Anal. Chem. 1996, 68, 199-202
Variable Path Length Transmittance Cell for Ultraviolet, Visible, and Infrared Spectroscopy and Spectroelectrochemistry Paul A. Flowers* and Sherry-Ann Callender
Department of Physical Science, Pembroke State University, Pembroke, North Carolina 28372
The design and characteristics of a transmittance cell for ultraviolet, visible, and infrared spectroscopy and spectroelectrochemistry are described. Through modification of a previously reported design, this cell employs threaded glass connectors as insertion ports for either quartz- or silicon-windowed tubes, thus permitting essentially continuous variation of the optical path length from ∼0.050 to 200 mm. Though the initial fabrication requires skillful glassblowing, once constructed, the cell’s simple design allows for rapid and reproducible disassembly/reassembly between experiments. The utility of the cell for a diversity of fluid samples is demonstrated through applications to water, aqueous ferricyanide, ferrocene in methylene chloride, and acetone vapor. Since the first reported spectroelectrochemical (SEC) study,1 apparatus and methodology have been developed for application to virtually every region of the electromagnetic spectrum, from γ rays to radio waves.2 To date, the vast majority of studies have been performed in the ultraviolet/visible (UV/vis) and infrared (IR) regions; consequently, reported efforts continue toward the design of useful SEC cells for these spectral regions.3 SEC cells for UV/vis and IR applications typically employ either transmittance or external reflectance sampling, the former being more common for examination of solution species4 and the latter for electrode surface studies.5 The optically transparent thin-layer electrode (OTTLE) is by far the most prevalent design among the reported transmittance cells. Despite their proven utility, there are several drawbacks associated with these types of cells. For example, they often require the use of adhesive materials that may be dissolved by common organic solvents, mechanical spacers that are subject to deformation and subsequent path length errors, and multiple gasket arrangements that result in complex assembly procedures and are prone to leakage. Work by the primary author has for several years been concerned with the development and application of spectroelectrochemical methods suitable for use with “low-temperature” molten salts (melting points below ∼200 °C). The generally corrosive and hygroscopic nature of these solvents imposes special experimental constraints, e.g., the use of inert cell materials and (1) Kuwana, T.; Darlington, R. K.; Leedy, D. W. Anal. Chem. 1964, 36, 2023. (2) Gale, R. J., Ed. Spectroelectrochemistry: Theory and Practice; Plenum Press: New York, 1988. (3) For example, see: Ryan, M. D.; Bowden, E. F.; Chambers, J. Q. Anal. Chem. 1994, 66, 360R-427R. (4) Heineman, W. R. J. Chem. Educ. 1983, 60, 305. (5) Ashley, K.; Pons, S. Chem. Rev. 1988, 88, 673. 0003-2700/96/0368-0199$12.00/0
© 1995 American Chemical Society
constant isolation of the salts from the atmosphere.6 IR analyses in these media are particularly difficult, primarily because of a lack of suitable window materials; diamond and silicon have been used most frequently.7 This paper reports the design and characteristics of a versatile transmittance cell for UV/vis and IR spectroscopy and spectroelectrochemistry. A modification of a previously reported IR cell design,8 this cell employs threaded glass connectors as insertion ports for either quartz- or silicon-windowed tubes, thus permitting continuous variation of the optical path length from ∼0.050 to 200 mm. Threaded glass connectors also serve as ports for electrodes and a stopcock assembly, resulting in a vacuum-tight vessel which facilitates solution degassing as well as anaerobic operation. The cell geometry permits close placement of reference and counter electrodes to an optically transparent working electrode (OTE) and thus minimizes ohmic potential drop during SEC experiments. Spectroscopic and spectroelectrochemical applications illustrating the versatility of this cell are described herein. EXPERIMENTAL SECTION Reagents. Reagent-grade acetone (Fisher Scientific), ferricyanide (Mallinkrodt), ferrocene (Eastman Kodak), dichloromethane (Aldrich), and potassium chloride (Fisher Scientific) were used as received from the vendor. Tetrabutylammonium hexafluorophosphate (TBAH, Aldrich) was dried in vacuo prior to use. Aqueous solutions were prepared using distilled/deionized water (R > 16 MΩ). Cell Construction. A diagram of the cell is shown in Figure 1. The windowed tubes were constructed by torch-sealing either silicon or quartz disks (10 mm × 2 mm, Spectra-Tech, Inc.) into the ends of Pyrex or quartz tubes, respectively. This procedure is most successful when the window outside diameter and tube inside diameter are closely matched; the tube is then plugged with the window at one end and heated with a torch to create the seal. When sealing Si windows to Pyrex tubes, the window/tube assembly is heated within a larger quartz tube under a constant nitrogen purge to minimize oxidation of the window faces. The sealing success rate is high, and tubes that leak may be resealed by repeating the heating procedure described above. Leakage is detected by evacuation of the tubes; successful seals will hold vacua to less than 10 mTorr. (6) Norvell, G. E.; Mamantov, G. In Molten Salt Techniques; Lovering, D. G., Gale, R. J., Eds.; Plenum Press: New York, 1983; Vol. 1, Chapter 7. (7) Gilbert, B. In Molten Salt Chemistry: An Introduction and Selected Applications; Mamantov, G., Marassi, R., Eds.; D. Reidel Publishing Co.: Boston, MA, 1987; p 201. (8) Flowers, P. A.; Mamantov, G. Anal. Chem. 1989, 61, 190.
