2272

Short Communication

Detection of Aluminum Chlorohydrate Content in Antiperspirant Deodorants Using Screen-Printed Silver Electrodes by One Drop Analysis Jyh-Myng Zen,b Ting-Hao Yang,b Annamalai Senthil Kumar,d Yu-Ju Chen,c Jaw-Cherng Hsu,c Ying Shiha* a

Department of Cosmetic Science, Providence University, Taichung 433, Taiwan *e-mail: [email protected] b Department of Chemistry, National Chung Hsing, University, Taichung 402, Taiwan c Department of Applied Cosmetology, Hung Kuang University, Taichung 433, Taiwan d Department of Chemistry, Vellore Institute of Technology University, Vellore 632 014, India Received: March 20, 2009 Accepted: June 12, 2009 Abstract Aluminum chlorohydrate (Al2(OH)5Cl · 2H2O, ACH) is an active ingredient in many antiperspirants and deodorants formulation to reduce the body odors (mainly sweat) through interaction with apocrine sweat glands to produce insoluble aluminum hydroxide and free chloride, which then plugs the sweat gland that stops the flow of sweat to the skins surface. We demonstrated here an one drop (50 mL) electrochemical sensing of the ACH using an in-built three screen-printed electrodes assembly containing Ag as working and pseudo reference and carbon as counter electrode system (AgSPE). The free Cl ion librated from ACH/H2O reaction was detected at AgSPE surface at 0.072 V vs. pseudo Ag reference electrode system in pH 2 phosphate solution by Cyclic voltammetric Technique. Under optimal working condition the AgSPE shows a linear calibration plot in the window of 30 – 2000 ppm of ACH with sensitivity and regression values of 0.104 mA/ppm and 0.998 respectively. Calculated detection limit is 3.03 ppm. RSD values of intra- and interassays were 0.19% and 2.79% respectively. Finally, real sample (antiperspirant deodorant lotions) assays were successfully demonstrated with results comparable to the predicted values. Keywords: Aluminum chlorohydrate, Screen-printed silver strip, Antiperspirant DOI: 10.1002/elan.200904670

Body odor is a universal experience, and even revered in some cultures as an expression of the persons inner psyche. Generally, however, people prefer not to be too smelly and manufacturers offer plenty of solutions to reduce body odor. The trick is to find solutions which are completely safe. Many manufactured antiperspirants, and some deodorants, contain aluminum chlorohydrate (ACH) usually in range from 15% to 20% of the products [1 – 3]. An antiperspirant aims to reduce the amount of sweat the body produces, while deodorant aims mainly to perfume the odor released by the combination of sweat and bacteria. ACH is a soluble aluminum complex that consists of complex basic aluminum chloride having the general formula Aln(OH)mCl(3n–m) [4]. This complex is polymeric, loosely hydrated and encompasses a range of aluminum to chloride ratios from 2.1 – 1.9. [4] In the United States, ACH may be used as an active ingredient in over-the-counter (OTC) drug products. The active ingredient precipitates inside the apocrine sweat glands to produce insoluble aluminum hydroxide which then plugs the sweat gland that stops the flow of sweat to the skins surface [3 – 5]. This compound is preferred in some cases because the pH value can be varied according to the exact values chosen for the subscripts, n and m in the pattern formula. When used as an Electroanalysis 2009, 21, No. 20, 2272 – 2276

active ingredient the pH must be balanced as basic enough not to cause irritation to the skin, but acidic enough to form the solid plugs in the sweat ducts. ACH is also used as a flocculent in water purification, where it is usually called as polyaluminum chloride. Recent research indicated that ACH may have negative effects for the body, such as increasing the risk of breast cancer or Alzheimers disease [6 – 9].The fate of the aluminum applied to the skin as ACH is currently unknown. Hence, selective and sensitive detection of the ACH is paramount of interest in many research fields. Very limited analytical measurements were available for the ACH [10]. Atomic absorption based detection technique was often used for the Aluminum containing real sample analyses [11 – 13]. To our best of our knowledge, there is no electroanalytical detection approach developed for the ACH so far. Major challenge with Aluminum based electrochemical sensing is the high redox potential based operation of the Al/Al3þ ion system (~ 2 V vs. RHE), which is practically not viable in aqueous solution due to effective O2 gas evolution reaction at 1.5 V vs. RHE prior to the redox reaction [14]. In this work, we have successfully demonstrated effective analysis of the ACH sensitively and selectively by using screen-printed silver electrode (AgSPE)  2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

