Enhancement of the photodegradation of N-nitrosodimethylamine in water using amorphous and platinum manganese oxide catalysts

Article abstract of DOI:10.1016/j.jphotochem.2010.10.021

The enhancement of the photodegradation of toxic N-nitrosodimethylamine (NDMA) in water using amorphous manganese oxide (AMO) and platinum manganese oxide (Pt/Mn₃O₄) catalysts was investigated. Characterization of the catalysts was carried out using XRD, FESEM, TEM, EDXS, BET, XPS, and AOS. Pt/Mn₃O₄ and its precursor AMO, synthesized by a redox reaction of Mn²⁺ and Mn⁷⁺, showed similar morphologies. High surface area AMO was confirmed to be amorphous, whereas Pt/Mn₃O₄ was a mixture of two crystalline structures. The optimum catalyst loading was 25 mg per 100 mL NDMA solution for which the photocatalytic activity was maximized. The average hydrodynamic particle size of a given amount of catalyst increased due to aggregation as a result of an increase in temperature during UV illumination (λ = 254 nm). Photocatalytic studies showed that NDMA degraded according to zero-order kinetics under air saturation at pH 7.0. AMO and Pt/Mn₃O₄ showed photostability and comparable activities with pure TiO₂ and platinized TiO₂. Mixed valencies of Mn and the presence of O ₂ on the surface of the catalysts, which reacts with photogenerated electrons to form reactive oxygen species (hydroxyl and superoxide anion radicals), played significant roles in the enhancement of the photodegradation of NDMA in water.

Full text of DOI:10.1016/j.jphotochem.2010.10.021

Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Enhancement of the photodegradation of N-nitrosodimethylamine in water
using amorphous and platinum manganese oxide catalysts
Homer C. Genuinoa, Eric C. Njagia, Evan M. Benbowa, George E. Hoagb, John B. Collinsb, Steven L. Suiba,∗
a Department of Chemistry, University of Connecticut, 55 North Eagleville Rd., Unit 3060, Storrs, CT 06269, USA
b VeruTEK Technologies, Inc., 65 West Dudley Town Rd, Suite 100, Bloomfield, CT 06002, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 27 July 2010
Received in revised form 12 October 2010
Accepted 21 October 2010
Available online 28 October 2010
The enhancement of the photodegradation of toxic N-nitrosodimethylamine (NDMA) in water using
amorphous manganese oxide (AMO) and platinum manganese oxide (Pt/Mn3O4) catalysts was investi-
gated. Characterization of the catalysts was carried out using XRD, FESEM, TEM, EDXS, BET, XPS, and AOS.
Pt/Mn3O4 and its precursor AMO, synthesized by a redox reaction of Mn2+ and Mn7+, showed similar mor-
phologies. High surface area AMO was confirmed to be amorphous, whereas Pt/Mn3O4 was a mixture of
two crystalline structures. The optimum catalyst loading was 25 mg per 100 mL NDMA solution for which
the photocatalytic activity was maximized. The average hydrodynamic particle size of a given amount
of catalyst increased due to aggregation as a result of an increase in temperature during UV illumination
(ꢀ = 254 nm). Photocatalytic studies showed that NDMA degraded according to zero-order kinetics under
air saturation at pH 7.0. AMO and Pt/Mn3O4 showed photostability and comparable activities with pure
TiO2 and platinized TiO2. Mixed valencies of Mn and the presence of O2 on the surface of the catalysts,
which reacts with photogenerated electrons to form reactive oxygen species (hydroxyl and superoxide
anion radicals), played significant roles in the enhancement of the photodegradation of NDMA in water.
© 2010 Elsevier B.V. All rights reserved.
Keywords:
NDMA
AMO
Pt/Mn3O4
Photodegradation
Hydroxyl and superoxide anion radicals
1. Introduction
Treatment technologies such as direct UV photolysis, advanced
oxidation/photooxidation, and reduction processes have been sug-
In recent years, N-nitrosodimethylamine (NDMA) has become
a priority pollutant under the United States Environmental Protec-
tion Agency (US EPA) and has been classified by the World Health
Organization as a potential carcinogen in humans [1–3]. NDMA
can naturally be formed in the environment as a result of various
chemical and biological processes involving precursor substances
such as tertiary amines, dimethylamine (DMA), and natural organic
matter [4,5]. Anthropogenic sources of NDMA such as release from
agricultural, dye, cosmetic, food processing, polymer, rocket-fuel,
and wastewater industries can largely contribute to the presence
of NDMA in the environment [6–8]. The widespread presence of
its precursors, resistance to biodegradation, high thermal stabil-
ity, and excellent solubility in water make NDMA persistent and
difficult to destroy once formed [5,8,9].
gested to degrade NDMA in water. Direct UV photolysis is effective
but is not economical on a large scale due to high intensity
input requirements [11,12]. Advanced oxidation treatments using
UV/pure and surface modified TiO2, UV/O3, and UV/H2O2 producing
hydroxyl radicals (•OH) as primary oxidant may be limited by the
amount of •OH they produce and the presence of •OH scavengers
in water [13–15]. Further, addition of H2O2 does not significantly
enhance the performance of medium-pressure UV lamps in NDMA
degradation [16]. Recent studies on reductive destruction of NDMA
in water include the use of bimetallic catalysts such as nickel–boron
[17], nickel–iron [18], and palladium–indium [19]. However, these
catalysts have not been extensively applied to groundwater and
wastewater because of matrix effects associated with the complex-
ity of these samples which could interfere in the determination of
NDMA.
Manganese oxide catalysts are generally known as oxidants
due to the presence of active oxygen species (in the lattice or
adsorbed on the surface) [20] and are proven active for thermally
and photochemically catalyzed reactions. This study describes the
use of two Mn oxide catalysts, AMO and Pt/Mn3O4, to enhance
the photodegradation of NDMA present in high concentrations in
water at neutral pH. AMO catalysts prepared using KMnO4 and
oxalic acid as a reducing agent are known to demonstrate pho-
Removal of NDMA in the environment as a chemical con-
taminant [7] remains a challenge. The utilization of natural
light to degrade NDMA is promising due to NDMA’s weak UV
absorption (ꢀmax = 332 nm) [10–12], but the rate of photolysis is
slow and limited by the availability of sunlight in some areas.
