Rapid Prototyped Optically Transparent Thin-Layer ElectrodeHolder for Spectroelectrochemistry in Bench-TopSpectrophotometers

Robert A. Wilson, Tatyana S. Pinyayev, Nellymar Membreno, William R. Heineman*

Department of Chemistry, University of Cincinnati, 301 Clifton Court, Cincinnati, OH 45221-0172, USA*e-mail: heinemwr@ucmail.uc.edu

Received: April 21, 2010;&Accepted: May 26, 2010

AbstractA new optically transparent thin layer electrode (OTTLE) cell and holder have been designed to facilitate spectroe-lectrochemical measurements in standard bench-top absorbance and fluorescence spectrophotometers. The use ofrapid prototyping for the OTTLE cell holder combined with the selection of inexpensive OTE materials results in apractical, low-cost spectroelectrochemical cell. The cell was characterized by thin-layer cyclic voltammetry and coul-ometry of ferricyanide/ferrocyanide. Spectroelectrochemistry of tris-(2,2’-bipyridine) ruthenium(II) chloride (Ru-(bpy)3Cl2) and 1-hydroxypyrene (1-pyOH) was done with commercially available bench-top absorbance and fluores-cence spectrophotometers. The good correlation between the results obtained and the known properties of eachcompound demonstrate that the OTTLE cell and holder provide an effective means for making spectroelectrochem-ical measurements in bench-top absorbance and fluorescence spectrophotometers.

Keywords: Cyclic voltammetry, Fluorescence spectroscopy, Spectroelectrochemistry

DOI: 10.1002/elan.201000267

1. Introduction

Thin-layer spectroelectrochemistry is a proven techniquefor studying electrochemically generated species by ab-sorption and fluorescence [1–4] spectroscopy. It couplesthe ability to change the oxidation state of a compoundconfined in a thin layer of solution next to an opticallytransparent electrode (OTE) with the detection capabili-ties of spectroscopy.

The first use of an optically transparent thin layer elec-trode (OTTLE) for optical monitoring of an electrochem-ically generated species used a gold minigrid sandwichedbetween two microscope slides [5]. The transparency ofthis OTE is due to the physical holes in the grid and isdependent on the quality of micromesh selected (general-ly 100–200 wires/inch). Consequently, the optical range ofthe OTTLE is determined only by the choice of materialfor the cover slide. Alternatively, OTEs consisting of athin layer (100 to 5000 �) of optically transparent andconductive material such as Au, Pt, carbon, or tin-dopedindium oxide (ITO) can be used [6–13]. The transparentproperties of these OTEs depend on the thinness of theconductive layer and the material selection. Thinner con-ducting layers increase OTE transparency, which is bene-ficial for the spectroscopy, but have an adverse effect onresistance, which is detrimental to electrolysis.

The main advantage of an OTTLE is the fast and com-plete electrolysis with simultaneous optical monitoring

due to restricted diffusion in the thin layer of solution.Depending on the area of the electrode exposed to solu-tion and the optical path length, a relatively small volumeof solution is electrolyzed (50 mL or less), which usuallyrequires only a few minutes or less for complete electroly-sis.

Since the initial OTTLE cell, several cells have beendesigned based on the conventional cuvette for use withstandard spectrophotometers [14–22]. A few cells haveincorporated an OTTLE into a standard quartz cuvettewith the optical beam perpendicular to the electrode[14,15]. One such design places the OTTLE cell at a 458angle within a standard cuvette [14]. By placing the cellat a 458 angle to the excitation and emission slits, fluores-cence measurements can be obtained, but the cell suffersfrom high variability between experiments due to incon-sistencies in cell positioning. Another approach to incor-porating an OTTLE into a standard cuvette involves cut-ting away the majority of three sides of the cuvette, leav-ing only the base intact to hold the OTTLE [15]. Whilethis modification is easy to make, the resulting cell re-mains unsuitable for making fluorescence measurementsin spectrophotometers that measure at a 908 angle. An-other spectroelectrochemical cell is the Thin LayerQuartz Glass Spectroelectrochemical cell made by Bio-analytical Systems, Inc. [16]. This cell is similar in size toa conventional cuvette and is designed for use with theUV-absorbance spectrophotometers commonly available

2162 � 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Electroanalysis 2010, 22, No. 19, 2162 – 2166

Full Paper

in research facilities. However, this design is not compati-ble with standard bench-top fluorescence spectrophotom-eters that measure emission at an angle of 908.