Analytical Chemistry, Vol. 68, No. 1, January 1, 1996 199
Figure 2. Infrared spectra of liquid water (upper, ∼0.050 mm path length) and 700 ppm acetone vapor (lower, 198 mm path length).
Figure 1. Schematic diagram of the adjustable path length cell showing side (A) and end (B) views.
The cell body was constructed from threaded glass connectors and a Teflon stopcock assembly purchased from Ace Glass, Inc. This work employed Viton O-rings, though alternate materials appropriate to the chemical nature of the sample may be used (e.g., Teflon-coated rubber). For most path lengths, the windowed tubes were inserted face-in as depicted in Figure 1; after positioning as desired, the tubes were sealed into place by tightening the connector bushings. For maximal path lengths (i.e., greater than the cell body length), the tubes were inserted face-out. The minimum liquid sample volume of the cell is ∼10 mL when the minimum path length configuration is employed. This means of varying optical path length is conceptually similar to that reported by Shu and Wilson9 for a cell with one fixed window and a movable light pipe. The use of two adjustable windows as described here, however, offers several advantages: (1) the path length may be varied over a much greater range; (2) the same cell body may be used in both the UV/vis and IR spectral regions as both windows may be exchanged; and (3) both windows may be removed for cleaning or polishing if necessary. A 10 mm × 13 mm platinum screen (Aesar, 80-mesh) welded to a Pt lead at the end of a sealed Pyrex tube was employed as the optically transparent working electrode (OTE) in spectroelectrochemical studies. Platinum auxiliary and quasi-reference electrodes were isolated in fritted Pyrex tubes (Ace Glass, C porosity). Instrumentation. Infrared spectra were acquired using a BioRad FTS-40 Fourier transform infrared spectrometer configured (9) Shu, F. R.; Wilson, G. S. Anal. Chem. 1976, 48, 1676-1679.
200 Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
for standard mid-IR analysis. A Hewlett-Packard 8452A diode array spectrometer was used to measure ultraviolet/visible spectra. Electrochemical measurements were performed using a BioAnalytical Systems CV-27 voltammograph. Voltammograms were recorded on a Houston Omnigraphic Model 2000 X-Y recorder. SEC Procedure. Solutions were prepared, loaded in the cell, and thoroughly degassed by bubbling nitrogen prior to SEC analysis. The auxiliary and reference electrode chambers were filled with supporting electrolyte solution. The infrared absorbance spectra are presented in the “potential difference” mode, i.e., calculated from single-beam spectra of the sample solution obtained with the OTE at different applied potentials. These spectra exhibit positive and negative features due to species produced and consumed, respectively, as the working electrode potential is changed. An attractive aspect of this mode of data presentation is that only those features due to electroactive species are observed; absorptions due to the solvent, inactive solutes, cell windows, and atmospheric water and carbon dioxide are effectively eliminated. RESULTS AND DISCUSSION Path Length Limits. The minimum cell path length was estimated to be ∼0.050 mm on the basis of the 1960 cm-1 absorbance of neat benzene.10 This path length is sufficiently short to permit transmittance sampling in aqueous media, as shown by the infrared spectrum of neat water in the upper portion of Figure 2. The maximum absorbance around 3400 cm-1 is only (10) Willard, H. H.; Merritt, L. L., Jr.; Dean, J. A.; Settle, F. A., Jr. Instrumental Methods of Analysis, 6th ed.; Wadsworth Publishing Co.: Belmont, CA, 1981; p 206.