∗
Corresponding author. Tel.: +1 860 486 2797; fax: +1 860 486 2981.
E-mail address: Steven.Suib@uconn.edu (S.L. Suib).
1010-6030/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jphotochem.2010.10.021

H.C. Genuino et al. / Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
285
tocatalytic activities for the oxidation of isopropanol to acetone
with excellent selectivity at ambient temperature [21]. AMO has
also been found to be an active photocatalyst in the oxidation
of methyl bromide, methyl chloride, and methyl iodide [22] and
in the decomposition of dimethyl methylphosphonate [23]. AMO
may as well be applicable for the photodegradation of an organic
pollutant such as NDMA. Noble metals such as Au, Ag, Pd, and
Pt are known to enhance the photocatalytic efficiency of TiO2 in
degrading pollutants in water by retarding recombination of pho-
togenerated electrons and holes [24]. A number of studies were
conducted using pure and platinized TiO2 (Pt/TiO2) for UV degra-
dation of pollutants including NDMA [13,25]. Investigating the
catalytic activity of a platinum-containing Mn oxide with different
properties from AMO toward NDMA photodegradation is therefore
interesting.
X-ray spectroscopy (EDXS) analyzer at an accelerating voltage of
20 kV. A standardless EDXS ZAF quantification method was used to
obtain the compositions of the catalysts as atomic % (average value
at three random locations).
The specific surface areas of the catalysts were determined
using the Brunauer–Emmett–Teller (BET) method by acquiring N2
adsorption and desorption isotherms at 77 K using a Micromerit-
ics ASAP 2010 instrument system. The catalysts were pretreated by
degassing them at 120 ◦C under vacuum for 2 h prior to BET analysis.
APerkinElmer Physical Electronics Model5300 X-ray photoelec-
tron spectrometer (XPS) equipped with a monochromator Al K␣
X-ray power source (250 W) with a pass energy of 50 eV and hemi-
spherical analyzer was used to study the binding energies (BE) of
Mn in the catalysts. The signal from adventitious carbon with BE of
284.6 eV was used as a reference.
The preparation and characterization of AMO and Pt/Mn3O4 are
described here. Results from spectroscopic, chromatographic, and
voltammetric studies are interpreted to study the role of the cata-
lysts and the reactive oxygen species (•OH and •O2−) that form, and
the kinetics of catalyzed photodegradation of NDMA. This seems to
be the first time that Mn oxide catalysts have been used to enhance
the photodegradation of NDMA in water. The high photocatalytic
activity, recyclability, hydrophobicity, relatively low cost, and low
toxicity are some of the advantages of Mn oxide catalysts for large-
scale environmental applications.
The determination of average oxidation state (AOS) of Mn in the
catalysts was carried out by potentiometric titrations [26]. First,
the total Mn content was determined by digesting the sample with
HCl(aq) to form Mn2+ in solution. To determine the concentration of
Mn, the Mn2+
was then titrated with standard MnO4
con-
−
(aq)
(aq)
taining Na4P2O7·10H2O(aq) forming Mn(H2P2O7)3−(aq). The total
MnO2 content of the sample was also determined by reacting
the sample with excess Fe[NH4]2(SO4)2(aq) under N2 atmosphere
followed by back-titration of the excess Fe2+
with standard
(aq)
−
MnO4 (aq). The number of moles of oxygen in the form of MnO
was obtained by the difference in the number of moles of Mn2+ and
of MnO2. On the basis of the total amount of charge due to oxygen
in the compound, the AOS was then calculated.
2. Materials and methods
2.1. Preparation of the catalysts
2.3. Sample preparation and photocatalytic studies
AMO was synthesized via
a redox sol–gel method by
adding 50 mL of 1.2 M KMnO4 dropwise to 50 mL of 2.55 M
Mn(C2H3O2)2·4H2O. The solution was stirred for 24 h and the
black precipitate formed was washed with 1000 mL ultrapure
water, dried in a vacuum at ambient temperature, and ground into
powder. The Pt/Mn3O4 was synthesized by dissolving 2 g of pre-
pared AMO to a solution made by dissolving 20 g of powdered
polyvinylpyrollidone in ultrapure water. K2PtCl4 of 0.6220 g was
then added to the mixture. The resulting mixture was purged
with Ar gas for 1 h and then bubbled with H2 gas. The reac-
tion vessel was sealed and kept in an oven at 50 ◦C for 12 h and
then washed. Finally, the solid was calcined in air at 150 ◦C for
5 h.
In a typical experiment, 25 mg of the catalyst was added to a
beaker containing 100 mL of 100 or 1000 ␮M NDMA solution. High
concentrations of NDMA were used not only to allow better quan-
tification but also to explore the applicability of this study in highly
contaminated water. The heterogeneous mixture was then stirred
for 30 min to attain equilibrium and sonicated for 15 min to uni-
formly disperse the catalyst particles prior to UV illumination. All
mixtures were prepared using ultrapure water buffered with 4 mM
bicarbonate and adjusted to pH 7.0 0.1 by acid dilution using
H2SO4. The pH of the mixture was measured using an Accumet
pH meter 25.
Sixteen 15-W UV-C lamps emitting mainly monochromatic light
with 99% emission at ꢀ = 254 nm installed on the Rayonet® pho-
tochemical reactor were used as the constant UV source. The
prepared mixture was transferred to a quartz cylindrical vessel
with dimensions of 4.5 cm i.d. × 24 cm length and positioned at
the center of the photoreactor (∼10 cm away from the lamps).
The mixture was then purged either with compressed air or Ar
gas (ultra-high purity grades) for 30 min through an inlet port
through a removable top. The UV lamps were stabilized for 5 min
prior to illumination. The fan controlled the temperature inside
the photoreactor fairly constant. UV illumination times were at
60 min intervals from 0 to 240 min performed with continuous
stirring of the mixture. NDMA was extracted using continuous
liquid–liquid extraction prior to gas chromatographic analysis
as described in US EPA Method 3520C [27]. In another run, a
desired volume (∼10 mL) for spectroscopic analysis was with-
drawn through a rubber septum using a glass syringe at various
time intervals. Suspended catalysts were easily recovered from the
mixture by centrifugation of the mixture for 30 min at 8000 rpm,
subsequently washed and filtered through a Millipore filter (pore
size 0.45 ␮m), and dried under vacuum at ambient temperature.
