Anal Bioanal Chem (2003) 377 : 316–321
DOI 10.1007/s00216-003-2051-6
O R I G I N A L PA P E R
R. M. García-Rey · R. Quiles-Zafra ·
M. D. Luque de Castro
New methods for acceleration of meat sample preparation prior
to determination of the metal content by atomic absorption spectrometry
Received: 26 March 2003 / Revised: 29 April 2003 / Accepted: 9 May 2003 / Published online: 19 July 2003
© Springer-Verlag 2003
Abstract Focused microwave-assisted digestion and ul-
trasound leaching have been applied for the extraction of
Pb, Cd, Cr, Cu, Fe, Zn, Ca, and Mg from raw meat. Semi-
membranous muscle (SM) of raw pig ham was used for
optimizing both the digestion and extraction steps by mul-
tivariate approaches. The detection and quantification lim-
its were 0.5 and 0.9 µg kg–1 for Pb, 0.06 and 0.1 µg kg–1
for Cd, 0.2 and 1.2 µg kg–1 for Cr, 0.4 and 3 µg kg–1 for
Cu, 0.04 and 0.1 mg kg–1 for Fe, 0.012 and 0.017 mg kg–1
for Zn, 0.3 and 0.4 mg kg–1 for Ca, and 0.01 and 0.03 mg kg–1
for Mg. The precision, expressed as relative standard de-
viation (RSD), ranged between 2.5 and 9.6% for focused
microwave-assisted digestion and between 3.5 and 10.6%
for ultrasound leaching. The methods were then compared
with a reference method and applied to a certified refer-
ence material (bovine muscle 184, from the BCR). The
t-test, applied to the results obtained from focused micro-
wave-assisted digestion, revealed that they are in agree-
ment (p>0.01) with the certified and estimated values in
the case of Pb, Cd, Cr, Cu, Fe, Ca, Mg, and Zn but not in
that of Fe. In the case of ultrasound leaching, only the ex-
traction of Pb, Cu, and Ca was quantitative. The method
based on microwave digestion provides more accurate
and precise results than ultrasound leaching. These new
procedures have many advantages with regards to con-
ventional methods, namely, reduction of the extraction
time, simplification of the process, avoidance of chemical
emissions to the atmosphere, and no losses of metals by
volatilization.
Keywords Meat analysis · Pork · Sample preparation ·
Focused microwave-assisted digestion · Ultrasound
R. M. García-Rey · M. D. L. de Castro (✉)
Department of Analytical Chemistry, Marie Curie Building,
Campus de Rabanales, University of Córdoba,
14071 Córdoba, Spain
e-mail: QA1LUCAM@uco.es
R. Quiles-Zafra
Department of Environmental Protection and Waste Management,
Junta de Comunidades de Castilla La Mancha, Toledo, Spain
leaching · Metal determination · Atomic absorption
spectrometry · Multivariate optimisation
Introduction
The determination of trace elements in food is of great rel-
evance because the minerals are essential in vital pro-
cesses in humans [1]. In fact, knowledge of the relation-
ship between the mineral content of the diet and some dis-
eases like hypertension, osteoporosis, etc., has increased
interest in the mineral content of food [2, 3, 4, 5, 6]. Meat
and meat products constitute an important part of the hu-
man diet [7] and their metal content influences the quality
of the final product and thus acceptance by consumers.
An aspect to be taken into account by manufacturers of
meat products (particularly dry cured-ham producers) is the
relationship between the content of some metals (namely,
Ca, Mg, Zn, Fe, etc.) and the appearance of quality de-
fects such as pasty textures or defective colour. Further-
more, one of the most important consequences of envi-
ronmental pollution on humans is the intake of toxic ele-
ments through the diet. Metals are essential in small
amounts and toxic at high levels (Fe, Cu, Zn, etc.) and the
intake of food containing metals such as Pb and Cd has
proved cumulative character [8, 9, 10]. Special pro-
grammes have been carried out in many countries with the
purpose of avoiding the distribution of foodstuffs that
could be a risk to human health if consumed [11].
Most methods proposed to determine the metal content
in meat products use atomic absorption spectrometry. De-
struction or removal of the organic matter is mandatory
prior to determination of trace elements. The decomposi-
tion or mineralization of the sample can be carried out by
different methods, namely: calcination [12], wet digestion
with acid mixtures [13] in either open vessels [14] or closed
systems [15], etc. These methods involve some shortcom-
ings, such as long manipulation time [12, 16, 17] large con-
sumption of chemicals, losses of elements by volatilization
and fume emissions. The use of a domestic microwave
oven with containers of either stainless steel or PTFE

(polytetrafluoroethylene) equipped with a sealing pres-
sure-resistant system [18, 19] provides short lixiviation
times and reduces emission of gases, but involves serious
safety problems [15, 20]. With a view of overcoming these
shortcomings, procedures based on either focused micro-
wave-assisted digestion or ultrasound-assisted leaching
have been proposed [21]. The use of focused-microwave
digestors and ultrasound probes endows with key features
the methods for solid sample preparation. Firstly, reduc-
tion of the time required for the digestion or the extraction
step; secondly, simplification of the process; and, thirdly,
avoidance of chemicals emissions to the atmosphere [22,
23]. This last aspect fulfils the present policy for reducing
environmental contamination. Furthermore, ultrasound
leaching diminishes matrix effects because the analyte is
removed from non-extractable and potentially interfering
matrix components [24].
The aim of this research was to check the potential of
two alternative energies, namely focused microwaves [25]
and ultrasound [26, 27] for pig meat treatment prior to
monitoring the metal content.
Experimental
Apparatus and instruments
A Soxwave-100 focused microwave digester (Prolabo, Fontenay-
sous-Bois, France) with a maximum irradiation power of 300 W,
controlled by a TX 32 device (Prolabo) was used.
Ultrasonic irradiation was applied by means of a Branson 450
sonifier (20 kHz, 400 W) equipped with a cylindrical titanium al-
loy probe (12.7 mm diameter), which was immersed in a water
bath in which the extraction cell was placed. An extraction cham-
ber consisting of a stainless steel cylinder (10 cm in length and
3 mm i.d.), closed with screws at either end, was used.
A Hobersal model HD-150 (maximum temperature 1200 °C)
furnace (Forns Hobersal, Barcelona, Spain) was used to carry out
the reference method.
An atomic absorption spectrometer with deuterium background
correction and an SPS-5 autosampler from Varian (Spectraa 110,
USA) was used to determine the metal content. The analyses of
Pb, Cd, Cr and Cu were carried out by graphite furnace atomic ab-
sorption spectrophotometry (GFAAS). Pyrolytic platforms were
used for Pb and Cd measurements, which were unnecessary in the
case of Cr and Cu. Zn, Fe, Ca and Mg were determined by flame
atomic absorption spectrophotometry (FAAS) with a 10 cm burner
head and standard air–acetylene flame. Varian hollow cathode
lamps were used as radiation sources in all cases.
Table 1 Values of the instrumental parameters for the determination of Pb, Cd, Cr, and Cu by GFAAS
Parameter Pb Cd
λ (nm) 217 228.8
A (mA) 5 4
SW (nm) 1 0.5
Background correction Yes Yes
T dry stage (°C) 95 110 140 95 110
Time (s) 5 80 5 5 50
T pyrolysis stage (°C) 700 650
Time (s) 27.5 30
T atomisation stage (°C) 2150 1800
Time (s) 3 3.5
317
Reagents and standards
High purity de-ionized water (resistivity 18.2 mOhm cm) obtained
from a Milli-Q water purification system (Millipore, Bedford, MA,
USA) was used throughout. All glassware and containers used
were cleaned in 10% HNO3 for a minimum of 24 h and rinsed with
ultrapure water before use. Analytical reagent grade 65% HNO3
(Panreac, Barcelona, Spain) was used. Stock solutions (1 g L–1
each metal) were prepared using 99.5% granular Pb (Aldrich, Mil-
waukee, WI, USA), Cd powder (Merck, Darmstadt, Germany),
K2Cr2O7 for analysis (Panreac), CuSO4 for analysis (Panreac),
granular Zn (Aldrich), granular Fe (Aldrich), CaCO3 for analysis
(Panreac) and granular Mg (Aldrich).
The mixture of 1.8 mL 0.2% NH4(H2PO4)+200 µL 1% Mg(NO3)2
was used as a chemical modifier for Pb and Cd determinations by
GFAAS. Sr(NO3)2 for analysis at a final concentration of 2 g L–1
was used as releasing agent in Ca and Mg determinations.
Argon, (Carburos Metálicos, Barcelona, Spain) was used as a
protective gas in GFAAS, and air and acetylene (Carburos Metáli-
cos) as oxidizing and fuel gas, respectively, in FAAS.
Samples
Semimembranosus muscle (SM) – raw pig ham – was used for de-
velopment of the methods. Three samples of SM from different
pigs were then used for comparison with the reference method.
Bovine muscle, a CRM (BCR-184), was used for validating the
methods.
Sample preparation
Focused microwave-assisted digestion
SM sample (1 g) was introduced into the extraction vessel together
with 10 mL HNO3 (50% v/v). The vessel was connected to a distil-
lation column and then irradiated with a microwave power of 300 W
for 3 min. The suspension thus obtained was filtered through Albet
no. 135 paper and diluted to 25 mL in a volumetric flask with de-
ionized water. Potential losses by filtering were checked by the
standard addition method and no losses were detected.
Ultrasound leaching
SM sample (1 g) was introduced into the extraction chamber to-
gether with 10 mL HNO3 (15% v/v) and then immersed into a wa-
ter bath at room temperature. The solid sample was subject to ul-
trasound irradiation for 10 min (duty cycle 0.1 s, output amplitude
70% of the converter nominal amplitude, with the probe placed
1 mm from the top surface of the extraction cell). The extract was
filtered through Albet no. 135 paper and diluted to 25 mL in a vol-
umetric flask with deionized water.
Cr Cu
357.9 327.4
7 4
0.2 0.5
No No
135 85 95 120 85 95 120
10 5 30 10 5 40 10
1000 800
17 8
2600 2300
3.2 3
318
Table 2 Values of the instrumental parameters for the determina-
tion of Zn, Fe, Mg, and Ca by FAAS
Parameter Zn Fe Mg Ca
λ (nm) 213.9 248.3 285.2 422.7
A (mA) 5 5 4 10
SW (nm) 0.1 0.1 0.1 0.1
Background correction Yes Yes Yes No
Reference method
SM sample (2 g) was ashed in a furnace at 450 °C for 12 h, using
the AOAC method [28]. The ash was dissolved in 2.5 mL concen-
trated nitric acid and diluted to 50 mL with deionized water.
Determination procedure
After extraction Pb, Cd, Cr, and Cu were determined by GFAAS
(instrumental conditions in Table 1). A final stage at higher tem-
perature was applied in all cases in order to remove all extract
residues from the tube, thus avoiding possible memory effects be-
tween both samples and replicates. Zn, Fe, Mg, and Ca were deter-
mined by FAAS (instrumental conditions in Table 2). All measure-
ments were the average of at least three replicates.
Results and discussion
After checking the determination procedure for the target
analytes (linear range of the calibration curve, precision,
etc.) the research was focused on establishing a method
for removal of the metals from meat which was faster,
cleaner, and required less consumption of reagents than
those presently used. Two types of energy were used to
achieve a drastic reduction of both the time and analyte
losses common in long sample manipulation.
Optimisation of focused microwave-assisted digestion
A preliminary study involving different concentrations and
volumes of HNO3, and different sample amounts and di-
gestion times was performed. Nitric acid was selected as
extractant because it is the most frequently used acid for
mineralization by a microwave oven [20]. This acid be-
haves ideally under microwave energy irradiation, has a
relatively low boiling point, can be obtained in high pu-
rity, and lacks of the hazards and inconveniences of other
acids such as perchloric acid. The results showed that
concentrations of HNO3 higher than 55% must not be
used because fume production. The extractant volume is
not significant, so the minimum value tested (10 mL) was
used. One gram of sample gave the best results, requiring
a digestion time of 3 min.
Then, a multivariate approach was designed for opti-
misation of variables such as HNO3 concentration, diges-
tion time, and H2O2. The addition of H2O2 was assayed
because of its positive effect on the digestion of meat
products [29, 30]. A full factorial design 23 involving eight
randomized runs plus three centred points was built for a
Table 3 Tested ranges in (a) the first and (b) the second factorial
design and optimal values for the variables related to the extraction
step assisted by focused microwaves
Variable Tested range Optimum
value
(a) First (b) Second
factorial factorial
design design
HNO3 concn (%) 20–40 30–50 50
Digestion time (min) 5–15 3–9 3
H2O2 (%) 0–10 0 0
screening study of the behaviour of the main factors af-
fecting the extraction process. The upper and lower values
given to each factor were selected from the available data
and experience gathered in the preliminary experiments.
The ranges over which the variables were studied and the
optimum values found are listed in Table 3a.
The conclusions were that both the HNO3 concentra-
tion and digestion times were significant and positive but
their interaction was antagonist. The addition of H2O2 was
not significant.
In view of these results, a full two level factorial de-
sign was performed for HNO3 concentration and digestion
time only. This second design involved a total of 22=4 ex-
periments plus three centred points. Table 3b summarizes
the values used for the design.
The study showed a positive and significant effect for
HNO3 concentration and a positive but non-significant ef-
fect for digestion time. Thus, the highest HNO3 concen-
tration without fume production (50%) and the shortest
digestion time providing full digestion (3 min) were the
optimum values.
Optimisation of ultrasound leaching
A preliminary study of the HNO3 concentration and ex-
traction time was performed. HNO3 concentrations higher
than 20% and extraction times longer than 20 min did not
improve the extraction yield. One-gram portions of sam-
ple were used in all instances with the view of subsequent
comparison of this extraction method with digestion based
on microwave assistance.
The variables which affect ultrasound-assisted extrac-
tion are probe position, ultrasound radiation amplitude,
percentage of duty cycle of ultrasound exposure, sonica-
tion time, acid concentration, and extractant volume. The
probe position was defined as the distance between the tip
horn of the ultrasound probe and the top surface of the ex-
traction cell.
A half fractionated factorial design 26–1 type V resolu-
tion allowing 3 d.f. and involving 32 randomized runs
plus three centred points was built for a screening study of
the behaviour of the main factors affecting the extraction
process. The ranges for HNO3 concentration and extrac-
tion time were selected from the preliminary experiments.
The values used for the rest of the variables covered the
 319
Table 4 Tested ranges in (a) the first and (b) the second factorial ative but non-significant; the interaction with the volume
design and optimal values for the variables related to the extraction extractant was negative and significant. In view of these
step using ultrasound
results, the highest extractant volume (10 mL) and the
Variable Tested range Optimum shortest both sonication time (10 min) and duty cycle
value (10%) were selected and used for further experiments.
(a) First (b) Second
factorial factorial
design design
Features of the methods
Probe position (mm) 1–20 1 1
Radiation amplitude(%) 10–70 70 70 The detection and quantification limits were calculated by
Duty cycle (%) 30–90 10–30 10 use of the equations LD=µbl±3sbl and LC=µbl±10sbl, re-
Sonication time (min) 1–10 10–20 10 spectively, where µbl is the mean and sbl the standard de-
HNO3 concentration (%) 5–15 15 15 viation of the blank responses obtained from the analyses
Extractant volume (mL) 2–7 5–10 10 of 10 blanks. The values obtained for the detection and
quantification limits were 0.5 and 0.9 µg kg–1 for Pb, 0.06
and 0.1 µg kg–1 for Cd, 0.2 and 1.2 µg kg–1 for Cr, 0.4 and
whole range provided by the ultrasound system (Table 4a). 3 µg kg–1 for Cu, 0.04 and 0.1 mg kg–1 for Fe, 0.012 and
The conclusions of this screening study were that the 0.017 mg kg–1 for Zn, 0.3 and 0.4 mg kg–1 for Ca, and 0.01
ultrasound radiation amplitude and the acid concentration and 0.03 mg kg–1 for Mg.
had positive effects but they were not statistically influen- The precision, expressed as relative standard deviation
tial factors. Thus, the highest values tested for these vari- (RSD) of five independent analyses of the same raw meat
ables (namely, 70 and 15%, respectively) were selected for sample (SM), was studied for the two proposed methods
subsequent experiments. The probe distance had a nega- and the reference method (Table 5). For the digested sam-
tive but non-significant effect; so the minimum value tested ple obtained with microwave-assistance the precision for
(1 mm) was selected for subsequent experiments. The ex- the metals determined by FAAS ranged between 3.4 and
traction time, the volume of extractant, and the duty cycle 5.7% and that of the metals determined by GFAAS ranged
were significant factors in the ranges under study. The ex- between 2.5 and 9.6%. For the extracts obtained with ultra-
traction time and the volume of extractant had a positive sounds-assistance, the precision for the metals determined
influence; meanwhile the effect of the duty cycle was neg- by FAAS ranged between 3.5 and 8.6% and that of the met-
ative. Therefore, a new design was performed in order to als determined by GFAAS ranged between 5.4 and 10.6%.
evaluate the effect of using higher values of both extrac-
tion time and extractant volume and lower values of the
duty cycle. Validation of the methods
A full two-levels factorial design involving a total of
eleven experiments (23=8 experiments plus three centred To check the validity of the proposed methods for the de-
points) was selected as a second experimental design. The termination of Pb, Cd, Cr, Cu, Fe, Zn, Ca, and Mg in raw
ranges within which the variables were studied are shown pig meat, they were compared with a reference method.
in Table 4b. The analysis of the design showed that ex- The results obtained by applying them to three samples
traction time and extractant volume were significant and are shown in Table 6. Each result is the mean of three repli-
positive and their interaction was significant but antago- cates. As can be seen, the focused microwave-assisted di-
nist. The duty cycle had a negative but non-significant ef- gestion provides more accurate and precise results than
fect, and its interaction with the extraction time was neg- ultrasound leaching for the eight metals.

Table 5 Repeatability as rela-

tive standard deviation (%RSD) Metal Method Mean %RSD Metal Method Mean %RSD
of the reference (RM), micro- (n=5a) (n=5a)
wave (MWM) and ultrasound Zn RM 20.5 1.7 Pb RM 0.14 9.5
(USM) methods for each metal
MWM 24.4 4.3 MWM 0.14 9.6
USM 15.5 7.6 USM 0.21 5.4
Fe RM 25.9 5.1 Cd RM 0.039 8.8
MWM 26 3.4 MWM 0.037 5.6
USM 16.7 8.6 USM 0.042 10.6
Mg RM 206. 1.9 Cr RM 0.37 6.6
MWM 227.9 4.6 MWM 0.36 6.3
USM 126.8 6.3 USM 0.57 10.4

aFive
Ca RM 99.2 2.4 Cu RM 2.15 6.4
independent digestions/
MWM 106.9 5.7 MWM 2.17 2.5
extractions of the same raw
pork meat sample (SM) USM 81.5 3.5 USM 2.24 8.7
320
Table 6 Metal contents of three samples of raw pork meat (SM) using the reference (RM), microwave (MWM) and ultrasound (USM)
methods (the mean and standard deviation are from three independent digestions/extractions of each sample)
Sample 1 2 3
Metal
RM MWM USM RM MWM USM RM MWM USM
Pb 0.13±0.01 0.140±0.001 0.17±0.01 0.18±0.02 0.15±0.01 0.13±0.009 0.094±0.01 0.093±0.002 0.085±0.004
Cd 0.015±0.002 0.015±0.002 0.022±0.003 0.016±0.001 0.015±0.001 0.019±0.002 0.0096±0.0007 0.010±0.002 0.013±0.002
Cr 0.49±0.03 0.55±0.06 9.1±1.9 0.52±0.01 0.54±0.04 2.3±0.2 0.56±0.02 0.52±0.01 1.7±0.2
Cu 1.9±0.1 2±0.1 1.8±0.4 2.2±0.7 2.3±0.1 1.9±0.5 2.5±0.6 2.54±0.01 2.1±0.3
Zn 22.9±3 25.9±1 19.4±2.7 22.8±1.4 24.9±2 24.7±2.4 21.9±1.7 24.9±2.6 26.3±3.6
Fe 24.2±0.4 27.5±0.4 16.6±1.2 27.5±0.4 26.8±0.8 12.5±0.5 31.8±0.5 30.7±1.3 18.9±1.4
Ca 170±2 171±2 166±11 188±4 187±4 170±1 160±1 162±5 150±6
Mg 223±1 213±0.3 219±3 215±0.8 212±1 189±5 225±2 224±3 199±10

Table 7 Validation of the methods (values obtained as the average of three independent digestions/extractions)
CRM 184 Certified value Estimated Found value Efficiency Found value Efficiency
value (MW) (%) (US) (%)
Pb (mg kg–1) 0.239±0.011 – 0.253±0.036a 106 0.251±0.017a 105
Cd (mg kg–1) 0.013±0.002 – 0.015±0.001a 112 0.016±0.001a 121
Cr (mg kg–1) – 0.076 0.076±0.003a 99 – –
Cu (mg kg–1) 2.36±0.06 – 2.55±0.3a 108 2.22±0.09a 94
Fe (mg kg–1) 79±2 – 92±2.6a 116 64±2a 80
Zn (mg kg–1) 166±3 – 150±4a 91 101±3a 61
Ca (mg kg–1) – 150 156±9a 104 153±4a 102
Mg (mg kg–1) – 1020 1008±16a 99 652±10a 64
aStandard deviation obtained for n=3

The optimized conditions for both methods were ap-
plied to 1 g of a CRM (bovine muscle, 184 from the BCR).
The extraction efficiencies for the eight metals are listed
in Table 7. The t-test was applied to the results obtained
with focused microwave-assisted digestion and it revealed
that they are in agreement (p>0.01) with the certified and
estimated values in all cases, except for Fe. The results
obtained with ultrasound leaching are in agreement (p>
0.05) with the certified and estimated values in the case of
Pb, Cd, Cu, and Ca, but not for Fe, Zn, and Mg, for which
the extraction efficiency ranged between 61 and 80%.
Contamination of the extract by metals from the sample
chamber was checked by blank extractions. The chamber
causes contamination only from Cr.
Conclusions
Focused microwave-assistance allows fast, easy, and
clean digestion of the raw pig ham. The results obtained
with the proposed method and with dry digestion were
similar, thus allowing the substitution of tedious and/or
error-prone handling methods for extracting metals from
meat. There are no differences between the results ob-
tained with the proposed method and the reference method
for fresh meat samples for all the metals under study, non-
significant differences between the content found and the
certified value of the lyophilized CRM 184 exist for all
the metals except Fe. The extraction efficiency ranges
from 91 to 116% for the eight metals. These reasons make
focused microwave-assisted digestion of raw meat an ex-
cellent method for sample preparation prior to atomic ab-
sorption monitoring of these metals.
The slightly different behaviour of the CRM as com-
pared with the raw meat samples could be due to the dif-
ferent texture and humidity of the natural samples and the
lyophilized CRM.
Ultrasound-assisted leaching provides acceptable re-
sults for the extraction of Pb, Cd, Cu, and Ca – extraction
efficiencies of 105, 94, and 102%, respectively – but not
in the case of Cr, Cd, Fe, Zn, and Mg. This method re-
duces the extraction time, avoids chemicals emission and
losses of metals and diminishes matrix interference, so it
is useful for extraction of Pb, Cu and Ca from raw meat.
Acknowledgements Financial support from Spanish Comisión
Interministerial de Ciencia y Tecnología (CICyT) is gratefully ac-
knowledged (Proyect BQU2002–01333). The R&D and Quality
Department of Campofrío Alimentación S.A., especially Juan
Ángel García, Julio Tapiador, Roberto Rodríguez, and Jesús Ro-
dríguez, are thanked for supplying the samples and CRM.
References
1. Forero-Vizcaino FJ, Venegas-Taracena M, Mataix-Verdú J,
Planells del Pozo E, Céspedes Sánchez F, Moreno Rojas R,
León Crespo F (2001) I Ham Mundial Congress, Córdoba,
pp 395–404
2. Alcaide-Castiñeira E, Gómez R, Carmona-González MA, Fer-
nández-Salguero J (1995) Alimentaria, May, 63–67

3. Davis GK, Mertz W (1987) In: Mertz W (ed) Trace elementes
in human and animal nutrition, 5th edn, vol 2. Academic Press,
Orlando, FL
4. Mataix J (1998) Tablas de composición de alimentos es-
pañoles, 3rd edn. Universidad de Granada
5. National Research Council (1989) Diet and health. Implica-
tions for reducing chronic disease risk. National Academy
Press, Washington, DC
6. Committee on Dietary Allowance Food and Nutrition Board
(1989) Recommended dietary allowances (RDA), 10th edn,
National Academy of Science, Washington, DC
7. Jiménez-Colmenero F, Carballo J, Cofrades S (2001) Meat Sci
59:5–13
8. Navas MJ, Herrador AM, Jiménez AM, Nuevo AG (1984) Al-
imentaria, May, 61–66
9. Blanco Jiménez MC, López de Sa Fernández A, Cirugeda Del-
gado ME (1991) Alimentaria, June, 25–31
10. Sola Azcoiti S, Martín Pérez A (1996) Alimentaria, June, 31–
37
11. López Alonso M, Benedito JL, Miranda M, Castillo C, Hernán-
dez J, Shore RF (2000) Food Addit Contam 17:447–457
12. Castro-Ferreira M, Morgano MA, Do Nascimento de Queiroz
SC, Bassi Mantovani DM (2000) Food Chem 69:259–265
13. Demirbas A (1999) Food Chem 67:27–31
14. Catalá R, Montoro R, Ibañez N (1983) Rev Agroqui Tecnol Al-
iment 23:202–216
15. Yang Q, Penninckx W, Smeyers-Verbeke J (1994) J Agric
Food Chem 42:1948–1953
16. Ybáñez N, Cervera ML, Montoro R, Catalá R (1987) Rev
Agroquim Tecnol Aliment 27:590–598
321
17. López-Alonso M, Miranda M, Castillo C, Hernández J, Bened-
ito JL (2000) Eurocarne 86:71–85
18. Lizondo F, Vidal MT, De La Guardia M (1991) Analysis
19:136–138
19. Oles PJ, Wande MG (1991) J Assoc Off Anal Chem 74:812–
814
20. De la Fuente MA, Guerrero G, Juárez M (1995) J Agric Food
Chem 43:2406–2410
21. Nascentes CC, Korn M, Arruda MAZ (2001) Microchem J
69:37–43
22. Luque-García JL, Morales-Muñoz S, Luque de Castro MD
(2002) Chromatographia 55:117–122
23. Ruiz-Jiménez J, Luque-García JL, Luque de Castro MD (2003)
Anal Chim Acta 480:231–237
24. Méndez H, Alava F, Lavilla I, Bendicho C (2002) Anal Chim
Acta 452:217–222
25. Herrera MC, Luque de Castro MD (2002) J Anal At Spectrom
17:1530–1533
26. Luque-García JL, Luque de Castro MD (2003) Trends Anal
Chem 22:41–47
27. Al-Merey R, Al-Masri MS, Bozou R (2002) Anal Chim Acta
452:143–148
28. AOAC (1980) Official Methods of Analysis, 13th edn. Associ-
ation of Official Analytical Chemists, Washington, DC
29. Wolinsky I, Lane HW, Warren DC, Whaley B (1988) J Agric
Food Chem 36:749–752
30. Montoro R, Ybañez N, Bueno A (1983) Rev Agroquim Tecnol
Aliment 23:510–520