Food Chemistry 135 (2012) 2988–2993

Contents lists available at SciVerse ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Analytical Methods

Optimisation for subcritical fluid extraction of 17-methyltestosterone with

1,1,1,2-tetrafluoroethane for HPLC analysis
Yuqian Han ⇑, Qinchuan Ma, Jie Lu, Yong Xue, Changhu Xue
Department of Food Science and Engineering, Ocean University of China, P.O. Box 266003, Qingdao, China

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

Article history: A simple, rapid and sensitive method was developed for determination of 17a-methyltestosterone in
Received 13 April 2012 aquatic products by extraction with subcritical 1,1,1,2-tetrafluoroethane (R134a) extraction and high
Received in revised form 4 July 2012 performance liquid chromatography (HPLC). Response surface methodology (RSM) was adopted to opti-
Accepted 13 July 2012
mise extraction pressure, temperature and co-solvent volume. The optimum extraction conditions pre-
Available online 25 July 2012
dicted within the experimental ranges were as follows: pressure 5 MPa, temperature 31 °C, and co-
solvent volume 3.35 ml. The analysis was carried out on XDB-C18 column (4.6 mm  250 mm, 5 lm) with
Keywords:
the mobile phase acetonitrile–water (55:45, v/v), flow rate 0.8 ml/min, temperature 30 °C and wave-
Determination
Subcritical R134a extraction
length 245 nm. Good linearity of detection was obtained for 17a-methyltestosterone between concentra-
17a-Methyltestosterone tions of 50–250 ng/ml, r2 = 0.999. The method was validated using samples fortified with 17a-
Aquatic products methyltestosterone at levels of 10, 30 and 50 ng/g, the mean recovery exceeds 90%, and the RSD values
were less than 10%.
Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.

1. Introduction
The synthetic androgen 17a-methyltestosterone is commonly
used in newly hatched tilapia fry for sex reversal (Gale, Fitzpatrick,
Lucero, Contreras-Sanchez, & Schreck, 1999; Kuwaye et al., 1993;
Pandian & Sheela, 1995). In tilapia aquaculture, an all-male popu-
lation is desired because males grow faster and larger than do fe-
males (Chu, Lopez, Serfling, Gieseker, & Reimschuessel, 2006).
And 17a-methyltestosterone can also promote the anabolism of
protein so as to quicken animal’s growth and improve the conver-
sion rate of feed. It has been proven by practice that the fish pro-
duction could increase by 30% when 17a-methyltestosterone is
used in fish farming (Nie, Zhu, Zhu, He, & Pan, 2008). However,
traces of 17a-methyltestosterone residue in animals used for food
could endanger the health of consumers, such as early pregnant
reaction, fetal anomaly, and disrupting hormone balance. In many
countries, the use of 17a-methyltestosterone is prohibited in food-
producing animals, and could not be detected in food. 17a-Methyl-
testosterone is a kind of cumulative and persistent compound in
biota matrixes and as liposoluble compound it is difficult to be
metabolised by living beings. Analytical methods are therefore
needed to monitor 17a-methyltestosterone residues in food
animals.
A large number of analytical procedures have been described
for the detection and quantification of 17a-methyltestosterone
residues in the plant and animal tissues (Daeseleire, De Guesquiere,
⇑ Corresponding author. Tel./fax: +86 053282031629.
E-mail address: hanyuqian@ouc.edu.cn (Y. Han).
& Van Peteghem, 1991a, 1991b; Casademont, Perez, & Garcia-
Regueiro, 1996; Deng, Kurosu, & Pounder, 1999; Jansen, Van den
Berg, Zomer, & Stephany, 1985; Marwah, Marwah, & Lardy,
2005). The methods of radioimmunoassay (Daeseleire et al.,
1991a, 1991b), chemiluminescence immunoassay (Jansen et al.,
1985; Van Peteghem, van Look, & De Guesquiere, 1989), gas chro-
matography-mass spectrometry (GC–MS), and liquid chromatogra-
phy (LC) were applied in these analytical and separation
approaches. Although these methods have been improved, most
studies rely on time-consuming procedures. And these methods
are labour-intensive and require HPLC fractionation and handling
of radioactive materials. Supercritical fluids have both gas and li-
quid like properties. They possess gas like mass transfer properties
and solvation characteristics of liquids. Their high diffusivity al-
lows them to penetrate solid materials, and their liquid like densi-
ties enable them to dissolve analytes from a solid matrix
(Demirbas, 2001). The supercritical fluid extraction (SFE) can be
selective to some extent by controlling the density of the medium
and its solubility varies with the density of the medium. The ex-
tracted materials are easily recovered by simply depressurising,
allowing the supercritical fluid to return to gas phase and evapo-
rate, leaving no or little solvent residues. CO2 is non-flammable,
non-toxic, and inexpensive with moderate critical parameters.
Supercritical CO2 extraction has been proposed as a pretreatment
technology for residue analysis. However, it typically requires the
pressure up to 300 bar for satisfactory extraction. In view of the
economic and environmental needs, alternative SFE solvents that
enable operation at less intense conditions are needed to be ex-
plored to take place of SFECO2.