Reacted catalysts were reused twice and characterized using XRD
and AOS.
2.2. Characterization of the catalysts
Structural characterization of the catalysts was carried out using
powder X-ray diffraction (XRD). The XRD patterns of the catalysts
were recorded on a Scintag 2000 PDS diffractometer using Cu-K␣
˚
radiation (1.5418 A wavelength), a beam voltage of 45 kV and a
current of 40 mA. The sample scans were done at 2 degree 2ꢁ min−1
.
The morphology of the catalysts was studied using field emis-
sion scanning electron microscopy (FESEM) and transmission
electron microscopy (TEM). FESEM images were obtained using
a Zeiss DSM 982 Gemini microscope equipped with a Schot-
tky emitter at a beam current of 1 ␮A and accelerating voltage
of 10–20 kV. Sample suspensions were prepared in ethanol and
applied as a monolayer on gold-coated Si wafers. TEM images
and Selected Area Electron Diffraction (SAED) patterns of the
catalysts were obtained using a JEOL 2010 FasTEM microscope
at an accelerating voltage of 200 kV. Sample suspensions were
prepared using 2-propanol and deposited on carbon-coated Cu
grids.
The compositions of the catalysts were determined using an
AMRAY/LICO Model 9800 SEM equipped with an energy dispersive

286
H.C. Genuino et al. / Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
(A) Amorphous manganese oxide
1000
500
0
5
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
2 Theta (degree)
Mn3O4
Pt
§
(B) Platinum manganese oxide
2000
1000
0
§
§
§
§
§
§
§
§
§
§
§
§
§
§
5
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
2 Theta (degree)
Fig. 1. Powder XRD patterns of sample (A) amorphous manganese oxide (AMO) and (B) platinum manganese oxide (Pt/Mn3O4) before photocatalytic reaction.
−
−
2.4. Analyses of reaction species and products
ysis of 1000 ␮M NDMA, NO3 and NO2 by a modified Nessler’s
method [28]. Dimethylamine and formaldehyde were analyzed
using Umbreit [29] and Hantzsch [30] reactions, respectively.
Ferrioxalate actinomety was performed for the determination of
quantum yields for NDMA photolysis by obtaining absorbances of
Fe(II)-1,10-phenanthroline solutions at 510 nm [31]. The UV–Vis
spectrophotometer was also used in the generation of absorp-
tion spectra of solutions containing 2.2 mM NDMA and 0.24 mM
tetranitromethane (TNM) with and without the catalysts after UV
illumination for the detection of superoxide anion radicals (•O2−).
External standard calibration method was used to correlate mea-
sured absorbance to concentration. Hydroxyl radicals produced in
The analyses of reaction intermediates and products were car-
ried out at pH 7.0 and ambient temperature. The concentrations
of dissolved O2 and temperatures of the mixture during UV illu-
mination were measured using a YSI 550A DO instrument meter.
Dynamic light scattering (DLS) experiments were performed using
a Zetasizer Nano ZS90 equipped with 4 mW He–Ne ion laser
(ꢀ = 633 nm) at a measurement angle of 90◦. This was utilized to
determine the Z-average particle size of AMO suspended in NDMA
solution. The V530 Jasco UV-Vis double-beam spectrophotometer
equipped with silicon photodiode detector was used in the anal-
Fig. 2. FESEM images of sample (A) AMO and (C) Pt/Mn3O4. TEM image and SAED pattern (inset) of sample AMO (B). Inset in (C) shows magnified FESEM images of sample
Pt/Mn3O4.

H.C. Genuino et al. / Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
287
(A)AMO
(B) Pt/Mn3O4
Mn 2p3/2
Mn 2p3/2
Mn 2p1/2
Mn 2p1/2
643.3 eV
639.6 eV
645 640
Binding energy (eV)
670 665 660 655 650 645 640 635 630
Binding energy (eV)
660
655
650
635
630
Fig. 3. Mn 2p3/2 XPS spectra of sample (A) AMO and (B) Pt/Mn3O4 before photocatalytic reaction.
water containing the catalysts after UV illumination were detected
using 0.10 mM terephthalic acid solution. The photoluminescence
(PL) spectra were measured on a Cary Eclipse Fluorescence spec-
trophotometer using excitation light of 320 nm.
about 38◦ and 67◦ in sample AMO shown in Fig. 1A confirmed that
AMO is almost completely amorphous. This is in agreement with
the SAED pattern of AMO (Fig. 2B inset) where no spot patterns
were observed and with diffraction rings that look fuzzy around the
direct beam. Compared to AMO, Pt/Mn3O4 has completely different
XRD patterns which show sharp and very intense peaks suggest-
ing well-formed crystalline materials. The XRD pattern of sample
Pt/Mn3O4 shown in Fig. 1B reveals diffraction peaks corresponding
to Pt (syn) (JCPDS No. 04-802) and Mn3O4 (Hausmannite) (JCPDS
No. 16-0154), thus the formula Pt/Mn3O4.
Fig. 2A presents the FESEM image of AMO which shows a non-
distinct morphology that is composed of small clusters that are
homogeneous throughout the material and with diameters rang-
ing from approximately 50 to 100 nm. To further investigate this
structure, TEM has been employed. The TEM image of AMO in Fig. 2B
shows that these clusters are made up of nanoparticles of 5–15 nm
sizes. A similar morphology to that of AMO is observed in the FESEM
image of Pt/Mn3O4 shown in Fig. 2C. However, Pt/Mn3O4 forms
larger aggregated clusters of smaller numerous nanoparticles that
are regular in shape and size due to calcination.
The Mn/K atomic % ratios of AMO and Pt/Mn3O4 are 8.1 and
17.1, respectively. The Mn/Pt atomic % ratio for Pt/Mn3O4 is 40.1.
The results of EDXS analyses are precise and are in good agreement
with the expected results.
Fig. 3 shows the Mn 2p spectra for sample AMO and Pt/Mn3O4
before photocatalysis. The photoelectron binding energies (BE) cor-
responding to the Mn 2p3/2 transitions in AMO and Pt/Mn3O4 are
643.3 eV and 639.6 eV, respectively. These results indicate mixed
valency of Mn (Mn2+, Mn3+, and Mn4+) in both catalysts. However, it
is difficult to accurately determine the oxidation state of Mn only by
the BE of Mn 2p3/2 since the BEs for Mn2+, Mn3+, and Mn4+ overlap
in the Mn 2p3/2 region.