Other designs have positioned the electrode parallel tothe optical beam inside a cuvette [17–20] or a cell of simi-lar dimensions [21]. Orientating the optical beam parallelto the electrode increases the optical path length andtherefore improves the optical sensitivity [22]. The major-ity of this work involves absorbance based measurementsin a long optical path thin-layer cell (LOPTLC) [17–22].However, pseudo-thin-layer, long optical path cells havealso been designed for use with standard spectrophotome-ters, including one designed for both absorbance and fluo-rescence measurements. This cell consists of narrow opti-cal channels drilled at 908 angles through a reticulatedvitreous carbon (RVC) electrode [23]. However, a 25–30 min equilibration time is required for each measure-ment due to the 2 mm diameter of the channel.

Here we introduce a new OTTLE cell to be used withconventional spectrophotometers for both absorbanceand fluorescence measurements. The cell consists of twoparts: an OTTLE and a unique cell holder. The cellholder is designed to position the OTTLE, a referenceelectrode, and an auxiliary electrode for both absorbanceand fluorescence measurements and to standardize thedistance between electrodes. The OTTLE cell and holderprovide a reproducible way to perform spectroelectro-chemical experiments in instruments compatible with astandard 1 cm cuvette.

The unique OTTLE cell holder (Figure 1B) was de-signed and fabricated using rapid prototyping technology.This technique was first introduced in the late 1980s andis now used extensively to produce small models andparts [24,25]. Rapid prototyping is based on taking acomputer-designed model and automatically transformingit into a physical object by the sequential delivery of ma-terial to specified points in space. This process creates aphysical model identical to the virtual design. The use ofcomputer-aided design (CAD) software makes creatingvirtual models of specified dimension quick and easy.CAD software and rapid prototyping technology allowsfor the creation of complex structures in an inexpensiveand timely manner.

Here we demonstrate the OTTLE and cell holder forabsorbance and fluorescence measurements in bench-topspectrophotometers.

2. Experimental

2.1. Reagents and Materials

The following chemicals were used without further purifi-cation: potassium ferrocyanide (Aldrich); potassium ni-trate, sodium chloride, sodium bicarbonate, sodium hy-droxide (all from Fisher Scientific); tris-(2,2’-bipyridine)ruthenium(II) chloride hexahydrate ([Ru(bpy)3Cl2·6H2O],GFS Chemicals); 1-hydroxypyrene ([1-pyOH] TorontoResearch Chemicals Inc.); and 200 proof ethyl alcohol

(Pharmco-AAPER). Fe(CN)63� and Ru(bpy)3

2+ solutionswere prepared by dissolving appropriate amounts in1.0 M KNO3 (prepared with deionized water from aBarnstead purification system). 1-pyOH was prepared bydissolving appropriate amounts in 1.0 M NaCl/0.2 MpH 10.5 carbonate buffer/20 % EtOH.

The OTTLE cell is constructed from a quartz glassslide (ESCO products) cut to (1.90 �1.00 cm), 0.018 cm-thick silicone spacers (Specialty Manufacturing Inc., Pine-ville, NC), and ITO-coated glass slides (Corning 1737Fand 7059, 11–50 W/square, 130-nm-thick film on 1.1-mmglass, Thin Film Devices, Anaheim, CA) with dimensionsof 4.00�1.00 cm. Silicone spacers cut to approximately1.90� 0.20 cm are placed onto the edges of the ITO glassslide and sandwiched between a quartz slide and the ITO(Figure 1A). Two-part quick-set epoxy (Loctite) is appliedalong the edges of the spacers and allowed to cure for 2 hto hold the components together. OTTLE cells made inthis manner were capable of approximately 8 hr of con-tinuous use. The exposed ITO above the quartz slide isused for electrical contact.