0308-8146/$ - see front matter Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2012.07.081
 Y. Han et al. / Food Chemistry 135 (2012) 2988–2993
Subcritical fluid extraction has been known for many years. H2O
at subcritical conditions has been shown to be an effective fluid for
the extraction of valuable substances (Gamiz-Garcia and Luque de
Castro, 2000). Besides water, some other solvents, such as ethane,
propane, ethylene, dimethyl ether, 1,1,1,2-tetrafluoroethane
(R134a), have also been recommended as solvents under sub-
and supercritical conditions for the extraction of valuable
substances.
Now, many researchers have been conducted on the potential of
using R134a in various fields such as food, polymer and cosmetics
and to address the shortcomings of existing isolation processes,
including supercritical CO2 processes (Mustapa, Manan, Mohd Azi-
zi, Nik Norulaini, & Mohd Omar, 2009). In the high-pressure and
low compressibility region, the solvating power of R134a is greater
than that of CO2 (Roth, 1996). The increase in commercial availabil-
ity and reasonable critical properties (101.1 °C, 4.06 MPa) of R134a
in combination with a permanent dipole moment (2.05 D) led to
the evaluation of its use as an alternative to scCO2 for the extrac-
tion of more polar analytes (Corr, 2002).
The objective of this work is to evaluate the method of subcrit-
ical R134a extraction as pretreatment technology for residue anal-
ysis of 17a-methyltestosterone in aquatic products. By optimising
the operating conditions (co-solvent volume, pressure and temper-
ature), a method of detecting 17a-methyltestosterone in aquatic
products with subcritical R134a extraction and liquid chromatog-
raphy was established.
2. Materials and methods
2.1. Materials and solvents
Tilapias were obtained from Tilapia breeding farm (Qingdao,
China). Liquefied R134a stored in a cylinder with a purity of
99.9% was purchased from INEOS (England). Chromatographic
grade methanol was purchased from Biao Shiqi (Tianjin, China).
17a-methyltestosterone standard was purchased from Dr. Ehren-
storfer (Germany). Solid phase extraction (SPE) cartridges (C18, 6
cc) were obtained from Waters Corporation (Milford, MA, USA).
2.2. Reference standard solutions
Standard solution of 17a-methyltestosterone in methanol was
prepared with concentration of 0.1 mg/ml and stored at 20 °C
as stock solution in the darkness. Working solutions, which were
prepared by a series of 10-fold dilutions of the stock solution, were
stored in the dark at approximately 4 °C.
2.3. Apparatus
The instrument used in the subcritical R134a extraction experi-
ments was made at home (Fig. 1). It consists of a high-pressure
pump (Hangzhou Zhijiang Petrochemical Equipment Co. Ltd.,
Hangzhou, China) capable of generating a maximum pressure of
350 bar and a maximum flow rate around 0.4 L/h. The instrument
has an oven and equipped with an extraction vessel
(120 mm  20 mm I.D.) and a separator (120 mm  20 mm I.D.)
that can be operated at pressure up to 300 bar. Liquid R134a is han-
dled by a high pressure metering pump with jacketed head for cool-
ing, and the flow rate could be regulated between 0.0 and 0.4 L/h.
2.4. Sample preparation
Tilapias were filleted, the skin and bones removed and muscles
blended in a laboratory blender and frozen at 20 °C, before anal-
ysis. Tilapia tissues were previously tested (Nie et al., 2008), the
2989
limit of detection of this method is 1.0 lg/kg. It showed that no
detectable residue of 17a-methyltestosterone was contained and
