Determination of alternative preservatives in cosmetic products by chromophoric derivatization followed by vortex-assisted liquid–liquid semimicroextraction and liquid chromatography

Author’s Accepted Manuscript Determination of alternative preservatives in cosmetic products by chromophoric derivatization followed by vortex-assisted liquid-liquid semimicroextraction and liquid chromatography Pablo Miralles, Ilianna Vrouvaki, Alberto Chisvert, Amparo Salvador www.elsevier.com/locate/talanta

PII: DOI: Reference:

S0039-9140(16)30160-6 http://dx.doi.org/10.1016/j.talanta.2016.03.033 TAL16414

To appear in: Talanta Received date: 12 February 2016 Revised date: 10 March 2016 Accepted date: 11 March 2016 Cite this article as: Pablo Miralles, Ilianna Vrouvaki, Alberto Chisvert and Amparo Salvador, Determination of alternative preservatives in cosmetic products by chromophoric derivatization followed by vortex-assisted liquidliquid semimicroextraction and liquid chromatography, Talanta, http://dx.doi.org/10.1016/j.talanta.2016.03.033 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Determination

of

alternative

chromophoric

derivatization

preservatives followed

by

in

cosmetic

vortex-assisted

products

by

liquid-liquid

semimicroextraction and liquid chromatography Pablo Miralles, Ilianna Vrouvaki, Alberto Chisvert, Amparo Salvador* Departamento de Química Analítica, Facultad de Química, Universitat de València, 46100 Burjassot, Valencia, Spain ABSTRACT An analytical method for the simultaneous determination of phenethyl alcohol, methylpropanediol, phenylpropanol, caprylyl glycol, and ethylhexylglycerin, which are used as alternative preservatives in cosmetic products, has been developed. The method is based on liquid chromatography with UV spectrophotometric detection after chromophoric derivatization with benzoyl chloride and vortex-assisted liquid-liquid semimicroextraction. Different chromatographic parameters, derivatization conditions, and sample preparation variables were studied. Under optimized conditions, the limits of detection values for the analytes ranged from 0.02 to 0.06 µg mL−1. The method was validated with good recovery values (84 – 118 %) and precision values (3.9 – 9.5 %). It was successfully applied to 10 commercially available cosmetic samples. The good analytical features of the proposed method besides of its environmentally-friendly characteristics, make it useful to carry out the quality control of cosmetic products containing the target compounds as preservative agents. Keywords: Cosmetic products; Alternative preservatives; Liquid chromatography * Corresponding author: Tel.: +34 963543175; Fax: +34 96544436 E-mail address: [email protected] (A. Salvador)

1. Introduction According to the European Regulation of Cosmetic Products (EC Regulation) [1], ‘cosmetic product’ means ‘any substance or mixture intended to be placed in contact with the external parts of the human body (epidermis, hair system, nails, lips

and external genital organs) or with the teeth and the mucous membranes of the oral cavity with a view exclusively or mainly to cleaning them, perfuming them, changing their appearance, protecting them, keeping them in good condition or correcting body odours’. Typical cosmetic formulation includes active principles, excipients, and additives (colorants, perfumes, preservatives, etc.). The Annex V of this regulation contains the list of the currently allowed preservative agents in cosmetics and the restrictions to be used, including their maximum concentration, in order to ensure the safety of consumers. After the recent prohibition of some parabens by the European Commission [2], the cosmetic industries are continuously looking for new compounds that can perform the preservative function effectively and safely. In fact, it is well-known that some cosmetic ingredients can play more than one role in a cosmetic formulation. In this sense, according to the Inventory of Cosmetic Ingredients (2006/257/EC) [3], phenethyl alcohol (2-phenylethanol, PA) is used in cosmetics as deodorant; methylpropanediol (2-methyl-1,3-propanediol, MP) and phenylpropanol (3-phenyl-1propanol, PP) are commonly employed as solvents; caprylyl glycol (1,2-octanediol, CG) acts as emollient, humectant and hair conditioning; ethylhexylglycerin (3-[(2ethylhexyl)oxy]-1,2-propanediol, EG) is used as skin conditioning. However, all of them show an important antimicrobial activity [4-13] and their use in the cosmetic industries is widespread due to their antimicrobial properties, being the subject of several patents [6,14-17]. Despite their preservative features, these compounds, whose chemical structures are shown in Fig. 1, are not listed as preservatives in the abovementioned Annex V of the European regulation. It is therefore necessary that the cosmetic companies have procedures to perform the analytical control of these alternative preservatives and to assure the quality of the final products containing them. However, there are not official methods to 2

quantify the target compounds in cosmetic samples.