2273

Detection of Aluminum Chlorohydrate

surface in aqueous solutions. In further, we applied the procedure for the sensing of the ACH in the antiperspirant or deodorant lotions using in-built three electrodes assembly of AgSPE in couple with cyclic voltammetry (CV) for convenient, fast and low sample volume analysis. Initial experiments are optimization of working electrode and solution phase conditions. AgSPEs pseudo reference was found to give fluctuation in the detection potential. Hence, following pretreatment procedure is adopted to prepare stable pseudo Ag reference electrode in this work. The Ag reference part is first chemically treated with 3 mL of 35% H2O2 to convert the Ag to AgO as in Figure 1, then 3 mL of 3 M KCl spiked on the surface for 15 seconds then wash out with double distilled water, Figure 1C. Such a chemical treatment procedure resulted to stable potential and current response in this work. Potential difference between the pseudo and standard, EAg/AgCl  Epseudo Ag is 0.23 V. In order to check the suitability of the screen-printed electrodes to detect ACH, experiments were performed using screen-printed carbon electrode (SPCE) and AgSPE in a pH 2 phosphate solution (PS) (10 mL cell, pH adjusted using phosphoric acid). Figure 2 shows typical response of the SPCE and AgSPE in a potential window of  0.28 – 0.16 V vs. pseudo Ag reference with 500 ppm of ACH at a scan rate (v) of 50 mV/s. As can be seen SPCE failed to shows any faradic response with the ACH (Fig. 2A), while the AgSPE yielded marked redox response centered (E8’) at 0.072 V vs. pseudo Ag reference with the ACH (Fig. 2B).

Fig. 1. Photographs of an in-built three electrodes assembly of AgSPE before (A) and after (C) chemical treatment with 3 mL of 35% H2O2 on the Ag pseudo reference electrode part (B) and an in-built three electrodes assembly of the AgSPE with one drop containing 50 mL of the working solution (D) and its enlarged version (E). WE: Working electrode, CE: Counter electrode, RE: Reference electrode.  2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

The selective redox peak response at AgSPE is due to the following reason: empirical formula for ACH is Al2(OH)5 Cl · 2H2O, however when dissolved, the basic unit of the ACH becomes [4]: Al2(OH)5Cl · 2H2O ! Al13O4(OH)24(H2O)127þ þ 7 Cl

(1)

The above reaction yielded Al13O4(OH)24(H2O) 7þ 12 colloidal complex precipitate along with qualitative amount of Cl in the solution, which then could be detected selectively on the AgSPE with the following redox reaction [14]: AgCl þ e $ Ag þ Cl (E8 ¼ 0.072 vs. pseudo Ag)

(2)

Hence the peak at AgSPE is solely due to Ag/AgCl redox couple in this work. Figure 3 is the effect of pH on the electrochemical detection of 100 ppm ACH at AgSPE. Experiments were carried out in pH 2 – 12 solutions. Marked redox peak was noticed up to pH 8 after that deterioration in the electrode response was observed. Calculated anodic peak potential, Epa (peak current, ipa) values are; 0.072 V (14.88 mA), 0.076 V (15.07 mA), 0.077 V (15.59 mA) and 0.092 V vs. pseudo Ag (15.58 mA) respectively for the pH 2, 4, 6 and 8 solutions (Figs. 3A – D). Meanwhile, optical densities (OD) were also measured for the above pH solutions with ACH at 200 nm. Due to the formation of Al13O4(OH)24(H2O) 7þ 12 complex precipitate, OD of the ACHs varies with the solution pHs. The OD values found to increase simultaneously with decrease in the solution pHs (datas not enclosed). Amongst all, ACH dissolved in the pH 2 resulted in to clear solution with highest OD value, hence it was further chosen for all electroanalytical assays. Suitability of the pseudo reference electrode on the electrochemical detection of the ACH was demonstrated in response comparison with the standard Ag/AgCl reference electrode system in Figure 4. AgSPE shows a defined redox peak at E8’ for 0.3 V vs Ag/AgCl, while it was at 0.072 V against pseudo reference electrode system in pH 2 PS

