See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/40033558

Simultaneous Identification of Nine Carcinogenic Dyes from Textiles by Liquid

Chromatography/Electrospray Ionization Mass Spectrometry via
Negative/Positive Ion Switching Mode

Article  in  European Journal of Mass Spectrometry · December 2009

DOI: 10.1255/ejms.1032 · Source: PubMed

CITATIONS READS

15 2,801

3 authors, including:

Cheng Sun
Nanjing University
303 PUBLICATIONS   12,655 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Cheng Sun on 09 May 2017.

The user has requested enhancement of the downloaded file.

Y. Ding, C. Sun and X. Xu, Eur. J. Mass Spectrom. 15, 705–713 (2009) 705
Received: 27 August 2009 n Accepted: 23 September 2009 n Publication: 29 September 2009

European
Journal
of
Mass
Spectrometry

Simultaneous identification of nine

­carcinogenic dyes from textiles by liquid
chromatography/electrospray ionization
mass spectrometry via negative/positive ion
switching mode
Youchao Ding,a,b Cheng Suna,* and Xinhua Xub
a
State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, 22 Hankou Road,
Nanjing 210093, PR China. E-mail: envidean@nju.edu.cn
b
Textile Laboratory, Jiangsu Entry-Exit Inspection and Quarantine Bureau, Nanjing 210001, PR China

A method has been established by using liquid chromatography/electrospray ionization tandem mass spectrometry (LC/ESI/MS/MS) for
simultaneous separation and identification of nine carcinogenic dyes (Acid Red 26, Direct Blue 6, Direct Black 38, Direct Red 28, Basic
Red 9, Basic Violet 14, Disperse Blue 1, Disperse Orange 11 and Disperse Yellow 3) prohibited in textile materials under EU 2002/371/
EC decision. Three selected reaction monitoring (SRM) transitions and negative/positive ion switching mode in one single analysis was
applied to differentiate between the dyes of different classes. A lower limit of detection was achieved at 0.025–0.25 mg kg–1 of positive ESI
dyes and 0.005–0.025 mg kg–1 of negative ESI dyes. The accurate and sensitive identification of the nine dissimilar analytes was achieved
by combining the characterized ions and retention time of the standards, for example, the characterized ions of [M – xNa]x–, [M – Cl]+ and
[M + H]+, respectively were detected for the the sulfonated (acid and direct), basic and disperse dyes. The chromophore groups of azo,
triphenylmethane and anthraquinone in precursor ions were found to have fragmented in SRM mode. These results demonstrated a
sensitive, accurate and rapid identification.

Keywords: carcinogenic dyes, simultaneous identification, liquid chromatography, electrospray ionization mass spectrometry, textiles

Introduction
Synthetic dyes are widely used as coloring agents in commonly
used textile products, for example, clothes, bedding and deco-
rating materials. However, some dyes are toxic, mutagenic
and carcinogenic to life,1 and will transfer from the textile
to the skin of one who comes into contact with it. Therefore,
the safety of the textile dyes has become a great concern in
the world. The European Communities issued the Directive
1999/43/EC2 to prohibit the marketing and use of the carcino-
genic dyes in 1999 and the Decision 2002/371/EC3 to restrict
the use of nine carcinogenic dyes in eco-labeled textiles three
years later. The prohibited carcinogenic dyes are Direct Black
38 (DB38), Direct Blue 6 (DB6), Direct Red 28 (DR28), Acid Red
26 (AR26), Basic Red 9 (BR9), Basic Violet 14 (BV14), Disperse
Blue 1 (DB1), Disperse Orange 11 (DO11) and Disperse Yellow
3 (DY3), as shown in Figure 1. The International Association
for Research and Testing in the Field of Textile Ecology also
proclaimed that textile products would not be awarded with the
ecological label of Oeko-Tex if they contained one or more of

ISSN: 1469-0667 © IM Publications LLP 2009

doi: 10.1255/ejms.1032 All rights reserved
 706 Simultaneous Identification of Nine Dyes from Textiles

