Analytica Chimica Acta 733 (2012) 2833

Contents lists available at SciVerse ScienceDirect

Analytica Chimica Acta

journal homepage: www.elsevier.com/locate/aca

Enzymatic single-drop microextraction for the assay of ethanol in alcohol-free

cosmetics using microvolume uorospectrometry detection
Noelia Cabaleiro, Inmaculada de la Calle, Carlos Bendicho, Isela Lavilla
Departamento de Qumica Analtica y Alimentaria, rea de Qumica Analtica, Facultad de Qumica, Universidad de Vigo, Campus As Lagoas-Marcosende s/n, 36310 Vigo, Spain

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 An enzyme-based assay in drop

format is performed for ethanol
determination in alcohol-free cos-
metics.
 An aqueous drop containing ADH and
NAD is used as a probe to extract
ethanol.
 Fluorescence changes are monitored
by microvolume uorospectrometry.
 Matrix effects from the cosmetic
sample are not present using the
headspace-SDME mode.
 LODs better than those obtained by
other enzymatic-based methods are
reached.

a r t i c l e i n f o a b s t r a c t

Article history: A green assay based on the development of an enzymatic reaction in drop format under headspace single-
Received 27 February 2012 drop microextraction conditions is described for the rst time. An aqueous drop containing the enzyme
Received in revised form 27 April 2012 alcohol dehydrogenase and the cofactor -Nicotinamide adenine dinucleotide has been used as uo-
Accepted 29 April 2012
rescence probe for determining ethanol in alcohol-free cosmetics by microvolume uorospectrometry.
Available online 10 May 2012
Experimental parameters affecting the microextraction performance were carefully optimized. Under
the conditions employed, the contribution of other alcohols was found to be negligible. After 10 min of
Keywords:
microextraction, a detection limit of 0.04 g g1 ethanol, a repeatability, expressed as relative standard
Headspace-single-drop microextraction
Enzymatic probe
deviation, of 5.3% for a 0.05 mM ethanol standard and a preconcentration factor of 391, were reached.
Ethanol assay Accuracy of the proposed methodology was evaluated by comparison of calibration slopes correspond-
Alcohol-free cosmetics ing to external calibration with aqueous standards and standard addition calibration. The method was
Microvolume uorospectrometry successfully applied to different alcohol-free cosmetics (external calibration was carried out in all cases).
Green method Additional advantages such as simplicity and high sample throughput can be highlighted. The green-
ness prole of proposed methodology was established using NEMI criteria (US National Environmental
Methods Index).
2012 Elsevier B.V. All rights reserved.

1. Introduction
In the last years, sales of so called free-from cosmetics have dra-
matically increased. In general, consumers demand cosmetics more
Corresponding author. Tel.: +34 986 812291; fax: +34 986 812556.
E-mail address: isela@uvigo.es (I. Lavilla).
natural and healthier with the same effectiveness than traditional
products. In this framework, alcohol-free cosmetics are remark-
able, which are acquired by consumers in order to prevent possible
dryness and irritation of skin and hair. Approximately, 10% of cos-
metics are already labeled as alcohol-free [1].
These products do not contain ethanol, although may contain
other alcohols such as cetyl, stearyl, cetearyl or lanolin alcohol.
Ethanol is added to cosmetics due to its antimicrobial action