2.2. Instrumentation

For all experiments the electrochemical cell consisted of aPt wire auxiliary electrode, a miniature Ag/AgCl refer-ence electrode (3 M KCl, Cypress Systems), and anOTTLE. Thin layer cyclic voltammetry and coulometryof ferricyanide/ferrocyanide in 1.0 M KNO3 was per-formed on a BAS 100 Electrochemical Analyzer (Bioana-lytical Systems).

Nernst plots of absorbance and emission were con-structed by controlled potential electrolysis of Ru(bpy)3

2+

and 1-pyOH on an Epsilon Electrochemical Workstation(Bioanalytical Systems). Absorbance spectra of 1.00 mMRu(bpy)3

2+ in 1.0 M KNO3 were obtained with a VarianCary 50 Bio UV-Visible Spectrophotometer. Emissionspectra of 0.10 mM 1-pyOH (ex. 280 nm) were obtainedusing a Varian Cary Eclipse Fluorescence Spectropho-tometer with an excitation slit width for 5 nm, an emis-sion slit width for 1.5 nm, and a PMT voltage of 670 V.

Emission spectra of 10.0 mM Ru(bpy)32 + in 1.0 M

KNO3 were acquired using a custom-made system. Thissystem consists of a laser (441.6 nm HeCd model1K4153R-C, Kimmon Electric Co.), light control modules(shutter, attenuator, and focusing optics), a monochroma-tor (0.3 m focal length, triple grating turret), a photon-counting phototube (Acton Research Corp.), and a com-puter and control electronics (NCL and Spectra-Sensesoftware, Acton). Light from the laser was focused ontothe polished end of a 6-around-1 fiber optic bundle(RoMack Inc.). Laser power was attenuated to 0.5 mWand the sample was exposed to the laser light only duringdata acquisition to minimize photodegradation.

Electroanalysis 2010, 22, No. 19, 2162 – 2166 � 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.electroanalysis.wiley-vch.de 2163

Rapid Prototyped Optically Transparent Thin-Layer Electrode Holder

3. Results and Discussion

OTTLE cells have previously been shown to be useful foroptical characterization of electrochemical reactions. Themajority of these cells have focused on absorbance basedspectroelectrochemical detection. By incorporating thisunique cell holder (Figure 1B), created using SolidWorkssoftware and rapid prototyping technology, the OTTLEcell can be easily used for both absorbance and fluores-cence based measurements. The outer dimensions of theholder are identical to the standard cuvette commonlyused in spectrophotometers (1 �1 �4 cm). However,unlike a standard cuvette, all four side walls have win-dows for light passage. Two of the inside diagonal cornershave slots for the OTTLE cell. This design positions thethin layer cell at a 458 angle with respect to each of theholder�s walls. The remaining two corners are designed tohold a commercially available Ag/AgCl miniature refer-ence electrode and a Pt wire auxiliary electrode. Thebottom of the holder accommodates 1 mL of analyte so-lution. A detailed schematic created with SolidWorkssoftware is available as supporting material. The appara-tus is relatively inexpensive due to the low-cost materialsused for the holder (< $10), the commercial availabilityof inexpensive ITO thin film electrodes, and the low costof rapid prototyping compared to conventional machiningof the holder.

