J. Sep. Sci.

2014, 37, 2439–2445 2439

Juan Li1,2 Research Article

Xiaoming Ding1
Jiaxin Zheng1,2
Dandan Liu1 Determination of synthetic dyes in bean and
Fei Guo1
Hongmin Liu1,2 ∗ meat products by liquid chromatography
Yanbing Zhang1,2 with tandem mass spectrometry
1 School of Pharmaceutical
Sciences, Zhengzhou A sensitive and efficient method was developed for the simultaneous determination of eight
University, Zhengzhou, China synthetic dyes (Chrysoidin, Auramine O, Sudan(I–IV), Para Red, and Rhodamine B) in
2 New Drug Research &
bean and meat products using high-performance liquid chromatography with tandem mass
Development Center,
Zhengzhou University, spectrometry. A simple extraction procedure using acetonitrile has been applied for the
Zhengzhou, China extraction of these dyes from spiked bean and meat samples. Chromatographic separation
was achieved on a Waters XTerra C18 column (2.1 × 150 mm, 5 ␮m) with a multistep
gradient elution. Detection and quantification were performed using mass spectrometry
Received March 31, 2014
Revised May 12, 2014 in multiple reaction monitoring mode. Linear calibrations were obtained with correlation
Accepted May 27, 2014 coefficients R2 > 0.99. The limits of detection and quantification for the eight dyes were
in the ranges of 0.03–0.75 and 0.1–2.0 ␮g/kg depending on matrices, respectively. The
recoveries of these dyes in different food matrices were between 71.2 and 116.9% with
relative standard deviations <15.2%, suggesting that the developed method is promising
for the accurate quantification of the eight dyes at trace levels in bean and meat products.

Keywords: Bean products / Liquid chromatography with mass spectrometry /

Meat products / Multiresidue analysis / Synthetic dyes
DOI 10.1002/jssc.201400349

1 Introduction Different analytical methods have been reported in the

literature for the determination of synthetic dyes, including
Synthetic dyes are often added to food to compensate for the TLC [9, 10], solid-phase spectrophotometry [11], ion-pair liq-
loss of natural colors, which are destroyed during processing uid chromatography [9, 12, 13], HPLC with UV or diode array
and storage, and to provide the desired colored appearance [1]. detection [14–18], and LC–MS [19–23]. HPLC–MS/MS that
European Directive 1994/36/EC [2] lists the colors that can be combines the high separation performance with precursor
added to food: it also defines foodstuffs to which only certain and fragment information has been reported for the mul-
colorants may be added, their permitted maximum level, and tiresidue analysis in foods [24–26]. Under the multiple reac-
their use restrictions as well. Because most synthetic dyes tion monitoring (MRM) mode, it can distinguish and identify
contain azo functional groups and aromatic rings, they may targets from background matrix ions, which can increase sen-
have adverse effects on health including allergic and asth- sitivity due to the reduction of background signals [27–29].
matic reactions [3], DNA damage [4], and hyperactivity [5]. Prior to the detection of the synthetic dyes, pretreatment
Some synthetic dyes are even considered to be potentially of the real samples was necessary. Recently, one-step extrac-
carcinogenic and mutagenic to humans [6, 7]. Thus, the use tion by organic solvents was imported in the pretreatment of
of synthetic dyes in food products is strictly controlled in spices and chili products before HPLC detection of synthetic
many countries [8]. Many synthetic dyes, including Sudan dyes [17, 21]. We have previously studied the determination
(I–IV), Para Red, Rhodamine B, Chrysoidin, and Auramine of dye residues in chili products using acetonitrile/H2 O as the
O, are prohibited worldwide, and their presence in foods at extraction solvent [30]. So far, most of the reported HPLC-
any level is not permitted. Considering the potential effects based methods with one-step extraction procedure focused
on human health, it is necessary to develop a sensitive and on the determination of these dyes in chili samples. How-
accurate method for the determination of these dyes in foods ever, these methods are not suitable for determining dyes in
to ensure food safety. bean and meat matrix owing to the high protein content. The
high protein and salt content of such samples may induce se-
Correspondence: Professor Yanbing Zhang, School of Pharma- vere interference for LC–MS measurement, and lead to low
ceutical Sciences, Zhengzhou University, Zhengzhou, 450001, sensitivity and specificity of the developed method.
China
E-mail: zhangyb@zzu.edu.cn
Fax: +86-371-67781739
∗ Additional corresponding author: Professor Hongmin Liu,
Abbreviations: HLB, hydrophilic-lipophilic balance; MRM, E-mail: liuhm@zzu.edu.cn
multiple reaction monitoring Colour Online: See the article online to view Fig. 2 in colour.