0003-2670/$ see front matter 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.aca.2012.04.039
 N. Cabaleiro et al. / Analytica Chimica Acta 733 (2012) 2833
and its activity as topical penetration enhancer [2,3]. Neverthe-
less, side effects such as contact dermatitis or increase of the
transdermal absorption of toxic substances such as herbicides
have been observed when cosmetics with alcohol are used. Usu-
ally, ethanol control in cosmetics is made for scal or religious
reasons [4].
Apart from the above mentioned uses, ethanol is used along
different processes in cosmetics industry as solvent, viscosity
decreasing agent, antifoam agent, etc. Therefore, it is not surpris-
ing that many cosmetic products hide traces of ethanol in the
nal formulation. Then, in order to avoid possible labeling frauds,
analytical control is essential. Nevertheless, a scanty number of
analytical methods for ethanol determination in cosmetics have
been developed so far. Most of these methods include gas, liquid
chromatography or infrared spectrometry [57]. In general, good
results are obtained only if the ethanolic content of samples is
high (about 0.20.5%, v/v). Lack of specicity and matrix effects
have also been pointed out in some cases [7]. So, the improve-
ment of analytical methodologies for trace ethanol determination
in cosmetics by the implementation of recent developments is
advisable.
Miniaturized microextraction techniques have undergone an
impressive development in recent years. These can be considered as
a way for the development and improvement of more sustainable
analytical procedures into the framework of the Green Analytical
Chemistry (GAC) [8]. In addition, microextraction provides numer-
ous analytical advantages in the determination of analytes at trace
levels. In special, headspace techniques such as headspace single
drop microextraction (HS-SDME) allow a high sample clean-up for
complex matrixes with a high potential of analyte enrichment in
the extractant drop [9,10]. Organic solvents as hexane, octanol, hex-
adecane, dodecane, decane, decanol, toluene, ethylene glycol etc.
are commonly employed in the drop. In order to design total envi-
ronmentally friendly procedures, these traditional organic solvents
must be replaced with non-toxic ones. Room temperature ionic liq-
uids or supramolecular solvents have been proposed as greener
alternatives [8].
In spite of its unequivocal green nature, aqueous drops have
been scantily used in HS-SDME. This is because only volatile
or semi-volatile ionizable compounds can be preconcentrated in
aqueous drops. A chemical reaction needs to take place in the drop
in order to trap the analyte. Thus, Fragueiro et al. [11] used an
aqueous drop (3 L) containing 30 mg L1 Pd(II) in order to pre-
concentrate arsine for the determination of As by electrothermal
atomic absorption spectrometry (ETAAS). Aqueous drops con-
taining different concentration of Pd have been also used for
trapping other hydride forming elements [1215]. Other analytes
preconcentrated in aqueous drops are: phenols in 5 L of 1 M
NaOH [16], CN in 5 L of 0.1 M Ni(II) and 50 mM NH3 [17],
methamphetamine in 5 L of 0.05 M H3 PO4 [18], sulde in 2 L
of 750 mg L1 Zn(II) with 1 M NaOH [19], volatile N-oxides in a
Griess reagent-containing drop (3 L) [20] and formaldehyde in a
Hantzsch reagent-containing drop (3 L) [21]. Recently, Cabaleiro
et al. [22] have proposed a HS-SDME procedure for terpene screen-
ing in fragrance-free cosmetics. The drop is composed by an
aqueous solution of a uorescence probe formed by bovine serum
