154
M.J. López-Mun˜oz et al. / Catalysis Today 161 (2011) 153–162
tocatalytic reactions by means of luminescence bioassays using
Vibrio fischeri.
OAAT, OOAT and OOOT were carried out with a gradient of aque-
ous phosphoric acid buffer (pH 3)–acetonitrile (95:5, v/v) to (40:60,
v/v), at a flow rate of 0.8 ml min−1 and UV detection at 205 nm.
2.3.2. Toxicity measurements
The toxicity evaluation of samples was performed with a Micro-
tox Model 550 Toxicity Analyzer. The analysis is based on the
measurement of the ability of the sample to inhibit the natural bio-
luminescence of the marine bacterium Vibrio fischeri (strain NRLL
no. B-11177). The bacteria were in freeze-dried form (acquired
from Gomensoro) and activated prior to use by NaCl (2%) recon-
stitution solution. Samples were tested in solution containing 2%
sodium chloride in six dilutions. After 15 min of incubation at
15 ◦C light emission was recorded and compared with a toxic-free
control. The percentage of inhibition was obtained following the
established protocol using the Microtox calculation program.
2.1. Chemicals
Table 1 displays the chemical structure, nomenclature and
acronyms of the organics according to the nomenclature devel-
oped by Cook and Hütter [14] and Nélieu et al. [15] to identify
s-triazine compounds. All were of analytical grade and used as
received. Atrazine 99% (2-chloro-4-ethylamino-6-isopropylamino-
s-triazine), simazine 99.5% (2-chloro-4,6-diethylamino-s-triazine),
desisopropyl-atrazine 95% (2-chloro-4-ethylamino-6-amino-s-
triazine), desethyl-atrazine 99% (2-chloro-4-amino-6-isopropyl-
amino-s-triazine), hydroxy-atrazine 96% (2-hydroxy-4-ethyl-
amino-6-isopropylamino-s-triazine),
atrazine 95% (2-hydroxy-4-ethylamino-6-amino-s-triazine)
desisopropyl-2-hydroxy-
2.3.3. Factorial experimental design
Multivariate experimental design was carried out following the
methodology of response surface [13]. The software Statgraph-
ics Plus 5.0 was used to obtain the polynomial equations and the
response surfaces.
and hydroxy-simazine 99.5% (2-hydroxy-4,6-diethylamino-
s-triazine) were purchased from Riedel-de Häen. The
standard
desethyl-2-hydroxy-atrazine
99%
(2-hydroxy-4-
amino-6-isopropylamino-s-triazine) was obtained from Fluka.
Standards cyanuric acid 98% (2,4,6-trihydroxy-s-triazine) and
3. Results and discussion
desethyl-desisopropyl-atrazine
95%
(2-chloro-4,6-diamino-
s-triazine) were acquired from Aldrich. Ammeline 95%
(2-hydroxy-4,6-diamino-1,3,5-triazine) and ammelide (2,4-
dihydroxy-6-amino-1,3,5-triazine) were purchased from ABCR.
All standard solutions were prepared using organic-free deionized
water (Milli-Q). Titanium dioxide P-25 supplied by Degussa was
used as photocatalyst. It is a non-porous solid for which a BET
surface area of 50 m2 and mean particle size of ca. 30 nm were
measured. It contains anatase and rutile crystalline phases in
a ratio 4:1. In all experiments initial concentrations of 25 and
5 mg L−1 were set for atrazine and simazine, respectively.
Preliminary experiments were done without either titanium
dioxide or UV irradiation to evaluate the extent of adsorption and
photolysis of the herbicides under our experimental conditions.
Dark adsorption was negligible for both compounds as no sig-
nificant variations in concentration were found after 6 h. On the
contrary, photolysis was shown to occur to some extent. After a
time period of 1 h, degradation percentages of 15% and 25% were
respectively observed for atrazine and simazine. Our results are
significantly different from those reported by Héquet et al. [8] who
found a half-life time of 2.3 min for the photolytic atrazine degrada-
due to the differences in the irradiation set-up as we employed
a combination of Pyrex and copper sulphate solution as filtering
system instead of the quartz filter they used. On the other hand,
the photolysis extent found for simazine is in accordance with the
results obtained by Chu et al. [10] who also report a significant
range.
The kinetics of atrazine and simazine degradation was signifi-
cantly enhanced by irradiation in the presence of titanium dioxide
as catalyst following a pseudo-first order decay, in agreement with
previous reports [6,9,10]. The effect of pH and titania concentration
on the kinetics of the reaction were investigated by experimental
design methodology as it allows to obtain the optimal conditions for
the photocatalytic procedure from a minimum set of experiments.
Based on preliminary experiments carried out in a wide domain
of TiO2 concentration from 0.05 to 3 g L−1 at the natural pH of the
herbicide, a catalyst concentration ranging from 0.05 to 1.95 g L−1
and initial pH from 3 to 9 was selected for atrazine. For simazine,
a TiO2 concentration from 0.05 to 0.25 g L−1 and a pH interval 3–8
was fixed. Natural pH, i.e. 6.0 for atrazine and 5.5 for simazine, was
chosen as central point in both cases.
2.2. Photocatalysis procedure
Photocatalytic reactions were carried out in a batch Pyrex reac-
tor of 1 L effective solution volume. Irradiations were carried out
using a 150 W medium pressure mercury lamp (Heraeus TQ-150),
placed inside a Pyrex jacket and provided with a cooling tube
through which an aqueous solution of copper sulphate (0.01 M)
was circulated to prevent the overheating of the suspension and to
cut off the radiation below 300 nm. The initial pH values of the her-
bicides solutions were adjusted with H2SO4 0.1 M or NaOH 0.1 M.
Prior to irradiation, the suspension of triazine and catalyst was air
saturated and equilibrated by magnetic stirring for 20 min in the
dark to ensure complete equilibration of adsorption/desorption of
the organic compound on the catalyst surface. Continuous air bub-
bling and stirring were maintained through the reaction. Aliquots
were taken at time intervals, following filtration through 0.22 m
Nylon filters in order to remove the suspended TiO2 particles before
being analyzed.
2.3. Analysis
2.3.1. Liquid chromatography
In a first approach, a two-level factorial design consisting in four
experiments (22) at the limits of the pH and TiO2 intervals and
three experiments at the central values of the two factors to deter-
mine the experimental error and any possible effects of curvature
in the response surface were carried out. The analysis of the data
obtained showed that the curvature was significant, indicating that
a high-order model or response surface study was needed in order
to uncover the behaviour of the significant factors [13]. Therefore,
four experiments were performed in addition at the midpoints to
High performance liquid chromatography (Varian Polaris
Prostar 320) system equipped with UV–VIS detector was used to
identify and quantify the products obtained in the photocatalytic
reactions. The separation was performed using a Synergi 4 MAX-
RP column. CEIT, CEET, ACIT and ACET were eluted with a gradient
of aqueous phosphate buffer (pH 7)–acetonitrile (55:45, v/v) to
(35:65, v/v), at a flow rate of 0.5 ml min−1 and detection at 210 nm.
Elution of the other components, EOIT, EEOT, AOIT, AEOT, CAAT,