“Cosmetic electrochemistry”: the facile production of graphite microelectrode ensembles

Abstract

The facile and rapid production of microelectrode ensembles is shown to be possible using off-the-shelf cosmetic products and is exemplified with the electrochemical sensing of a toxic metal offering a novel fabrication methodology.

---

Microelectrode arrays are a unique tool in the electrochemist’s arsenal due to their well identified advantages such as large current densities, high spatial resolution, reduced capacitive-charging currents and find beneficial use in a range of applications such as clinical chemistry, electroanalysis, biosensing, environmental sensing and electrophysiology.^1–8 In microelectrode arrays, microelectrodes are at a fixed distance from their nearest neighbour and an alternative is to have a random array of microelectrodes which are termed as ensembles due to no regular spacing between neighbouring microelectrodes.^3,9 It has been shown by simulations that a random array can produce the same (but never greater) current–potential response as that of a regular array of equal macroscopic coverage¹⁰ and, due to their nature, random arrays are generally easier to fabricate. The fabrication of microelectrode arrays traditionally employs lithography but other approaches are possible such as sealing thousands of microelectrodes in epoxy resin, injection of liquid conductors into porous insulators, ion beam milling and attaching polymeric membranes onto electrode surfaces.^3,4,11–14 Inspired by the technological importance of microelectrode arrays/ensembles in a plethora of areas, we explore the novel and rapid methodology for producing graphite microelectrode ensembles.

Screen printed macroelectrodes (3 mm diameter) were fabricated as reported previously^15,16 and their electrochemical performance examined in 1 mM potassium ferrocyanide–1 M potassium chloride¹⁷ as depicted in Fig. 1. The heterogeneous rate constant, k^o, was evaluated from fitting of the voltammetric peaks over a range of scan rates with a simulation package which was found to correspond to 0.7 × 10⁻³ cm s⁻¹ indicating a quasi-reversible electron transfer process in agreement with previous studies.^15,16 The screen printed electrode was then ‘sprayed’ with a commercial deodorant, “Nivea For Men Antiperspirant”. The modified macroelectrode was then examined using potassium ferrocyanide, and, as depicted in Fig. 1, a change in the voltammetric profile is clearly evident. Increasing the spray time has a dramatic effect compared to the bare macroelectrode where the peak-to-peak separation increases and the magnitude of voltammetric peaks decreases, the response of which is consistent with that of a partially blocked electrode.¹⁸ The heterogeneous rate constant observed at the partially blocked electrode surface, k^o_obs, is related to the fractional coverage (θ) and the heterogeneous rate constant observed at the unmodified (bare) electrode k^o_barevia:¹⁸

k^o_obs = k^o_bare(1 − θ)

Note that for an overlapping random distribution, the real coverage is given by:¹⁹

θ_R = 1 − e^−θ

Consequently the real fractional coverage, θ_R, of the partially blocked electrode from increasing spray times was deduced to be 0.3 (±0.1) and 0.5 (±0.12). In the case where quasi-reversible voltammetric profiles are obtained in the accessible range of scan rates, as found here, information for the diffusion domain sites and their sizes is unable to be gathered. However an approximate value of the active site radius (R_a) may be estimated (the exposed working electrode area). The maximal value for R₀ (which is the radius of the diffusion domain, blocking part) can be determined by:¹⁸
where for disc-type active sites:

B(1 − θ_R) = 0.3(1 − θ_R)^−½

In the above expressions, F is the Faraday constant, D is the diffusion coefficient of the electro-active species (6.5 × 10⁻⁶ cm² s⁻¹), T is the temperature, R is the gas constant and ν is the scan rate. The average radius of the active sites, R_a, which is the underlying substrate, can then be estimated from:

R_a = R₀(1 − θ_R)^½

It follows that R_a ≤ 2.0 (±0.2) μm and thus the surface now consists of randomly distributed graphite domains which have a radius smaller than 2 microns.

Fig. 1 (A) Cyclic voltammetric profiles obtained at a bare screen printed electrochemical platform (dotted line) and then following modification with the cosmetic product at a distance of 200 mm for 2, 5, 8 and 12 seconds. (B) shows the profile of the 8 seconds modification. All scans recorded at 0.1 V s⁻¹vs. SCE.