C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
2440 J. Li et al. J. Sep. Sci. 2014, 37, 2439–2445

Figure 1. Chemical structures of

the eight synthetic dyes.

In this paper, we developed an HPLC–MS/MS method 2.3 Sample Pretreatment

for the determination of eight synthetic dyes in high protein
content matrices (bean and meat). The acetonitrile extrac- In order to validate the method, six different types of food
tion method developed here provides an efficient and conve- samples (dried beancurd, sausage, yuba, yellow croaker,
nient way of removing protein and other interferences from stewed chicken, and bean sauce) were purchased from a local
the food samples. The extract solution can be directly ana- market and stored at 4⬚C until they were processed. Dye-
lyzed with LC–MS/MS, thus, eliminate the complicated SPE spiked samples were prepared in a 15 mL centrifuge tube by
cleanup step. The method was validated by evaluating se- mixing 2 g of each matrix with a series of the eight-dye mixed
lectivity, linearity, recovery, and repeatability, and has been standard solutions at various concentrations, homogenized
applied for analysis of real samples. for 2 min, and kept at ambient temperature for 30 min before
extraction. Then, sample extraction was achieved by adding
6 mL acetonitrile solution. The tube was shaken by hand for
2 Materials and methods 30 s in order to allow the sample to mix thoroughly with
the solvent, and the extraction continued for 30 min in an
2.1 Chemical and reagents ultrasonic bath. The extraction solution was centrifuged at
12 000 rpm for 10 min to sediment the solids. Finally, the
HPLC-grade methanol and acetonitrile were purchased from extraction supernatant was filtered through 0.22 ␮m nylon
J. T. Baker (USA). HPLC-grade formic acid and ammonium membrane prior to LC–MS analysis. Blank samples were pre-
acetate were from Sigma–Aldrich (Poole, UK). Ethyl acetate treated in the same manner.
and acetone were of analytical reagent grade from Yongda
(Tianjin, China). The water was purified and deionized by
a water purification system (Millipore, Bedford, MA, USA). 2.4 Cleanup efficiency of SPE
The solvents for HPLC were filtered through a 0.45 ␮m nylon
membrane (Whatman, UK) and degassed in an ultrasonic In order to eliminate the co-extractives, Waters Oasis HLB
bath. Oasis HLB SPE cartridges (where HLB is hydrophilic- cartridge (hydrophilic–lipophilic balanced RP) was chosen.
lipophilic balance; 3 mL, 60 mg) were purchased from Waters For RP materials, the extracted supernatant was diluted with
(Milford, MA, USA). Standards of Sudan (I–IV), Para Red, water until the content of acetonitrile was no >5%. The su-
Rhodamine B, Chrysoidin and Auramine O were purchased pernatant was passed through the reconditioned cartridges,
from Sigma–Aldrich (St. Louis, MO, USA). The purities of all and then 2 mL of water was passed. The analytes were eluted
standards were >98%. The molecular structures of the eight with 2 mL of methanol at a rate of about 1–2 drops/s and then
synthetic dyes are shown in Fig. 1. evaporated under a stream of nitrogen gas at 45⬚C. Finally,
the residue was reconstituted with 500 ␮L of acetonitrile and
filtered through a 0.22 ␮m nylon membrane prior to LC–MS
2.2 Preparation of standard solutions analysis.