Besides, to the best of our

knowledge, there are not published analytical methods regarding to their determination. Only a few articles in which PA was determined in alcoholic beverages, such as wine [18-20], beer [21,22] and alcoholic distillates [23] by chromatographic techniques have been published. Vortex-assisted extraction procedures have been successfully applied for sample preparation in the analysis of water [24-26], food and drinks [27-29], and also cosmetic products [30,31] with good analytical features. Other applications of vortexassisted techniques can be found in some recent reviews [32-35]. For the first time, a liquid chromatography (LC) method with UV/Vis spectrophotometric detection is proposed here for the determination of these alternative preservatives, i.e., PA, MP, PP, CG, and EG, in cosmetic samples. Owing to the low UV/Vis absorbance of the target compounds, a derivatization step is proposed to convert them to more absorptive derivatives, in order to enhance their analytical response during the LC analysis. Benzoyl chloride is a common derivatizing agent used to perform chromophoric derivatizations through a simple and effective esterification reaction under soft conditions in basic medium [36-41]. In this sense, the chromophoric derivatization combined with vortex-assisted liquid-liquid semimicroextraction (VALLsME) make possible their simultaneous determination with good sensitivity and low reagents and solvents consumption. 2. Experimental 2.1. Apparatus An Agilent 1220 Infinity LC system including a degasser, a binary pump, an autosampler with up to 100 µL injection volume, a thermostated column oven, and a UV/Vis detector was employed. The column was a Purospher® STAR RP-18

3

endcapped (12.5 cm length, 4 mm I.D., 5 µm particle size) from Merck (Darmstadt, Germany). A ZX3 vortex mixer from VELP Scientifica (Usmate Velate, Italy) was used to ease the vortex-assisted liquid-liquid extraction in the preparation of cosmetic samples and an EBA 21 centrifuge from Hettich (Tuttlingem, Germany) was used for phase separation. A hotplate from Stuart Scientific (Staffordshire, United Kingdom) was used for the evaporation of the organic solvent prior the reconstitution of the extract. 2.2. Reagents and samples PP (3-phenyl-1-propanol) 98%, CG (1,2-octanediol) 98%, PA (2-phenylethanol) 99%, MP (2-methyl-1,3-propanediol) 99%, all from Aldrich (Steinheim, Germany), and EG

(3-[(2-ethylhexyl)oxy]-1,2-propanediol)

98%

from

Schülke&Mayr

GmbH

(Norderstedt, Germany) were used as standards. Deionized water obtained using a NANOpure II ultrapure water system from Barnstead (Boston, USA) was used as solvent to prepare the sample and standard solutions and the aqueous mobile phase. LC-grade n-hexane 96% from Scharlab Chemie (Barcelona, Spain) was used as extraction solvent in the VALLsME procedure. LC-grade acetonitrile from VWR International (Pennsylvania, USA) was used as solvent to reconstitute the extracts after the VALLsME and as organic modifier of the mobile phase. Extra-pure glacial acetic acid from Scharlab Chemie (Barcelona, Spain) was used to prepare the acetic acid solution used as aqueous mobile phase. LC-grade absolute ethanol from Scharlab Chemie (Barcelona, Spain) was also tested as organic modifier of the mobile phase.

4

Benzoyl chloride 99% from Aldrich (Steinheim, Germany) was used as chromophoric derivatizing agent and reagent-grade sodium hydroxide (NaOH) from Scharlab Chemie (Barcelona, Spain) was used to prepare the basic solutions needed to carry out the derivatization reaction. Ten commercial cosmetic samples with different formulations, including creams (samples A-D, moisturizing creams; samples E and F, sunscreen creams) and gels (samples G-J, bath gels), were purchased at local stores. 2.4. Proposed method 2.4.1. Preparation of sample solutions and standards Samples were prepared by weighing approximately 0.1 g into a 5 mL volumetric flask and diluted up to the line with deionized water. The solution was centrifuged and/or filtered when needed to remove the non-soluble compounds and an aliquot of 0.5 mL was placed into a 5 mL volumetric flask. Then, 1 mL of NaOH 5 M and 20 µL of benzoyl chloride were added and the flask was stirred vigorously for 3 min using a vortex mixer (40 Hz) in order to perform the chromophoric derivatization. Then, deionized water was used to complete the total volume of the volumetric flask. After that, the solution was transferred into a polypropylene conical-bottom tube and 1 mL of n-hexane was added to carry out the VALLsME by vortex mixing for 30 seconds. Then, both phases were separated by centrifugation (6000 rpm, 2 min). The upper organic extract was collected and transferred into a LC injection vial and it was then evaporated to dryness under a fume hood using a hotplate set approximately at 60 ºC. Finally, the extract was reconstituted adding 400 µL of acetonitrile. Regarding with the standard solutions, a stock solution containing all the analytes at 500 µg mL-1 was prepared using deionized water as solvent. From this solution, the calibration solutions (1 – 25 µg mL-1) were prepared daily and were subjected to the same derivatization and extraction processes than samples.