albumin and uorescein. A list of additional applications of aqueous
drop based HS-DME can be found in the review by Jain and Verma
[23].
Despite enzymatic reactions have been extensively used in ana-
lytical chemistry in order to achieve both increased selectivity and
sensitivity, to the best of our knowledge, aqueous drops contain-
ing enzymes in HS-SDME have not been employed so far. The
implementation of enzymes in an aqueous drop seems a promis-
ing strategy in HS-SDME, since it provides a high sample clean-up
together with increased selectivity and sensitivity. In this work,
29
the enzyme alcohol dehydrogenase (ADH) and the cofactor -
Nicotinamide adenine dinucleotide (NAD) have been conned in
an aqueous drop and used as a uorescent probe for ethanol deter-
mination in alcohol free cosmetic samples. Fluorescence changes
in the drop in the absence and presence of ethanol have been mon-
itored by microvolume uorospectrometry.
2. Experimental
2.1. Apparatus
A Thermo Scientic NanoDrop 3300 Fluorospectrometer
(Thermo, Wilmington, USA) was used for microvolume uores-
cence measurements. A 2 L drop is placed onto the pedestal of the
uorospectrometer. The measurements were carried out at 446 nm
using an UV Light Emitting Diode as excitation source at 365 nm.
Headspace single-drop microextractions were carried out by
means of a 10-L Hamilton syringe (Hamilton Company, Nevada,
USA) with a guided-PTFE plunger. 40 mL glass vials with a silicone
rubber septum were used for performing the HS-SDME procedure.
A microbalance MC5 Sartorius (Sartorius, Goettingen, Germany)
was used for weighing ADH and NAD.
2.2. Reagents and samples
High-purity deionised water obtained from a PETLAB ultrapure
water system (Peter Taboada, Vigo, Spain) was used through-
out the work. Ethanol, methanol, butanol and geraniol were
purchased from Prolabo (Fontenaysous-Bois, France). 2-Propanol
was purchased from Merck (Darmstadt, Germany). Yeast alco-
hol dehydrogenase (300 U mg1 protein) and -Nicotinamide
adenine dinucleotide were purchased from Sigma (Steinheim,
Germany) and Fluka (Steinheim, Germany), respectively. A 0.02 M
tris-hydroximethyl-aminomethane (TRIS, Sigma) buffer solution at
pH 9 was used to prepare daily the working solution of ADH and
NAD used in the drop.
Different kinds of cosmetics were analyzed: hands cream, mois-
turizing cream, anti irritant cream, deodorants, facial cleansing gel,
nappy cream, hair and body gel and foaming gel. No ethanol, dena-
tured ethanol or other low chain alcohols (C1 C4 ) were indicated
in the labels.
2.3. HS-SDME procedure
About 1 g of cosmetic sample was accurately weighed in a 40 mL
glass vial containing a stirring bar (10 mm 3 mm). After adding
10 mL of deionised water, the vial was closed and thermostated
at 40 C. Then, a 2.5 L-aqueous drop composed of 5 mM NAD,
4000 U mL1 ADH and 0.02 M TRIS buffer at pH 9 was exposed to the
headspace above of the sample stirred at 1100 rpm for 10 min. Then,
the drop was retracted back into the microsyringe. A 2 L volume of
this solution was placed on the pedestal of the uorospectrometer
for measurement. Blanks were treated in the same way.
3. Results and discussion
3.1. Enhancing ADH substrate specicity with the HS-SDME
procedure
In general, alcohols can react with NAD in the presence of ADH
to yield reduced NAD (NADH) and the corresponding aldehyde
or ketone. These reactions can be followed by spectrouorome-
try since initial uorescence of NAD increases as a result of the
conversion to NADH.
30 N. Cabaleiro et al. / Analytica Chimica Acta 733 (2012) 2833