OH NH 2
Direct Black 38
N=N NH2
N=N CAS NO.:
N=N
NaO 3S
SO3Na
1937-37-7
NH 2 C.I. 30 235
M.W. 781.73
N aO 3 S SO 3 N a N aO 3 S SO 3 N a Direct Blue 6
CAS NO.:2602-46-2
N =N N =N
C.I. 22 610
NH2 OH OH NH2
M.W. 934.76
CG:azo

SO 3 N a SO 3 N a Direct Red 28
CAS NO.:573-58-0
N =N N =N C.I. 22 120
NH2 NH2 M.W. 696.67
CG:azo
Disperse Blue 1
CH 3 OH SO 3Na Acid Red 26 NH2 O NH2
CAS NO.:2475-45-8
CAS NO.:
CH 3 N=N C.I. 64 500
3761-53-3
M.W. 268.27
C.I. 16 150
CG:anthraquinone
M.W. 480.42 NH2
O
NH2
SO 3Na
O NH 2 Disperse Orange 11
CH3
O
Disperse Yellow 3
CH 3 CAS NO.:82-28-0
CAS NO.:2832-40-8
N=N NH C CH3 C.I. 60 700
C.I. 11 855
M.W. 237.26
OH M.W. 269.30 O
CG:anthraquinone
CG:azo
NH2
NH2 Basic Violet 14
Basic Red 9 CH3
CAS NO.:632-99-5
CAS NO.:569-61-9
C.I. 42 510
C.I. 42 500
M.W. 337.85
M.W. 323.83 H2N C NH2 Cl
H2N C NH2 Cl CG:
CG:
Figure 1. The name, chemical structure, CAS registered number, Color Index, Molecular Weight (MW) and Chromophore Group(CG)of
the nine prohibited carcinogenic dyes.