While numerous OTEs can be used, ITO has become apopular choice due to its low cost, good optical transpar-ency over the visible range, and durability. The major dis-advantages of ITO OTEs are the large resistance result-ing from the thin layer of indium tin oxide and low trans-parency in the UV region. The resistance of the electrodebecomes a larger issue when used in the construction ofan OTTLE because thin layer cells already have a pro-nounced resistance due to the relatively thin solutionlayer. The resistance problem is further exacerbated bythe large current resulting from the large electrode areatypically used. In order to diminish this effect theOTTLE cell�s height was reduced to the minimum al-lowed by our bench-top spectrophotometer (1.90 cm), thescan rate was reduced to 2 mV/s, and a high concentrationof supporting electrolyte was used (1.0 M).

A typical cyclic voltammogram for the model ferri/fer-rocyanide reversible couple in the 1.90 cm cell can be

seen in Figure 2. Voltammograms recorded at a 2 mV/sscan rate resulted in a peak separation of 195 mV. Thisvalue is primarily the result of the uncompensated iRdrop between the reference and ITO working electrodesand the resistance resulting from the ITO thin film. Theformal potential of the ferri/ferrocyanide redox reactionwas determined to be 283 mV from Figure 2. This resultis slightly more positive than a literature value [26].

The volume of the cell was determined by coulometryof a 4.00 mM K3Fe(CN)6 in 1.0 M KNO3 solution. Thetotal charge (QT) was measured by complete electrolysisof ferricyanide to ferrocyanide by a potential step from600 mV to 0 mV. Similarly, the background charge (QB)was measured in the supporting electrolyte solution andsubtracted from QT to give the Faradaic charge (QF). Sub-stituting QF and other known values into Faraday�s lawallows for the determination of the volume of the thinlayer cell.

QF ¼ QT�QB ¼ nFVC ð1Þ

Experimentally, QF was found to be 1.07 �10�2 C, re-sulting in a thin-layer solution volume of 27.7 mL. Usingthis volume and the measured area of the cell, the thick-ness of the cell was determined to be 1.46�10�2 cm. Toverify this result a simple calculation of a newly madecell�s thickness was done using absorbance measurementsand Beer�s Law, giving a path length of 1.60�10�2 cm.This path length results in a calculated cell volume of30.3 mL. The discrepancy in these results is most likelydue to the uncertainty in determining the point of com-plete electrolysis of ferricyanide to ferrocyanide duringthe single step, coulometry experiment.

The cell was evaluated spectroelectrochemically by re-cording spectra of the model analyte Ru(bpy)3

2+. Figure 3shows the absorbance spectra of Ru(bpy)3

2+ in anOTTLE for a series of applied potentials. The spectrum

Fig. 1. OTTLE cell and holder. (A) Assembly of cell front andside view. (B) Cell holder, 3D and top down view.

Fig. 2. Thin-layer cyclic voltammogram of 2.00 mM Fe(CN)64�,

1.0 M KNO3, 2 mV/s versus Ag/AgCl.

2164 www.electroanalysis.wiley-vch.de � 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Electroanalysis 2010, 22, No. 19, 2162 – 2166

Full Paper R. A. Wilson et al.

with the highest absorbance value was recorded after ap-plication of +900 mV for 4 min, which caused completereduction of Ru(bpy)3

3+ ([O]/[R]<0.001). The spectrumwith the lowest absorbance value was recorded after ap-plication of +1200 mV for 4 min, causing complete oxida-tion of Ru(bpy)3

2 + ([O]/[R]>1000). Intermediate spectracorrespond to the intermediate values of Eapplied. For eachvalue of applied potential, the equilibrium value of theratio [Ox]/[Red] in the thin solution layer was calculatedfrom the appropriate spectra using the previously report-ed method [27]. A Nernst plot of Eapplied vs. log([Ox]/[Red]) at 454 nm gave an E8’ value of 1.07�0.01 Vvs. a Ag/AgCl reference electrode, which is in closeagreement with previously reported values [28,29]. Theslope of the plot is �59.3�0.3 mV (R2 =1.00), which cor-responds to an n value of 0.995. The slope of the plot isonly slightly higher than the theoretical value of 59.2 mVand gave an n value of ~1, which corresponds nicely tothe actual value.