Standard stock solutions (100 ␮g/mL in acetonitrile) were

separately prepared for eight synthetic dyes, and the stan- 2.5 LC–MS/MS Conditions
dard solutions were stored in darkness at −20⬚C until use.
Mixed standard stock solution was prepared by mixing and A Waters HPLC system (Milford) including a 2695 separa-
diluting the standard stock solution of each dye with acetoni- tion module was controlled by Micromass Masslynx V4.1
trile to a final concentration of 10 ␮g/mL (Sudan (I–IV), Para software. Chromatographic separation was carried out by a
Red, Chrysoidin) and 1 ␮g/mL (Rhodamine B, Auramine Waters XTerra C18 RP column (2.1 × 150 mm, 5 ␮m) and
O). Standard working solutions were daily prepared in the a binary gradient, which included 5 mM ammonium acetate
determination process by acetonitrile/H2 O (50:50, v/v). buffer solution at pH 3.0 (mobile phase A) and methanol


C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2014, 37, 2439–2445 Other Techniques 2441

Figure 2. MRM chromatograms of the analytes

obtained from dried beancurd spiked standard
mix solution: (a) Sudan I (30 ␮g/kg); (b) Sudan II
(30 ␮g/kg); (c) Sudan III (30 ␮g/kg); (d) Sudan IV
(30 ␮g/kg); (e) Para Red (30 ␮g/kg); (f) Chrysoidin
(30 ␮g/kg); (g) Auramine O (3 ␮g/kg); (h) Rho-
damine B (3 ␮g/kg).

(mobile phase B). The linear gradient elution was pro- 3 Results and discussion
grammed as follows: 0–10 min, 45–100% B; 10–20 min, 100%
B; 20–25 min, 100–45% B. The flow rate was set at 0.2 mL/min 3.1 LC–MS/MS analysis
while the column temperature was maintained at 35⬚C. The
injection volume was 5 ␮L. Satisfactory separation and responses for all analytes were ob-
A triple quadrupole mass spectrometer (Quattro-Micro R
, tained under the gradient elution conditions described above.
Waters, Manchester, UK) equipped with an electrospray MRM chromatograms of the analytes in spiked bean sample
source operating in a positive-ion mode was used for de- are shown in Fig. 2. The cone voltages and collision energies
tection in MRM mode. The optimum conditions of the ESI used for MRM transitions of each compound were optimized
interface for all target analytes were as follows: a capillary to increase sensitivity, and the results are listed in Table 1.
voltage of 3.0 kV, a source temperature of 120⬚C, and a dry The product-ion mass spectra of the eight dyes are shown
temperature of 300⬚C. High purity N2 was used as both dry- in Fig. 3. The product ions were selected to be as specific as
ing gas with a flow rate of 50 L/h and nebulizing gas with possible, avoiding the use of common losses to prevent false
a flow rate of 450 L/h. The analysis was performed in just positives in the analysis of such complex matrices.
one time segment during the entire LC–MS run, which al-
lowed obtaining sufficient scans for each peak. The average
number of data points recorded across the chromatographic 3.2 Optimization of sample pretreatment
peaks of the analytes is 32. Each dye was analyzed using two
MRM transitions in order to improve accuracy. One tran- The development of an efficient sample extraction procedure
sition was used for quantification and qualification while is critical for accurate determination of dyes in food sam-
the other was used as a supplemental data for qualification. ples. Four extract solvents were evaluated in this experiment
For quantification, the most intense MRM transition was based on chemical properties of the eight dyes. The extract
selected. solvents include ethyl acetate, methanol, acetonitrile, and


C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
2442 J. Li et al. J. Sep. Sci. 2014, 37, 2439–2445

Table 1. Values of the instrumental settings optimized for the eight dyes tested

Compound Retention time Molecular weight Quantitative Qualitative Dwell Cone Collision
(min) (g/mol) transition (m/z) transition (m/z) time (s) voltage (V) energy (eV)