the nine carcinogenic dyes.4 Thus, development of a universal
analytical technique which is able to accurately identify the
nine dyes in textiles is of great importance.
Thin-layer chromatography (TLC) was historically applied
for the analysis of dyes.5 However, it is relatively insensitive
and lacks reproducibility. These limitations are significantly
alleviated by application of high-performance liquid chroma-
tography (HPLC). The use of HPLC for the analysis of textile
dyes was introduced in the 1980s6,7 and gained rapid develop-
8 of 15
ment thereafter. It is reported that HPLC, coupled with diode
array detector (DAD), had been used to detect the nine carci-
nogenic dyes.8 However, it is not sufficient for their structural
identification by spectrophotometry because of much spectral
interference arising from intermediates, homologs and auxil-
iary agents of dyes. Mass spectrometry (MS) is a powerful
detector which offers accurate masses of precursors and
fragments.9 Combing the high performance separation and
unambiguous mass identification, liquid chromatography/
tandem mass spectrometry (LC/MS/MS) has been applied to
the detection of sulfonated10 azo11 and anthraquinone12 and
triphenylmethane13 dyes in food and environment in recent
years. Furthermore, the compounds, as shown in Figure1,
include four anionic sulfonated (three direct and one acid) dyes,
two cationic basic dyes and three non-ionic disperse dyes,
differing greatly in molecular structure, chemical behavior
and dyeing application. Also, the dyes can be distinguished by
their special chromophore groups: AR26, DR28, DB6, DB38
and DY3 are azo dyes, BR9 and BV14 are triphenylmethanes,
and DB1 and DO11 are anthraquinones. However, to the best
of our knowledge, there was no report about the ­simultaneous
 Y. Ding, C. Sun and X. Xu, Eur. J. Mass Spectrom. 15, 705–713 (2009) 707
detection of the nine compounds of interest by LC/MS/MS. The
current LC/MS/MS methods for analysis of dyes mainly focus
on one class of dye, and simultaneous detection of different
classes of dye has proved to be difficult and seldom available.
The objective of this work was to develop a confirmatory
method for simultaneous identification of three different
(anionic, cationic and non-ionic) classes of a total nine carcino-
genic textile dyes. The successful separation and identification
of nine dyes in textiles was achieved by LC/MS/MS interfaced
with negative/positive ion-switching electrospray ionization
(ESI) in selected reaction monitoring (SRM) mode which was
applied to perform a single run analysis. MS parameters were
optimized and particular attention was paid to the ESI ioniza-
tion behavior and the selected MS/MS fragmentation behavior
of dyes.
Experimental
Chemicals and materials
Ultra-pure water, prepared using a Milli-Q purification system
(Millipore, Boston, USA), and HPLC grade solvents (methanol
and acetonitrile) (Merck, Darmstadi, Germany) were used in all
procedures. Ammonium acetate (Merck, Darmstadi, Germany)
used to prepare buffer solution was of HPLC grade. All nine
carcinogenic dyes were purchased from Dr Ehrenstorfer
GambH (Augsburg, Germany) as standard reference materials
with certificated purity.
Instrumentation
A Thermo Scientific LC/MS/MS system (ThermoFisher, CA,
USA) consisting of an Accela pump with on-line degasser, an
Accela autosampler and a TSQ Quantum Access tandem mass
spectrometer equipped with a standard ESI source was used.
Separation was accomplished at 50°C on a Hypersil GOLD
column 5 µm × 100 mm × 2.1 mm i.d. column (ThermoFisher,
CA, USA) protected with a C18 pre-column (Phenomenex, CA,
USA) at a flow rate of 0.3 mL min–1. The mobile phase composi-
tion was 5 mM aqueous ammonium acetate (A) and acetrontrile
(B), the gradiant program was: 5% of B holding for 1 min, then
increased to 80% B within 4 min and kept at 80% of B for
5 min, then returned to the starting condition within another
5 min. The injection volume was 10 µL. The time-scheduled
mass acquisition was performed in SRM mode via ESI oper-
ated alternatively in negative and positive modes in a single
period of time containing two segments. In the first segment
(up to 5.65 min) ESI was set to negative mode for detection
of the anionic dyes: AR26, DB38, DB6 and DR28, while in the
second segment (after 5.65 min) ESI was set to positive mode
for detection of both cationic and non-ionic dyes: BR9, BV14,
DB1, DO11 and DY3.
In order to optimize the MS operation parameters, standard
solutions of dye at 1.0 mg L–1 were directly introduced into the
mobile phase at a flow rate of 5 µL min–1. The spray voltage
was set at –4 kV for ESI– and +4.5 kV for ESI+ mode. The
optimal velocity of sheath and auxiliary gases was 16.0 L min–1
and 3.2 L min–1 for the entire period. The capillary temperature
was set at 350°C. Three SRM transitions were monitored for
the identification of each analyte. The transitions and their
corresponding collision voltages were reported in Table 1. The
inter-channel delay was 5 ms and the dwell time was set to
0.2 s.
Sample preparation
The sample was cut into small pieces (below 1 cm × 1 cm), of
which 0.5 g was weighed into a glass tube. Ten mL of methanol
was added into the tube, which was then sealed and placed
into a (70 ± 2)°C water bath (AquaWave Ultrasonic Cleaner
Elma, Stuttgart, Germany) for sonication. The bath released
ultrasonic energy at 40 kHz and 420 W for 30 min. A portion
of the centrifuged extract (0.5 mL) was mixed with 1.5 mL
water and analyzed by LC/MS/MS after filtration by using
­polytetrafluoroethylene (PTFE) membrane.
Preparation of standard solutions and
reference samples
Standard reference materials were accurately weighed to
prepare 200 µg mL–1standard stock solutions in methanol,
which were kept refrigerated at 2–4°C and could be used
for six months. Due to the different charge state in solution,
the basic dyes could not coexist with the acid or direct dyes.
According to the detection mode of each dye for ESI, two
groups of working standard solutions were prepared: (I) AR26,
DB38, DB6 and DR28 and (II) BR9, BV14, DB1, DO11 and DY3,
and diluted with methanol: water (1 : 3, v / v) to construct six-
point instrumental calibration curves. To evaluate the matrix
effect on the LC/MS/MS analysis, the matrix-matching stan-
dards were prepared by adding the standard solutions into
blank sample extracts which were at the same concentrations
and were analyzed under identical instrumental conditions to
construct matrix-matching calibration curves.
Reference samples were provided by Shanghai Dyes
Research Institute of China, with no expected concentration