5

2.4.2. Chromatographic analysis Twenty microliters of the standard or sample solution were injected into the column set at 30 ºC. The elution was performed at 1 mL min−1 flow rate with a mixture of acetonitrile:aqueous 0.5 % acetic acid solution as mobile phase, following the elution gradient program shown in Table 1. The detection wavelength for signal monitoring was fixed at 235 nm and the runtime was completed with 14 min. 3. Results and discussion 3.1. Study of the chromatographic variables Different variables may affect the chromatographic determination, such as the mobile phase composition and the column temperature. With the purpose of obtaining a good chromatographic resolution in a time as short as possible, the effect of the mobile phase composition was studied. A solution of acetic acid in deionized water (0.5 %, v/v) was used as aqueous mobile phase. Although the analytes are in their neutral form in the whole range of column pH compatibility, according to their predicted pKa values (ranged from 13.6 to 15.2), the use of an aqueous acetic acid solution as mobile phase allowed us to improve the chromatographic peak profile, increasing the symmetry and reducing the tailing of the chromatographic peaks. Moreover, acetonitrile and ethanol were tested as organic modifiers, obtaining the best results when acetonitrile was used, due to its lower absorbance at the selected wavelength for signal monitoring (235 nm). Different programs of gradient elution were tested using acetonitrile and aqueous 0.5 % acetic acid solution as mobile phases, obtaining the best results when using the elution gradient program shown in Table 1.

6

The effect of the column temperature was also studied. Oven temperatures from 30 ºC to 50 ºC were tested. No significant differences were observed, so 30 ºC was selected for further analysis in order to set the system at fixed conditions, although the analysis can be performed at room temperature without temperature control. With those conditions, a good chromatographic separation of the five derivatives was achieved in less than 14 min, as can be seen in Fig. 2.

3.2. Study of the sample preparation variables 3.2.1. Study of the chromophoric derivatization variables A chromophoric derivatization was carried out in order to enhance the analytical response of the target compounds. In the preparation of standard and sample solutions, benzoyl chloride was chosen as derivatizing reagent and a solution of NaOH in deionized water (5 M) was chosen as solvent to get the basic medium needed to perform the derivatization. Different parameters such as the volumes of the reagents and the reaction time were studied. The following tests were performed employing standard solutions containing the target analytes at 25 µg mL-1. 10, 20, and 30 µL of benzoyl chloride combined with 500 and 1000 µL of NaOH 5 M were tested, obtaining the best analytical signal when 20 µL of benzoyl chloride and 1000 µL of NaOH 5 M were used. In order to ease the derivatization, the reagents were shaken using a vortexmixer and the effect of the reaction time was studied. 0.5, 1, 2, 3, and 5 min were tested as reaction time, obtaining the best results with 3 min of vortex mixing. 3.2.2. Study of the vortex-assisted liquid-liquid semimicroextraction (VALLsME) variables After the derivatization reaction, VALLsME, using n-hexane as extraction solvent and followed by reconstitution with acetonitrile, was performed in order to 7

