Talanta 62 (2004) 912917

Flow injection analysis of zinc pyrithione in hair care products on a cobalt phthalocyanine modied screen-printed carbon electrode
Ying Shih a, , Jyh-Myng Zen b, , Annamalai Senthil Kumar b , Pei-Yen Chen b
b a Department of Applied Cosmetology, Hung Kuang University, Taichung 433, Taiwan, ROC Department of Chemistry, National Chung Hsing University, 250 Kuo-Kuang Road, Taichung 40217, Taiwan, ROC

Received 6 August 2003; received in revised form 21 October 2003; accepted 21 October 2003

Abstract Zinc pyrithione (ZPT) is an antibacterial and antifungal reagent that is often utilized for the antidandruff activity in hair-care shampoos with a composition level up to 1% in the formulation. It has some adverse effects to human and animal if consumed orally. A disposable type of cobalt phthalocyanide modied screen-printed carbon electrode (CoPc/SPE) in couple with ow injection analysis (FIA) was developed for easy and selective analysis of ZPT in commercial hair-care products. Under the optimized FIA conditions, the CoPc/SPE yielded a linear calibration plot in the window of 6576 M with sensitivity and detection limit of 1.65 nA M1 and 0.9 M (i.e. 1.42 pg in 5 l sample loop), respectively, in 0.1 M KOH solution at an applied potential of 0.3 V versus Ag/AgCl. Since the approach is simple, easy, selective, and inexpensive, it offers a potential application of daily ZPT analysis in hair-care products. 2003 Elsevier B.V. All rights reserved.
Keywords: Zinc pyrithione; Cobalt phthalocyanine; Screen-printed carbon electrode; Flow-injection analysis

1. Introduction Zinc pyrithione (ZPT, zinc bis(2-pyridylthio)-N-oxide) (Scheme 1) is an active ingredient in most of daily hair care products (shampoos, conditioner, hair-rinses, etc.). Its effective bactericide, fungicide and algaecide functions were often utilized for the antidandruff and antifouling activities in the products [15]. The recommended maximum amount of ZPT in hair care items is 1% (with preservatives), according to the scientic committee on cosmetic products and non-food products intended for consumer (SCCNFP) [6]. However, the toxicity of ZPT remains uncertain. The adverse effect and toxicity of ZPT in the hair-care products have been investigated for many decades [7]. A lethal dosage of ZPT in the range of 1100 and 25 mg kg1 per day, respectively, for dogs and monkeys would result in emesis, pupil dilation, paralysis, blindness, and diarrhea, etc [69]. Teratologic studies with rat and rabbits with an oral dosage of 15 mg kg1 per day of ZPT show increased incidence of fused or foked ribs [6,8]. It was also found ZPT producing
Corresponding authors. Fax: +886-422-862547. E-mail addresses: jmzen@dragon.nchu.edu.tw, krystle@sunrise.hk.edu.tw (Y. Shih). 0039-9140/$ see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2003.10.039

strong inhibition of cell growth in vitro in rabbit corneal cells and human broblasts [10]. In addition, throw away of ZPT-containing hair-care products to water and/or soil may have signicant impact on the environment and, in turn, to food chain contamination. Recent embrotoxicological study postulates the uptake of ZPT by sh samples [11]. All the above data indicate the requirement for rapid detection and controlling analysis of ZPT in the hair-care formulations. Several analytical methods have been reported previously, including thin-layer chromatography (TLC) [12] and high performance liquid chromatography (HPLC) [1317], titration with Ti(III) [18], and electrochemical approach for the ZPT assays [19]. Even though the titrimetric methods apply a simple way for ZPT determination [18], poor sensitivity appears to be less attractive for practical analysis; while the HPLC and TLC methods are too cumbersome. Owing to the interaction of Zn2+ with stainless steel components, HPLC is not a very appropriate procedure for the direct ZPT analysis [18,19]. In some cases the strong interaction can even damage the stainless-steel body. In addition, due to the low absorptivity of ZPT, uorescence based tagging reagents (e.g. N-dansylaziridine) [16] were always required for sensitive analysis. Alternatively, ZPT in commercial products can be determined by trans-chelation to the Cu2+ complex

Y. Shih et al. / Talanta 62 (2004) 912917

913

addition, the CoPc/SPE based working systems also shows good selectivity for the ZPT containing hair-care products. Typical real sample analysis was successfully veried with the present approach with less operation time and without any sample pretreatment procedures.

2. Experimental Cobalt phthalocyanine (-form, Aldrich, Milwaukee, WI, USA), potassium hydroxide (Showa, Japan), zinc pyrithione (Sigma, USA) and all other compounds (ACS-certied reagent grade) were used without any further purication. Aqueous solutions are prepared with doubly distilled deionized water by reverse osmosis technique. A fresh 0.1 M KOH stock solution was prepared daily. Voltammetric measurements and FIA coupled owthrough experiments were carried out with a CHI 621 electrochemical workstation (Austin, TX, USA). A threeelectrode cell assembled with a bare SPE or the CoPc/SPE working, an Ag/AgCl reference (BAS RE-5), and platinum disc auxiliary electrodes were used. Since oxygen did not interfere the analysis at the detection potentials, no deaeration was performed in this study. The FIA system consisted of a carrier reservoir, a Cole Parmer Masterex microprocessor pump drive, a Rehodyne 7725 sample injection valve (5 l loop), interconnecting Teon tubing, and a BAS CC-5 thin layer electrochemical detector with a BAS electrochemical detector. The BAS ow-through system consists of the working SPE in-between the gasket and the closing holder as reported earlier [27,28]. The SPE with a working area of 0.196 cm2 and a conductive track radius of 2.5 mm was purchased from Zensor R&D (Taichung, Taiwan). The measured average resistance was 85.64 2.10 cm1 . The CoPc/SPE was prepared by drop-coating with 3 l (90%:10% (w/w)) of carbon ink and CoPc mixed solution, prepared in 5:5 volume ratio of dichloromethane and acetonitrile. It was then dried in an oven at 60 2 C for 23 min. Net loading of CoPc on the working SPE is about 1.5 mg cm2 . The CoPc/SPE was equilibrated in a carrier solution at 0.3 V versus Ag/AgCl until the current becomes constant in FIA, which normally took less than 4 min. The quantication of ZPT was achieved by measuring the oxidation peak current in FIA at room temperature (25 C). After experiments, the electrodes were washed thoroughly with distilled water and stored in a close container at normal atmosphere. Branded commercial shampoo samples with (Type 1; #14) and without ZPT (Type 2; $1) were purchased from a local supermarket. The Type 1 products are formulated with ammonium laureth sulfate, sodium chloride, glycol distearate, dimethicone, cetyl alcohol, hydrogenated polydecene, sodium citrate, sodium benzoate, citric acid, and benzyl alcohol in addition with 1% ZPT. The ingredients are almost the same except without ZPT for Type 2 shampoos. Each sample were weighed carefully in the range of

Scheme 1. Structure of zinc pyrithione.

using the normal-phase HPLC [20]. In that process, the central metal ion Zn2+ in the ZPT complex was exchanged to Cu2+ ions as CPT complex and was further used for detection purpose (the CPTs absorption coefcient is about three times higher over ZPT) [2022]. This clearly stresses the search for an easy, simple, and less complicated analytical assay for ZPT in real samples. Electrochemical methods offer convenient, easy, and direct quantication methods in analytical chemistry [23]. However, so far there are very few reports regarding ZPT analysis. Recently metal oxide based carbon paste electrode (CPE) was reported for the ZPT determination assays [19]. A comparative differential pulse voltammetric studies (DPV) with SnO2 , CeO2 , Cr2 O3 , PbO, CdO, and CuO systems resulted in the best performance to SnO2 /CPE in pH 12.5 (0.1 M tetrabutylammonium hydroxide) with a detection anodic peak potential of ca. 0.6 V versus SCE. Even though the method is useful for the practical analysis, the preparation, cost, and time factors made it less appreciable. Moreover, the practical analysis requires extensive off-line sample pretreatment procedures. For example, the real samples were rst treated with 10 ml of 99% methanol to separate the ZPT from the hair-care shampoos, followed by dissolving in dimethylformamide (DMF) for the quantication assays [19]. It is noteworthy that most of the metal oxide based electrochemical systems showed profound oxidation behavior to methanol in alkaline medium, leading to unreliable data. In this work, a disposable type cobalt phthalocyanine (CoPc) modied screen-printed carbon electrodes (CoPc/SPE) was developed as a simple ZPT analytical assay without any sample pretreatment procedures by ow injection analysis (FIA) in 0.1 M KOH solution at an applied potential of 0.3 V versus Ag/AgCl. The CoPc/SPE based FIA shows several advantages for the ZPT detection over the SnO2 /CPE based DPV assays, such as low over-potential and low working volume, etc. Note that the CoPc based screen-printed carbon electrodes have already been successively used for the application of some thiols containing molecules in alkaline pHs [24,25]. Meanwhile, a recent review by Honeychurch and Hart describes many useful metal-ions analysis by SPEs [26]. The high stability, low cost, and easy to be prepared nature of the CoPc/SPE is an ultimate success of the present approach for FIA. In

914

Y. Shih et al. / Talanta 62 (2004) 912917

0.4600.480 g and then dissolved in 250 ml, 0.1 M KOH. All solutions were ltered before regular FIA injection.

the oxidation mechanism based on the following simplied equations.For the diffusion controlled process [ ipa v1/2 ]: 1 log(ipa ) = constant + log(v); 2 log(ipa ) slope(m) = = 0.5 log(v) For the adsorption controlled mechanism [ ipa v]: log(ipa ) = constant + log(v); log(ipa ) slope(m) = =1 log(v) The reaction mechanism can usually understand from the simple plots of either ipa versus v1/2 or ipa versus v for the process of diffusion or adsorption controlled reaction, respectively [29,30]. However, this approach is much difcult for the case of mixed adsorption and diffusion controlled system. On the other hand, the double logarithmic plot of ipa versus v gives exact information about the nature and extend of the reaction process, where the slope (m = log(ipa )/ log(v)) of 0.5 or 1.0 corresponding to the diffusion and adsorption controlled processes [30,31]. For the case of mixed adsorption and diffusion controlled mechanism the m value lies in the window of 0.5 < m > 1.0. Interestingly, a m value of 0.7 was obtained from the log(ipa ) versus log(v) plot in this study, which further reminiscent the ZPT oxidation as mixed adsorption and diffusion controlled reaction mechanisms on the CoPc/SPE. 3.2. Flow injection analysis Initial FIA studies are focused to the optimization of solution phase and experimental conditions for the ZPT analysis. Fig. 2 shows the typical FIA response of 60 M of ZPT (5 l loop) under increasing applied potential (Eapp ) and ow rate (Hf ). As can been seen in Fig. 2A, the ZPT detection currents increase with Eapp . It is important to notice that the stability of the experiments decreases dramatically as Eapp > 0.3 V (insert gures in Fig. 2A). It is expected that oxygen gas evolution reaction at Eapp > 0.3 V may complicate the ZPT detection scheme in the FIA. To achieve a stable ZPT analysis, Eapp = 0.3 V was chosen as optimal for the rest of the analysis. Fig. 2B shows a peak like trend FIA response of 60 M ZPT at different Hf at an Eapp of 0.3 V. The Hf at 0.4 ml min1 was found to have a maximum FIA detecting current signal. It is expected that both the adsorption and diffusion mechanisms can compete with each other under the working Hf window. Under the optimized FIA condition, three different types of alkaline base electrolytes of KOH, NaOH, and NaOH + KCl of the ionic strength 0.1 M was tested for 60 M ZPT in FIA. The obtained respective current signals values are 99.7, 41.3 and 42.3 nA. It is obvious that KOH solution found can inuence the current enhancement. Exact mechanism for such effect is unknown for us now.

3. Results and discussion 3.1. Basic electrochemical studies Cyclic voltammetric (CV) response of the CoPc/SPE and bare SPE with and without 144 M of ZPT in 0.1 M KOH solution at a scan rate of 50 mV s1 were shown in Fig. 1. As can be seen, no obvious faradic response to ZPT in the entire potential range was observed at the bare SPE (Fig. 1A). On the other hand, the CoPc/SPE showed profound oxidation characteristic with a feeble anodic peak at about 0.3 V versus Ag/AgCl (Fig. 1B). This observation clearly demonstrates the electrocatalytic mediation of the CoPc towards the ZPT oxidation reaction. Previous study towards the H2 S gas oxidative sensing reaction at CoPc in alkaline medium also resulted in a similar feeble peak behavior at 0.4 V versus Ag/AgCl [24]. The observation further supports the CoPc redox mediation effect in the present work with the following type of reaction scheme: xCo1+ Pc/SPE xCo2+ Pc/SPE (at 0.10.5 V) (1) (2)

xCo2+ Pc/SPE+yZPT xCo1+ Pc/SPE+yZPTO

In the above scheme, the higher oxidation state of Co2+ (i.e. Co2+ Pc/SPE) was reversibly electrogenerated at 0.10.5 V and was directly participated in the irreversible ZPT oxidation to ZPTO (oxidized form of the complex). Further experiments with increasing scan rate (v) from 5 to 500 mV s1 showed a regular increase in the ZPT oxidation current. A double logarithmic plot of anodic peak current (ipa ) versus v was further adopted to understand the exact nature of