Similar results were obtained (not shown) for the emis-sion spectra of 10 mM Ru(bpy)3

2+ using the custom opti-cal system previously described. Utilizing the potentialsfrom Figure 3, a Nernst plot was obtained for the data re-sulting in a log ([Ox]/[Red]) between ~1 and �1. Thisplot of Eapplied vs. log ([Ox]/[Red]) for the emission dataresults in a slope of �58.0�4 mV (R2 =0.968), which cor-responds to an n value of 1.02. The E8’ value determinedwas 1.07�0.01 V vs. Ag/AgCl. These values are in goodagreement with the values obtained from the absorbancebased measurements and from the literature [28,29].

The main advantage of this cell and holder design is itsability to be used with any optical setup that measures1.00 cm above the base of the cuvette. Here we haveshown the use of this design with a model analyte on astandard absorbance spectrophotometer and a custom

fluorescence spectrophotometer. In order to show theOTTLE cell and holder�s versatility, the more complex ir-reversible electrochemical oxidation of 1-pyOH wasmonitored using a bench-top Cary Eclipse fluorescencespectrophotometer. 1-pyOH undergoes a one-electronelectrochemical oxidation at a pH-dependent potential,ultimately resulting in an ECE mechanism [30]. This ana-lyte is a good representative of analytes that do not ex-hibit the ideal reversible electrochemistry shown bymodel compounds.

At pH 10.5, 1-pyOH is in the deprotonated 1-pyO�

form, which results in a broad emission spectrum from~400 to 500 nm (Figure 4). 1-pyO� was incrementallyconverted from its reduced to its oxidized form. Spectrafor the fully reduced, fully oxidized, and intermediateforms of 1-pyO� are shown (Figure 4). A Nernst plot cor-responding to this data gave an E8’ of 162�1 mV vs. aAg/AgCl reference electrode. The slope of the plot was�58.1�2.4 mV (R2 =0.992), which results in an n valueof 1.02. Though in a different solution matrix, this data isconsistent with previously published results [30].

4. Conclusions

The design of a new rapid prototyped OTTLE holder forspectroelectrochemical measurements in standard bench-top absorbance or fluorescence spectrophotometers hasbeen described. The performance of this cell holder hasbeen evaluated by spectroelectrochemical detection ofmultiple analytes in a variety of optical setups. The abilityto couple this cell with fluorescence detection offers thepossibility of lower detection limits than similar cells thatare only capable of absorbance detection. The rapid andcomplete electrolysis of the analyte in the OTTLE andthe versatility provided by the rapid prototyped holdermake this design a useful method for spectroelectrochem-

Fig. 3. Thin-layer absorbance spectra and a plot of Eapplied vs.log ([O]/[R]) at 454 nm of 1.00 mM Ru(bpy)3

2+ , 1.0 M KNO3 fordifferent values of Eapplied. Cell thickness 1.60� 10�2 cm. From topto bottom 900, 1020, 1050, 1062, 1074, 1086, 1098, 1110, 1122,1134, 1200 mV versus Ag/AgCl.

Fig. 4. Thin-layer emission spectra (ex. 280 nm) and a plot ofEapplied versus log ([O]/[R]) at 452 nm of 1.00� 10�4 M 1-pyOHfor different values of Eapplied. From top to bottom 0, 145, 150,160, 170, 180, 190, 200, 210, 400 mV versus Ag/AgCl.

Electroanalysis 2010, 22, No. 19, 2162 – 2166 � 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.electroanalysis.wiley-vch.de 2165

Rapid Prototyped Optically Transparent Thin-Layer Electrode Holder

ical detection in laboratories with commercial spectro-photometers.

References

[1] W. R. Heineman, Anal. Chem. 1978, 50, 390A.[2] W. R. Heineman, C. W. Anderson, H. B. Halsall, M. M.