preconcentrate the target compounds, to avoid possible matrix interferences and to prevent the degradation of the derivatized analytes (see Section 3.2.3). The effect of the extraction volume, the extraction time, and the reconstitution volume were studied. As extraction volume, 0.5, 1, and 2 mL of n-hexane were tested. No significant differences were observed, so 1 mL of n-hexane was chosen for further analysis in order to reduce the consumption of organic solvent and to ease the subsequent separation of the extract (upper phase). In order to increase the contact surface between phases, improving the extraction effectiveness, the mixture was shaken using a vortex-mixer. 10, 30, 60, and 90 seconds were tested as extraction time, obtaining the best results with 30 seconds of vortex mixing. Regarding to the reconstitution volume, 100, 250, 400, and 500 µL of acetonitrile were tested to reconstitute the extract after the evaporation to dryness. Since all the tested volumes provided the required sensitivity, a 400 µL volume was finally chosen because it provided the best compromise between sensitivity and precision. 3.2.3. Study of the stability of derivatives In an early stage of the method development, the sample and standard solutions were measured after the derivatization reaction without performing the VALLsME. Although the concentrations of the analytes in the solution after the derivatization were enough to quantify them, it was observed that their stability was low in the basic aqueous medium, due to their chemical structure as carboxylate esters. The VALLsME allowed us to increase the sensitivity of the proposed method but also it was a way to stabilize the derivatized analytes. As an example, the relative stability (analytical signal / initial analytical signal) of PA is shown in Fig. 3, without

8

performing the VALLsME and performing it. Similar results were observed with the other target compounds.

3.3. Analytical figures of merit of the proposed method Quality parameters of the proposed method were evaluated under the selected conditions (see Section 2.4) with standard solutions containing the target compounds. These results are shown in Table 2.

The linearity studied reached at least to 25 µg mL−1 for all the analytes, obtaining a high level of linearity (R2 > 0.9990). The instrumental limits of detection values ranged from 0.02 to 0.06 µg mL-1 and the limits of quantification values ranged from 0.06 to 0.19 µg mL-1 (3 sy/x/b and 10 sy/x/b criteria respectively, being sy/x the residual standard deviation and b the slope of the calibration line). These low limits of detection and quantification in addition to the linear range observed, allowed us to determine the target compounds in a wide range of concentrations. The precision, expressed as relative standard deviation (RSD), was evaluated by applying the entire proposed method to ten replicates of standard solutions containing the analytes at 1 and 10 µg mL-1 in the same day (intra-day precision) and in different days (inter-day precision). The intra-day precision values ranged from 3.9 to 6.9 % at 1 µg mL-1, and from 4.7 to 7.0 % at 10 µg mL-1. The inter-day precision values ranged from 6.7 to 9.5 % at 1 µg mL-1, and from 6.9 to 9.3 % at 10 µg mL-1. These results reveal that good precision was achieved. In order to evaluate both the enrichment factor and the extraction yield, standard solutions containing the analytes (10 µg mL−1) were analyzed both directly and after performing the VALLsME procedure under the selected conditions (see 9

Section 2.4.1). The analytical signals were compared and the obtained values for the enrichment factor (analytical signal with VALLsME / analytical signal without VALLsME) ranged from 10.7 to 13.6 (see Table 2). Considering the experimental procedure, these results shown that the extraction yield achieved for the target compounds could be considered quantitative in all the cases. 3.4. Analysis of commercial samples Samples were prepared by triplicate as specified in the proposed method (see Section 2.4) and injected into the chromatographic system under the selected conditions. A good chromatographic resolution was obtained for all the tested samples. The results obtained for the 10 analyzed cosmetic samples are shown in Table 3. Fig. 4 shows, as an example, the chromatograms obtained for some of them.

In order to study the matrix effects, four cosmetic samples were prepared by triplicate according to the proposed method, and then spiked with the target analytes at two concentration levels (1 and 10 µg mL-1). Finally, the recoveries were evaluated by determining the concentration of the analytes in both spiked and non-spiked samples, applying the proposed method. The obtained recoveries (Table 4) ranged from 84 to 118%, thus showing that matrix effect was negligible and conventional calibration can be used as described in the proposed method.

4. Conclusions A new LC method with chromophoric derivatization followed by vortex-assisted liquid-liquid semimicroextration (VALLsME) is proposed for the determination of five alternative preservatives in cosmetics products: phenethyl alcohol, methylpropanediol,

10

phenylpropanol, caprylyl glycol and ethylhexylglycerin. These alternative preservatives are not yet included in the current European Regulation on cosmetic products. The method allows the quantification of the target compounds in both fat- and water-soluble cosmetic samples with good analytical features, such as accuracy and precision, as well as secondary figures of merit such as simplicity and affordable procedure, making the proposed method useful for the quality control of cosmetic products containing the target compounds as preservative agents. In addition, the use of small volumes of organic solvents, such as acetonitrile or n-hexane, makes the proposed method safe for both the operator and the environment, according to the principles of the so-called Green Analytical Chemistry. Acknowledgements Authors wish to acknowledge the Spanish ‘Ministerio de Economía y Competitividad’ for the financial support (Project CTQ-70301-R). P.M. also would like to thank the Spanish ‘Ministerio de Educación, Cultura y Deporte’ for his predoctoral grant.