it can be used as negative controls. Screening experiments were
carried out on tilapia spiked with 17a-methyltestosterone. The
standard 17-methyltestosterone was spiked into homogenised tis-
sue samples. One kilogramme of tilapia muscle tissue was spiked
with a final concentration of 3 lg/g 17a-methyltestosterone for
use in the evaluation of the subcritical extraction.
The optimisation procedure was applied to the analysis of real
samples of tilapia. Tilapias with an average weight of 200 g were
obtained from a commercial fish farm. After entering our experi-
mental facilities, tilapias were stocked in the experimental aquaria
for an acclimation period of one month. The experiment was con-
ducted in 160 L aquaria. The tank was connected to a recirculation
system equipped with a sedimentation tank, a trickling biofilter, a
sump and a pump. During the experiments the water temperature
was 25 ± 1 °C and photoperiod 12 h, from 7:00 am–7:00 pm. At the
beginning of the experiment, tilapias were fed on a diet containing
17a-methyltestosterone twice a day with the feeding level of 50 g/
kg. After the acclimation period of 30 days, tilapias were taken
from the pool, euthanised and stored at 20 °C for later
determination.
2.5. Extraction procedure
Optimisation experiments were carried out on a 5-g sample of
tilapia tissue. The tilapia tissues were mixed with 15-g quartz sand
and 5-g fibreglass to disperse the sample fully. In each experiment,
the sample was placed in the middle of an extraction cell, and co-
solvent was added. The bottom and top of the extraction cell were
filled with glass wool to prevent entrainment of the material and
blockages in the system.
The extraction cell was placed in a heating bath to maintain an
operating temperature within ±1 °C of the set-point temperature
for each run. R134a was delivered via a metering pump with a flow
rate of 10 g/min. The system pressure was controlled by a back-
pressure regulator that was adjusted to maintain pressure in the
range of 20–150 bar. To improve the extraction efficiency, a static
period of 20 min was allotted to promote contact between the
sample and the subcritical R134a fluid. The static period was sub-
sequently followed by dynamic extraction for 40 min. Next, the
stream of subcritical R134a fluid containing 17a-methyltestoster-
one was depressurised through a pressure restrictor, and the ex-
tracts were collected in a separator with 10 ml methanol.
2.6. SPE clean-up
To decrease interferences and increase sensitivity, the acquisi-
tion of clean extracts is highly desirable, especially for the analysis
of biological samples. The methanol solution containing the ex-
tracts was evaporated to dryness under reduced pressure at
40 °C, diluted with 2.0 ml of a methanol:water (1:1) solution, and
applied to a preconditioned solid-phase extraction (SPE) cartridge
(C18, 6 cc, preconditioned with methanol (2.0 ml) and water
(2.0 ml)). After washing with 3 ml of water, 17a-methyltestoster-
one was eluted with 5 ml of methanol. The eluent was evaporated
to dryness, diluted to volume with the mobile phase. About 1.0 mL
of the final solution was transferred into a 2.0 ml autosampler vial,
and 20 ll was injected into the HPLC for analysis.
2.7. Chromatographic conditions
An Agilent 1100 series LC system was used in this study. The
HPLC system consisted of a model G1311A HPLC pump with a
model G1313A autosampler. The separation was carried out on a
XDB-C18 column (5 lm, 4.6 mm  250 mm, Agilent, USA) and
2990 Y. Han et al. / Food Chemistry 135 (2012) 2988–2993