Gas chromatographic analysis was performed on a Hewlett-
Packard Model 5890A Series II system equipped with a mass
selective detector (Hewlett-Packard Model 5871 Series). The
0.18 mm (i.d.) × 20 m (length) × 0.18 ␮m (film size) Agilent DB-
17MS analytical column on (50%-phenyl)-methylpolysiloxane
support was used for separation. A 1 ␮L sample was fed into the
column using a 10-␮L microsyringe. Helium carrier gas was set
at a flow rate of about 10 cm3 min−1. The injection port tempera-
ture and oven temperature range were set at 200 ◦C and 30–270 ◦C
(rate = 10 ◦C min−1), respectively. The retention time for NDMA was
about 5.3 min. Quantitative analyses based on peak areas were per-
formed. The concentrations of the NDMA were calculated following
linear least square methods.
Cyclic voltammetric (CV) techniques were optimized and uti-
lized for the oxygen reduction experiments performed on a
computer-controlled CV-50W BAS Model 100B electrochemical
analyzer with and without UV light. A three-electrode cell con-
sisted of a Mn oxide catalyst slurry as the working electrode, a
Ag/AgCl as the reference electrode, and a Pt wire as the counter
electrode. The catalyst slurry was prepared by mixing 10 mg carbon
powder, 10 mg AMO or Pt/Mn3O4 (10 mg carbon powder as blank),
and 5 drops of polytetraflouroethylene binder solution in 5.0 mL
ultrapure water. The resulting mixture was sonicated for 5 min and
20 ␮L of this was applied on the pyrolytic graphite electrode and
dried overnight. A 20 mL volume of 1.0 M KOH aqueous solution
(electrolyte solution) was purged either with pure O2 or N2 gas via
a Teflon tube for 15 min. A quartz beaker was used as a cell compart-
ment for experiments performed under UV light for 10–15 min. The
scan ranges and scan rates used were +0.2 to −1.0 V and 50 mV s−1
,
The AOS values of Mn in AMO and Pt/Mn3O4 before photocat-
alytic reaction are 3.65 and 3.87, respectively, suggesting that Mn
exists in mixed oxidation states. AMO has a BET surface area of
312 m2 g−1, significantly higher than that of Pt/Mn3O4 (97 m2 g−1).
respectively. For CV experiments in non-aqueous aprotic solvents,
saturated calomel electrode (SCE) was used as the reference elec-
trode. A 0.1 M solution of tetrabutylammonium perchlorate (TBAP)
was used as electrolyte solution. This was prepared by dissolv-
ing the proper amount of TBAP in dimethylsulfoxide (DMSO) or
in dimethylformamide (DMF). The scan ranges and scan rates used
were +0.20 to −1.2 V or −1.5 V and 50 mV s−1, respectively.
3.2. Direct UV illumination
NDMA showed strong (ꢂ → ꢂ* transition) and weak (n → ꢂ*
transition) absorptions in the 210–250 nm and 300–360 nm wave-
length regions, respectively. UV illumination of 1000 ␮M NDMA
solution using 254 nm UV lamps during the first 180 min decreases
NDMA concentration by 50%. However, the absorbance read-
ings in the weak absorption band centered at 330 nm remain
almost unchanged. The irradiance spectrum of natural sunlight
(>∼300 nm) overlaps within NDMA’s weak absorption range [10]
3. Results
3.1. Characterization of the catalysts
The XRD patterns of the catalysts before photocatalytic reac-
tion are shown in Fig. 1. The presence of weak and broad peaks at

288
H.C. Genuino et al. / Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
Fig. 4. Time course photodegradation of NDMA and formation of products (DMA, formaldehyde, NO2−, and NO3−) using (A) UV alone, and UV with (B) AMO, (C) Pt/Mn3O4,
(D) TiO2, and (E) Pt/TiO2 catalysts under air-saturation at pH 7.0.
so a preliminary study using three UV-A/BL lamps (ꢀ = 350 nm) was
conducted. Results show that after 180 min of UV illumination, the
concentration of NDMA in the weak absorption band decreased by
almost 30%. The uncatalyzed UV illumination of 1000 ␮M NDMA
targeting either strong or weak absorption bands follows first-order
kinetics. A quantum yield of ∼0.30 mol einstein−1 was determined
for uncatalyzed NDMA photodegradation under air and Ar condi-
tions at pH 7.0.
3.4. Enhanced NDMA photodegradation and product formation
The kinetic studies show that catalyzed and uncatalyzed pho-
todegradation of 100 ␮M NDMA (Fig. 4A) obey zero-order kinetics
consistent with the literature results [11]. The use of AMO (Fig. 4B)
and Pt/Mn3O4 (Fig. 4C) in the presence of O2 from air effectively
enhances the photodegradation of NDMA as a function of time.
Mixed-valent AMO and Pt/Mn3O4 have a significant difference in
BET surface areas (215 m2 g−1) and demonstrate photocatalytic
activities. In the first 60 min, 34% of 100 ␮M NDMA are already
degraded by AMO and 43% by Pt/Mn3O4, both with UV light.
After 180 min, AMO and Pt/Mn3O4 achieve 95% and 100% NDMA
photodegradation, respectively. Since both catalysts exhibit pho-
toactivites, TiO2 (Fig. 4D) and Pt/TiO2 (Fig. 4E) were also tested for
comparison purposes. Commercially available pure TiO2 (Degussa
P25) was used as received whereas Pt/TiO2 was synthesized follow-
ing the procedure in the literature [32]. Results reveal that Pt/TiO2
is a better catalyst than pure TiO2 and AMO but is comparable with
Pt/Mn3O4 toward enhancing photodegradation of NDMA.
3.3. Effect of catalyst loading on NDMA photodegradation and the
Z-average diameter of AMO
The optimum amount of AMO and Pt/Mn3O4 for NDMA
photodegradation is 25 mg per 100 mL NDMA solution. DLS exper-
iments were then performed to determine the Z-average diameter
(intensity weighted average hydrodynamic size) of AMO particles
suspended in NDMA aqueous solution at 25 and 100 mg catalyst
loading. In general, the Z-average diameter of AMO particles in
solution is at least 4× larger compared to its dehydrated particle
size (TEM size). The Z-average diameters also increase with catalyst
loading.