Sudan I 15.62 248.1 249.3/127.9 249.3/127.9 0.1 30 20

249.3/231.1
Sudan II 17.25 276.1 277.3/156.0 277.3/156.0 0.1 25 20
277.3/121.0
Sudan III 18.30 352.1 353.3/196.2 353.3/196.2 0.1 32 22
353.3/156.0
Sudan IV 20.72 380.2 381.3/224.2 381.3/224.2 0.1 32 23
381.3/209.2
Para Red 14.75 293.1 294.2/247.1 294.2/247.1 0.1 30 20
294.2/156.0
Chrysoidin 3.87 212.1 213.2/121.0 213.2/121.0 0.1 30 20
213.2/196.1
Auramine O 7.36 267.2 268.3/147.1 268.3/147.1 0.1 35 28
268.3/252.1
Rhodamine B 11.00 478.2 443.4/399.2 443.4/399.2 0.1 45 42
443.4/413.1

Table 2. Effect of different solvents on the extraction of eight syn-

thetic dyes from dried beancurd spiked at 10 ␮g/kg

Compound Recovery (%) ± SD

Ethyl acetate Acetonitrile Acetone Methanol

Sudan I 86.1 ± 0.2 103.9 ± 0.2 70.1 ± 0.2 89.5 ± 0.2

Sudan II 77.3 ± 0.1 110.4 ± 0.1 44.3 ± 0.1 79.2 ± 0.2
Sudan III 60.1 ± 0.2 110.6 ± 0.3 9.4 ± 0.1 35.5 ± 0.1
Sudan IV 47.3 ± 0.3 80.4 ± 0.1 9.7 ± 0.1 13.4 ± 0.1
Para Red 69.7 ± 0.2 95.7 ± 0.1 64.9 ± 0.2 87.3 ± 0.1
Chrysoidin 11.3 ± 0.1 112.0 ± 0.2 5.9 ± 0.1 124.3 ± 0.2
Auramine O 2.4 ± 0.1 102.5 ± 0.2 35.5 ± 0.1 137.7 ± 0.3
Rhodamine B 49.6 ± 0.1 98.1 ± 0.1 79.4 ± 0.3 124.2 ± 0.2

n = 5.

acetone. For bean and meat samples, acetonitrile showed

good extraction efficiency in terms of recovery and repro-
ducibility for the eight dyes. Most of the interference proteins
present in food samples were precipitated in the presence of
acetonitrile. When acetone was used for extraction solvent,
the recoveries of Sudan III, Sudan IV, and Chrysoidin from
the bean sample spiked at 10 ␮g/kg were <10%. The reason
may be that acetone caused agglomeration and precipitation
of the sample, and thus resulted in incomplete extraction of
certain dyes from the sample. The extraction efficiency of
different solvents is shown in Table 2.
In this experiment, further purification of the extract
solution with SPE (Waters Oasis HLB cartridge) was also
considered. With the extra SPE purification step, the chro-
Figure 3. The product-ion mass spectra of (a) Sudan I;
matograms showed little improvement, impurities observed
(b) Sudan II; (c) Sudan III; (d) Sudan IV; (e) Para Red; (f) Chrysoidin; during the first 3.0 min of HPLC elution cannot be re-
(g) Auramine O; (h) Rhodamine B. moved with SPE. Since all analytes of interest eluted after
3.5 min, the existence of these low retention interferences did
not affect the detection. Moreover, these eight dyes possess


C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2014, 37, 2439–2445 Other Techniques 2443

Table 3. Linear regression parameters of eight synthetic dyes from dried beancurd calibration curves

Compound Concentration range (ng/mL) Standard calibration curve Matrix-matched calibration curve Matrix effect (%)a)

Slope y-intercept R2 Slope y-intercept R2

Sudan I 2–100 59.8977 27.6516 0.996367 58.5874 −21.5117 0.996572 97.8

Sudan II 2–100 90.4477 60.5585 0.999451 88.8436 −21.5494 0.996397 98.2
Sudan III 2–100 22.1337 12.8558 0.994318 16.9475 −0.801824 0.994640 76.6
Sudan IV 2–100 31.3871 29.4571 0.996360 26.5527 4.8716 0.993294 84.6
Para Red 2–100 15.9359 −3.60478 0.993556 14.5869 13.7045 0.996691 91.5
Chrysoidin 2–100 95.6412 −84.233 0.992412 106.0690 −88.8734 0.997853 110.9
Auramine O 0.2–10 336.2780 13.9875 0.997784 311.3110 −20.7922 0.994136 92.5
Rhodamine B 0.2–10 489.7240 39.8837 0.998603 419.6440 −12.7093 0.992537 85.7

a) Expressed as the ratio between the slope of matrix-matched and solvent calibration curves multiplied by 100. A value of >100% indicates
signal enhancement, and a value of <100% indicates signal suppression.