of dyes given. These references were employed for the method
application. To prepare reference samples, five pieces of color-
less fabric were uniformly dyed by commercial dyes in lab-
scale machines with DB38 on cotton (RS1), BV14 on acrylic
(RS2) and DY3 on polyester (RS3) fabric, respectively.
Results and discussion
ESI/MS spectra
In this study, it was found that the polar functional groups
in dye molecules affect their ionization behavior in ESI/MS
spectra. For sulphonated dyes, sodium-deducted molecules
of [M – xNa]x– with observed charges equal to the number of
sodium, were characterized as precursor ions, i.e. single-
charged m/z 217 of AR26, di-charged m/z 325 of DR28,
di-charged m/z 368 of DB38 and tetra-charged m/z 210 of
DB6. Figure 2 showed the negative ESI/MS spectra of DB38, in
which sodium-deducted ions were the typical ions. However,
 Figure 2
708 Simultaneous Identification of Nine Dyes from Textiles
m/z 105
Figure 2. Negative ESI/MS spectra of Direct black 38.
it was different from those in previous works in which the
typical ionization ions of a sodium sulphonated compound
was sulphonate-deducted molecules of [M – SO 3 – Na] –, 10
[M-SO2-Na]– or [M – SO3 + Na]+.14 The charges of ions can be
determined from the mass difference of the isotopic ions
determined by higher resolution scans with narrower scan
width. For example, the tetra-charged m/z 210 can distinguish
isotopes with four-charges difference as Dm/z = 0.25.
For chlorinated dyes, BR9 and BV14 were observed with
characteristic ions of [M – Cl]+, i.e. m/z 288 of BR9 and m/z
302 of BV14. For non-ionic compounds, the three disperse
dyes were observed with protonated-molecules of [M + H]+ for
each attaching a donated proton, i.e. m/z 269 of DB1, m/z 238
of DO11 and m/z 270 of DY3. In ESI/MS spectra, the negative
precursor ions of DB6, DB38 and DR28 had lower response
than the other positive ones, probably resulting from more
negative charges carried that were not easily nebulized nor
transferred through the capillary.15
It should be mentioned that nitrogen atoms are included
in the molecules of all analytes, which provides us with a
reason for interpreting mass behavior using the so-called
nitrogen rule. As it is known, the hard ionization technique
of electron ionization (EI) generally produces odd-electron
ions (OE•+) with unpaired electrons outside which are conven-
tionally regarded as molecular ions (M+). In contrast, the soft
ionization technique of ESI produces even-electron ions (EE+/–)
with paired electrons. As an explanation of the nitrogen rule
for EE+/–,16 odd molecular weight (MW) means even number
m/z 315
of nitrogen atoms. In this study, in ESI mode, AR26, DR28
and DB1 containing an even number of nitrogen atoms were
ionized to yield precursor ions with odd MW; however, DB38,
BR9, BV14, DO11 and DY3 containing odd numbers of nitrogen
atoms were ionized to yield precursor ions with even MW, all
of which followed the rule.
ESI/MS/MS spectra
ESI/MS/MS spectra of the nine dyes (Figure 3) were obtained
by further conducting collision-induced dissociation of the
selected precursor ions. When interpreting the spectra, it was
interesting to find the relationship between the fragmenta-
tion behavior and the presence of an individual chromophore
group. Three typical examples were illustrated in Figures 4(a),
(b) and (c) to explain the monitored transitions, resulting from
the cleavage of chromophore groups of azo, anthraquinone
and triphenylmethane, respectively.
The azo form, subjected to a hydrazo-azo tautomerization,
was likely to split at the C–N bond, while homolytic cleavage
of the azo bond occurs in the hydrazo form 9 of the dyes.
Figure 4(a) showed the fragmentation scheme for DR28, in
which di-charged m/z 325 yielded two single-charge radical
anions of m/z 416 and m/z 234 after the split of azo bond,
while the former product ion yielded another radical anion
7 of 14
of m/z 401 after the split of C–N bond. These radical anions
with unpaired electrons on the conjugated polyaromatic
system are regarded as odd-electron ions (OE•–) and follow
the so-called nitrogen rule, as another explanation, that odd
 Y. Ding, C. Sun and X. Xu, Eur. J. Mass Spectrom. 15, 705–713 (2009) 709
Figure 3. Product-ion spectrum of: (a) direct blue6; (b) acid red26; (c) direct red28; (d) direct black38; (e) disperse blue1; (f) basic red9;
(g) basic violet14; (h) disperse yellow3 and (i) disperse orange11.
MW automatically means odd number of nitrogen atoms in
the ­structure.16
In anthraquinone dye, hydrogen rearrangement takes place
because of the electron-donating amino group in 1-position on
the ring system.17 Besides, the typical i cleavage takes place
near carbonyl in the center ring, which is induced by the positive
charge and involves two-electron transference, as in the case of
DB1 shown in Figure 4(b), that EE+ m/z 269 yielded EE+ m/z 107.
An EE+ is preferentially split to yield another EE+ and neutral
molecule EE0.16 For example, EE+ m/z 269 yielded EE+ m/z 241
with a neutral loss of carbon monoxide (EE0 28 Da). M. Holcapek
had found this neutral loss of shortening of an aromatic ring by
one carbon atom for anthraquinone dyes.14 However, Karni and
Mandelbaum argued that EE+ dissociation could yield OE•+ and
OE•0 (odd-electron radicals).18 In this way, EE+ m/z 269 yielded
OE•+ m/z 253 with the loss of an amino radical (OE•0 16 Da).
The triphenylmethane group is a conjugated and coplanar
delocalized system. It was deemed that the positive charge
was added not on one atom but on the whole system of the
Figure 4
8 of 13
dye.13 For BR9, shown in Figure 4(c), EE+ m/z 288 yielded OE•+
m/z 272 with neutral loss of amino radical (OE•0 16 Da), and
concurrently yielded EE+ m/z 195 with neutral loss of anline
(EE0 93 Da). EE+ m/z 195 had a further cleavage to yield EE+ m/z
168 with neutral loss of hydrogen cyanide (EE0 27 Da).
Although MS and MS/MS in general are destructive, func-
tional groups have been proven to have active effects on the
fragmentation of analytes.10,12,14 From the discussion above in
this study, the chromophore groups of the dyes were fractured
under SRM. Besides, an EE+ of the dye was found with two
potential fragmentation pathways to produce product ions as:
another EE+ after loss of a neutral molecule EE0 and an OE•+
after loss of a neutral radical OE•0. Moreover, both EE+ and OE•+
followed their corresponding nitrogen rules with the reversed
explanation on MW and number of nitrogen atoms.
Liquid-column separation
Reversed-phase was utilized as the separation mode because
of its excellent performance for separating dyes in previous
 710 Simultaneous Identification of Nine Dyes from Textiles