Fig. 1. Study of the activity of the enzyme tested with different alcohols at a con-
centration of 0.5 mM.

ADH is an enzyme characterized by its specicity toward low

chain alcohols (primary and secondary alcohols), although it is
mostly active with ethanol (its activity diminishes as the size of
the alcohol increases or decreases) [24,25]. Then, the activity of the
enzyme was tested for different volatile alcohols with and without
HS-SDME: ethanol (vapour pressure 59.5 hPa at 20 C), methanol
(130.3 hPa), 2-propanol (43.2 hPa), butanol (5 hPa) and geraniol
(0.3 hPa). These alcohols are commonly used in cosmetics with dif-
ferent purposes, i.e. methanol is used as denaturing agent, butanol
as solvent, propanol is incorporated as reducing thickness agent,
reducing foam agent or as solvent and geraniol is added as fragrance
Fig. 2. Optimization of (A) NAD concentration in the drop using a xed ADH con-
ingredient in many types of cosmetics. centration of 1500 U mL1 and (B) ADH concentration in the drop for a xed NAD
The assays were made under usual conditions corresponding to concentration (5 mM).
commercial photometric or uorimetric assay kits [26,27]: 0.5 mM
aqueous solutions of the different alcohols, 1500 U mL1 of ADH,
9 mM NAD and pH 8.8. Reaction time (with and without microex- was obtained for a concentration in the range 57 mM. As a conse-
traction) was xed at 20 min in order to minimize its possible quence, 5 mM was employed for further experiments. Higher NAD
inuence (especially when HS-SDME is used). Results are shown in concentrations inuence the ADH activation causing a decrease in
Fig. 1. As can be seen, under these conditions, this reaction is very the signal [30]. The ADH concentration was studied in the range
selective for ethanol in both procedures. A high preconcentration 506500 U mL1 (Fig. 2B). An increase in the concentration was
of ethanol in the drop can be also observed with microextrac- observed up to 4000 U mL1 .
tion (ca. an order of magnitude under non-optimized conditions). In general, activity of ADH enzymes shows a remarkable depen-
Fluorescence signals for the other alcohols were only 7.5, 6.8, 4.5 dence on pH. The inuence of this parameter was studied by varying
and 1.4% of the ethanol signal for butanol, propanol, gerianol and the pH in the drop between 7 and 9.5 (0.02 M TRIS was adjusted
methanol, respectively, when the HS-SDME procedure is used. with HCl or NaOH). As can be seen in Fig. 3, the pH in the drop turns
Without microextraction, these percentages were 15.2, 13.3, 6.9 out to be a crucial parameter. Maximum ADH activity occurs in a
and 19.2% for butanol, propanol, gerianol and methanol, respec- narrow pH range (8.89.2). No activity of the enzyme is observed
tively. Hence, it can be concluded that HS-SDME contributes to in the range 78. Similar optimal values have been also reported
enhance the specicity of ADH. Probably, it is due to the different in the literature for yeast ADH [31]. Therefore, pH of the drop was
volatility and diffusion of these alcohols in the drop. maintained at 9 for further experiments.
In addition, the isolation of the enzyme in the drop allows to Extractant phase and sample volumes directly inuence the
overcome matrix effects from ADH inhibitors, i.e. heavy metals, - extraction efciency. Generally, the amount of extracted analyte
SH compounds and high molecular weight organic compounds (in increases with increasing drop volume, as a consequence of the
special proteins) [28,29]. Among these inhibitors, some common
ingredients of cosmetics such as SDS (sodium dodecyl sulphate),
acid ascorbic or metals are included [28]. Consequently, strong
matrix effects would occur if the enzyme were directly used for
cosmetic analysis.