different chemical characteristics, especially Chrysoidin, Table 4. LOD and LOQ values of the studied dyes in the different
which belongs to highly polar dyes; no single SPE column matrices

can achieve satisfactory retention and recovery performance

for each dye. Thus, acetonitrile was chosen as the extraction Analyte Matrix LOD LOQ
(␮g/kg) (␮g/kg)
solvent, and without SPE, the whole pretreatment process is
very convenient and effective.
Sudan I, Para Red Dried beancurd, bean sauce, 0.03 0.1
yuba, yellow croaker,
stewed chicken
3.3 Performance of the HPLC–MS/MS method Sudan II, Sudan III, Dried beancurd, bean sauce, 0.45 1.5
Sudan IV yuba, yellow croaker,
stewed chicken, sausage
3.3.1 Selectivity Chrysoidin Dried beancurd, bean sauce, 0.75 2.0
yuba, yellow croaker,
Selectivity was evaluated by analysis of the blank sample. No stewed chicken, sausage
interferences were observed in corresponding retention times Auramine O, Dried beancurd, bean sauce, 0.03 0.1
of target compounds by comparing chromatograms of spiked Rhodamine B yuba, yellow croaker,
samples and blank samples. From Fig. 2, it is indicated that stewed chicken, sausage
the present method has good selectivity for the studied dyes.

ment values for bean sample, which were in the range of

3.3.2 Linearity and matrix effects
76.6–110.9%. In view of these results, we can confirm that
these distributions depend not only on the matrix but also on
Linearity was evaluated using solvent and matrix-matched
the combination of the compound and the matrix. This fact
calibration curves at six concentration levels of 2–100 ng/mL
highlights the importance of performing quantification with
for Sudan (I–IV), Para Red, and Chrysoidin and 0.2–10 ng/mL
matrix-matched calibration curves.
for Auramine O and Rhodamine B. Calibration curves were
fitted by least-squares linear regression using 1/X2 as the
weighing factor. The linearity of the analytical response for 3.3.3 Detection and quantification limit
all studied dyes within the range of two orders of magnitude
was very good, with correlation coefficients (R2 ) >0.99 in all The LOD was determined as the sample concentration that
cases. produced a peak with three times the S/N, and the LOQ was
The matrix effect was also evaluated during the valida- calculated as the sample concentration that produced a peak
tion of the method, as it is known that signal suppression or with ten times the S/N. The LOD and LOQ values for the
enhancement as a result of matrix effect can severely compro- studied dyes in six matrices are shown in Table 4, which were
mise quantitative analysis of the compounds at trace levels as in the ranges of 0.03–0.75 and 0.1–2.0 ␮g/kg, respectively.
well as method reproducibility. The matrix effect was stud- The LOD values were equal or lower than those from the
ied by the ratio between the slope of matrix-matched and previous reports for Sudan (I–IV) and Para Red [19, 20], and
solvent calibration curves multiplied by 100 (signal suppres- lower than the LOD values for Auramine O and Rhodamine
sion enhancement, %). A value of >100% indicates signal B in refs. [21, 22, 28]. Better limit was only achieved by Liu
enhancement, and a value of <100% indicates signal sup- et al. [32], who used UHPLC–MS (LOQ of 0.04 ␮g/kg for
pression [31]. Table 3 shows the signal suppression enhance- Rhodamine B).