SO 3
-
SO 3
- a reversed-phase column.9,11 Some volatile ion-pair reagents
such as tetrabutylammonium acetate and dihexylammonium
(a)
N =N N =N
acetate were tried in this study. It was disappointing that these
N H2
cationic reagents competitively interfered with the cationic
N H2
m/z(6 5 0 /2 = 3 2 5 )
and non-ionic dyes in ionization, although chromatographic
S O 3- S O 3- retention of the anionic dyes was improved. Experiments
showed that a gradient mobile phase with 5 mM ammonium
N N N N
acetate being used as the buffer solution, and acetonitrile
H
SO 3
- NH m/z (6 5 0 /2 = 3 2 5 ) H
NH being used as elution solvent could successfully elute all the
S O 3-
investigated dyes from the stationary phase in sequence of
N =N NH N anionic, cationic and non-ionic dyes according to their polarity
N H2
from strong to weak, with good reproducibility of retention
N H m/z2 3 4
S O 3- m/z 4 1 6
times. However, the anionic sulfonated dyes could not be
eluted from a C18 column when methanol was used as the
N =N eluent, instead of acetonitrile. The reason may be attributed to
N H2 the fact that acetonitrile, in comparison with methanol, shows
m/z 4 0 1
(b) more basic characteristic and is more suitable for the elution
of compounds with sulfonate functionality.
The total ion chromatogram was deconvoluted by different
transitions to obtain the analyte peak in Figure 5. The chro-
matogram could be divided into two segments, where the
first included four anionic dyes and the second included two
cationic and three disperse ones. After LC separation, the
mass spectrometer detected the effluents in two segments
and changed the ESI mode from negative to positive at tR
5.65 min. Thus, the anionic, cationic and non-ionic dyes of
interest could be detected by a single instrumental method
and the analytical efficiency of multi-residue identification was
greatly improved.

Method detection limits and matrix effect

Method detection limits (MDLs) of the dyes in Table 1, calcu-
lated upon statistic criteria based on a signal-to-noise ratio of
3, was at 0.005–0.025 mg kg–1 for the basic and disperse dyes
(c)
detected under ESI+ mode and 0.025–0.25 mg kg–1 for the acid
and direct dyes detected under ESI– mode, which were much
lower than previously reported MDL using the HPLC-DAD
method (5 mg kg–1).8 So, the proposed LC/MS/MS method is
sensitive enough to identify the carcinogenic dyes at trace
level in textiles. The negative dyes had higher MDLs than
the positive ones because the former had lower ionization
response as discussed above.
Using least-square regression, the calibration function
was calculated to obtain the standard calibration curve with
good linearity (g > 0.995) for each dye, as shown in Table 1.
The matrix-matching calibration curves of each dye were
compared with instrumental calibration curves to evaluate
the matrix effect in LC/MS/MS detection. It was found that the
two calibration curves were nearly super-imposable, which
demonstrated that interferences resulting from the fiber
Figure 4. Scheme of the proposed fragmentation pattern of: (a)
DR28, (b) DB1 and (c) BR9.
substrate was negligible for this method.

Figure 5 Sample analysis

LC/MS/MS identification of the unknown compounds was
work.10–15,19 However, for anionic sulphonated dyes, an ion-pair based not only on the retention time, but also mass spectra
reagent was used to aid in the adsorption of ionic8 ofspecies
12 on in which the difference of relative intensities of product ions
 Figure 4

Y. Ding, C. Sun and X. Xu, Eur. J. Mass Spectrom. 15, 705–713 (2009) 711
Figure 5

Figure 5. LC-ESI-MS/MS chromatograms of the standard solutions of the nine dyes with selective reaction monitoring mode in (a)
negative ionization mode and (b) positive ionization mode.

between samples and standard solutions were compared.

Three reference samples were tested by the proposed method.
Comparing with analytical results of standard solutions, the
carcinogenic dyes were identified in reference samples with
the similar retention times and the relative differences in
the masses of the product ions less than 10%: DB38, BV14
and DY3 were identified in cotton, acrylic and polyester fiber
respectively, with re-constructed ion chromatogram and
selected product-ion spectra shown in Figure 6. Moreover, all
the nine compounds of interest were properly identified in the
recovery experiment.
Conclusions
The study focused on simultaneous identification of the
nine law-prohibited carcinogenic dyes in textiles. The
­developed LC/ESI/MS/MS with negative/positive ion-switching
electrospray ionization was proven to be a powerful approach
for identification of the nine dissimilar analytes (including
three basic, one acid, two basic and three disperse dyes)
in a single run analysis. On the one hand, monitoring the
retention time together with three SRM transitions for each
analyte minimizes the potential for misidentification of dyes

7 of 10
Table 1. LC-ESI-MS/MS detection parameters and analytical data for the nine dyes.

No. Analyte Detection Precursor ion Product ion Collision tR Linearity MDL
Mode (m/z) (m/z) energy (min) g  (mg kg–1)
(Charge state) (Relative intensity, %) (eV)
1 Direct blue 6 ESI– 210 [M – 4Na] 4– 158(40), 236(5), 248(100) 12, 10, 18 4.63 ± 0.08 0.9979 0.25
2 Acid red 26 ESI– 217 [M – 2Na] 2–
137(100), 142(25), 208(15) 21, 33, 14 4.98 ± 0.01 0.9993 0.025
3 Direct red 28 ESI– 325 [M – 2Na]2– 234(40), 401(20), 416(100) 18, 23, 20 5.18 ± 0.05 0.9991 0.05
4 Direct black 38 ESI– 368 [M – 2Na] 2–
164(100), 644(60), 322(15) 18, 30, 20 5.39 ± 0.02 0.9998 0.2
5 Disperse blue 1 ESI+ 269 [M + H]+ 107(20), 241(100), 253(95) 27, 27, 29 5.90 ± 0.02 0.9982 0.1
6 Basic red 9 ESI+ 288 [M – Cl] +
168(15),195(100), 272(5) 34, 29, 29 6.41 ± 0.07 0.9978 0.005
7 Basic violet 14 ESI+ 302 [M – Cl]+ 195(55), 209(100), 287(10) 34, 30, 22 6.62 ± 0.06 0.9994 0.005
8 Disperse yellow 3 ESI+ 270 [M + H] +
107(55), 122(30), 150(100) 30, 16, 21 6.92 ± 0.01 0.9999 0.025
9 Disperse orange 11 ESI+ 238 [M + H] +
105(15), 165(100), 223(65) 24, 32, 20 7.07 ± 0.01 0.9996 0.025
(e)disperse blue1; (f)basic red9; (g)basic violet14; (h)disperse yellow3 and (i)disperse orange11
Figure 4 Scheme of the proposed fragmentation pattern of: (a) DR28, (b) DB1 and (c) BR9.
Figure 5. LC-ESI-MS/MS chromatograms of the standard solutions of the nine dyes with selective reaction
monitoring
712 mode in (a) negative ionization mode and (b) positive ionization mode
Simultaneous Identification of Nine Dyes from Textiles
Figure 6 (a) Re-constructed ion chromatograms and (b) selected ion monitoring spectra of reference
samples

RT:6.91
RT:5.40 RT:6.63
Disperse Yellow
Direct Black Basic Violet 14

0 2 4 6 8 10 12 14
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
t / min
t / min t / min

209.0 107.0
164.0
Direct Black Basic Violet 14 Disperse Yellow

195.0
644.1

150.0

322.4
287.0 122.0

150 200 250 300 350 100 150 200 250 300
100 200 300 400 500 600 700 800
m/z m/z m/z

Figure 6. (a) Re-constructed ion chromatograms and (b) selected ion monitoring spectra of reference samples.

and the selective MS/MS technique provides higher detec- and use of certain dangerous substances and prepara-
tion sensitivities and less matrix effects, compared with the tions, Official Journal of European Communities L166/87,
published HPLC/DAD technique. On the other hand, a one- (1999).
time extraction of dyes from different kinds of textile fibers 3. Council Decision 2002/371/EC, establishing the ecologi-
with ultrasound-assisted methanol combining with a single cal criteria for the award of the Community eco-label to
instrumental injection with short run time can curtail test textile products, Official Journal of European Communities
time, which is important for large batches of samples. This L133/29, (2002).
sensitive, accurate and rapid assay is suitable for lab routine 4. Oeko-Tex Standard 100: General and special condi-
analysis to identify the carcinogenic dyes in textiles products. tions. International Association for Research and
Testing in the Field of Textile Ecology, Switzerland,
2007. Available: http://www.oeko-tex.com/xdesk/

Acknowledgments ximages/470/16459_100def2007.pdf
5. J.C. West, “Extraction and analysis of disperse dyes on
This work was financially supported by Project 2006BAK10B03-3 polyester textiles”, J. Chromatogr. 1208, 47 (1981). doi:
sponsored by the Ministry of Science and Technology of PR 10.1016/S0021-9673(00)87958-3
China. Special thanks are given to Shanghai Dyes Research 6. J. Wouters, “High performance liquid chromatography
Institute for the kind preparation of dyed fabrics as reference of anthraquinones”, Stud. Conserv. 30, 119 (1985). doi:
samples. 10.2307/1505927
7 of 9
7. J. Wouters and A. Verhecken, “The coccoid insect
dyes: HPLC and computerized diode-array analy-