Fig. 1. Schematic diagram of the R134a extraction apparatus. A: R134a cylinder; B: filter; C: cooler; D: high pressure pump; E: heat exchanger; F: extraction cell; G: heating
bath; H: back pressure regulator; I: collecto; and J: rotameter.

equipped with a guard column (Agilent, USA). The column temper- Table 1
ature was maintained at 30 °C. The mobile phase, consisting of ace- Uncoded and coded independent variables used in RSM design.
tonitrile:water (55:45, v/v), was pumped at 0.8 ml/min. 17a- Symbols Independent variable Coded levels
Methyltestosterone was detected at 245 nm (bandwidth 4 nm)
1 0 1
for quantitation using a diode array detector (DAD, G1315A).
X1 Pressure (MPa) 2 8 14
X2 Temperature (°C) 30 35 40
2.8. Method validation X3 Co-solvent volume (ml) 2 3 4

Method validation was performed to meet the criteria specified

on preliminary experimental results. The experimental design
by European Commission Decision 2002/657/EC (2002). Tilapia tis-
sue samples fortified at different levels (5–50 ng/g) were extracted, was based on the Box Behnken design with three central points
as shown in Table 2. The experimental data obtained were ana-
and linearity and sensitivity were checked by injecting the analytes
at different concentration levels into the HPLC. Calibration curves lysed by the response surface regression procedure using the fol-
lowing second-order polynomial equation:
of the spiked blank fish samples for each 17a-methyltestosterone
were calculated by using a least squares linear regression analysis X
3 X
3 X
3

then plotting the peak area of each analyte versus the analyte con- Y ¼ b0 þ bi X i þ bii X 2i þ bij X i X j
centration. The limit of detection (LOD) was based on the 3s crite- i¼1 i¼1 i–j¼1

rion using a series of 10 solutions containing a low concentration of
17a-methyltestosterone. Finally, the limit of quantification (LOQ)
was selected as the lowest concentration used in the calibration
curve (FDA, 2001).
In addition to LOD and LOQ, accuracy and precision were studied.
The fish samples that were fortified with 17a-methyltestosterone at
levels of 10, 30 and 50 ng/g were extracted and analysed on 3 differ-
ent days. Each group was repeated four times. The muscle tissue of
the tilapias that were fed a diet containing 17a-methyltestosterone
was extracted using subcritical R134a to evaluate this method.
To show the applicability of the method to other species, a
study was also completed for other aquatic products. Cyprinus car-
pio (Carp), Silver carp (Chub), Ctenopharyngodon idellus (Grass carp),
Carassius auratus (Crucian), Pseudosciaena crocea (yellow crock),
Japanese sea perch (Sea bass), Metapenaeus ensis (De Hann), Pena-
eus vannamei Boone (White shrimp), Penaeus (Marsupenaeus)
japonicus Bate (Penaeus japonicus) from the local market were
analysed.
2.9. Experimental design
Many factors could affect the efficiency of subcritical R134a
extraction: including pressure, temperature, co-solvent volume,
flow rate, extraction mode and extraction time. So multiple vari-
ables may influence the extraction efficiency, the response surface
methodology (RSM) is an effective technique for optimising the
process (Bas and Boyaci, 2007). RSM was applied to evaluate the ef-
fects of pressure, temperature, co-solvent volume on the extraction
efficiency of 17a-methyltestosterone. The coded and uncoded
independent variables used in the RSM design were listed in Ta-
ble 1. And the levels of the independent parameters were based
where Y is the response (extraction rate%), b0 is a constant, bi, bii and
bij are the linear, quadratic and interactive coefficients, respectively.
Xi and Xj are the levels of the independent variables. Three-dimen-
sional surface response plots were generated using the fitted model
by varying two variables within the experimental range and holding
the other constant at the central point. The coefficients of the re-
sponse surface equation were estimated by using the software of
Design Expert 7.1.6. The test of statistical significance was based
on the total error criteria with a confidence level of 95.0%.
3. Results and discussion
3.1. Fitting the models
The extraction rates of 17a-methyltestosterone obtained from
all the experiments are listed in Table 2. The experimental data
were used to calculate the coefficients of the second-order polyno-
mial equation and Table 3 summarises the regression coefficients
obtained. The coefficient of determination (R2) and adjusted coeffi-
cient of determination (Adj.R2) were 0.9768 and 0.9352, respec-
tively, indicating adequate accuracy and general availability of
the polynomial model. The application of RSM yielded the follow-
ing regression equation which was an empirical relationship be-
tween extraction rate (Y) and the test variables in coded units:
Y ¼ 88:39  8:11X 1  5:99X 2 þ 9:90X 3  7:81X 21  3:45X 22
 7:24X 23 þ 1:59X 1 X 2 þ 4:72X 1 X 3 þ 3:05X 2 X 3
ANOVA was also used to evaluate the significance of the coeffi-
cients of the models (Table 3). For any of the terms in the model, a
large regression coefficient and a small p-value would indicate a
 Y. Han et al. / Food Chemistry 135 (2012) 2988–2993 2991