Fig. 2. CV responses of SPCE (A) and AgSPE (B) without (dotted lines) and with 500 ppm (solid lines) of ACH in 10 mL of pH 2 phosphate solution (PS) (I ¼ 0.1 M) at v ¼ 50 mV/s.

www.electroanalysis.wiley-vch.de

Electroanalysis 2009, 21, No. 20, 2272 – 2276

2274

J.-M. Zen et al.

Fig. 3. Typical CV responses of in-built three electrode assembly containing 100 ppm ACH in various pH PS at v ¼ 50 mV/s (pH adjusted using dilute phosphoric acid or NaOH solutions). Condition ¼ 10 mL working volume and pseudo Ag reference electrode.

(Fig. 4A). However, the ipa values found to be similar against both the reference electrodes. Figures 4B and C are comparative datas for the responses of the AgSPE for the repeated detection (n ¼ 10) of 100 ppm of ACH against commercial Ag/AgCl, untreated- and H2O2 treated- pseudo Ag reference electrodes under identical working condition. Respective Epa relative standard deviation (RSD) values are: 2.99, 10.74 and 0.88%. The H2O2 treated pseudo Ag reference electrode system shows very stable Epa response with a lowest RSD value of 0.88%. Meanwhile, drastic change in the ipa responses were noticed with the commercial Ag/AgCl (Figure 4C), due to leaching of the KCl

solution from the reference electrode unit unlike the pseudo reference case (RSD ¼ 0.44%). These results indicate appreciable suitability of the pseudo reference in the present work. Further experiments are one drop analysis of the ACH sample solution on the AgSPEs three electrode assembly unit. Analysis with 100 ppm ACH containing 50 mL (one drop) or 10 mL solutions were compared (Table 1). It was found that at the 95% confidence level, no significant difference in the ipa values with respect to the volumes (RSD values 0.63% and 0.49% respectively, n ¼ 10). In aim to work with less quantity, one drop containing 50 mL test solution is uniformly taken for further electroanalytical measurements. Figures 1D and E show typical arrangement of in-built three electrodes assembly containing AgSPE with one drop (50 mL) of the test sample solution. Typical calibration plot of ipa vs. [ACH] under an optimal working condition was linear up in the range of 30 – 2000 ppm with sensitivity and regression values of 0.104 mA/ppm and 0.998 respectively at AgSPE. Calculated detection limit is (S/N ¼ 3) 3.03 ppm. Intra- and inter-assays were also carried out with the AgSPEs (n ¼ 7) for the detection of 100 ppm of ACH. Respective RSD values were 0.19% and 2.79%. These results conclude stable electroanalytical sensing of the ACH at the AgSPE system. Final experiments are detection of aluminum chlorohydrate content in antiperspirant deodorant lotion real samples (#1 and #2) under optimal one drop analysis condition. Standard addition method [7, 9] was adopted for the assays. Figure 5 show linear scan response for the antiperspirant real sample #1 without and with added ACH standard in pH 2 PS. Table 2 summarizes the analytical data for the real sample analysis. Detected ACH values are 99.23 and 101.91 ppm respectively for the sample #1 and #2 (without dilution factor). After the dilution factor correction, corresponding values are 17.38% and 12.21%, which is closer to the labeled values of 17.5% and 12.0% with accuracy of 99.31% and 98.25% respectively for the above