Dark experiments involving catalysts with and without O2
showed no significant NDMA degradation. In the presence of UV
light, however, AMO and Pt/Mn3O4 still showed better photocat-

H.C. Genuino et al. / Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
289
(A) AMO
2.10
1.0 M KOH
0.5
0.4
0.3
0.2
0.1
0
1.80
1.50
1.20
0.90
0.60
0.30
0.00
DMF
DMSO
0.2
0
-0.2 -0.4 -0.6 -0.8
-1
-1.2
-0.1
-0.2
-0.3
0.3 0.2 0.1
0
-0.1 -0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1 -1.1
-0.30
Potential (V)
C
-0.60
Potential (V)
A
B
D
E
(B) Pt/Mn3O4
2.10
1.80
1.50
1.20
0.90
0.60
0.30
0.00
1.0 M KOH
0.5
0.4
0.3
0.2
0.1
0
DMF
DMSO
0.1 -0.1 -0.3 -0.5 -0.7 -0.9 -1.1 -1.3 -1.5
-0.1
-0.2
0.3 0.2 0.1
0
--00.3.10 -0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1 -1.1
-0.3
-0.60
Potential (V)
C"
Potential (V)
A"
B"
D"
E"
Fig. 5. Cyclic voltammograms of sample (A) AMO and (B) Pt/Mn3O4 in 1.0 M KOH and 0.1 M tetrabutylammonium perchlorate in DMSO and DMF solvents. (For CV studies in
1.0 M KOH: A curves: fresh catalysts, O2-saturated; B curves: reused catalysts, O2-saturated; C curves: reused catalysts, N2-saturated; D curves: reused catalysts, N2-saturated,
UV light; E curves: reused catalysts, O2-saturated, UV light.)
−
alytic activities than UV alo−ne even in the absence of O2 from air.
Formaldehyde and NO3 are the major oxidation photoprod-
protonated peroxide ion, HO2 (Eq. (1)) [33]:
−
O2(g) + H2O(l) + 2e− → HO2 (aq) + OH−
(noUVlight)
(1)
−
(aq)
ucts formed using AMO −and Pt/Mn3O4. Dimethylamine and NO2
−
are also quantified. NO2 and NO3 form simultaneously and do
Minimal electroactivities are observed when the electrolyte solu-
tion was saturated with N2 (C curves) for both catalysts. However,
in the presence of UV light, small increase in the peak currents are
observed (D curves). A much better electroactivity is restored when
the electrolyte solution was repurged with O2, indicating stability
of the catalysts toward O2 reduction. A significant increase in the
reduction peak current and a shift toward more negative potential
are observed for both catalysts when the solution was repurged
with O2 in the presence of UV light (E curves). Curves B and E for
both catalysts fit perfectly from +0.20 V to −0.30 V suggesting that
a different type of electron-transfer process, in addition to Eq. (1),
occurs in the presence of UV light.
not interfere with NDMA during UV photolysis. In the presence of
AMO or Pt/Mn3O4, format−ion of NO3− is faster than−NO2−. However,
the concentration of NO2 is 2× higher than NO3 in the absence
of the catalysts after 240 min UV light illumination. There was no
significant difference in the average concentration of NDMA befor−e
and after−the mixture was spiked with 50 and 100 ␮M of both NO2
and NO3 after 180 min of UV illumination.
3.5. O2 gene−ration, oxygen reduction studies, and detection of
•OH and •O2 species
In O2 saturated non-aqueous (aprotic) solutions, a 1-electron
reduction of O2 to O2 is observed using the same Mn oxide cata-
lysts as working electrodes (Eq. (2)) [34]:
−
O2 is generated in an Ar-saturated solution of NDMA in the pres-
ence of AMO or Pt/Mn3O4 catalyst and UV light. However, unlike
in AMO, the concentration of dissolved O2 using Pt/Mn3O4 gradu-
ally decreases upon further reaction for more than 120 min. Dark
experiments and/or absence of the catalysts showed no detectable
dissolved O2.
Parallel studies by electrochemical reduction of O2 using CV
were conducted to further investigate the role of the catalysts,
and thus of reactive oxygen species, in the photodegradation of
NDMA. Fig. 5 shows that under O2 saturation, both catalysts (fresh)
(A curves) are electroactive in the potential window employed and
have higher O2 reduction peaks than the reused catalysts (B curves).
The reduction peak potentials (∼−0.30 V) versus Ag/AgCl reference
electrode for both catalysts in alkaline aqueous media are in the
region corresponding to a 2-electron reduction step of O2 to singly
O2 + e− ↔ O2
(inaproticmedia, noUVlight)
(2)
−
The CV obtained using electrolyte solutions prepared using DMSO
and DMF solvents, and TBAP for AMO and Pt/Mn3O4 are shown in
Fig. 5, respectively, showing reduction (∼−0.86 V) and oxidation
(∼−0.65 V) peak potentials versus SCE.
An indirect detection of •OH and •O2 species formed during
−
UV illumination of water was performed using terephthalic acid
and tetranitromethane (TNM) as chemical indicators, respectively.
Fig. 6 presents the PL spectra of UV illuminated aqueous solutions of
terephthalic acid in the presence of the catalysts. The visible emis-
sions appear as strong and broad peaks with maxima at 427 nm,
similarly observed for AMO and Pt/Mn3O4. The fluorescence inten-

290
H.C. Genuino et al. / Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
550
500
450
400
350
300
250
200
150
100
50
species, and photogenerated electrons and holes. Direct UV illumi-
50 min
AMO
nation of NDMA can also be blocked by excess catalysts through
absorption. For DLS experiments, AMO was tested on the basis of
its high surface area and small particle size compared to Pt/Mn3O4.
Loadings of AMO greater than 25 mg could also reduce the avail-
ability of free space within which the catalyst particles can move
in NDMA solution. This could be imparted by the tendency of the
particles to aggregate in solution at a high loading (i.e. 100 mg
per 100 mL) during the course of UV illumination as a result of an
increase in temperature. Therefore, restricted diffusion due to AMO
aggregation not only impacts mobility but could also explain why
low percent NDMA degradation was observed for 100 mg AMO.