Fig. 1. Cyclic voltammetric response of SPE (A) and the CoPc/SPE (B) with and without 144 M of zinc pyrithione (ZPT) in 0.1 M KOH at a scan rate of 50 mV s1 .

Y. Shih et al. / Talanta 62 (2004) 912917

915

Fig. 4. Continuous ow injection response on the detection of 60 and 576 M of ZPT at the CoPc/SPE. Other conditions are the same as Fig. 3. Fig. 2. Effects of applied potential (A) and ow rate (B) on the detection of 60 M of ZPT at the CoPc/SPE by ow injection analysis using 0.1 M KOH carrier solution.

Fig. 3 shows the FIA calibration response for ZPT on the CoPc/SPE. The FIA responses are systematically increased with [ZPT]. A linear range of 6576 M with sensitivity and regression coefcient of 1.65 nA M1 and 0.9989, respec-

tively, was obtained (Fig. 3B). The detection limit (S/N = 3) is 0.9 M (i.e. 1.42 pg per 5 l loop), which is lower than those reported earlier by HPLC and SnO2 /CPE systems [19,22]. Continuous injections (n = 13) of 60 and 576 M

Fig. 3. Calibration response (A) and graph (B) for the ZPT by ow injection analysis at an applied potential of 0.3 V vs. Ag/AgCl using 0.1 M KOH carrier solution. Flow rate: 0.4 ml min1 . Sample loop: 5 l.

Fig. 5. Flow injection analysis of the ZPT assays in two kinds of shampoo products (Types 1 and 2) at the CoPc/SPE. All the standard ZPT samples were prepared as a mixture with real samples. Other conditions are the same as Fig. 3.

916

Y. Shih et al. / Talanta 62 (2004) 912917

Table 1 Results for the assay of ZPT in commercial shampoos by FIA Samplea #1 Weight mass (mg) 467 Original value (M) 55.51 0.79 (0.97%) Injectb (M) 60 120 240 60 120 240 60 120 240 60 120 240 60 120 240 Detected value after spike (M) 112.86 1.84 176.74 0.27 293.92 0.67 126.94 0.51 184.7 0.88 305.19 4.95 121.23 0.54 182.32 0.75 301.23 0.43 126.32 0.60 183.24 0.45 306.21 1.02 60.35 0.32 120.12 0.20 239.27 0.77 Recovery (%) 95.58 2.00 101.02 0.83 99.33 1.03 102.03 0.60 99.18 0.94 99.78 4.96 103.30 0.70 102.55 0.87 100.82 0.62 101.31 0.81 98.09 0.71 100.28 1.16 100.58 0.32 100.10 0.20 99.69 0.77

#2

460

65.72 0.33 (0.991%)

#3

480

59.25 0.45 (0.994%)

#4

458

65.53 0.55 (1.02%)

$1

472

a b

Samples #14 and $1 are Types 1 and 2, respectively. Injected samples mixture as composed of standard ZPT with real sample in 0.1 M KOH.

ZPT shows low R.S.D. values of 3.87 and 3.54%, respectively, indicating the excellent stability and reproducibility of the present system (Fig. 4). In addition, since the electrode is disposable, it can offer for bulk preparation and routine analytical assays. Finally, the real sample assays were demonstrated for two different types of shampoo products (Types 1 and 2). Fig. 5 shows typical FIA real sample analysis by standard addition methods. The standard ZPT amounts were prepared as a mixture with the real samples. As can be seen, the Type 2 shampoo (in absence of ZPT) did not show any detectable current signal in FIA (Fig. 5B), while the Type 1 sample (with ZPT) showed good FIA signals (Fig. 5A). This particular experiment validates the present approach to be practical. In addition, there is no interference of the matrix effects from other ingredients in the hair-car products. Table 1 showed quantitative data based on the present approach. The obtained values are very close to the labeled values, which further veried the quality of the commercial hair-care products.