Fig. 4. CV responses of AgSPE with 100 ppm of ACH against standard Ag/AgCl reference electrode (dash line) and the pseudo reference electrode (solid line) at a v ¼ 50 mV/s. (A). The stability of Epa (B) and ipa (C) during ten continuous detections. Electroanalysis 2009, 21, No. 20, 2272 – 2276

www.electroanalysis.wiley-vch.de

 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

2275

Detection of Aluminum Chlorohydrate Table 1. Effect of working solution volume on the continued detection of 100 ppm ACH by AgSPE (n ¼ 10). Vol. [a]

50 mL 10 mL

Ipa (mA ) by different experiments

Analytical data

1

2

3

4

5

6

7

8

9

10

Mean

SD

RSD

13.45 13.23

13.47 13.27

13.28 13.21

13.47 13.27

13.52 13.30

13.46 13.21

13.57 13.31

13.55 13.39

13.57 13.36

13.45 13.36

13.48 13.39

0.085 0.065

0.63% 0.49%

[a] One drop (50 mL ) analysis or by conventional 10 mL (immersed system) based analysis. WE ¼ Ag-, RE ¼ pseudo Ag reference and CE ¼ carbon screenprinted electrodes.

Table 2. Real sample analyses for the detection of ACH content in antiperspirant deodorants using AgSPE by one drop method. Real sample [a]

Original detected value (ppm) [b]

Spike (ppm)

Detected value after spike (ppm)

Recovery (%)

#1

99.23

#2

101.91

100 200 300 200 400 600

202.04 300.00 393.67 309.23 505.69 696.81

102.80  2.5 100.39  0.7 98.14  3.3 103.66  1.3 100.94  0.8 99.15  0.5

[a] Dilution factors and labeled values are 1751 and 1198 and 17.5% and 12.0% respectively. [b] Corresponding detected values after dilution factor correction are 17.38% and 12.21%.

samples. Recovery values were in the window of 98.14 – 103.66%. These results clearly indicate appreciable electrochemical sensing of the ACH by the AgSPE in this work. In conclusion, one drop analysis of ACH sensitively and selectively using in-built three electrodes assembly containing AgSPE strip configuration by CV was successfully performed. It was found that the liberated free chloride ions from ACH could be able to show a well defined redox peak at 0.072 V vs. pseudo Ag reference at AgSPE and its was further used as a tool for the selective sensing of the ACH in the test solution. Electrode and solution phase parameters were systematically optimized. A pseudo Ag reference electrode fabricated by treating with H2O2 gave stable electrochemical sensing unlike to the case of Ag/AgCl reference with fluctuative response. ACH real sample

analyses were done using the standard addition method which gave appreciable results with comparable to that of the predicted labeled values. The proposed method of analysis will remain challenging to other real samples also. Experimental Aluminum chlorohydrate/aluminum hydroxychloride (ACH) is purchased from (SUMMIT, NY, USA) and all chemical compounds (ACS certified reagent grade) were all used without any further purification. Aqueous solutions were prepared with doubly deionized water by reverse osmosis technique. Unless otherwise stated the base electrolyte solution used in this work was 0.1 M pH 2 phosphate solution (PS), pH adjusted by using phosphoric acid.

Fig. 5. Typical linear sweep responses of AgSPE to detect the standard ACH (A) and antiperspirant deodorant lotion sample (#1) (B) by one drop analysis method. Working volume ¼ 50 mL, solution ¼ pH 2 PS and v ¼ 50 mV/s.  2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.electroanalysis.wiley-vch.de

Electroanalysis 2009, 21, No. 20, 2272 – 2276

2276

J.-M. Zen et al.