The combined effect of solvation and the ability of AMO particles
to aggregate with temperature as the catalyst loading is increased
explain why the Z-average diameter of AMO is much larger than its
TEM size. AMO (25 mg) is beneficial for the purpose of photodegrad-
ing thermally stable NDMA in water. Catalyst aggregation will affect
particle transport and could be a crucial factor for catalyst recov-
ery from suspension and recyclability especially when applied in
aquatic environments or in ground and in wastewater.
Pt/Mn O4
30 min
20 min
10 min
0
350 375
400 425
450
475 500 525 550
Wavelength (nm)
Fig. 6. Photoluminescence spectra of 0.10 mM terephthalic acid solution (adjusted
to pH 7.0) with the catalysts using 320 nm excitation light. UV illumination
(ꢀ = 254 nm) times are 0, 10, 20, 30, and 50 min at pH 7.0 under air saturation.
Excitation and emission slit widths are set at 5 nm.
Mixed valency is very common in Mn oxide materials and an
important property in redox catalysis [35]. The AOS of Mn in AMO
and Pt/Mn3O4 before reaction with NDMA under UV light sug-
gests that the Mn in these samples exists in mixed oxidation states,
most likely a mixture of Mn4−+, Mn3+, and possibly Mn2+, produced
from the reduction of MnO4 (Mn7+) during synthesis [22]. These
results are in agreement with the Mn 2p3/2 XPS spectra of AMO and
Pt/Mn3O4 (Fig. 3). Potassium ions counterbalance the charge as a
result of the reduction of Mn4+. The XRD pattern of Pt/Mn3O4 shows
the existence of Pt and Mn3O4 (Hausmannite) crystal structures and
also supports mixed valency of Mn in this material. The differences
in the Mn3+ to Mn4+ ratio of AMO and Pt/Mn3O4 could enhance
their photocatalytic activities since these Mn species have different
electronic structures that have different effects on the band gaps of
the Mn oxides. The significant increase in the AOS of Mn in AMO
after reaction with NDMA (AOSafter = 3.90) suggests that there is a
re-oxidation of Mn in the lower oxidation states (Mn2+ or Mn3+) to
Mn4+ by some regenerated O2 species (Eq. (3)) during UV illumi-
nation. Although the amorphous structure of AMO did not change
after reaction with NDMA and UV light, the significant change in
the AOS of Mn due to re-oxidation of Mn in AMO is an important
contributing factor as to its high activity in enhancing NDMA pho-
todegradation. The XRD patterns of AMO and Pt/Mn3O4, and the
AOS of Mn in Pt/Mn3O4 after photocatalytic reaction did not sig-
nificantly change which suggest that UV illumination did not cause
change in the original structures and therefore the same numbers of
excited electrons and positive holes are possibly consumed during
reaction and/or recombination. Further, the photocatalytic activi-
ties of AMO and Pt/Mn3O4 could be mainly due to the presence of
surface adsorbed O2.
sities are higher in the presence of the catalysts (higher in AMO
than in Pt/Mn3O4) and increase with longer UV illumination times.
Fig. 7 shows the variations in the weak absorption band of NDMA
solution centered at 330 nm (A) in the presence of TNM with and
without the catalysts after UV illumination. In the presence of TNM
and UV light for 10 min (optimum time) under Ar saturation, the
NDMA peak shifts from 330 nm to 337 nm and increases absorbance
by 30% (ꢃA330 = 0.076) (B). However, broad and strong absorption
bands centered at 345 nm are observed in the presence of TNM,
UV light, and catalysts (C and D). Continuous UV illumination up
to 30 min significantly decreases the absorption accompanied by a
slight peak shift to a longer wavelength (350 nm) (E and F).
4. Discussion
In this study, absorption of incident photons by an optimum
amount of suspended catalyst particles appears to enhance the
photodegradation of NDMA. This was observed for both AMO and
Pt/Mn3O4 even though AMO has much larger surface area (higher
number of active sites) than Pt/Mn3O4. This suggests that there are
other factors that govern the photocatalytic activities of AMO and
Pt/Mn3O4 aside from surface area.
At constant NDMA concentration, excess loading of catalyst
(>25 mg per 100 mL NDMA) generally brought about lower % NDMA
photodegradation as observed for both AMO and Pt/Mn3O4. This
could be due to hindered penetration of UV light by the excess cat-
alyst particles with surfaces unbound to any NDMA molecules and
thus do not take part in the reaction of NDMA, reactive oxygen
When the reaction is carried out in an environment devoid of O2,
O2 is evolved from the catalysts themselves to the NDMA solution
containing partially reduced Mn oxide catalyst upon absorption of
UV light. Eq. (3) shows the suggested sequence of O2 release from
the catalyst [21]. Surface adsorbed O2 can accept electrons from
photogenerated electrons (and/or reduced Mn species) and become
negatively charged, O2− (bulk). Upon further absorption of UV light,
O2− (bulk) react with electrons to form active oxygen radicals (O−).
O− species react with electrons to form bulk O2−. Oxygen evolution
may be then due to migration of O2− (bulk) to the surface of the
catalyst as O2 (regenerated) [21]:
A
NDMA solution only
C
400
350
300
250
200
150
100
50
10 min UV photolysis:
B
C
D
no catalyst (under Ar
AMO
Pt/Mn O
)
D
30 min UV photolysis:
E
F
AMO
Pt/Mn O
B
F
A
A
E
0
280 290 300 310 320 330 340 350 360 370 380 390 400 410
hꢄ
hꢄ
hꢄ
hꢄ
hꢄ
+ e−−→ O2− −→ 2O−−→ O2− −→ 2O−−→
O2
(3)
Wavelength (nm)
O2
−
−
(bulk)
e
e
(surface adsorbed)
(bulk)
(on surface,regenerated)
Fig. 7. UV–Vis absorption spectra of 2.2 mM NDMA aqueous solutions with 0.24 mM
tetranitromethane indicator with and without the catalysts. UV illumination
(ꢀ = 254 nm) times are 0, 10, and 30 min at pH 7.0 under air saturation.
Some regenerated O2 could be transformed into reactive oxygen
species as peroxides or as oxygen radicals (•OH and •O2−) which

H.C. Genuino et al. / Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
291
−
Fig. 8. Hypothetical mechanism for the photodegradation of NDMA forming NO3 and HCOH as major photoproducts using the Mn oxide catalysts [21,41].
−
−
due to instability may remain on the surface for a short period of
time. Regenerated O2 can be exhausted which leads to a slower
rate of NDMA photodegradation or loss of activity. Spent catalysts
can easily be rejuvenated in an O2 environment which makes them
reusable.
of NO2 and NO3 compared with that of NDMA and presumably
partly due to their dark colors imparted by the Mn oxide catalysts. A
similar study supported this observation which showed−that radical
−
intermediates that were possibly generated from NO2 and NO3
during the complete UV illumination of NDMA at 254 nm did not
interfere in its photodegradation [41].
−
Significant electroactivity of AMO for O2 reduction to HO2
could be associated with high concentrations of lattice defects
and active sites of the amorphous material. Peroxide species can
undergo decomposition to form hydroxide (Eq. (4)) [36] or super-
oxide species using metal−oxides [37–39]. To explain the CV curves
D and E in Fig. 4, the HO2 formed from Mn oxide-catalyzed elec-
trochemical reduction of O2 in protic solvent has likely undergone
hydrolysis and decomposition into •OH and •O2− through a dispro-
portionation reaction in the presence of UV light. The •OH species
were successfully detected and the visible emissions observed in
the PL studies are attributed to the fluorescent product, monohy-
droxyterephthalate, formed by the reaction between terephthalic
acid and photogenerated •OH (Fig. 6). As expected, the visible emis-
sion increases as •OH are produced with continued UV illumination.
Considerably lower visible emission for Pt/Mn3O4 could be due to
less structural defects in crystalline Pt/Mn3O4 than in AMO:
The enhanced rate of NDMA photodegradation using AMO
and Pt/Mn3O4 significantly produced HCOH and NO3 as the
−
major oxidation photoproducts. A hypothetical mechanism for the
enhancement of NDMA photodegradation using Mn oxide cata-
lysts has therefore been proposed on the basis of experimental
results and of previous studies suggesting photolytic pathways
for the degradation of N-nitrosamines in water (Fig. 8) [41,43,44].
NDMA is adsorbed at the surface of the Mn oxide catalysts. Upon
absorption of photons by the catalyst bound to NDMA, the elec-
trons are excited from the valence band to their conduction bands
leaving positive holes in the valence band. In Pt/Mn3O4, the elec-
trons further transfer to Pt and the separation efficiency of the
photoinduced electrons and holes is enhanced. The photogener-
ated electrons in the conduction band of Mn oxide react with
dissolved O2 in aerated NDMA solution to form reactive oxygen
species such as •OH and •O2−. Under Ar saturation, the source of
reactive oxygen species is UV illumination of regenerated O2 (from
physisorbed O2) from the catalyst as depicted in Eq. (3). Initial
absorption of UV light by NDMA also yields photoexcited NDMA
[41,44]. The N-N bond in the photoexcited NDMA is photocleaved
and in the presence of regenerated O2, forms organic fragment as
N-methylidenemethylamine (CH2 NHCH3+) and NO fragment as
•NO. Reactive oxygen species formed further oxidize •NO to pro-
duce a more stable product, NO3− and then desorbed from catalyst
surface. N-methylidenemethylamine undergoes hydrolysis even-
tually forming formaldehyde and methylamine in 1:1 ratio.
−
HO2 (aq) + H2O(l) + 2e− → 3OH−
(4)
(aq)
Previous study has shown that •O2 are formed from UV illu-
mination of alkaline elec−trolyte solution in the presence of O2 [40].
In the present work, •O2 are also formed in addition to •OH radi-
−
cals. In the CV experiments using 1.0 M KOH, although •O2 were
−
not directly detected due to poor stability of superoxide species in
aqueous solution, reactions in aprotic solvents provided clear evi-
dence that superoxide anions are formed from reduction of O2 using
the catalysts. Ultimately, •O2 were spectroscopically detected
−
using TNM indicator. Eq. (5) shows the photoreduction of TNM,
C(NO2)4, in the presence of •O2− to form nitroform anion, C(NO2)3
−
[41]. C(NO2)3− in solution has a strong absorption band centered at
350 nm which explains the observed peak shifts and increase in UV
absorbance of NDMA (Fig. 7). In the absence of any catalyst under
Ar saturation, the slight peak shift and increase in the UV absorp-
tion could be explained by the direct UV illumination of NDMA and
photoreduction of TNM to C(NO2)3− [41]. In the presence of the cat-
alysts, the photoreduction of TNM is altered significantly (increase
or decrease of ε) which could be due to regeneration of O2 from the
catalysts with continuous UV illumination:
5. Conclusion
Morphologically uniform and nanostructured amorphous man-
ganese oxide (AMO) and crystalline Pt/Mn3O4 have been
successfully synthesized and applied as catalysts for the enhance-
ment of NDMA photodegradation with environmental relevance.
Mixed valency of Mn in the catalysts was observed before pho-
tocatalytic reaction. AMO and Pt/Mn3O4 have provided significant
improvements for the enhancement of photodegradation of NDMA
in water in high concentrations. These catalysts can be good alter-
natives for established photocalysts such as TiO2 and Pt/TiO2 since
they have comparable catalytic activities. AMO and Pt/Mn3O4
are stable toward O2 reduction and UV light, and are reusable.
Physisorbed O2 reacts with photogenerated electrons to form reac-
tive oxygen species and facilitates the trapping of photogenerated
electrons and improves photocatalytic activity. This study provides
−
−
C(NO2)4(aq) + •O2 (aq) → C(NO2)3 (aq) + NO2(g) + O2(g)
(5)
−
−
Previous studies showed that NO2 and NO3 produced as
NDMA is photodegraded could possibly be competing with NDMA
for absorption of UV light [11,42]. Here, interference studies suggest
that almost all incident photons of ꢀ = 254 nm are absorbed by the
NDMA–catalyst mixtures due to low molar absorption coefficients

292
H.C. Genuino et al. / Journal of Photochemistry and Photobiology A: Chemistry 217 (2011) 284–292
an important groundwork for future investigations of the enhance-
ment of photodegradation of NDMA in water. Future studies should
require experimental data from the application of these Mn oxide
catalysts in the actual removal of NDMA from real and complex
environmental samples.
[13] J. Lee, W. Choi, Environ. Sci. Technol. 39 (2005) 6800–6807.
[14] C. Lee, W. Choi, J. Yoon, Water Res. 41 (2005) 581–590.
[15] C. Lee, C. Schmidt, J. Yoon, U. von Gunten, Environ. Sci. Technol. 41 (2007)
2056–2063.
[16] C.M. Sharpless, K.G. Linden, Environ. Sci. Technol. 37 (2003) 1933–1940.
[17] A.J. Frierdich, C.E. Joseph, T.J. Strathmann, Appl. Catal. B: Environ. 90 (2009)
175–183.
[18] L. Gui, R.W. Gillham, R.W. Gillham, Environ. Sci. Technol. 34 (2000) 3489–3494.
[19] M.G. Davie, K. Shih, F.A. Pacheco, J.O. Leckie, M. Reinhard, Environ. Sci. Technol.
42 (2008) 3040–3046.
Acknowledgments
[20] S. Jana, S. Praharaj, S. Panigrahi, S. Basu, S. Pande, C.-H. Chang, T. Pal, Org. Lett.
9 (2007) 2191–2193.
The authors thank Heng Zhang for collecting the XPS data. This
research was supported by VeruTEK Technologies and the U.S.
Department of Energy, Office of Basic Energy Sciences, Division of
Chemical, Biological and Geosciences.
[21] H. Cao, S.L. Suib, J. Am. Chem. Soc. 116 (1994) 5334–5342.
[22] J. Chen, J.C. Lin, V. Purohit, M.B. Cutlip, S.L. Suib, Catal. Today 33 (1997) 205–214.
[23] S.R. Segal, S.L. Suib, Chem. Mater. 11 (1999) 1687–1695.
[24] A. Mill, S. Le Hunte, J. Photochem. Photobiol. A: Chem. 108 (1997) 1–35.
[25] B. Kraeutler, A.J. Bard, J. Am. Chem. Soc. 100 (1978) 2239.
[26] D. Glover, B. Schumm Jr., A. Kozowa, Handbook of Manganese Dioxides Battery
Grade, International Battery Materials Association, Cleveland, OH, USA, 1989.
[27] US EPA Method 3520C, Continuous Liquid–Liquid Extraction, Revision 3, 1996.
[28] G.T. Richardson, J.A. Davies, J.G. Edwards, Fresenius J. Anal. Chem. 343 (1992)
473–474.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jphotochem.2010.10.021.
[29] G.J. Hanna, in: S. Siggia (Ed.), Instrumental Methods of Organic Functional
Group Analysis, Wiley-Interscience, New York, USA, 1972.
[30] T. Nash, J. Biochem. 55 (1953) 416.
References
[31] C.G. Hatchard, C.A. Parker, Proc. R. Soc. Lond. A 235 (1956) 518–536.
[32] J. Lee, H. Park, W. Choi, Environ. Sci. Technol. 36 (2002) 5462–5468.
[33] L. Mao, D. Zhang, T. Sotomura, K. Nakatsu, N. Koshiba, T. Ohsaka, Electrochim.
Acta 48 (2003) 1015–1021.
[1] C.T. Gombar, H.M. Pylypiw Jr., G.W. Harrington, Cancer Res. 47 (1987) 343–347.
[2] E.C. Fleming, J.C. Pennington, R.A. Howe, M.R. Colsman, K.E. Garrett, B. Wachob,
Hazard. Ind. Wastes 26 (1994) 285–292.
[34] D.L. Maricle, W.G. Hodgson, Anal. Chem. 37 (1965) 1562–1565.
[35] F. Kapteijn, L. Singoredjo, A. Andreini, J.A. Moulijn, Appl. Catal. B: Environ. 3
(1994) 173–189.
[36] Holleman-Wiberg, The Chalcogen Group. Inorganic Chemistry, Academic Press,
New York, 2001, pp. 473–475.
[37] Y. Ono, T. Matsumura, N. Kitajima, S.I. Fukuzumi, J. Phys. Chem. 81 (1977) 1307.
[38] N. Kitajima, S. Fukuzumi, Y. Ono, J. Phys. Chem. 82 (1978) 1505.
[39] S. Do, B. Batchelor, H. Lee, S. Kong, Chemosphere 75 (2009) 8–12.
[40] Y. Ilan, J. Rabani, A. Henglein, J. Phys. Chem. 80 (1976) 1558.
[41] C. Lee, W. Choi, J. Yoon, Environ. Sci. Technol. 39 (2005) 9702–9709.
[42] C. Lee, J. Yoon, J. Photochem. Photobiol. A: Chem. 189 (2007) 128–134.
[43] J. Polo, Y.L. Chow, J. Natl. Cancer Inst. 56 (1976) 997–1001.
[44] C. Lee, W. Choi, Y.G. Kim, J. Yoon, Environ. Sci. Technol. 39 (2005) 2101–2106.
[3] N-nitrosodimethylamine in drinking-water, Background document for devel-
opment of World Health Organization (WHO) Guidelines for Drinking-water
Quality, Geneva, Switzerland, 2006.
[4] Z. Chen, R.L. Valentine, Environ. Sci. Technol. 40 (2006) 7290–7297.
[5] A.C. Gerecke, D.L. Sedlak, Environ. Sci. Technol. 37 (2003) 1331–1336.
[6] W.A. Mitch, D.L. Sedlak, Environ. Sci. Technol. 36 (2002) 588–595.
[7] W.A. Mitch, J.O. Sharp, R.R. Trussell, R.L. Valentine, L. Alvarez-Cohen, D.L. Sedlak,
Environ. Eng. Sci. 20 (2003) 389–404.
[8] W.A. Mitch, D.L. Sedlak, Environ. Sci. Technol. 38 (2004) 1445–1454.
[9] I. Najm, R. Trussell, J. Am. Water Works Assoc. 93 (2001) 92–99.
[10] M.H. Plumlee, M. Reinhard, Environ. Sci. Technol. 41 (2007) 6170–6176.
[11] M.I. Stefan, J.R. Bolton, Helv. Chim. Acta 85 (2002) 1416–1426.
[12] Orange County Water District takes a proactive stance on newly regulated
compound N-nitrosodimethylamine: OCWD recommends taking two drinking
water wells out of service, Press Release from Orange County Water District,
Fountain Valley, CA, USA, 2000.