Table 2 technology for gases and fluids (De Castro, Valcarcel, & Tena,
Experimental points of the Box Benken design and the experimental data. 1994; Taylor, 1996; Turner, King, & Mathiasson, 2001). The pres-
Experiment number X1 X2 X3 Extraction rate (%) sure increasing around subcritical state causes the density of
Experimental Predicted R134a solvent increasing (and hence, its solvating power)
(Mustapa et al., 2009). Besides, as the density of solvent increases,
1 1 1 0 94.96 92.82
2 1 1 0 70.61 73.43
the inter-molecular interactions of solutes also rise. As a result,
3 1 1 0 80.49 77.67 17a-methyltestosterone and solvent dissolution were promoted,
4 1 1 0 62.48 64.62 thereby increasing the 17a-methyltestosterone extracted. At
5 1 0 1 74.09 76.47 high-pressure levels, however, the extraction rate decreased, per-
6 1 0 1 53.39 5081
haps because the state of fluid was far from the subcritical state
7 1 0 1 84.26 86.84
8 1 0 1 82.45 80.06 when the extraction pressure was high. The change of state would
9 0 1 1 77.29 77.04 affect the extraction ability of subcritical fluid. In addition, the
10 0 1 1 58.53 58.96 viscosity also increases with pressure, which would prevent the
11 0 1 1 91.17 90.74
solute from spreading to fluid.
12 0 1 1 84.62 84.87
13 0 0 0 88.85 88.39
Note that the co-solvent was added into extraction cell before
14 0 0 0 88.01 88.39 dynamic extraction. High extraction pressure may result in low
15 0 0 0 88.32 88.39 polarity of subcritical fluid, because high density R134a fluid
(which rises with increasing pressure) would decrease the percent-
age of co-solvent. So the extraction rate decreased with increasing
pressure, and this effect can be seen clearly from Fig. 2a and b.
The solubility of a compound depends on a complex balance be-
Table 3 tween the subcritical fluid density and solute vapour pressure,
Estimated coefficients of second order response model.
which are both controlled by fluid temperature and pressure. For
Regression Extraction rate (Y) F- Probability 17a-methyltestosterone, there is a competition between its vola-
coefficient value (p) tility (which rises with increasing temperature) and its solubility
Regression Standard
coefficients error in subcritical R134a (which decreases as the temperature in-
b0 88.39 1.84 creases). So, the solubility of 17a-methyltestosterone is likely to
Linear decrease or keep constant, or increase with the temperature rising
b1 8.11 1.12 51.98 0.0008 at constant pressure, which depends on whether the solvent den-
b2 5.99 1.12 28.36 0.0031
sity or the solute vapour pressure is the predominant one. In this
b3 9.90 1.12 77.51 0.0003
study, it is shown that temperature has a negative linear effect, be-
Quadratic
cause temperature rising would decrease the fluid density, conse-
b11 7.81 1.59 22.24 0.0053
b22 3.45 1.59 4.34 0.0916 quently, the solvency of R134a decreases.
b33 7.24 1.59 18.09 0.0081 In order to improve the extractability of 17a-methyltestoster-
Interaction one, the effect of co-solvent volume on the extraction rate was
b12 1.59 1.66 0.99 0.3646 investigated. As presented in Fig. 2b and c, the extraction rate dras-
b13 4.72 1.66 8.81 0.0312 tically increased when methanol was added. Methanol could en-
b23 3.05 1.66 3.69 0.1129
hance the polarity of subcritical R134a fluid, and owing to its
R2 0.9768
Adj.R2 0.9352
polarity, the co-solvent favoured dissolution of 17a-methyltestos-
terone. Note that co-solvent is likely to be volatile under high tem-
perature. So a higher extraction rate was obtained at low
temperature when the co-solvent volume was large. However, it
more significant effect on the respective response variables (Li & should be noted that large amounts of co-solvent will change the
Fu, 2005). Thus, the variable with the largest effect on the extrac- critical parameters of the mixtures (Crowther & Henion, 1985)
tion rate was the linear term of co-solvent volume (p < 0.001), fol- and the extraction rate drastically increased with the amount of
lowed by the linear terms of pressure (p < 0.001) and temperature co-solvent volume, suggesting that the range of this variable exam-
(p < 0.01). The quadratic terms of pressure and co-solvent volume, ined in this study was probably not broad enough to observe its
as well as the interactions between pressure and co-solvent vol- influence on 17a-methyltestosterone extraction rate.
ume, had significant effects on the extraction rate.
3.3. Optimisation of extraction condition

