References

Fig. 5. (A) Cyclic voltammograms (scan rate 10 mV s−1 ) for the reduction of tetraethyl-ethylenetetracarboxylate with a concentration of (i) 0, (ii) 10, (iii) 20, and (iv) 40 × 10−3 mol dm−3 in 100 ␮L acetonitrile in contact to aqueous in 0.1 mol dm−3 PBS pH 7 + 1 mol dm−3 Na2 SO4 . (B) Plot of reduction peak current (without stirring) vs. tetraethyl-ethylenetetracarboxylate concentration. (C) Conversion data (filled circles) obtained during electrolysis (with stirring) of 80 × 10−3 mol dm−3 tetraethyl-ethylenetetracarboxylate in acetonitrile in contact to 0.1 mol dm−3 PBS pH 7 + 1 mol dm−3 Na2 SO4 at a potential of −1.25 V vs. SCE as a function of time. Conversion data obtained by 1 H NMR. (D) Plots of the logarithm of electrolysis current for reduction of an 80 × 10−3 mol dm−3 tetraethyl-ethylenetetracarboxylate solution (100 ␮L, at −1.25 V vs. SCE) and the charge vs. time. The dashed line indicates the initial first order current decay line (see text).

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]

T. Fuchigami, T. Tajima, Electrochemistry 74 (2006) 585. C.P. Andrieux, F. Gonzalez, J.M. Saveant, J. Electroanal. Chem. 498 (2001) 171. T. Tajima, T. Fuchigami, J. Synth. Org. Chem. Japan 64 (2006) 406. M.D. Otero, B. Batanero, F. Barba, Tetrahedron Lett. 47 (2006) 8215. J. Yoshida, S. Suga, A. Nagaki, J. Synth. Org. Chem. Japan 63 (2005) 511. C.A. Paddon, M. Atobe, T. Fuchigami, P. He, P. Watts, S.J. Haswell, G.J. Pritchard, S.D. Bull, F. Marken, J. Appl. Electrochem. 36 (2006) 617. H. Kurihara, T. Fuchigami, T. Tajima, J. Org. Chem. 73 (2008) 6888. D. Horii, M. Atobe, T. Fuchigami, F. Marken, Electrochem. Commun. 7 (2005) 35. J.D. Watkins, S.M. MacDonald, P.S. Fordred, S.D. Bull, Y.F. Gu, K. Yunus, A.C. Fisher, P.C. Bulman-Page, F. Marken, Electrochim. Acta 54 (2009) 6908. J.D. Wadhawan, F. Marken, R.G. Compton, Pure Appl. Chem. 73 (2001) 1947. S.M. MacDonald, J.D. Watkins, S.D. Bull, I.R. Davies, Y. Gu, K. Yunus, A.C. Fisher, P.C.B. Page, Y. Chan, C. Elliott, F. Marken, J. Phys. Org. Chem. 22 (2009) 52. J.D. Watkins, S.D. Bull, F. Marken, J. Phys. Chem. C 113 (2009) 15629. J.D. Watkins, F. Amemiya, M. Atobe, P.C. Bulman-Page, F. Marken, Electrochim. Acta 55 (2010) 8808. C.E. Banks, T.J. Davies, R.G. Evans, G. Hignett, A.J. Wain, N.S. Lawrence, J.D. Wadhawan, F. Marken, R.G. Compton, Phys. Chem. Chem. Phys. 5 (2003) 4053. L. Petrik, P. Ndungu, E. Iwuoha, Nanoscale Res. Lett. 5 (2010) 38. B.A. Kakade, V.K. Pillai, J. Phys. Chem. C 112 (2008) 3183. P.G. Whitten, G.M. Spinks, G.G. Wallace, Carbon 43 (2005) 1891.

6770

J.D. Watkins et al. / Electrochimica Acta 56 (2011) 6764–6770

[18] B.P. Binks, Z.G. Cui, P.D.I. Fletcher, Langmuir 22 (2006) 1664. [19] S.M. MacDonald, P.D.I. Fletcher, Z.G. Cui, M. Opallo, J.Y. Chen, F. Marken, Electrochim. Acta 53 (2007) 1175, and references cited therein. [20] J. Matsui, K. Yamamoto, N. Inokuma, H. Orikasa, T. Kyotani, T. Miyashita, Mol. Cryst. Liq. Cryst. 491 (2008) 9. [21] F. Marken, M.L. Gerrard, I.M. Mellor, R.J. Mortimer, C.E. Madden, S. Fletcher, K. Holt, J.S. Foord, R.H. Dahm, F. Page, Electrochem. Commun. 3 (2001) 177. [22] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Wiley, New York, 2001, p. 424.

[23] A.D. Clegg, N.V. Rees, O.V. Klymenko, B.A. Coles, R.G. Compton, J. Electroanal. Chem. 580 (2005) 78. [24] C.A. Paddon, G.J. Pritchard, T. Thiemann, F. Marken, Electrochem. Commun. 4 (2002) 825. [25] C.R. Wilke, P. Chang, AIChE J. 1 (1955) 264. [26] E. Bak, M. Donten, M. Skompska, Z. Stojek, J. Phys. Chem. B 110 (2006) 24635. [27] M. Donten, E. Bak, M. Gniadek, Z. Stojek, F. Scholz, Electrochim. Acta 53 (2008) 5608.