References sis of dyed yarns”, Stud. Conserv. 34, 189 (1989). doi:
10.2307/1506286
1. R. Anliker, “Ecotoxiocology of dyestuffs: a joint effort 8. J.P. Wang, H.W. Su and C.Y. Hong, “Testing method
by industry”, Ecotoxicol. Environ. Saf. 3, 59 (1979). doi: for disperse dyestuffs in textiles”, Chinese Dyeing and
10.1016/0147-6513(79)90060-5 Finishing 33, 32 (2007) (in Chinese).
2. Council Directive 1999/43/EC, amending for the 17th 9. T. Reetsma, “ Liquid chromatography-mass
time Directive 76/769/EEC, on the approximation of the ­spectrometry and strategies for trace-level analysis
laws, regulations and administrative provisions of the of polar organic pollutants”,J. Chromatogr. A 1000, 477
Member States relating to restrictions on the marketing (2003). doi: 10.1016/S0021-9673(03)00507-7
 Y. Ding, C. Sun and X. Xu, Eur. J. Mass Spectrom. 15, 705–713 (2009) 713
10. D. Vanerkova, A. Sakalis, M. Holcapek, P. Jandera and
A. Voulgaropulos, “Analysis of electrochemical degra-
dation products of sulphonated azo dyes using high-
­performance liquid chromatography/tandem mass
spectrometry”, Rapid Commun. Mass Spectrom. 20, 2807
(2006). doi: 10.1002/rcm.2656
11. S.D. Richardson, “Mass spectrometry in ­environmental
sciences”, Chem. Rev. 101, 211 (2001). doi: 10.1021/
cr990090u
12. L. Rafaelly, S. Heron and W. Nowik, “Optimisation
of ESI-MS detection for the HPLC of anthraquinone
dyes”, Dyes Pigments 77, 191 (2008). doi: 10.1016/j.
dyepig.2007.05.007
13. D. Arroyoa, M.C. Ortiz, L.A. Sarabia and F. Palacios,
“Advantages of PARAFAC calibration in the determi-
nation of malachite green and its metabolite in fish
by liquid chromatography–tandem mass spectrom-
etry”, J. Chromatogr. A 1187, 1 (2008). doi: 10.1016/j.
chroma.2008.02.040
14. M. Holcapek, K. Volna and D. Vanerkova, “Effects of
­functional groups on the fragmentation of dyes in
­electrospray and atmospheric pressure chemical
­ionization mass spectra”, Dyes Pigments 75,156 (2007).
doi: 10.1016/j.dyepig.2006.05.040
View publication stats
15. C. Rafols and D. Barcelo, “Determination of mono- and
disulphonated azo dyes by liquid chromatography-
atmospheric pressure ionization mass spectrometry”,
J. Chromatogr. A 777, 177 (1997). doi: 10.1016/S0021-
9673(97)00429-9
16. F.W. McLafferty, Interpretation of mass spectra,3rd Edn.
University Science Books, Mill Valley, California, USA
(1980).
17. W.J. Epolito, Y.H. Lee and L.A. Bottomley,
“Characterization of the textile anthraquinone dye
Reactive Blue 4”, Dyes Pigments 67, 35 (2005). doi:
10.1016/j.dyepig.2004.10.006
18. M. Karni and A. Mandelbaum, “The even electron
rule”, Org. Mass Spectrom. 15, 53 (1980). doi: 10.1002/
oms.1210150202
19. T. Storm, T. Reemtsma and M. Jekel, “Use of volatile
amines as ion-pairing agents for the highperformance
liquid chromatographic–tandem mass spectromet-
ric determination of aromatic sulfonates in industrial
wastewater”, J. Chromatogr. A 854, 175 (1999). doi:
10.1016/S0021-9673(99)00525-7