3.2. Analysis of response surfaces
In order to visualise the effect of the independent variables on
the dependent ones, drawing surface response plots of the model
is usually regarded as a good way, which were done by varying
two variables within the experimental range and holding the other
constant at the central point (Xu, Gao, Liu, Wang, & Zhao, 2008).
The response surface was generated by fitting the quadratic poly-
nomial equation obtained from regression analysis, holding one
parameter at a constant value and changing the other two
variables.
Fig. 2a shows the effect of extraction pressure and temperature
on the extraction rate at 3 mL of co-solvent. Pressure had a small
positive linear effect on extraction rate at low-pressure levels. This
can be explained by using the established principles of SFE
The subcritical R134a extraction conditions would be consid-
ered optimum if the extraction rate reached maximum value. From
the solutions predicted by the model, the experimental conditions
set at the pressure of 5.03 MPa, temperature 30.909 °C, and co-sol-
vent volume 3.35 mL could give an extraction rate of 94.64%. Due
to the limited conditions, the experiments were conducted at the
temperature of 31 °C, pressure 5 MPa and co-solvent volume
3.35 mL, and giving the extraction rate of 94.72%, which was iden-
tical to the model prediction.
3.4. Results of method validation
The method was validated using the developed conditions
(temperature, 31 °C; pressure, 5 MPa; co-solvent volume,
3.35 ml). The calibration curves were linear (r2 = 0.999) over the
2992 Y. Han et al. / Food Chemistry 135 (2012) 2988–2993
Fig. 2. Surface plot of the 17a-methyltestosterone extraction rate.
Table 4
Extraction recoveries of 17a-methyltestosterone from tilapia.
Analyate Fortification Recovery
level (ng/g) (%)
17a-Methyltestosterone 10 93.65
30 94.72
50 92.97
range of 50–250 ng/ml. The limit of quantification for the method,
as determined from the lowest standard on the calibration curve
(50 ng/ml), was 10 ng/g (S/N = 3). The accuracy and precision of
the method were determined using fish samples fortified with
17a-methyltestosterone at levels 10, 30 and 50 ng/g. All experi-
ments were repeated four times. The mean recovery of 17a-meth-
yltestosterone was above 90%, and the RSD values were below 10%
(Table 4).
3.5. Analysis of real samples
The validity of this method was studied using real tilapia sam-
ples. Initially, the tilapias were fed a diet containing 50 g/kg of
17a-methyltestosterone twice a day. After an acclimation period
of 30 days, the tilapias were taken from the aquarium for analysis.
The results indicate that the tilapia tissues contained 0.135 lg/g of
17a-methyltestosterone. Besides, the following aquatic products
were also subjected to the extraction procedure: C. carpio (Carp),
S. carp (Chub), C. idellus (Grass carp), C. auratus (Crucian), P. crocea
(yellow crock), Japanese sea perch (Sea bass), M. ensis (De Hann), P.
vannamei Boone (White shrimp), Penaeus (Marsupenaeus) japoni-