C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
2444 J. Li et al. J. Sep. Sci. 2014, 37, 2439–2445

3.3.4 Precision, recovery, and repeatability

30 ␮g/kg 120 ␮g/kg

12 ␮g/kg
91.6 (3.5)

71.2 (5.4)
79.1 (6.4)
89.8 (2.8)

79.7 (3.4)

96.5 (6.5)

80.3 (8.6)
86.4 (6.2)
The precision of the method was verified by measuring re-
coveries from spiked blank samples of the different matrices

88.3 (6.4)

78.2 (2.7)
91.4 (4.7)
97.9 (1.0)

77.6 (2.8)

79.1 (9.1)
84.7 (1.9)
90.4(3.2)
3 ␮g/kg
investigated (dried beancurd, sausage, yuba, yellow croaker,
R (%); RSD (%)
stewed chicken, and bean sauce) at three concentration levels,

102.8 (11.2)

0.75 ␮g/kg
90.4 (13.2)

94.1 (10.6)

112.7 (3.2)
7.5 ␮g/kg 7.5, 30, 120 ␮g/kg for Sudan (I–IV), Para Red, and Chrysoidin;
Sausage

93.2 (9.2)

93.8 (9.1)
93.8 (4.3)

87.3 (3.3)
and 0.75, 3, 12 ␮g/kg for Auramine O and Rhodamine B.
112.3 (11.9) Mean recovery data and RSDs are given in Table 5, which
120 ␮g/kg

102.0 (8.5)

74.5 (15.2)
expressed the repeatability of the extraction method. Recov-

12 ␮g/kg
78.9 (4.5)
94.9 (9.3)

78.8 (8.6)

83.1 (3.3)

79.3 (7.4)
eries were between 71.2 and 116.9% for all analytes in the
studied matrices, the RSDs were calculated in terms of intra-
113.3 (11.5)

105.2 (12.3)
86.9 (14.7)

94.1 (14.5)

74.2 (14.9)
day repeatability and interday reproducibility (three alterna-
30 ␮g/kg

86.9 (5.4)
80.2 (8.0)

79.3 (3.2)
3 ␮g/kg
tive days). Intraday and interday RSDs were between 1.0 and
Stewed chicken

R (%); RSD (%)

15.2% and 2.3 and 14.9%, respectively. The overall precision

101.4 (11.9)
101.4 (10.4)

116.1 (11.9)
0.75 ␮g/kg

and accuracy of the method were sufficient for the quantifi-

106.3 (9.6)
90.6 (10.8)
87.6 (12.8)
7.5 ␮g/kg

79.5 (2.7)

79.7 (6.9)

cation of the eight dyes in real samples.

120 ␮g/kg

91.1 (12.4)

12 ␮g/kg
72.3 (1.5)
94.8 (4.6)
73.2 (4.8)
88.4 (6.8)
90.0 (7.0)
85.5 (3.3)

78.2 (3.9)

3.4 Analysis of real samples

102.0 (14.4)
89.6 (14.4)

88.5 (10.8)

Thirty bean and meat products collected from different mar-

30 ␮g/kg

86.9 (6.9)
95.2 (4.5)

85.6 (4.2)
91.8 (7.8)
89.4 (7.3)
3 ␮g/kg

kets of local city were applied for validating the proposed

Yellow croaker

R (%); RSD (%)

method. To identify and confirm the analytes, two MRM

0.75 ␮g/kg
116.9 (8.2)
95.2 (12.9)
82.2 (11.5)
87.8 (12.7)

92.3 (13.9)

114.1 (8.4)

transitions for each compound were monitored during LC–

120 ␮g/kg 7.5 ␮g/kg

97.4 (8.5)

89.5 (3.5)

MS/MS analysis. No peaks of target analytes were detected

in the studied samples, which demonstrates the safety of the
83.9 (11.5)
101.3 (2.2)
100.4 (2.5)

foods tested.
12 ␮g/kg
82.2 (5.5)

62.8 (1.8)
82.2 (3.5)
94.6 (5.8)

80.3 (7.7)
110.7 (10.3)
109.7 (4.2)
93.8 (12.3)
89.1 (10.5)

88.9 (11.2)
108.2 (4.6)

4 Concluding remarks
30 ␮g/kg

99.2 (2.9)

87.7 (9.4)
3 ␮g/kg
Table 5. Recovery values and RSDs of the target compounds in different matrices (n = 5)

R (%); RSD (%)