References [1] Regulation (EC) No. 1233/2009 of the European Parliament and of the Council of 30 November

2009

on

cosmetic

products.

http://eur-lex.europa.eu/legal-

content/EN/TXT/?uri=celex:32009R1223 Last accessed on March 8, 2016 [2] Commission Regulation (EU) No. 358/2014 of 9 April 2014 amending Annexes II and V to Regulation (EC) No. 1233/2009 of the European Parliament and of the Council of 30 November 2009 on cosmetic products. http://eur-lex.europa.eu/legalcontent/EN/TXT/?uri=celex:32014R0358 Last accessed on March 8, 2016 [3]

Commission

Decision of

9 February 2006

amending

Decision

96/335/EC

establishing an inventory and a common nomenclature of ingredients employed in cosmetic

products

(2006/257/EC).

http://eur-lex.europa.eu/legal-

content/EN/TXT/?uri=celex%3A32006D0257 Last accessed on March 8, 2016

11

[4] S. Silver, L. Wendt. Mechanism of action of phenethyl alcohol - Breakdown of cellular permeability barrier, J. Bacteriol. 93 (1967) 560-566. [5] S. Fraud, E.L. Rees, E. Mahenthiralingam, A.D. Russell, J.Y. Maillard. Aromatic alcohols and their effect on Gram-negative bacteria, cocci and mycobacteria, J. Antimicrob. Chemother. 51 (2003) 1435-1436. [6] I. Slominski, G. Stoll,B. Meyer, W. Schneider, M. Waldmannlaue, B. Mayer, Synergistic antimicrobial mixt.|comprises aromatic alcohol and 1,2- or 1,3-di:ol; useful esp. as preservative in cosmetic compsns, Patent EP524548-A1. (1993). [7] R.M. Richards, R.J. McBride, M.A. Gunn. Preliminary investigation of the preservative properties of 3-phenylpropanol. J. Pharm. Pharmacol. 24 (1972) 158-159. [8] J.A. Goodall, J. Lyall, R.J. Mcbride, J.B. Murray, G. Smith. A comparison of phenylpropanol (0.5-percent) with chloroform (0.25-percent) and an admixture of methyl hydroxybenzoate (0.06-percent) and propyl hydroxybenzoate (0.03-percent) as preservatives in liquid medicines, J. Clin. Hosp. Pharm. 5 (1980) 323-327. [9] E.Y. Choi. Effect of phenoxyethanol and alkane diol mixture on the antimicrobial activity and antiseptic ability in cosmetics, Korean J. Aesthetic Cosmetol. 13 (2015) 213-220. [10] W. Johnson, W.F. Bergfeld, D.V. Belsito, R.A. Hill, C.D. Klaassen, D. Liebler, J.G. Marks, R.C. Shank, T.J. Slaga, P.W. Snyder, F.A. Andersen. Safety assessment of 1,2-glycols as used in cosmetics, Int. J. Toxicol. 31 (2012) 147-168. [11] M. Leschke, S. Wüstermann. A reliable alternative for traditional preservative systems, SÖFW-Journal. 132 (2006) 2-4. [12] T.A. Gaonkar, I. Geraldo, M. Shintre, S.M. Modak. In vivo efficacy of an alcoholbased

surgical

hand

disinfectant

containing

a

synergistic

combination

of

ethylhexylglycerin and preservatives, J. Hosp. Infect. 63 (2006) 412-417. [13] W. Beilfuss, M. Leschke, K. Weber. A new concept to boost the preservative efficacy of phenoxyethanol, SÖFW-Journal. 131 (2005) 2-7.

12

[14]

H.

Okamoto,

F.

Okada.