3.2. Optimization of the HS-DME procedure

Different experimental parameters that could inuence the ef-

ciency of the HS-SDME procedure were exhaustively studied. A
solution containing 0.05 mM ethanol in water was used for opti-
mization purposes. All experiments were carried out in triplicate.
The oxidation reaction used in this work critically depends on
both the enzyme and coenzyme concentrations. For this reason, the
effect of NAD concentration on the reaction was examined in the
range 215 mM (concentrations higher than 15 mM caused some Fig. 3. Effect of the drop pH on the uorescence response using 5 mM NAD and
turbidity in the drop). As can be seen in Fig. 2A, maximum signal 4000 U mL1 of ADH.
 N. Cabaleiro et al. / Analytica Chimica Acta 733 (2012) 2833
Fig. 4. Study of microextraction time on the uorescence response.
larger surface area [10]. Thus, the inuence of the drop volume was
studied in the range 23.5 L (S1). A slight increase in the uo-
rescence response was observed up to 2.5 L. Volumes larger than
3 L caused drop dislodgement. As a consequence, 2.5 L were used
as drop volume. The effect of the sample volume was studied in
the range 520 mL (S2). An increase in the analytical signal was
observed up to a volume of 10 mL, which was nally selected.
Agitation of the sample turns into an important parameter for
an efcient mass transfer of ethanol to the headspace. For this pur-
pose, magnetic stirring of the sample was studied in the range
5001400 rpm (S3). An increase in the extraction efciency was
obtained for stirring rates up to 1100 rpm. Higher values did not
cause a signicant improvement on the analytical signal. Moreover,
there is a risk of drop fall by spattering. Accordingly, 1100 rpm were
used for further experiments.
Higher concentrations of the analyte in the headspace can be
generally obtained as a result of the increase in temperature. On
the other hand, a worsening of the precision due to evaporation
of the drop can occur during the microextraction procedure as the
temperature increases [22]. In addition, when an enzymatic reac-
tion is used, an excessive increase in temperature could also lead
to a decrease in the enzyme activity. Yeast ADH allows working
with temperatures up to 50 C [31]. In order to test the effect of
the temperature on the microextraction, the sample vial was ther-
mostated at different temperatures between 20 and 45 C (S4). An
increase in temperature showed a benecial effect on the analyt-
ical signal up to 40 C. Therefore, sample vials were thermostated
at this temperature.
The salting out effect has been considered benecial for most
liquid phase microextraction approaches, since an increase in the
ionic strength of the sample solution allows achieving a more ef-
cient mass transfer. Addition of salt as NaCl to the sample solution
was studied in the range 030% (w/v) (S5). No signicant increase
in the analytical signal was observed, and consequently, no salt was
added for further experiments.
HD-SDME critically depends on the microextraction time [10].
In general, maximum sensitivity is achieved when the equilibrium
among the three phases (sample solution, headspace and drop) is
reached. In order to test whether the equilibrium was reached,
microextraction time was varied between 2 and 15 min. As can
be seen in Fig. 4, equilibrium among the three involved phases
was reached from 10 min of microextraction. When the equation
of MichaelisMenten is considered, it allows conrming that the
reaction in the drop takes place in approximately 30 s, thus being
the limiting factor in the microextraction of ethanol.
3.3. Effect of other alcohols on the analytical signal
To evaluate the contribution of other alcohols on the analytical
signal obtained under optimized conditions, the microextraction
31
Fig. 5. Inuence of the sample mass on the concentration of ethanol found (0.05 mM
spiked ethanol in the anti-irritant cream 0% alcohol).
procedure was carried out using 0.05 mM butanol and methanol
solutions. Analytical signals for butanol and methanol were 5.9%
and 1.3% respectively, with regard to that obtained for ethanol. This
conrms that ADH activity toward other alcohols different from
EtOH is very low in the HS-SDME procedure developed.
3.4. Cosmetic sample pre-treatment
Due to the volatile nature of ethanol, a simple sample pre-
treatment can be accomplished even when analysing complex
samples such as cosmetics [22]. In order to achieve an efcient
extraction of ethanol from the sample matrix, no pre-treatment and
a simple dilution with deionized water were considered. Exper-
iments were carried out with (i) 0.5 g of anti-irritant cream 0%
alcohol spiked with 10 g of ethanol in the vial and (ii) 0.5 g of
anti-irritant cream 0% alcohol spiked with 10 g of ethanol and
10 mL of water as solubilizing agent. The recovery found without
pre-treatment was 91 4% (n = 3) (i) A recovery of 101 2% (n = 3)
was obtained with the simple dilution procedure (ii) As a result, the
addition of 10 mL of water to the cosmetic sample was selected as
the only sample pre-treatment.
Sample mass can play an important role when enhanced sen-
sitivity is required. However, a high increase in sample mass
can negatively affect the extraction efciency. In these cases, the
magnetic stirring can be inefcient to achieve the homogeneous
distribution of the sample into the 10 mL of water [22]. Conse-
quently, the effect of the sample mass on the ethanol extraction
was examined in the range 0.51.5 g using the anti-irritant cream
0% alcohol sample (Fig. 5). Extraction was found to be quantitative
for a sample mass up to 1 g.
3.5. Validation of the proposed methodology
A typical equation of external calibration was:
Y = 38575X + 159.14, r 2 = 0.997
where Y is the uorescence intensity and X is the ethanol concen-
tration expressed in mM.
Analytical characteristics of the proposed methodology were
obtained. Limit of detection (LOD) and quantication (LOQ), lin-
ear range, repeatability and reproducibility are shown in Table 1.
LOD and LOQ were determined following the 3 and 10 criterion,
respectively, where  is standard deviation of 10 blank consecutive
measurements and taking into account the sample preparation per-
formed (i.e., procedural LOD and LOQ). Instrumental detection and
quantication limits were found to be 9 105 and 3 104 mM,
respectively. Procedural LOD and LOQ were 0.04 and 0.13 g g1 ,
respectively.
32 N. Cabaleiro et al. / Analytica Chimica Acta 733 (2012) 2833