cus Bate (P. japonicus) were bought from the local market. All
experiments were repeated three times. It is found that the Crucian
tissues had a level of 17a-methyltestosterone of 0.082 lg/g
(RSD = 5.32%), and no 17a-methyltestosterone was found in other
aquatic products, which suggests that this method is valid to deter-
mine 17a-methyltestosterone in aquatic products.
4. Conclusion
A subcritical extraction method has been developed for the
analysis of 17a-methyltestosterone in aquatic products, giving
acceptable recovery and repeatability. In addition, this method
takes about only 1 h. Under the improved conditions (31 °C,
5 MPa, co-solvent volume 3.35 mL), the recovery of 17a-methyl-
testosterone from tilapia exceeds 90%, and the RSD was less than
5%, proving that the subcritical R134a extraction is a feasible sam-
ple pretreatment technology for the analysis of 17a-
methyltestosterone.
Acknowledgement
This work was supported by the National Natural Science
RSD
(%, n = 4)
Funds, Project Number: 31071541.
2.78
1.26
2.22 References
Bas, D., & Boyaci, I. H. (2007). Modeling and optimization I: Usability of response
surface methodology. International Journal of Food Engineering, 78, 836.
Casademont, G., Perez, B., & GarciaRegueiro, J. A. (1996). Simultaneous
determination, in calf urine, of twelve anabolic agents as heptafluorobutyryl
derivatives by capillary gas chromatography-mass spectrometry. Journal of
Chromatography B: Biomedical Sciences and Applications, 686, 189.
Chu, P. S., Lopez, M., Serfling, S., Gieseker, C., & Reimschuessel, R. (2006).
Determination of 17a-methyltestosterone in muscle tissues of tilapia,
rainbow trout, and salmon using liquid chromatography–tandem mass
spectrometry. Journal of Agricultural and Food Chemistry, 543, 3193.
Corr, S. (2002). 1,1, 1, 2-Tetrafluoroethane; from refrigerant and propellant to
solvent. Journal of Fluorine Chemistry, 118, 55.
Crowther, J. B., & Henion, J. D. (1985). Supercritical fluid chromatography of polar
drugs using small-particle packed columns with mass spectrometric detection.
Analytical Chemistry, 57, 2711.
Daeseleire, E., De Guesquiere, A., & Van Peteghem, C. (1991a). Derivatization and gas
chromatographic–mass spectrometric detection of anabolic steroid residues
isolated from edible muscle tissues. Journal of Chromatography B: Biomedical
Sciences and Applications, 562, 673.
Daeseleire, E., De Guesquiere, A., & Van Peteghem, C. (1991b). Combined high-
performance liquid chromatography and radioimmunoassay for the
screening of 19-nortestosterone and methyltestosterone residues in meat
samples. Journal of Chromatography B: Biomedical Sciences and Applications,
564, 445.
De Castro, L., Valcarcel, M. D., & Tena, T. M. (1994). Analytical Supercritical Fluid
Extraction. German: Springer-Verlag.
Demirbas, A. (2001). Supercritical fuid extraction and chemicals from biomass with
supercritical fuids. Energy Conversion and Management, 42, 279.
Deng, X. S., Kurosu, A., & Pounder, D. J. (1999). Detection of anabolic steroids in head