This work developed a sensitive and effective HPLC–MS/MS

Bean sauce

109.8 (13.6)

0.75 ␮g/kg
95.9 (13.7)

103.2 (3.6)
7.5 ␮g/kg

method with acetonitrile extraction that provides satisfactory

94.9 (2.7)
93.5 (8.7)

77.8 (2.5)
81.2 (4.9)

78.7 (3.7)

specificity and sensitivity for the simultaneous determina-

tion of eight synthetic dyes in foods. Analysis of dye-spiked
112.0 (13.9)
120 ␮g/kg

102.5 (5.7)

101.8 (4.1)

samples with the simple extraction procedure developed here

12 ␮g/kg
81.4 (3.6)
88.9 (4.0)
97.7 (2.6)
80.9 (1.8)

93.8 (2.5)

showed that the recoveries ranged from 71.2 to 116.9%, which

can meet regular analytical requirements. And the method
92.4 (13.3)

86.6 (12.5)

has a good repeatability and high accuracy with low detection

7.5 ␮g/kg 30 ␮g/kg

85.7 (5.7)
97.3 (3.6)
92.3 (3.8)
78.5 (4.5)
84.5 (7.2)

90.2 (2.1)
3 ␮g/kg

limits. The proposed HPLC–MS/MS method is effective for

R (%); RSD (%)

detecting illegal dyes in bean and meat products.

0.75 ␮g/kg
93.5 (12.6)

98.0 (12.3)
96.8 (12.3)
88.3 (5.2)
94.9 (8.6)

79.8 (5.1)
86.3 (4.2)

93.1 (3.2)
Yuba

This work was financially supported by the National Natural

Science Foundation of China (grant no. 81273393), the Natural
113.9 (12.2)
120 ␮g/kg

83.6 (12.2)
12 ␮g/kg

Science Foundation of Henan Province (grant no. 12A350009),

80.3 (8.4)
79.2 (4.2)

93.6 (5.8)
79.4 (7.1)
81.2 (5.0)

90.1 (2.9)

and the Scientific and technological project of Henan Province

(grant no. 132102110093).
30 ␮g/kg

89.6 (6.3)
88.2 (2.8)
81.7 (9.2)
78.5 (1.8)

77.2 (1.5)
80.7 (5.0)
78.9 (7.2)

80.6 (3.2)
3 ␮g/kg
Dried beancurd

R (%); RSD (%)

The authors have declared no conflict of interest.

0.75 ␮g/kg
112.2 (7.3)
94.4 (12.9)
106.6 (4.7)
7.5 ␮g/kg

Auramine O 77.6 (7.4)

85.9 (9.1)
94.5 (1.7)

78.3 (0.8)

Rhodamine 81.6 (1.6)

5 References
Chrysoidin
Para Red
Sudan III
Sudan IV
Sudan II
Sudan I

[1] Vidotti, E. C., Cancino, J. C., Oliveira, C. C., Anal. Sci.

Dyes

2005, 21, 149–153.


C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2014, 37, 2439–2445 Other Techniques 2445