Table 1
Comparison of the analytical characteristics obtained in the proposed method with other analytical methodologies for the determination of ethanol in different samples.

Method Matrix Total estimated sample LOD Linear range RSD (%) Reference
treatment + analysis
time (min)
HS-SDME microvolume Cosmetics 10 + 1 0.00009a 0.00030.1a 5.3 This work
uorospectrometry 0.04b 0.1345b
Microextraction-based approaches
HS-SDMEGCFID Beer 15 + 36 0.2a <33a 5.810 [32]
Dynamic-HS-SDMEGCMS Beer 9.5 + 27 0.0014a 0.0040.43a 9.3 [33]
MHSSDME-GCFID Solid pharmaceuticals (180c + 20) + 9.4 0.1b 2.515d 4.13 [34]
HS-SPMEGCFID Solid pharmaceuticals 25 + 35.5 0.0005b Not given 16.5 [35]
IL-HS-SPMEGCFID Aqueous solutions 13 + 24.7 0.0020.009a 0.0022.2a 1.16 [36]

Enzymatic-based approaches
SEyADHSFuorospectrometry Gasoline, breath, liquor Not given Semiquantitative 101000a [37]
samples
FI-immobilized enzyme reactors- Wine 3+1 0.34a 8.5785.7a 2.3 [38]
multi-site spectrophotometry
a a
ADH-enzymatic Alcoholic beverages 15 + 1 1 15 7.2 [31]
assay-spectrophotometry
IE-EFCA-uorospectrometry Alcoholic beverages 20 + 1 24a 43326a 1.9 [30]
SI-LOV enzymatic Alcoholic beverages <2.2 + 1 0.510.69a < 6.9a 0.75 [39]
assay-spectrophotometry
HS-EB Apple juices (1.1523.5) + 1 0.0050.015a 0.130a 3 [40]
ADH-Graphene modied electrode Alcoholic beverages 0.17 + 1 0.025a 0.221.0a 2.86.4 [41]

Other approaches
VPGPLSFTIR Alcoholic beverages 0.9 + 1 36a 428.73858.3a 2.7 [6]
and colognes

[Acronym denition]: GCFID, gas chromatographyame ionization detection; GCMS, gas chromatographymass spectrometry detection; MHSSDME, multiple headspace
single-drop microextraction; HS-SPME, headspace solid phase microextraction; IL-HS-SPME, ionic liquid based-headspace single drop microextraction; SEyADHSF,
silicateencapsulated yeast alcohol dehydrogenaseuorospectrometry; FI, ow injection; IE-EFCA, immobilized enzyme uorescence capillary analysis; SI-LOV, sequential
injection-lab-on-valve; HS-EB, headspace electrochemical biosensor; VPGPLSFTIRS, vapor phase generationpartial least-squaresFourier transform infrared spectrometry.
a
Values expressed as mM.
b
Values expressed as g g1
c
Previous time to microextraction.
d
Values expressed as g.

A comparison between the characteristics of the proposed
method and those of different methodologies reported in the lit-
erature for ethanol determination in different matrices is shown
in Table 1. In general, good LODs are obtained with regard to other
reported procedures, especially those involving enzymatic analysis.
Although better sensitivity is achieved in the HS-SPME-GCFID pro-
cedure [35], in this case the long time needed per analysis (30 min
for extraction, 5 min for desorption and 1520 min for measure-
ment), the short life of the SPME ber and the much more complex
instrumentation can be emphasized. The developed methodology
provides improved detection limits by a factor of at least 50 times
as compared to those obtained by other enzymaticbased methods
not involving preconcentration.
A wide linear range (0.00050.1 mM of ethanol) can also be
highlighted. The precision, expressed as relative standard devia-
tion (RSD), was evaluated in terms of repeatability. RSD from ve
independent replicates was found to be 5.3% for a 0.05 mM ethanol
standard. Reproducibility was studied during ve consecutive days
and was found to be 7.5%.
The enrichment factor, dened as the ratio between the nal
analyte concentration in the extractant aqueous drop and the initial
sample concentration, was found to be 391.
In addition, the proposed methodology is characterized by its
high sample throughput. After 10 min of microextraction time,
uorescence measurements are carried out in less than 5 s. Longer
analysis times have been reported in other procedures involving
preconcentration in combination with GCFID or GCMS (between
20 and 60 min). Enzymatic-based procedures usually report times
from few seconds to 23.5 min, corresponding to the reaction time
[40,41].
Accuracy of the proposed methodology was evaluated by com-
parison of calibration slopes corresponding to external calibration
with aqueous standards and standard addition method for different
cosmetic samples. The results are shown in Table 2. The comparison
of slopes is presented as percentage of change (aqueous standard
calibration/calibration by standard addition). As can be seen, the
absence of matrix effects can be conrmed, since the change in
slope was lower than 7.5% in all cases assayed.
3.6. Greenness prole of proposed methodology
The greenness prole of the proposed methodology was estab-
lished in concordance with the suggestions of the NEMI (US
National Environmental Methods Index) [42]. The presence of PBT
(persistent, bioaccumulative and toxic chemicals), hazardous and
corrosive reagents as well as the generated waste were considered.
In this case, the amount of wastes generated is <50 g, the pH is no

Table 2
Comparison of slopes from external calibration (aqueous standards) and standard addition calibration method for different cosmetic samples.

