2194
Environ. Toxicol. Chem. 20, 2001
M. Blanco et al.
The negative charge of the molecule facilitated the separation
of the degradation products by capillary zone electrophoresis
(CZE).
identified by comparing the spectra and migration times of the
peaks with those for pure standards.
Solutions containing 10 mg/L DSBP and 1 mg/L concen-
tration of the transition metals (buffered with sodium carbonate
at pH 9) were used in the fluorescence experiments. They were
allowed to circulate in the absence of hypochlorite for 1,000
s, the time during which the initial fluorescence was measured,
then the hypochlorite was added and the fluorescence decay
with time recorded.
EXPERIMENTAL
Reagents
The optical brightener E,E(4,4
E,E-DSBP) and its E,Z and Z,Z isomers—2-sulfobenzalde-
hyde (aldehyde 1), 4,4 -bisaldehyde biphenyl (aldehyde 2), 2-
sulfobenzoic acid (acid 1), 4,4 -biphenyldicarboxylic acid
(acid 2), and 3-benzaldehyde-2 -sulfonic acid stilbene—were
Ј-bis[2-sulfostyryl]biphenyl)
(
Ј
Ј
RESULTS AND DISCUSSION
Ј
Preliminary experiments were carried out, monitoring the
UV spectra for DSBP solutions exposed to hypochlorite buff-
ered at pH 9 (washing baths are normally alkaline). The UV
spectrum for the DSBP solution exhibited an absorption max-
imum at 348 nm, and the addition of hypochlorite gave rise
to a band at 290 nm (the UV maximum for hypochlorite). This
spectrum evolved to complete disappearance of DSBP band
and the appearance of a new band peaking at 280 nm. These
changes expose the instability of DSBP against hypochlorite
and suggest the formation of new species.
supplied by Ciba Specialty Chemicals (Basel, Switzerland) and
used as received. The purity of all was greater than 95% with
the exception of the E,E-DSBP isomer, which was only 82%
pure and accompanied by the inorganic salts used to precipitate
it.
Ammonium iron (II) sulfate, sodium tetraborate, and so-
dium carbonate (all proanalysis reagents from Merck, Darm-
stadt, Germany) were also used. Milli-Q water from a Mil-
lipore Water Purification system (Millipore, Molsheim,
France) was used throughout.
The initial tests were conducted to identify the DSBP ox-
idation products by hypochlorite involved preparing a 50-mg/
L solution containing DSBP and NaClO (mole ratio 1:6.5, pH
9) that was stored at room temperature. Aliquots of this so-
lution were injected into the capillary electrophoresis system
on different days over a three-month period (results not
shown). These tests revealed the oxidation of DSBP to involve
the formation of various intermediates that evolve very slowly.
However, the derivatives aldehyde 1 and aldehyde 2 were suc-
cessfully identified from among the other degradation inter-
mediates. In order to expedite the process, the sample was
Apparatus
Electrophoretic separations were conducted on a 3Dcapillary
electrophoresis instrument from Hewlett-Packard (Waldbronn,
Germany) equipped with a diode array detector, an autoinjec-
tor, and a capillary thermostat. Samples were injected in the
hydrodynamic mode, using a pressure of 50 mbar on the in-
jection vial for 5 s. Direct polarity (with the cathode next to
the detector) was used throughout, and electropherograms were
recorded at wavelengths where the analytes exhibited strong
absorption (198, 270, 279, 292, and 348 nm). The electro-
phoretic separation was carried out in a 56-cm-long (effective
subjected to accelerated aging by storage at 60ЊC in a stove.
This increased temperature resulted in significantly simpler
electropherograms than those obtained at room temperature
and also in a faster evolution of the intermediates.
length) by 50-
m-i.d. fused silica capillary thermostated at
25 C. A 20-mM sodium tetraborate solution adjusted to pH
Њ
9.3 with 0.1 M NaOH was used as background electrolyte,
and a voltage of 30 kV was applied between both capillary
ends. All pH measurements were made with a micropH 2001
pH-meter from Crison (Alella, Spain).
Two degradation tests at 60ЊC in the absence of light and
metals were then performed: one at a 1:8 DSBP/hypochlorite
mole ratio (112 mg/L DSBP, 119 mg/L NaClO) and the other
at a 1:24 ratio (112 mg/L DSBP, 357 mg/L NaClO).
Fluorescence measurements were made on a Perkin-Elmer
LS50 fluorimeter (Norwalk, CT, USA), using a continuous-
flow system consisting of a peristaltic pump, a Hellma flow-
cell of 1 cm light path, and a 30% beam attenuator at the
detector gate. The sample, maintained under continuous stir-
ring in the dark, was propelled to the measuring cell, where
its fluorescence was excited at 348 nm and its emission mea-
sured at 430 nm. The beam attenuator was used in order to
avoid excessive dilution of DSBP because the brightener pos-
sesses very strong fluorescence.
The electropherogram obtained for the initial 1:8 DSBP/
hypochlorite solution approximately 30 min after preparation
exhibited two small peaks (peaks 1 and 2) and three overlapped
peaks (Fig. 1) that were assigned to the initial brightener spe-
cies (E,E-DSBP) in addition to two intermediate species with
maximum absorption at 270 and 320 nm, respectively. Sub-
sequently, the species with maximum absorption at 320 nm,
which was designated intermediate I, evolved to the one with
maximum absorption at 270 nm (intermediate II). Subsequent
electropherograms (Fig. 2) showed that intermediate II also
evolved and that it yielded the derivatives aldehyde 1 and
aldehyde 2, which were then partially oxidized to their cor-
responding acid derivatives (acid 1 and acid 2). No change in
these species was apparent from electropherograms recorded
at longer times.
Procedure
Solutions containing 112 mg/L DSBP and different
amounts of hypochlorite in the presence and absence of iron
(II) were kept at room temperature or 60ЊC as required, away
from light in order to avoid photodegradation. Several aliquots
of these solutions were withdrawn at different times after prep-
aration and injected into the electrophoretic capillary system,
which was previously conditioned by passing the background
electrolyte for 10 min and equilibrated by applying a potential
of 30 kV for 10 min. After each separation, the capillary was
reconditioned by passing 1 M NaOH for 5 min and then 0.1
M NaOH for a further 10 min. Degradation products were
A degradation test conducted in the presence of a large
excess of hypochlorite revealed the formation of intermediate
II to be much faster. A freshly made solution yielded a main
peak corresponding to intermediate II and weak signals cor-
responding to residual amounts of E,E-DSBP and intermediate
I. Subsequent electropherograms revealed that the degradation
end products were the acid derivatives, which were accom-
panied by residual amounts of the aldehyde derivatives. How-