Electrochemistry was performed on a CHI 621A electrochemical workstation (Austin, TX, USA). In built three electrodes combined SPE assembly contains a carbon(SPCE) or Ag- screen printed disc (AgSPE) as a working (3 mm diameter), an outer carbon-ring as a counter and an outer Ag printed layer as a pseudo reference electrode as given in Figure 1. Apart from that, standard Ag/AgCl (3 M KCl) reference electrode with 10 mL conventional electrochemical cell is also used in the initial optimization experiments. Preparation and characterization of the AgSPE were as per the previous reports [15, 16]. Since dissolved oxygen did not interfere with the working potential window, no deaeration was performed. Electrochemical analyses were carried out by using 50 mL of working solution spiked on the in-built three electrodes assembly AgSPE (optimal). Comparative experiments were also performed in conventional 10 mL working solution by directly immersing the AgSPE in the respective solution. Base line corrected anodic peak current (ipa) values were taken for all the quantitative analysis. Antiperspirant deodorant lotions (dual purpose samples), #1 and #2 were purchased from local supermarket. ACH contents in the products were labeled as 17.5% (w/w) and 12.0% (w/w) respectively. The ingredients present are: aqua, alcohol, ceteareth-12, ceteareth-30, Bis-PEG-18 methyl ether dimethyl silane, perfume, hydroxyethyl cellulose, tocopheryl acetate, sodium hydroxide, stearyl ETO (20 M) alcohol, isopropyl palmitate, dicapryl adipate, POP (15 M) stearyl ether, POE (2 M) stearyl alcohol and fragrance oil. Real sample solutions were prepared by dissolving 0.05 g of the antiperspirant deodorant lotion in 0.1 M pH 2 PS with suitable dilution, and then ultrasonication was done for 5 min. 50 mL of the solution was used for the one drop electroanalysis.

Electroanalysis 2009, 21, No. 20, 2272 – 2276

Acknowledgements The authors gratefully acknowledge financial supports from The National Science Council of The Republic of China. References [1] H. E. Jass, in Antiperspirants and Deodorants (Eds: K. Laden, C. B. Felger), Marcel Dekker, New York, 1988, pp. 33 – 66. [2] Handbook of Nonprescription Drugs, 10th ed. (Ed: T. R. Covington), American Pharmaceutical Association 1990. [3] D. L. Teagarden, S. L. Hem, J. L. White, J. Soc. Cosmet. Chem. 1982, 33, 281. [4] D. L. Teagarden, J. F. Kozowski, J. L. White, S. L. Hem, J. Pharm. Sci. 1981, 70, 758. [5] R. P. Quatrale, in: Antiperspirants and Deodorants (Eds: K. Laden, C. B. Felger), Marcel Dekker, New York 1988, pp. 89 – 118. [6] A. B. Graves, E. White, T. D. Koepsell, B. V. Reifler, G. Van Belle, E. B. Larson, J. Clin. Epidemiol. 1990, 43, 35. [7] P. D. Darbre, J. Inorg. Biochem. 2005, 99, 1912. [8] P. D. Darbre, Best Practice and Research Clinical Endocrinology and Metabolism 2006, 20, 121. [9] E. Christopher, L. M. Charles, L. Barr, C. Marti, A. Polawart, P. D. Darbre, J. Inorg. Biochem. 2007, 101, 1344. [10] R. Flarend, C. Keim, T. Bin, D. Elmore, S. Hem, M. Ladisch, J. Inorg. Biochem. 1999, 76, 149. [11] J. Koma´rek, R. Cˇervenka, T. Ru˚zˇicˇka, V. Kuba´nˇ, J. Pharm. Biomed. Anal. 2007, 45, 504. [12] P. C. Kruger, P. J. Parsons, Spectrochim. Acta B 2007, 62, 288. [13] M. I. S. Veri´ssimo, M. T. S. R. Gomes, Anal. Chim. Acta 2008, 617, 162. [14] A. J. Bard, R. Parsons, J. Jordan, Standard Potentials in Aqueous Solutions, IUPAC, Marcel Dekker, New York 1983. [15] J.-M.Zen, A. S. Kumar, S.-C.Lee, Y. Shih, Electroanalysis 2007, 19, 2369. [16] J.-M.Zen, A. S. Kumar, Screen-Printed Electrochemical Sensor, Encyclopedia of Sensors (Eds: A. G. Craig, E. C. Elizabeth, V. P. Michael), American Scientific Publishers, California, 2006, pp. 33 – 52.

www.electroanalysis.wiley-vch.de

 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim