J. Sep. Sci. 2006, 29, 1607 – 1612 Y. Zu et al.

1607

Yuangang Zu Original Paper

Chunjian Zhao
Chunying Li
Lin Zhang A rapid and sensitive LC-MS/MS method for
Key Laboratory of Forest Plant
determination of coenzyme Q10 in tobacco
Ecology, Ministry of Education, (Nicotiana tabacum L.) leaves
Northeast Forestry University,
Harbin, PR China
A rapid and sensitive liquid chromatography-tandem mass spectrometry method
with multiple reaction monitoring has been proposed for the analysis of coenzyme
Q10 in (CoQ10) tobacco leaves. The method used electrospray ionization with detec-
tion in positive ion mode. Sample pretreatment involved ultrasonic extraction of
fresh tobacco leaves with anhydrous ethanol for 15 min and followed by extraction
of the supernatant with hexane. The separation of CoQ10 was performed on a Sym-
metry Shield RP18 column with a mixture of acetonitrile and isopropanol (8 : 7, v/v)
containing 0.5% formic acid as mobile phase. Quantification of CoQ10 was performed
by the standard addition method. The limit of detection and limit of quantitation
of CoQ10 were, respectively, 1.2 ng/mL (S/N = 3) and 4.0 ng/mL (S/N = 10). The relative
standard deviations of peak area were 0.91% and 1.21% for intra-day and inter-day,
respectively. The recoveries of CoQ10 ranged from 98.2 to 99.3% and the correspon-
ding RSDs were less than 2.4%. Analysis took 5 min, making the method suitable for
rapid determination of CoQ10 in tobacco leaves. The proposed method has been suc-
cessfully applied to the analysis of CoQ10 in the leaves from eight varieties of
tobacco.
Keywords: Coenzyme Q10 / LC-MS/MS / MRM / Tobacco leaves / Ultrasonic extraction /
Received: January 22, 2006; revised: March 30, 2006; accepted: March 31, 2006
DOI 10.1002/jssc.200600047

1 Introduction tobacco leaf extracts [9,10]. Since CoQ10 can be used as a

Coenzyme Q10 (CoQ10), also known as ubiquinone 50, is
widely present in many organisms. Tobacco (Nicotiana
tabacum L.) belongs to the Solanaceae family; the leaves of
the plant are considered to be a good source of a large
number of bioactive substances. CoQ10 is an important
bioactive ingredient which occurs naturally in tobacco
[1]. It has potent antioxidant and cell membrane stabiliz-
ing effects and has also been used as a protective agent in
acute cardiac ischemic syndromes, breast cancer, hepatic
ischemia reperfusion injury, diabetes, and atherosclero-
sis [2 – 8]. CoQ10 is currently sold as a dietary supplement
in the United States, with a majority of these products
being derived from the fermentation of carbohydrates or
Correspondence: Professor Yuangang Zu, Box 332, Northeast
Forestry University, Harbin, 150040, PR China.
E-mail: zygorl@vip.hl.cn
Fax: +86-451-82102082
Abbreviations: APCI, atmospheric pressure chemical ioniza-
tion; CE, collision energy; CEM, channel electron multiplier;
CoQ10, coenzyme Q10; CXP, collision cell exit potential; DP, de-
clustering potential; ECD, electrochemical detection; EP, en-
trance potential; FP, focusing potential; IS, ionspray voltage;
MRM, multiple reaction monitoring
food supplement or as an adjunctive therapy in many
kinds of diseases, it is very important to be able to deter-
mine the content of CoQ10.
Many analytical methods including spectrophotometry,
voltammetry, and liquid chromatography coupled, for
example, with ultraviolet or fluorescence detectors were
formerly used for the quantitative determination of
CoQ10 [11 – 14]. In the above methods, the LOD was typi-
cally 0.247 lg/mL, 10 – 7 mol/L, 0.05 lg/mL, and 29 nmol/
L, respectively. At present, LC-MS is being adopted by
more and more people as a useful method for identifica-
tion and determination of compounds. In particular, it is
very effective in the analysis of CoQ10 in complex samples
because of its low detection limit and high sensitivity.
Teshima et al. established a sensitive and simple quantifi-
cation method for ubiquinone-9 (CoQ9) and ubiquinone-
10 (CoQ10) in rat thigh muscle and heart using LC-MS/MS
[15]. Hansen et al. developed two new sensitive and selec-
tive LC-MS methods for the quantification of the total
coenzyme Q10 concentration in human blood serum and
both methods were shown to be more selective and to
have comparable or better sensitivity than the HPLC-ECD
method [16]. Strazisar et al. developed LC and LC-MS meth-

i 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com

1608 Y. Zu et al.
ods for the quantitative determination of coenzyme Q10
in dairy products. The results showed that MS detection
is more sensitive than UV detection [17].
It should be borne in mind that signal suppression or
enhancement of the target extracts by matrix compo-
nents is a common phenomenon in LC-MS/MS analysis.
Moreover, interfering matrix components can affect the
accuracy of the proposed method and may lead to some
compromising or erroneous results [18, 19]. In order to
avoid problems related to matrix effects, some authors
have undertaken optimization of sample preparation
and of the chromatographic system and MS/MS detection
[20 – 22]. In addition, some workers have adopted the
standard addition method as a way of eliminating
matrix effects [23, 24].
The aim of this study is to propose a validated LC-MS/MS
method with multiple reaction monitoring (MRM) for
separation and determination of CoQ10 in tobacco leaves.
So far, no report has been published on the determina-
tion of CoQ10 in tobacco leaves by the LC-MS/MS method.
Because the matrix components in tobacco leaves are
complex, in order to avoid problems arising from the
matrix effect each step mentioned above (sample pre-
paration, chromatographic system, and MS/MS detec-
tion) was carefully optimized in the present study. At the
same time, the standard addition method was used as a
quantitation method to further minimize the matrix
effect. Based on this work, the contents of CoQ10 in the
leaves of eight varieties of tobacco were determined and
compared.
2 Experimental
2.1 Equipment
LC-MS/MS analysis was performed on an API3000
(Applied Biosystems, Canada) triple-stage quadrupole
mass spectrometer equipped with an electrospray ioniza-
tion (ESI) interface and an Agilent 1100 series HPLC from
, filter (Q3). Finally, all MS parameters were manually fine-
Agilent technologies (Agilent, CA). A Model 11’ single
syringe pump (Harvard Apparatus Inc., Holliston, USA)
was also used.
The Agilent HPLC system consisted of a G1312A HPLC bin-
ary pump, a 7725i manual injector, and a G1379A degas-
ser. A reverse phase HPLC column (Symmetry Shield
RP18, 5 lm, 93
/66% porosity/3.96150 mm; Waters,
USA) was used.
2.2 Reagents and materials
Acetonitrile and isopropanol were of HPLC grade (Kracke-
ler Scientific, Albany, USA). Formic acid was of HPLC
grade (DIMA Technology, USA). Methanol, ethanol, and
hexane were of analytical grade (Beijing Chemical
Reagents Company, China).
i 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Sep. Sci. 2006, 29, 1607 – 1612
CoQ10 standard (98%) was purchased from Sigma (USA). A
stock solution of CoQ10 was prepared by accurately
weighing 5.4 lg of powder and dissolving it in 10 mL of
the mobile phase. The stock solutions were stored in the
dark at – 208C and were stable for at least one year.
Tobacco leaves were collected from the arboretum in
Northeast Forestry University of China and were stored
at 48C. The eight varieties of tobacco are Hong Hua Da Jin
Yuan, Long Jiang 911, Yun Yan 87, Piao He 1, Yun Yan 85,
K346, K326, and NC89, respectively.
2.3 LC-MS/MS conditions
Chromatographic analysis was carried out on a Symme-
try Shield RP18 column (5 lm/93
/66% porosity/
3.96150 mm). The column temperature was maintained
at 258C. The mobile phase was a mixture of acetonitrile
and isopropanol (8 : 7, v/v) containing 0.5% formic acid.
Elution was performed at a flow rate of 1 mL/min with
the split ratio set at 7 : 3 (the real flow rate used was
0.3 mL/min). The injection volume was 5 mL. For opera-
tion in MS/MS mode, a mass spectrometer fitted with an
electrospray ion source interface was used for analysis.
The source was operated at a temperature of 3508C. The
infusion experiment was performed using a Harvard
,
Model 11’ single syringe pump. The mass spectrometer
was operated in the positive ion mode with the following
parameters: declustering potential (DP), 120 V; focusing
potential (FP), 400 V; entrance potential (EP), 10 V; colli-
sion cell exit potential (CXP), 8 V; collision energy (CE),
38 V; ionspray voltage (IS), 4000 V; channel electron mul-
tiplier (CEM), 2000 V. Nitrogen was used for the nebuliz-
ing (setting 12), curtain (setting 10), collision (setting 5),
and auxiliary gas (setting 6 L/min). CoQ10 was monitored
using the MRM mode and MRM was performed with
150 ms dwell time. The mass spectrometer was pro-
grammed to monitor the protonated molecule [M+H]+ at
m/z 864.0 via the first quadrupole filter (Q1), the product
ion at m/z 197.0 was monitored via the third quadrupole
tuned to obtain the highest MRM signals. The MRM tran-
sition m/z 864.0/197.0 was monitored for the detection of
CoQ10. Under the above optimum conditions, the ion
source was thermally stabilized for 30 min before injec-
tion. The peak areas of CoQ10 obtained from the MRM
were utilized for the quantitation. Data were processed
by Analyst Software version 1.4.
2.4 Preparation of standard solutions
Working standards were prepared from the stock solu-
tion (0.54 mg/mL) by dilution with mobile phase to a
final concentration of 8.4/16.9/33.8/67.5/135.0/270.0/
540.0 ng/mL, respectively. All standard solutions were fil-
tered through a 0.45-lm membrane filter (Millipore) and
injected directly.
www.jss-journal.com
J. Sep. Sci. 2006, 29, 1607 – 1612
2.5 Preparation of sample solution
The fresh tobacco leaves were ground to a fine powder in
liquid nitrogen. Extraction was carried out using 2.0 g of
the fine powder with 20 mL of anhydrous ethanol at
258C in an ultrasonic extraction device for 15 min, and
the procedure was repeated three times. The combined
supernatant was extracted with hexane. The hexane
phase was then concentrated to dryness and the dried
extract was dissolved in the chromatographic mobile
phase. After filtering, the supernatant was injected
directly.
2.6 Recovery studies
The recovery experiment of CoQ10 was performed by add-
ing CoQ10 standards to the extract solution of tobacco
leaves which was treated according to the procedure
described in Section 2.5. The extract solution was divided
into four portions (one portion as control group). The
four portions of extract solution were respectively spiked
with the same volume of CoQ10 standard solution at four
different mass concentrations (0, 100, 350, 600 ng/mL).
All samples were Vortex-mixed and filtered through a
0.45-lm Millipore filter. Samples were determined three
times by LC-MS/MS. The recovery was calculated as fol-
lows:
Recovery (%) = (A – B)/C6100% (1)
where A is the amount detected, B is the amount of
extract without added standard, C is the added amount
of the standard.
3 Results and discussion
3.1 Optimization of sample preparation
Fresh tobacco leaves were used as extraction material.
Fresh tobacco leaves contain water, so methanol, 95%
ethanol, and anhydrous ethanol were chosen as extrac-
tion solvent. At the same time, different extraction meth-
ods (ultrasonic, reflux, and Soxhlet) were compared.
Fresh tobacco leaves were extracted using these extrac-
tion methods; the extracts were centrifuged at
12 000 rpm for 8 min. The extraction procedures were
repeated three times; the supernatants were combined,
filtered, and extracted with hexane. A 1000-lL volume of
the hexane phase was concentrated to dryness under a
stream of nitrogen. The dried extracts were dissolved in
the chromatographic mobile phase. After filtering
through a filter paper and a 0.45-lm membrane filter
(Millipore), the extracts obtained, respectively, by the
above three extraction methods were injected. The
extraction yields of CoQ10 in tobacco leaves were deter-
mined under the LC-MS/MS conditions of Section 2.3. The
results are shown in Fig. 1.
i 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
LC-MS/MS determination of coenzyme Q10 in tobacco leaves 1609
Figure 1. Effect of different sample preparation methods
on extraction yields of CoQ10 in tobacco leaves.
As seen in Fig. 1, the extraction yield of CoQ10 obtained
with 95% ethanol is obviously low; that obtained with
methanol is higher and that with anhydrous ethanol is
the highest. Therefore, in this work, anhydrous ethanol
was considered a safe and more effective solvent for
extraction of CoQ10 from tobacco leaves.
It can also be seen from Fig. 1 that the extraction yield of
CoQ10 obtained by ultrasonication is highest; that
obtained by Soxhlet is lower and that by reflux is
obviously lowest. Ultrasonic extraction needs a shorter
time compared to reflux and Soxhlet extraction. With
regard to extraction yields and time, anhydrous ethanol
as extraction solvent and ultrasonic extraction were con-
sidered to provide effective conditions for the extraction
of CoQ10 from tobacco leaves and were consequently used
in the following tests.
3.2 Optimization of chromatographic conditions
It is vital to select an appropriate mobile phase to ensure
that the peak of CoQ10 could be separated and well
resolved within a reasonable analysis time. In order to
assure optimal chromatographic conditions for CoQ10,
the mobile phase systems were optimized. The organic
solvent methanol or acetonitrile is commonly used as
one component of the mobile phase. Since CoQ10 readily
dissolves in isopropanol, various proportions of metha-
nol-isopropanol and acetonitrile-isopropanol were initi-
ally tested as mobile phases. The results showed that the
CoQ10 responses increased with increasing percentage of
methanol or acetonitrile. However, because of acetoni-
trile's lower viscosity as compared to methanol, acetoni-
trile was chosen from the standpoint of pressure as one
component of the mobile phase. In succession, mixtures
of acetonitrile and isopropanol in different ratios were
tested. Eventually, it was found that acetonitrile-isopro-
www.jss-journal.com
1610 Y. Zu et al.
panol (8 : 7, v/v) gave the best separation of CoQ10 offset
only by a slightly off-maximum sensitivity for CoQ10.
In addition, the pH value of the mobile phase has the
greatest effect on the separation, peak shape, and detec-
tion sensitivity of CoQ10. To enhance the separation and
ionization efficiency, different percentages of formic
acid were used in the mobile phase. The results showed
that CoQ10 was completely separated and the peak shape
of CoQ10 was better when the concentration of formic
acid was in the range of 0.1 – 1%. However, when the con-
centration of formic acid was below 0.5%, the ionization
was poor. Considering both resolution and ionization,
the optimized formic acid concentration was 0.5%. As a
result, a mixture of acetonitrile and isopropanol (8 : 7,
v/v) containing 0.5% formic acid was confirmed as the
optimum mobile phase. Under these conditions, the
retention time of CoQ10 was 2.91 l 0.1 min.
3.3 Optimization of MS/MS detection conditions
Electrospray ionization (ESI) and atmospheric pressure
chemical ionization (APCI) techniques were respectively
tested in positive and negative ion mode. The results
showed that ESI in positive mode was superior to ESI in
negative mode and to APCI in positive and negative
mode. CoQ10 was determined with much better sensitiv-
ity on using the ESI source in positive mode. Thus, the
ESI source in positive mode was chosen for the detection
of CoQ10. Infusion experiments were carried out to exam-
ine ionization and fragmentation patterns of the CoQ10
standards using a syringe pump.
First, a standard solution of CoQ10 was chosen to obtain a
constant signal in the Q1 scan mode. A full scan spec-
trum of the CoQ10 standard was acquired with a scan-
range of 700 – 900 amu, a dwell time of 1.5 s, and a step
i 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Sep. Sci. 2006, 29, 1607 – 1612
Figure 2. LC-MS/MS chromato-
gram of CoQ10.
size of 0.1 amu. The DP was optimized using the quanti-
tative optimization function of Analyst 1.4 to achieve the
highest signal response. [M + H]+ (at m/z 864.0) ions were
observed when the data were acquired in the Q1 scan
mode and the ion was chosen as precursor ion of CoQ10.
Secondly, we used product ion scans to look for the most
abundant product ion. The collision energy (CE) was opti-
mized to achieve highest sensitivity. The product ion
spectrum obtained, which is shown in Fig. 2, consisted of
an ion at m/z 197.0 and a second ion at m/z 234.9 when CE
was 38 V. The two base peaks at 197.0 and 234.9 corre-
spond to tropylium and pyrylium ions, respectively.
Among the product ions, that at m/z 197.0 was the most
abundant and was therefore chosen for CoQ10 quantita-
tion.
Finally, the precursor/product ion pair of m/z 864.0/197.0
was chosen for the MRM scan. MRM was performed with
150 ms dwell time. Peak areas obtained from the MRM of
CoQ10 standards were utilized for quantitation.
Sample solutions of tobacco leaf extract (prepared
according to the procedure described in Section 2.5)
were injected directly, separated, and detected under the
optimum condition mentioned earlier (Section 2.3). The
MRM chromatogram of coenzyme Q10 in tobacco leaves is
shown in Fig. 2.
It can be seen from Fig. 2 that the retention time of CoQ10
was 2.93 min. The analysis procedure can be finished in a
shorter analysis time. The results show the proposed
method to be very fast.
3.4 Method of standard addition
Quantification was based on the standard addition
method with the tobacco extract solution spiked with
CoQ10 standards of three different concentrations (50,
www.jss-journal.com
J. Sep. Sci. 2006, 29, 1607 – 1612 LC-MS/MS determination of coenzyme Q10 in tobacco leaves 1611

Table 1. Recoveries of CoQ10 (n = 3). method is highly sensitive. Meanwhile, a LOQ of 4.0 ng/
mL has been obtained for CoQ10 (S/N = 10).
CoQ10 added Peak area Found Recovery RSD of peak
(ng) (counts) (ng) (%) current (%)

0 2662 146 – 2.4 3.5.2 Precision and recovery

50 3552 192 98.2 2.3 The intra-day and inter-day precisions (expressed as the
175 5778 319 99.3 1.8 RSD) for the peak area were determined by repeated anal-
300 8003 440 98.6 1.5
ysis (n = 5). The result showed that intra-day and inter-
day RSDs for the peak area were, respectively, 0.91 and
100, 150 ng/mL). The detector response can then be 1.21%. The intra-day and inter-day RSDs for the peak area
plotted against the added concentration of CoQ10. This is are both quite low and the precisions are good.
referred to as a standard addition curve. The unknown To validate the proposed method, the recoveries were
CoQ10 concentration can be found by extrapolating the obtained by the procedure described in Section 2.6 and
best fit line to the x-axis intercept. That intercept will be the results are listed in Table 1. It was found that the
the unknown CoQ10 concentration. If the equation of the recoveries of CoQ10 were close to 100% and the corre-
best fit line is written in the form y = mx + q (y, peak area, sponding RSDs were all no more than 2.4%.
counts; x, added CoQ10 concentration, ng/mL) then the x-
axis intercept is equal to the y-axis intercept (q) over the
slope (m). 3.6 Determination of CoQ10 in tobacco leaves
The leaves from eight varieties of tobacco grown in the
3.5 Method validation same location were prepared and analyzed by the proce-
3.5.1 Linearity and detection limit dure described in Sections 2.5 and 2.3. The contents of
CoQ10 calculated by the standard addition method
Seven standard solutions of CoQ10 with concentrations
described in Section 3.4 are summarized in Table 2.
between 8.4 and 540.0 ng/mL were determined by LC-MS/
MS with MRM. Each individual standard with a certain The results listed in Table 2 show that the contents of
concentration was consecutively injected three times. CoQ10 are different for different varieties of tobacco and
The regression equation for CoQ10 (peak area, A, counts, that the content of CoQ10 is highest in the K326 variety.
vs. concentration, C, ng/mL) was as follows: The content of CoQ10 in tobacco leaves is seen to be dis-
tinctly related to the variety.
A = 17.802 C + 63.121 (R2 = 0.9996) (2)

The regression equation was found to be linear in the

range of 8.4 – 540.0 ng/mL with a good correlation coeffi-
cient (R2 = 0.9996). The LOD of CoQ10 was 1.2 ng/mL
(S/N = 3). Compared with other analytical methods, the
LOD obtained by the LC-MS/MS method proposed in the
text was 205.8, 71.9, 41.7, and 208.6 times lower than the
LOD obtained by spectrophotometric, voltammetric, and
chromatographic methods with ultraviolet, fluorescence
detection [11 – 14]. The results show that the proposed
4 Concluding remarks
In the present work, a LC-MS/MS method for the determi-
nation of CoQ10 in tobacco leaves has been presented. The
method permits analysis in a short time with good recov-
eries, precision, and sensitivity. The contents of CoQ10 in
the leaves of eight varieties of tobacco were analyzed and
compared using the method. This method also provides
a reference for the analysis of CoQ10 in other samples.

Table 2. Determination of CoQ10 in the leaves from eight varieties of tobacco (n = 3).

Variety of tobacco Standard addition Regression coefficient Content of CoQ10 RSD (%)
curvea) (R2) (lg/g)

Hong Hua Da Jin Yuan y = 17.82x + 7765.13 0.9996 11.5 1.2

Long Jiang 911 y = 17.63x + 6409.14 0.9996 10.9 1.1
Yun Yan 87 y = 17.96x + 6398.01 0.9997 10.7 2.3
Piao He 1 y = 17.91x + 5912.55 0.9997 9.9 1.7
Yun Yan 85 y = 17.57x + 7675.65 0.9991 13.1 1.9
K346 y = 17.57x + 6385.73 0.9995 10.9 0.8
K326 y = 17.88x + 9001.97 0.9996 15.1 1.9
NC89 y = 18.04x + 6013.87 0.9995 10.0 2.5
a)
y, peak area (counts); x, added CoQ10 concentration (ng/mL).

i 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com

1612 Y. Zu et al.
This work was financially supported by the Project of Scientific
Research Conditions Upgrade from the Ministry of Science and
Technology, China (JG-2004-09).
5 References
[1] Griffiths, W. T., Threlfall, D. R., Goodwin, T. W., Eur. J.
Biochem. 1968, 5, 124 – 132.
[2] Alleva, R., Tomasetti, M., Battino, M., Curatola, G., et al.,
Proc. Natl. Acad. Sci. USA 1995, 92, 9388 – 9391.
[3] Stocker, R., Bowry, W. V., Frei, B., Proc. Natl. Acad. Sci. USA
1991, 88, 1646 – 1650.
[4] Niibori, K., Yokoyama, H., Crestanello, J. A., Whitman,
G. J., J. Surg. Res. 1998, 79, 141 – 145.
[5] Lockwood, K., Moesgaard, S., Yamamoto, T., Folkers, K.,
Biochem. Biophys. Res. Commun. 1995, 212, 172 – 177.
[6] Portakal, O., Inal-Erden, M., Clin. Biochem. 1999, 32, 461 –
466.
[7] Hodgson, J. M., Watts, G. F., Playford, D. A., Burke, V.,
Croft, K. D., Eur. J. Clin. Nutr. 2002, 56, 1137 – 1142.
[8] Kttner, A., Pieper, A., Koch, J., Enzmann, F., Schrder,
S., Int. J. Cardiol. 2005, 98, 413 – 419.
[9] Kitano, M., Hosoe, K., Fukutomi, N., Hidaka, T., et al.,
Food Chem. Toxicol. 2004, 42, 1817 – 1824.
[10] Hendler, S. S., Rorvik, M. S. (Eds.), PDR for Nutritional Sup-
plements, Medical Economics, Montvale, New Jersey,
2001, p. 103.
i 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Sep. Sci. 2006, 29, 1607 – 1612
[11] Karpilska, J., Frankowska, R., Instrum. Sci. Technol. 2004,
32, 281 – 290.
[12] Litescu, S. C., David, I. G., Radu, G. L., Aboul-Enein, H. Y.,
Instrum. Sci. Technol. 2001, 29, 109 – 116.
[13] Kommuru, T. R., Khan, M. A., Ashraf, M., Kattenacker,
R., Reddy, I. K., J. Pharm. Biomed. Anal. 1998, 16, 1037 –
1040.
[14] Molyneux, S. L., Lever, M., Clin. Chim. Acta 2005, 358,
198 – 200.
[15] Teshima, K., Kondo, T., Anal. Biochem. 2005, 338, 12 – 19.
[16] Hansen, G., Christensen, P., Tuchsen, E., Lund, T., Analyst
2004, 129, 45 – 50.
[17] Strazisar, M., Fir, M., Golc-Wondra, A., Milivojevic, L., et
al., J. AOAC Int. 2005, 88, 1020 – 1027.
[18] Zllner, P., Jodlbauer, J., Lindner, W., J. Chromatogr. A
1999, 858, 167 – 174.
[19] Benijts, T., Dams, R., Lambert, W., De Leenheer, A., J.
Chromatogr. A 2004, 1029, 153 – 159.
[20] Ohkawa, T., Ishida, Y., Kanaoka, E., Takahashi, K., et al., J.
Pharm. Biomed. Anal. 2003, 31, 1089 – 1099.
[21] Lagerwerf, F. M., van Dongen, W. D., Steenvoorden, R. J.
J. M., Honing, M., Jonkman, J. H. G., Trends Anal. Chem.
2000, 19, 418 – 427.
[22] Mullins, F. G. P., J. Sep. Sci. 2001, 24, 593 – 598.
[23] Espeja, E., Concepcin Garca, M., Luisa Marina, M., J.
Sep. Sci. 2001, 24, 856 – 864.
[24] Basilicata, P., Miraglia, N., Pieri, M., Acampora, A., et al.,
J. Chromatogr. B 2005, 818, 293 – 299.
www.jss-journal.com