Matrix Slope (stand. addition) Change (%)a

Hands cream atopic skin 40,749 5.3

Moisturizing cream sensitive/atopic skin 41,713 7.5
Anti-irritant cream 0% alcohol 39,221 1.6
0% alcohol deodorant 1 40,653 5.1
Nappy cream for delicate or atopic cream 37,601 2.6
a
Change with respect to slope of the external calibration (38,575).
 N. Cabaleiro et al. / Analytica Chimica Acta 733 (2012) 2833
Table 3
Recovery study and analytical results for the determination of ethanol in cosmetics.
Sample Labeling
Hands cream Atopic skin
Moisturizing cream Sensitive/atopic skin
Anti-irritant cream 0% alcohol
Deodorant 1 0% alcohol
Deodorant 2 0% alcohol
Cleansing gel Without alcohol
Nappy cream Delicate or atopic cream
Hair and body gel Ethanol not labeled
Foaming gel Delicate skin
a
For a 0.05 M ethanol spike.
b
Average value standard deviation (n = 3).
corrosive, PBTs and hazardous solvents were not used. As a result,
the proposed methodology was found to have a total green prole.
3.7. Sample analysis
Different cosmetic samples were analyzed following the pro-
posed procedure. Calibration with aqueous standards was carried
out in all cases. As can be seen in Table 3, the ethanol concentra-
tion was found to be below the LOD in only three samples. Besides,
six samples were found to have ethanol concentrations in the range
3510 g g1 . In view of these results and in order to prevent label-
ing fraud, specic legislation in the matter should be developed
4. Conclusions
In this work, a total green miniaturized assay for ethanol in
alcohol-free cosmetic samples has been developed. The use of HS-
SDME with an aqueous drop containing an enzyme (ADH) and a
co-enzyme (NAD) in combination with a microvolume uorospec-
trometer allows achieving excellent analytical characteristics. ADH
and NAD conned in the drop act as a selective uorescence probe
obtaining high selectivity, thus making negligible the contribution
from other alcohols different than ethanol. A high preconcentration
factor can be highlighted due to the extractant phase-to-sample
volume ratio. Low limits of detection and quantication and good
repeatability and reproducibility were also obtained under opti-
mized conditions. A high sample throughput can be highlighted in
comparison with other existing methods used for ethanol deter-
mination. A methodology with total greenness prole has been
reached.
Acknowledgements
Financial support from the Spanish Ministry of Economy and
Competitiveness (Project CTQ2009-06956/BQU) and the Vigo Uni-
versity (Contract for Reference Research Groups 09VIA08) is
gratefully acknowledged. N. Cabaleiro thanks the Vigo University
for a predoctoral contract. I. De La Calle also thanks the Xunta de
Galicia for nancial support as a researcher of the Maria Barbeito
program.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.aca.2012.04.039.
References
[1] http://www.thend.com/beauty/info-alcohol-free-cosmetics, 2012 (accessed
February 2012).
33
Recoverya Found ethanol (g g1 )b
104 51 1
108 8 0.4
102 3 0.9
105 <LOD
95 <LOD
93 62
98 <LOD
104 510 2
96 61 1
[2] A.C. Williams, B.W. Barry, Adv. Drug Deliv. Rev. 56 (2004) 603618.
[3] D. Hassen, A.H. Bennasir, S. Sridhar, J. Pharm. Res. 2 (2009) 17321734.
[4] L. Gagliardi, D. De Orsi1, S. Dorato, in: A. Salvador, A. Chisvert (Eds.), Analysis
of Cosmetic Products, Elsevier, Netherlands, 2007, p. 49.
[5] G.L.P. Grant, J. AOAC 57 (1974) 565567.
