2.2. Experimental procedures
The photochemical experiments were performed in a 700-mL,
magnetically-stirred cylindrical borosilicate glass reactor with a
quartz tube in the center, in which a low pressure mercury lamp
(Heraeus GPH 212T5L/4, 10 W) was placed. The temperature was
maintained at 25(±0.2) C. The UV photon flux (I0) entering the
solution was determined using iodide/iodate chemical actinometry
(Bolton et al., 2011) and calculated to be 0.477 mE s1
. The effective
path length (L) was determined to be 2.43 cm by measuring the
kinetics of the photolysis of dilute H2O2 (shown in Fig. S1) (Garoma
and Gurol, 2005). The corresponding average UV fluence rate (Ep
0
)was estimated to be 0.89 mW cm2
.
A 700 mL testing solution containing 10 mM TMP and 2 mM
phosphate buffer was dosed with the NaOCl stock solution to
achieve an initial chlorine concentration of 20e100 mM and
simultaneously exposed to UV irradiation. Samples were collected
at different time intervals and quenched with NH4Cl at a molar ratio
of NH4Cl to chlorine of 4:1 to minimize the potential N-chloro TMP
intermediates transforming back to TMP (Dodd and Huang, 2007).
In addition, another aliquot of each sample was quenched with
Na2SO3 at the same mole concentration as NH4Cl to investigate the
formation of N-chloro TMP intermediates in the UV/chlorine and
chlorination processes. The control, which involved exposure to
either UV irradiation or chlorination alone was carried out in a
similar manner and quenched by NH4Cl. Another set of tests was
conducted with DAP and TMT as the target compounds to evaluate
the degradation of the sub-structure moieties of TMP in the UV/
chlorine process. Also, tests were conducted to differentiate the
responsible radicals for the degradation of TMP using NB as the
radical probe.