Microbicidal

antiseptic,

useful

in

cosmetics,

pharmaceuticals and foods, is a synergistic mixture of 1,2-alkanediol and tea tree oil, Patent FR2857222-A1. (2005). [15] J.J. Merianos, P. Garelick, S.M. Lindstrom, Antimicrobial composition useful in personal care product comprises 1,2-diol selected from 1,2-pentanediol, 1,2-hexanediol or 1,2-octanediol, phenoxyethanol and co-biocide, Patent EP1734922-A1. (2006). [16] Johnson & Johnson Consumer Co Inc, Combination, useful as an antimicrobial preservative system for cosmetic/personal care formulations, comprises caprylyl glycol and hinokitiol, Patent RD516001-A. (2007). [17] R. Pillai, J. Roeding, S. Lange, G. Schmaus, Mixture for e.g. preservation or antimicrobial treatment of perishable products comprises 1,2-hexanediol, 1,2octanediol, potassium sorbate, parabens, iodopropynyl butylcarbamate and optionally 1,2-pentanediol and/or 1,2-decanediol, Patent EP1804580-A1. (2007). [18] M. Vilanova, C. Martinez. First study of determination of aromatic compounds of red wine from Vitis vinifera cv. Castanal grown in Galicia (NW Spain), Eur. Food Res. Technol. 224 (2007) 431-436. [19] E. Coelho, R. Perestrelo, N.R. Neng, J.S. Camara, M.A. Coimbra, J.M.F. Nogueira, S.M. Rocha. Optimisation of stir bar sorptive extraction and liquid desorption combined

with

large

volume

injection-gas

chromatography-quadrupole

mass

spectrometry for the determination of volatile compounds in wines, Anal. Chim. Acta. 624 (2008) 79-89. [20] A. Angioni, G. Pintore, P. Caboni. Determination of wine aroma compounds by dehydration followed by GC/MS, J. AOAC Int. 95 (2012) 813-819. [21] J. Hrivnak, D. Smogrovicova, P. Nadasky, J. Lakatosova. Determination of beer aroma compounds using headspace solid-phase microcolumn extraction, Talanta. 83 (2010) 294-296. [22] M. Li, Z. Yang, J. Hao, L. Shan, J. Dong. Determination of tyrosol, 2-phenethyl alcohol, and tryptophol in beer by high-performance liquid chromatography, J. Am. Soc. Brew. Chem. 66 (2008) 245-249.

13

[23] J.H. Kahn, H.A. Conner. Rapid gas-liquid chromatographic determination of phenethyl alcohol in alcoholic distillates, J. AOAC Int. 55 (1972) 1155. [24] C. Jia, X. Zhu, J. Wang, E. Zhao, M. He, L. Chen, P. Yu. Extraction of pesticides in water samples using vortex-assisted liquid-liquid microextraction, J. Chromatogr. A. 1217 (2010) 5868-5871. [25] A. Papadopoulou, I.P. Roman, A. Canals, K. Tyrovola, E. Psillakis. Fast screening of perfluorooctane sulfonate in water using vortex-assisted liquid-liquid microextraction coupled to liquid chromatography-mass spectrometry, Anal. Chim. Acta. 691 (2011) 5661. [26] P. Gonzalez-Hernandez, V. Pino, J.H. Ayala, A.M. Afonso. A simplified vortexassisted emulsification microextraction method for determining personal care products in environmental water samples by ultra-high-performance liquid chromatography, Anal. Methods. 7 (2015) 1825-1833. [27] Z. Yang, Y. Lu, Y. Liu, T. Wu, Z. Zhou, D. Liu. Vortex-assisted surfactantenhanced-emulsification liquid-liquid microextraction, J. Chromatogr. A. 1218 (2011) 7071-7077. [28] G. Leng, W. Chen, M. Zhang, F. Huang, Q. Cao. Determination of phthalate esters in liquor samples by vortex-assisted surfactant-enhanced-emulsification liquid-liquid microextraction followed by GC- MS, J. Sep. Sci. 37 (2014) 684-690. [29] D. Moreno-Gonzalez, J.F. Huertas-Perez, A.M. Garcia-Campana, L. GamizGracia. Vortex-assisted surfactant-enhanced emulsification liquid-liquid microextraction for the determination of carbamates in juices by micellar electrokinetic chromatography tandem mass spectrometry, Talanta. 139 (2015) 174-180. [30] P. Miralles, A. Chisvert, A. Salvador. Determination of hydroxytyrosol and tyrosol by liquid chromatography for the quality control of cosmetic products based on olive extracts, J. Pharmaceut. Biomed. 102 (2015) 157-161. [31] P. Miralles, R. Bellver, A. Chisvert, A. Salvador. Vortex-assisted emulsification semimicroextraction for the analytical control of restricted ingredients in cosmetic products: determination of bronopol by liquid chromatography, Anal. Bioanal. Chem. (2016) DOI 10.1007/s00216-016-9306-5.