SEM images are shown in Fig. 2 where comparison of a standard electrochemical platform and a spray modified electrochemical platform is observed to cover the electrode surface and a decrease in the ‘webbed' aspect is clearly evident indicating that the polymer in the cosmetic spray has filled in these gaps and thus reduces the amount of accessible graphite which is attributable for the reduction in the voltammetric profiles. Noticeable is the presence of smaller clumps of polymer and spherical objects with EDAX indicating that these are aluminium which is a common component in cosmetic ‘antiperspirant’ sprays. We have found that the aluminium, which is commonly used in deodorant products, is not electrochemically active and consequently does not interfere with our electrochemical measurements. A recent paper by Zen et al.²⁰ reported on an electroanalytical method to determine the aluminium content of deodorant products and also reported that electrochemical signals are only possible on silver screen printed electrodes and not graphite screen printed electrodes, confirming our observations.

Fig. 2 SEM images of the unmodified (top) and spray modified (bottom) screen printed electrodes

From modification with a cosmetic product, a macroelectrode has been turned into a partially blocked electrode with numerous graphite microdomains. We can estimate the number of graphite domains from the following:

where A is the electrode area of the macroelectrode. From this we can estimate the number of graphite domains to be of the order ∼4 × 10⁵. The voltammetric profiles indicate diffusional overlap between neighbouring microdomains such that there is no regular spacing and thus an ensemble³ of graphite microdomains has been fabricated. Other ways of fabricating ensembles include covering the electrode surface with inert materials,¹⁹ sealing microelectrodes into epoxy resin,³ covering the electrode surface with a polymer,²¹ and sonochemical fabrication.²² Clearly our approach allows the fabrication of microdomains within seconds and given the low cost of the underlying electrode substrate and the cosmetic modifier these electrodes are very economical. We believe that through the controlled modification with the cosmetic product this is a potential manufacturing route for these devices.

Last to explore the electroanalytical utility of our fabricated microdomain electrode, we turn to showing proof-of-concept for the electrochemical detection of lead using cathodic stripping voltammetry.²³ The cathodic stripping is, due to its very nature, highly selective, with chromium, nickel, cadmium and zinc having no measurable effect on the electroanalytical measurement. In the case of copper and iron, these may be detected by this methodology but their stripping peaks occur at well resolved potentials from that of lead.²³ Using a solution of 0.1 M nitric acid and a deposition potential of +1.65 V (vs. SCE) in accordance with previous studies, additions of lead were made over the range 0 to 300 μM using a 30 s accumulation time.

Fig. 3 depicts a typical voltammetric profile where a large and easily quantifiable peak is observed at +1.38 V (vs. SCE) which is in excellent agreement with literature reports.²³ Analysis of the peak height (I_H) plotted against added lead concentration (as shown in the inset ofFig. 3) reveals two distinct linear ranges, the first over the range 20 to 50 μM (I_H/A = 3.0 × 10⁻¹ A M⁻¹ − 6.2 × 10⁻⁶ A; R² = 0.987) and 75 μM to 200 μM (I_H/A = 8.6 × 10⁻² A M⁻¹ + 8.8 × 10⁻⁶ A; R² = 0.988). Based on the first linear part, the limit of detection (3σ) was found to correspond to (n = 3) 9.5 μM. This linear range and the un-optimised detection limit are comparable to ultrasound-assisted deposition using a boron-doped diamond electrode²³ employing a 60 s accumulation time. This comparability is due to the enhanced mass transport at the graphite microdomains.

Fig. 3 Square-wave voltammogram in the absence (dashed line) and presence (solid line) of 20 μM lead (in 0.1 M nitric acid) using the graphite microdomain ensemble. Parameters: square-wave voltammetry using a conditioning potential at −0.5 V (vs. SCE) for 20 s followed by +1.65 V for 30 s. Spraying conditions were 200 mm for 12 s. The inset shows the analysis of peak height from additions of lead.