Acknowledgements The authors gratefully acknowledge nancial supports from the National Science Council of Republic of China.

References
[1] S. Shuster, Br. J. Dermatol. 111 (1984) 235. [2] G.C. Priestley, J.C. Brown, Acta dermato-Venereologica 60 (1980) 145. [3] C.P. Franchimont, V. Gofn, F. Henry, I. Uhoda, C. Braham, G.E. Pierard, Int. J. Cosmet. Sci. 24 (2002) 249. [4] G. Mokawa, H. Shimizu, K. Okamoto, J. Soc. Cosmet. Chem. 33 (1982) 27. [5] R. Marks, A.D. Pearse, A.P. Walker, Br. J. Dermatol. 112 (1985) 415. [6] The scientic committee on cosmetic products and non-food products intended for consumers (SCCNFP), draft evolution and opinion on: zinc pyrithione (colioa no. P81), 17th December 2002, Brussels. (http://europa.eu.int/comm/food/fs/sc/sccp/out225 en.pdf). [7] F.H. Snyder, E.V. Buehler, C.L. Winek, Toxicol. Appl. Pharmacol. 7 (1965) 425. [8] G.A. Nolen, T.A. Dierckman, Food Cosmet. Toxicol. 17 (1979) 639. [9] D.W. Collom, C.L. Winek, J. Pharmcol. Sci. 56 (1967) 1673. [10] A.S. Maria, J.M. Pozuelo, J.M. Lopez, F. Sanz, Ecotoxicol. Environ. Saf. 20 (1990) 42. [11] K. Goka, Environ. Res. Sec. A 81 (1999) 81. [12] N. De Kruijf, M.A.H. Rijk, L.A. Pranoto-Soetardhi, A. Schouten, J. Chromatogr. 410 (1987) 395. [13] M.D. Seymour, D.L. Bailey, J. Chromatogr. 206 (1981) 301. [14] Y. Kondoh, S. Takano, J. Chromatogr. 408 (1987) 255. [15] H. Haiyung, R.R. Gadde, J. Chromatogr. 291 (1984) 434. [16] Y. Kondoh, S. Takano, J. Chromatogr. 408 (1987) 255. [17] N. Voulvoulis, M.D. Scrimshaw, J.N. Lester, Chemosphere 38 (1999) 3503. [18] S. Oliveri-Vigh, H.L. Karageozian, Anal. Chem. 48 (1976) 1001. [19] L.-H. Wang, Electroanalysis 12 (2000) 227. [20] R.J. Fenn, M.T. Alexander, J. Liq. Chromatogr. 11 (1998) 3408.

4. Conclusion The CoPc modied screen-printed carbon electrode was effective and selective for the ow injection analysis of zinc pyrithione (ZPT) in hair care products. The hydrodynamic ow injection parameters are systematically optimized and the obtained analytical performance with the CoPc/SPE is good. The approach allows the ZPT real sample analysis without any tedious pretreatment procedures as seen in conventional techniques. Since the approach is less expensive and easy to operate, it can be easily extended to a convenient portable type of ZPT analyzer. The work is now in progress in our laboratory.

Y. Shih et al. / Talanta 62 (2004) 912917 [21] K. Nakajima, T. Yasuda, H. Nakazawa, J. Chromatogr. 502 (1990) 379. [22] K.V. Thomas, J. Chromatogr. A 833 (1999) 105. [23] J.-M. Zen, A. Senthil Kumar, D.-M. Tsai, Electroanalysis 15 (2003) 1073. [24] J.P. Hart, A.K. Abass, Anal. Chim. Acta 342 (1997) 199. [25] A. Napier, J.P. Hart, Electroanalysis 8 (1996) 1006. [26] K.C. Honeychurch, J.P. Hart, TrAc. 22 (2003) 456. [27] J.-M. Zen, H.-H. Chung, A. Senthil Kumar, Analyst 125 (2000) 1633.

917

[28] J.-M. Zen, H.-P. Chen, A. Senthil Kumar, Anal. Chim. Acta 449 (2001) 95. [29] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Fundamentals and Applications, second ed., Wiley, New York, 2001. [30] F. Scholz (Ed.), Electroanalytical Methods, Guide to Experiments and Applications, Springer, Berlin, 2002. [31] J.-M. Zen, A. Senthil Kumar, J.-C. Chen, Anal. Chem. 73 (2001) 1169.