14

[32] V. Andruch, M. Burdel, L. Kocurova, J. Sandrejova, I.S. Balogh. Application of ultrasonic irradiation and vortex agitation in solvent microextraction, Trends Anal. Chem. 49 (2013) 1-19. [33]

M.

Moradi,

Y.

Yamini,

B.

Ebrahimpour.

Emulsion-based

liquid-phase

microextraction: a review, J. Iran. Chem. Soc. 11 (2014) 1087-1101. [34] A. Spietelun, L. Marcinkowski, M. de la Guardia, J. Namiesnik. Green aspects, developments and perspectives of liquid phase microextraction techniques, Talanta. 119 (2014) 34-45. [35] C. Bosch Ojeda, F. Sánchez Rojas. Vortex-assisted liquid-liquid microextraction (VALLME): Applications, Chromatographia. 77 (2014) 745-754. [36] A. Sinjewel, E.L. Swart, H. Lingeman, A.J. Wilhelm. LC determination of propylene glycol in human plasma after pre-column derivatization with benzoyl chloride, Chromatographia. 66 (2007) 103-105. [37] E.K. Paleologos, S.D. Chytiri, I.N. Savvaidis, M.G. Kontominas. Determination of biogenic amines as their benzoyl derivatives after cloud point extraction with micellar liquid chromatographic separation, J. Chromatogr. A. 1010 (2003) 217-224. [38] F. Aflaki, V. Ghoulipour, N. Saemian, M. Salahinejad. A simple method for benzoyl chloride derivatization of biogenic amines for high performance liquid chromatography, Anal. Methods. 6 (2014) 1482-1487. [39] M.A. Ahmed, I. Khan, J. Hashim, S.G. Musharraf. Sensitive determination of glycerol by derivatization using a HPLC-DAD method in biodiesel samples, Anal. Methods. 7 (2015) 7805-7810. [40] O. Ozdestan, A.Uren. A method for benzoyl chloride derivatization of biogenic amines for high performance liquid chromatography, Talanta. 78 (2009) 1321-1326. [41] S. Liu, H. Ruan. A highly sensitive quantification of phytosterols through an inexpensive derivatization, Chem. Phys. Lipids. 166 (2013) 18-25.

15

Highlights



The method allows the determination of five alternative preservatives in cosmetics



No analytical methods have been published to determine these compounds in cosmetics



VALLsME and LC-UV allow the determination in fat and water soluble samples



Simplicity and affordable procedure make it useful for quality control



The method is environmentally-friendly according to the Green Chemistry principles

Figure captions

Fig 1. Chemical structures of the target compounds.

Fig 2. Chromatographic separation obtained applying the proposed method to a standard solution containing the derivatized analytes at 10 µg mL-1.

Fig 3. Relative stability (analytical signal / initial analytical signal) of phenethyl alcohol performing the VALLsME (dotted line) and without performing the VALLsME (continuous line).

Fig. 4. Chromatographic separation obtained for four examples of the cosmetic samples analyzed: (a) sample C, (b) sample D, (c) sample E and (d) sample G.

16

Figure 1.

Phenethyl alcohol (PA)

Methylpropanediol (MP)

Phenylpropanol (PP)

(CAS 60-12-8)

(CAS 2163-42-0)

(CAS 1335-12-2)

Caprylyl glycol (CG)

Ethylhexylglycerin (EG)

(CAS 1117-86-8)

(CAS 70445-33-9)

Absorbance (mAU)

Fig 2.

Time (min)

Fig 3.

17

Absorbance (mAU)

Time (min)

Fig. 4 a)

b)

18

Relative stability

c)

d)

Time (min)

19

Table 1. Elution gradient program used in the liquid chromatography analysis. t (min)

Acetonitrile (%, v/v)

Aqueous 0.5 % acetic acid (%, v/v)

0

60

40

4

60

40

10

100

0

11

100

0

11.1

60

40

14

60

40 −1

Experimental conditions: injection volume = 20 µL; flow rate = 1 mL min ; oven temperature = 30 ºC; λ = 235 nm

20

Table 2. Analytical figures of merit of the proposed method.