[6] A. Prez-Ponce, F.J. Rambla, J.M. Garrigues, S. Garrigues, M. de la Guardia, Ana-
lyst 123 (1998) 12531258.
[7] M. Biniecka, S. Caroli, Trac-Trends Anal. Chem. 30 (2011) 17561770.
[8] F. Pena-Pereira, I. Lavilla, C. Bendicho, Trac-Trends Anal. Chem. 29 (2010)
617628.
[9] M.A. Jeannot, A. Przyjazny, J.M. Kokosa, J. Chromatogr. A 1217 (2010)
23262336.
[10] F. Pena-Pereira, I. Lavilla, C. Bendicho, Anal. Chim. Acta 669 (2010) 116.
[11] S. Fragueiro, I. Lavilla, C. Bendicho, Spectroc. Acta Pt. B-Atom. Spectr. 59 (2004)
851855.
[12] S. Gil, S. Fragueiro, I. Lavilla, C. Bendicho, Spectroc. Acta Pt. B-Atom. Spectr. 60
(2005) 145150.
[13] S. Fragueiro, I. Lavilla, C. Bendicho, Talanta 68 (2006) 10961101.
[14] S. Gil, M.T.C. de Loos-Vollebregt, C. Bendicho, Spectroc. Acta Pt. B-Atom. Spectr.
64 (2009) 208214.
[15] F. Pena-Pereira, I. Lavilla, C. Bendicho, Microchim. Acta 164 (2009) 7783.
[16] J. Zhang, T. Su, H.K. Lee, Anal. Chem. 77 (2005) 19881992.
[17] S. Jermak, B. Pranaityte, A. Padarauskas, J. Chromatogr. A 1148 (2007) 123127.
[18] Y. He, A. Vargas, Y.-J. Kang, Anal. Chim. Acta 589 (2007) 225230.
[19] I. Lavilla, F. Pena-Pereira, S. Gil, M. Costas, C. Bendicho, Anal. Chim. Acta 647
(2009) 112116.
[20] S. Senra-Ferreiro, F. Pena-Pereira, I. Lavilla, C. Bendicho, Anal. Chim. Acta 668
(2010) 195200.
[21] M. Senz, J. Alvarado, F. Pena-Pereira, S. Senra-Ferreiro, I. Lavilla, C. Bendicho,
Anal. Chim. Acta 687 (2011) 5055.
[22] N. Cabaleiro, I. de la Calle, C. Bendicho, I. Lavilla, Anal. Chim. Acta 719 (2012)
6167.
[23] A. Jain, K.K. Verma, Anal. Chim. Acta 706 (2011) 3765.
[24] H. Sund, H. Theorel, in: P.D. Boyer (Ed.), The Enzymes, 2nd ed., Academic Press,
New York, 1963, pp. 5783.
[25] K. Bendiskas, C. DiJiacomo, A. Krill, E. Vitz, J. Chem. Educ. 82 (2005) 10681070.
[26] Ethanol assay Genzyme Diagnostics. http://www.sekisuidiagnostics.
com/resources/PI/Ethanol.pdf, 2010 (accessed February 2012).
[27] Ethanol assay http://www.clonagen.com/clonagen/a244e16f-9f80-4383-
acd1-0e6a46ec4436/quantichrom ethanol assay kit product.aspx, 2012
(accessed February 2012).
[28] E.A. Cossins, L.C. Kopala, B. Blawacky, A.M. Sonk, Phytochemistry 7 (1968)
11251134.
[29] Sigma Aldrich product Information. http://www.sigmaaldrich.com/etc./
medialib/docs/Sigma/Product Information Sheet/a3263pis.Par.0001.File.tmp/
a3263pis.pdf, 2012 (accessed February 2012).
[30] Y.-S. Li, X.-F. Gao, Anal. Chim. Acta 588 (2007) 140146.
[31] J.P. Zanon, M.F.S. Peres, E.A.L. Gatts, Enzyme Microb. Technol. 40 (2007)
466470.
[32] A. Tankeviciute, R. Kazlauskas, V. Vickackaite, Analyst 126 (2001) 16741677.
[33] M. Saraji, J. Chromatogr. A 1062 (2005) 1521.
[34] Y. Yu, B. Chen, C. Shen, Y. Cai, M. Xie, W. Zhou, Y. Chen, Y. Li, G. Duan, J. Chro-
matogr. A 1217 (2010) 51585164.
[35] S. Legrand, J. Dugay, J. Vial, J. Chromatogr. A 999 (2003) 195201.
[36] S. Carda-Broch, M.J. Ruiz-Angel, D.W. Armstrong, A. Berthod, Sep. Sci. Technol.
45 (2010) 23222328.
[37] A.K. Williams, J.T. Hupp, J. Am. Chem. Soc. 120 (1998) 43664371.
[38] A.O.S.S. Rangel, I.V. Tth, Anal. Chim. Acta 416 (2000) 205210.
[39] S.S.M.P. Vidigal, I.V. Tth, A.O.S.S. Rangel, Talanta 77 (2008) 494499.
[40] M. Hmmerle, K. Hilgert, M.A. Horn, R. Moos, Sens. Actuator B-Chem. 158 (2011)
313318.
[41] K. Guo, K. Qian, S. Zhang, J. Kong, C. Yu, B. Liu, Talanta 85 (2011)
11741179.
[42] L.H. Keith, L.U. Gron, J.L. Young, Chem. Rev. 107 (2007) 26952708.