To conclude, we have demonstrated that microelectrode behaviour can be conferred on a macroelectrode through the use of readily available commercial cosmetics. That is, by taking a cosmetic product which is then used to spray a macroelectrode surface, a partially blocked electrode is rapidly and easily fabricated which exhibits steady-state type voltammetric behaviour. The polymer contained within the cosmetic product rapidly binds to the electrode surface and sets within seconds coating the electrode surface but leaving the underlying graphite electrode exposed in the form of graphite micron-sized sites; this new application of commercial products with electrochemistry is aptly termed “cosmetic electrochemistry”.

Notes and references

1. J. Gou and E. Lindner, Anal. Chem., 2009, 81, 130–138.
2. X.-J. Huang, A. M. O'Mahony and R. G. Compton, Small, 2009, 5, 776–788.
3. S. Fletcher and M. D. Horne, Electrochem. Commun., 1999, 1, 502–512.
4. Y. H. Lanyon, G. De Marzi, Y. E. Watson, A. J. Quinn, J. P. Gleeson, G. Redmond and D. W. M. Arrigan, Anal. Chem., 2007, 79, 3048–3055.
5. M. Wei, C. Terashima, M. Lv, A. Fujishima and Z.-Z. Gu, Chem. Commun., 2009, 3624–3626.
6. M. Varshney and Y. Li, Biosens. Bioelectron., 2009, 24, 2951–2960.
7. S. Myllymaa, K. Myllymaa, H. Korhonen, J. Toyras, J. E. Jaaskelainen, K. Djupsund, H. Tanila and R. Lappalainen, Biosens. Bioelectron., 2009, 24, 3067–3072.
8. J. Orozco, C. Jimenez-Jorguera and C. Fernandez-Sanchez, Bioelectrochemistry, 2009, 75, 176–181.
9. S. J. Hood, D. K. Kampouris, R. O. Kadara, N. Jenkinson, F. Javier del Campo, F. Xavier Muñoz and C. E. Banks, Analyst, 2009, 134, 2301.
10. T. J. Davies and R. G. Compton, J. Electroanal. Chem., 2005, 585, 63.
11. B. Kralj and R. A. W. Dryfe, Phys. Chem. Chem. Phys., 2001, 3, 3156–3164.
12. A. Berduque, Y. H. Lanyon, V. Beni, G. Herzog, Y. E. Watson, K. Rodgers, F. Stam, J. Alderman and D. W. M. Arrigan, Talanta, 2007, 71, 1022–1030.
13. O. Ordeig, J. Del Campo, F. X. Munoz, C. E. Banks and R. G. Compton, Electroanalysis (N. Y.), 2007, 19, 1973–1986.
14. R. O. Kadara, N. Jenkinson and C. E. Banks, Sens. Actuators, B, 2009, 11, 1377.
15. R. O. Kadara, N. Jenkinson and C. E. Banks, Sens. Actuators, B, 2009, 138, 556.
16. R. O. Kadara, N. Jenkinson, B. Li, K. H. Church and C. E. Banks, Electrochem. Commun., 2008, 10, 1517.
17. The reference electrode was an SCE with a platinum wire as the counter electrode.
18. C. Amatore, J. M. Saveant and D. Tessier, J. Electroanal. Chem., 1983, 147, 39–51.
19. T. J. Davies, E. R. Lowe, S. J. Wilkins and R. G. Compton, ChemPhysChem, 2005, 6, 1340–1347.
20. J.-H. Zen, T.-H. Yang, A. S. Kumar, Y.-J. Chen, J.-C. Hsu and Y. Shih, Electroanalysis (N. Y.), 2009, 21, 2272.
21. L. Xiao, I. Streeter, G. G. Wildgoose and R. G. Compton, Sens. Actuators, B, 2008, 133, 118–127.
22. A. C. Barton, S. D. Collyer, F. Davis, D. D. Gornall, K. A. Law, E. C. D. Lawrence, D. W. Mills, S. Myler, J. A. Pritchard, M. Thompson and S. P. J. Higson, Biosens. Bioelectron., 2004, 20, 328–337.
23. A. J. Saterlay, C. Agra-Gutierrez, M. P. Taylor, F. Marken and R. G. Compton, Electroanalysis (N. Y.), 1999, 11, 1083–1088.

---