Hurst, J. M. Johnson, G. P. Kreishman, B. J. Norris, M. J.Simone, C.-H. Su, in Electrochemical and SpectrochemicalStudies of Biological Redox Components (Ed. K. Kadish),Advances in Chemistry Series, No. 201, ACS, Washington,DC 1982, pp. 1–21.

[3] W. R. Heineman, F. M. Hawkridge, H. N. Blount, in Electro-analytical Chemistry (Ed. A. J. Bard), Marcel Dekker, NewYork 1984, pp. 1–113.

[4] S. Dong, J. Niu, T. M. Cotton, Meth. Enzymol. 1995, 246,701.

[5] R. W. Murray, W. R. Heineman, G. W. O’Dom, Anal. Chem.1967, 39, 1666.

[6] W. Thornmann, P. van den Bosch, A. M. Bond, Anal. Chem.1985, 57, 2764.

[7] M. Samuelson, M. Armgarth, C. Nylander, Anal. Chem.1991, 63, 931.

[8] H. A. O. Hill, N. A. Klein, I. S. M. Psalti, N. J. Walton, Anal.Chem. 1989, 61, 2200.

[9] K. R. Wehmeyer, M. R. Deakin, R. M. Wightman, Anal.Chem. 1985, 58, 1913.

[10] C. E. Chidsey, B. J. Feldman, C. Lundgren, R. W. Murray,Anal. Chem. 1986, 58, 601.

[11] A. Yildiz, P. T. Kissinger, C. N. Reilley, Anal. Chem. 1968,40, 1018.

[12] O. Niwa, H. Xu, H. B. Halsall, W. R. Heineman, Anal.Chem. 1993, 65, 1559.

[13] T. P. DeAngelis, R. W. Hurst, A. M. Yacynych, H. B. Mark,W. R. Heineman, J. S. Mattson, Anal. Chem. 1977, 49, 1395.

[14] M. J. Simone, W. R. Heineman, G. P. Kreishman, J. ColloidInterf. Sci. 1982, 86, 295.

[15] C. W. Anderson, M. R. Cushman, Anal. Chem. 1982, 54,2122.

[16] http://www.basinc.com/products/ec/sec.html; accessed May2010.

[17] Y. Gui, S. A. Soper, T. Kuwana, Anal. Chem. 1988, 60, 1645.[18] Y. Zhangyu, G. Tlande, Q. Mei, Anal. Chem. 1994, 66, 497.[19] N. J. Simmons, M. D. Porter, Anal. Chem. 1997, 69, 2866.[20] D. Daniel, I. G. R. Gutz, Electroanalysis 2001, 13, 681.[21] J. Niu, S. Dong, Electroanalysis 1995, 7, 1059.[22] J. Zak, M. D. Porter, T. Kuwana, Anal. Chem. 1983, 55,

2219.[23] Y. F. Lee, J. R. Kirchhoff, Anal. Chem. 1993, 65, 3430.[24] P. K. Wright, 21st Century Manufacturing, Prentice-Hall,

New Jersey 2000.[25] P. K. Wright, 21st Century Manufacturing, Prentice-Hall,

New Jersey 2001.[26] I. M. Kolthoff, W. J. Tomsicek, J. Phys. Chem. 1935, 39, 945.[27] T. P. DeAngelis, W. R. Heineman, J. Chem. Edu. 1976, 53,

594.[28] G. Villemure, T. J. Pinnavaia, Chem. Mater. 1999, 11, 789.[29] A. F. Slaterbeck, T. H. Ridgway, C. J. Seliskar, W. R. Heine-

man, Anal. Chem. 1999, 71, 1196.[30] K. C. Honeychurch, J. P. Hart, N. Kirsch, Electrochim. Acta

2004, 49, 1141.

2166 www.electroanalysis.wiley-vch.de � 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Electroanalysis 2010, 22, No. 19, 2162 – 2166

Full Paper R. A. Wilson et al.