[2] European Parliament and Council Directive 94/36/EC of [17] Long, C., Mai, Z., Yang, X., Zhu, B., Xu, X., Huang, X.,
30 June 1994 on colours use in foodstuffs, Off. J. EC No. Zou, X., Food Chem. 2011, 126, 1324–1329.
L237/13 10/9/94. [18] Sun, S., Wang, Y., Yu, W. Z., Zhao, T. Q., Gao, S. Q., Kang,
[3] Dipalma, J. R., Am. Fam. Physician 1990, 42, 1347– M. Q., Zhang, Y. P., Zhang, H. Q., Yu, Y., J. Sep. Sci. 2011,
1350. 34, 1730–1737.
[4] Sasaki, Y. F., Kawaguchi, S., Kamaya, A., Ohshita, M., [19] Sun, H. W., Wang, F. C., Ai, L. F., J. Chromatogr. A 2007,
Kabasawa, K., Iwama, K., Taniguchi, K., Mutat. Res. 2002, 1164, 120–128.
519, 103–119. [20] Li, C., Wu, Y. L., Shen, J. Z., Food Addit. Contam. A 2010,
[5] McCann, D., Barrett, A., Cooper, A., Crumpler, D., Dalen, 27, 1215–1220.
L., Grimshaw, K., Lancet 2007, 370, 1560–1567. [21] Ferrer Amate, C., Unterluggauer, H., Fischer, R. J.,
[6] IARC, Monographs on the Evaluation of the Carcino- Fernández-Alba, A. R., Masselter, S., Anal. Bioanal.
genic Risk of Chemicals to Man: Some Aromatic Azo Chem. 2010, 397, 93–107.
Compounds, Vol. 8, Lyon, France 1975, pp. 224–231. [22] Zhang, H., Liang, L., He, Z., Shi, L., J. Chin. Mass. Spec-
[7] EFSA, EFSA J., 2005, 263, 1–71. trom. Soc. 2010, 31, 48–52.
[8] GB2760-2011, Hygienic standards for the use of food [23] Fang, G. Z., Wu, Y., Dong, X. M., Liu, C. C., He, S. Y.,
additives, Ministry of Health of the People’s Republic of Wang, S., J. Agric. Food. Chem. 2013, 61, 3834–3841.
China, P.R. China 2011. [24] Chen, D. M., Li, X. Q., Tao, Y. F., Pan, Y. H., Wu, Q. H.,
[9] Andrade, F. I. D., Guedes, M. I. F., Vieira, I. G. P., Mendes, Liu, Z. L., Peng, D. P., Wang, X., Huang, L. L., Wang, Y. L.,
F. N. P., Rodrigues, P. A. S., Maia, C. S. C., Avila, M. M. M., J. Chromatogr. B 2013, 939, 45–50.
Ribeiro, L. D., Food Chem. 2014, 157, 193–198. [25] Zou, T. T., He, P. L., Yasen, A., Li, Z., Food Chem. 2013,
[10] Soponar, F., Mot, A. C., Sârbu, C., J. Chromatogr. A 2008, 138, 1742–1748.
1188, 295–300. [26] Chen, G., Miao, S., J. Agric. Food Chem. 2010, 58, 7109–
[11] Capitan-Vallvey, L. F., Fernandez, M. D., de Orbe, I., 7114.
Avidad, R., Talanta 1998, 47, 861–868. [27] Makarov, A., Scigelova, M., J. Chromatogr. A 2010, 1217,
[12] Fuh, M. R., Chia, K. J., Talanta 2002, 56, 663–671. 3938–3945.
[13] Gianotti, V., Angioi, S., Gosetti, F., Marengo, E., J. Liq. [28] Feng, Y. C., Jia, L., He, Y. H., Wang, J. F., Liu, Y., Fan,
Chromatogr. Relat. Technol. 2005, 28, 923–937. X. J., Chin. J. Chromatogr. 2013, 31, 1021–1027.
[14] Zhu, Y. H., Zhao, B., Xiao, R. Q., Yun, W., Xiao, Z. J., [29] Jia, W., Chu, X. G., Ling, Y., Huang, J. R., Lin, Y. H., Chang,
Tu, D. W., Chen, S. Q., Food Chem. 2014, 145, 956– J., J. Sep. Sci. 2014, 37, 782–791.
962. [30] Li, J., Ding, X. M., Liu, D. D., Guo, F., Chen, Y., Zhang,
[15] Enrı́quez-Gabeiras, L., Gallego, A., Garcinuno, R. M., Y. B., Liu, H. M., J. Chromatogr. B 2013, 942, 46–52.
Fernández-Hernando, P., Durand, J. S., Food Chem. 2012, [31] Matuszewski, B. K., Constanzer, M. L., Chavez-Eng, C. M.,
135, 193–198 Anal. Chem. 2003, 75, 3019–3030.
[16] Bonan, S., Fedrizzi, G., Menotta, S., Elisabetta, C., Dyes [32] Liu, R. Y., Hei, W. J., He, P. L., Li, Z., J. Chromatogr. B
Pigments 2013, 99, 36–40. 2011, 879, 2416–2422.


C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com