A
B
Figure 3. (A) Cyclic voltammograms for 1.5 mM ferricyanide in 0.1 M KNO3 measured at scan rates of 1, 2, 3, and 4 mV/s. (B) Absorbance spectra for this system acquired at ∼10 s intervals after stepping the OTE potential from +0.1 to -0.4 V vs Pt QRE (path length ∼0.2 mm). Figure 4. Potential difference spectra for 10 mM ferrocene in 0.1 M TBAH/methylene chloride acquired after stepping the OTE potential from -0.5 to 0.0 V vs Pt QRE (lower traces) and from 0.0 to -0.5 V vs Pt QRE (upper traces) (path length ∼0.2 mm).
∼1.2, leaving the majority of the mid-IR region suitable for quantitative studies of solutes. Demonstration of the maximum cell path length (198 mm as measured with a ruler) is provided by the lower spectrum shown in Figure 2. This spectrum was measured after permitting a volume of liquid acetone at the bottom of the sealed cell to achieve its equilibrium vapor pressure, estimated to be 226 Torr at 24.5 °C.11 Assuming ideal behavior, the spectrum shown corresponds to ∼700 ppm acetone vapor. Spectroelectrochemical Applications. Cyclic voltammograms measured at various scan rates for ∼1.5 mM ferricyanide in 0.1 M KNO3 are shown in Figure 3. As expected, a rather severe edge effect is manifest in these voltammograms due to exposure of the thin layer to the bulk solution about its entire perimeter. Separation of the cathodic and anodic peaks is observed to increase from ∼50 to 100 mV as the scan rate is increased from 1 to 4 mV/s. This is likely a result of ohmic drop across the OTE due to the resistance of the thin solution layer. Despite these shortcomings, the quality of these voltammograms indicates that the cell design permits reasonably accurate control of the OTE potential. UV/vis absorbance spectra for this system are shown in Figure 3. Upon stepping the OTE potential from +0.1 to -0.4 V vs Pt QRE, absorbance decreases at 302 and 420 nm and an absorbance increase below 280 nm are exhibited, representing the conversion of ferricyanide to ferrocyanide. These data are in good agreement with previously reported results,12 demonstrating the utility of this cell for SEC studies in the UV/vis region.
The presence of the OTE between the cell windows limits the minimum optical path to ∼0.2 mm (the Pt mesh thickness) and thus precludes SEC measurements in aqueous media. An organic solvent-based system was therefore chosen to illustrate applications in this region. Shown in Figure 4 are potential difference spectra for 10 mM ferrocene in 0.1 M TBAH/methylene chloride.
(11) Dean, J. A., Ed. Lange’s Handbook of Chemistry; McGraw-Hill Book Co.: New York, 1985; p 10-37.
(12) Pons, S.; Datta, M.; McAleer, J. F.; Hinman, A. S. J. Electroanal. Chem. 1984, 160, 369-376.
Figure 5. Plot of absorbance at 1107 cm-1 versus time for the system of Figure 4 after stepping the OTE potential from 0.0 to -0.5 V vs Pt QRE.
Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
201
Despite strong absorptions by the solvent and supporting electrolyte, this mode of data presentation is effective in showing analyte features in three different spectral regions. Each of the three plots in Figure 4 shows potential difference spectra for both forward (lower spectra) and reverse (upper spectra) potential steps. Except for the expected difference in the sign of the absorbance changes (positive vs negative), the upper and lower traces are identical, indicating the chemical reversibility of the couple. These data are also in agreement with published work.13 The exhaustive electrolysis time of the cell was estimated by monitoring the absorbance change at 1107 cm-1 for the ferricinium/ferrocene couple after stepping the OTE potential from 0.0 to -0.5 V vs Pt QRE. As may be seen in Figure 5, the cell’s thin layer volume may be completely electrolyzed in ∼1 min. In summary, a simple and versatile cell suitable for spectroscopic and spectroelectrochemical studies in the ultraviolet, visible, (13) Bullock, J. P.; Boyd, D. C.; Mann, K. R. Inorg. Chem. 1987, 26, 30843086.
202
Analytical Chemistry, Vol. 68, No. 1, January 1, 1996
and infrared regions has been described. Among the cell’s generally beneficial traits are its continuously adjustable path length, high degree of chemical comptability, ease of assembly and disassembly, and wide spectra range. Regarding SEC applications, the cell permits adequate control of the OTE potential and exhibits a relatively short exhaustive electrolysis time. Acknowledgment is made to the Donors of the Petroleum Research Fund, administered by the American Chemical Society, for support of this research. The authors thank Elder Mellon of the University of Tennessee glass shop for invaluable assistance in the construction of the cell.
Received for review June 13, 1995. Accepted October 11, 1995.X AC950580W X
Abstract published in Advance ACS Abstracts, November 15, 1995.