Precision (RSD, %) Slope

Intercept

LOD 2 b

Analyte

R -1 a

(mAU min mL µg )

(mAU min)

sy/x

a

LOQ

c

Intra-day -1 d

(µg mL )

f

Inter-day

EF

g

-1 e

(µg mL )

1 µg mL

-1

-1

10 µg mL

1 µg mL

-1

-1

10 µg mL

PA

710 ± 10

30 ± 80

0.9995

8

0.03

0.11

3.9

4.7

6.7

6.9

13.1 ± 0.9

MP

2010 ± 40

240 ± 200

0.9990

40

0.06

0.19

6.2

6.5

8.7

8.4

13.6 ± 1.2

PP

750 ± 20

- 60 ± 100

0.9995

5

0.02

0.06

6.9

6.7

9.5

9.1

12.6 ± 0.7

CG

1650 ± 80

350 ± 60

0.9998

30

0.06

0.18

6.0

6.2

8.6

8.5

11.5 ± 0.4

EG

1250 ± 20

170 ± 50

0.9994

9

0.02

0.07

6.8

7.0

9.2

9.3

10.7 ± 0.6

a

Parameters of the calibration curve obtained by simple linear regression. Working range: 1 – 25 µg mL , n = 6. Expressed as the value ± standard deviation.

-1

b

Regression coefficient (R ) of the calibration curve.

c

Residual standard deviation (sy/x) of the calibration curve.

d

Limit of detection (LOD) estimated as 3 sy/x/b, being sy/x the residual standard deviation and b the slope of the calibration curve.

e

Limit of quantification (LOQ) estimated as 10 sy/x/b, being sy/x the residual standard deviation and b the slope of the calibration curve.

f

Precision expressed as relative standard deviation (RSD, %), n = 10.

g

Enrichment factor (EF). The values are expressed as the mean of three replicates ± standard deviation.

2

21

Table 3. Obtained concentrations (%, w/w) of the target compounds in the analyzed cosmetic samples.

Concentration (%, w/w) Sample

b

a

PA

MP

PP

CG

EG

A

n.d.

n.d.

n.d.

n.d.

n.d.

B

n.d.

n.d.

n.d.

n.d.

n.d.

C

n.d.

n.d.

n.d.

0.18 ± 0.01

n.d.

D

n.d.

n.d.

n.d.

0.065 ± 0.005

0.047 ± 0.003

E

0.76 ± 0.07

n.d.

n.d.

n.d.

n.d.

F

0.79 ± 0.03

n.d.

n.d.

n.d.

n.d.

G

n.d.

1.88 ± 0.07

0.078 ± 0.005

0.51 ± 0.07

n.d.

H

n.d.

1.95 ± 0.05

0.084 ± 0.006

0.62 ± 0.04

n.d.

I

n.d.

1.72 ± 0.09

0.087 ± 0.006

0.59 ± 0.02

n.d.

J

n.d.

1.84 ± 0.05

0.076 ± 0.004

0.55 ± 0.03

n.d.

a

Samples A-F: creams; Samples G-J: gels.

b

The values are expressed as the mean of three replicates ± standard deviation.

n.d.: Not detected, analytes were not included in sample formulation. See text for experimental details.

Table 4. Recovery values (%) obtained by applying the proposed method to four samples spiked with known amounts of the analytes. Recovery values (%) Analyte

Sample A 1 µg mL

-1

b

Sample B -1

10 µg mL

1 µg mL

-1

b

a

Sample C -1

10 µg mL

1 µg mL

-1

b

Sample G -1

10 µg mL

1 µg mL

-1

b -1

10 µg mL

PA

102 ± 5

105 ± 8

99 ± 4

86 ± 3

94 ± 6

110 ± 9

90 ± 7

93 ± 6

MP

87 ± 7

84 ± 3

90 ± 6

94 ± 4

92 ± 3

84 ± 7

n.c.

92 ± 5

PP

97 ± 5

107 ± 5

91 ± 3

106 ± 3

97 ± 4

108 ± 7

89 ± 3

112 ± 8

CG

85 ± 9

106 ± 3

118 ± 8

97 ± 6

108 ± 9

106 ± 5

98 ± 4

100 ± 5

EG

104 ± 4

115 ± 7

105 ± 8

92 ± 5

98 ± 5

100 ± 6

94 ± 6

103 ± 3

a

The values are expressed as the mean of three replicates ± standard deviation.

b

Sample A-C: creams; Sample G: gel.

n.c.: Not calculated, the analyte was present in sample G at high concentration.

22

A l t e r n a t i v e p r e s e r v a t i v e s