hair. Journal of Forensic Sciences, 44, 343.
FDA (2001). Guidance for Industry. Bioanalytical Method validation. U.S.
Department of Health and Human Services. Food and Drug Administration.
Rockville, MD, USA. <http://www.fda.gov/cder/guidance/4252fnl.pdf>
Accessed: 19.10.2009.
Gale, W. L., Fitzpatrick, M. S., Lucero, M., Contreras-Sanchez, W. M., & Schreck, C. B.
(1999). Masculinization of Nile tilapia (Oreochromis niloticus) by immersion in
androgens. Aquaculture, 178, 349.
Gamiz-Garcia, L., & Luque de Castro, M. D. (2000). Continuous subcritical water
extraction of medicinal plant essential oil: Comparison with conventional
techniques. Talanta, 511, 197.
Jansen, E. H., Van den Berg, R. H., Zomer, G., & Stephany, R. W. (1985). A
chemiluminescent immunoassay for 17 a-methyltestosterone. Food Additives
and Contaminants, 2, 47.
Kuwaye, T. T., Okimoto, D. K., Shimoda, S. K., Howerton, R. D., Lin, H. R., Pang, P. K. T.,
et al. (1993). Effect of 17 [alpha]-methyltestosterone on the growth of the
euryhaline tilapia, Oreochromis mossambicus, in fresh water and in sea water.
Aquaculture, 113, 137.
Li, Q. H., & Fu, C. (2005). Application of response surface methodology for
extraction optimization of germinant pumpkin seeds protein. Food Chemistry,
92, 701.
Marwah, A., Marwah, P., & Lardy, H. (2005). Development and validation of a high
performance liquid chromatography assay for 17 a-methyltestosterone in fish
feed. Journal of Chromatography B: Analytical technologies in the biomedical and
life sciences, 824, 107.
 Y. Han et al. / Food Chemistry 135 (2012) 2988–2993
Mustapa, A. N., Manan, Z. A., Mohd Azizi, C. Y., Nik Norulaini, N. A., & Mohd Omar, A.
K. (2009). Effects of parameters on yield for sub-critical R134a extraction of
palm oil. International Journal of Food Engineering, 95, 606.
Nie, J. R., Zhu, M. L., Zhu, M. L., He, W. B., & Pan, Y. S. (2008). Rapid determination of
methyltestosterone in aquaculture with HPLC–MS/MS. Guangdong Agricultural
Sciences, 6, 107.
Pandian, T. J., & Sheela, S. G. (1995). Hormonal induction of sex reversal in fish.
Aquaculture, 138, 1.
Roth, M. (1996). Thermodynamic prospects of alternative refrigerants as solvents
for supercritical fluid extraction. Analytical Chemistry, 68, 4474.
Taylor, L. T. (1996). Supercritical fluid extraction. New York NY: John Wiley & Sons, Inc.
2993
Turner, C., King, J. W., & Mathiasson, L. (2001). Supercritical fluid extraction and
chromatography for fat-soluble vitamin analysis. Journal of Chromatography A,
936, 215.
Van Peteghem, C., van Look, L., & De Guesquiere, A. (1989). Chemiluminescence
immunoassay for the detection and quantification of methylestosterone
residues in muscle tissue. Journal of Chromatography B: Biomedical Sciences
and Applications, 489, 219.
Xu, X., Gao, Y. X., Liu, G. M., Wang, Q. & Zhao, J. (2008). Optimization of supercritical
carbon dioxide extraction of sea buckthorn (Hippophae thamnoides L.) oil using
response surface methodology, LWT, 41, 1223.