The photooxidative degradation of N-methylpyrrolidinone in the presence of Cs3PW12O40 and TiO2 colloid photocatalysts

Article abstract of DOI:10.1021/es980912p

The mechanisms of the heterogeneous photocatalytic behavior of TiO₂ and Cs₃PW₁₂O₄₀ for the oxidative destruction of N-methylpyrrolidinone (NMP) in water are discussed. Both photocatalysts generate H-methylsuccinimide (NMS) and succinimide as the main intermediates from oxidation at the α-carbon and methyl positions. 2-Pyrrolidmone (2P) and 5-hydroxy-N-methylpyrrolidinone (5-HNMP) have also been identified as primary intermediates These intermediates are all broken down under further photooxidation. However, differences in product distribution indicate the possibility of several mechanistic pathways. Hydroxyl radical-like attack predominates for TiO₂- photocatalyzed degradation, whereas the polyoxotungstate appears to oxidatrvely destroy NMP by both OH radical attack and direct oxidation For TiO₂, the relative quantum yields for NMP oxidation (0.5 g L^-1 TiO₂, pH 3. 254 nm, 7 × 10^-6 ein sec^-1) increase from <0.01 at 10 ppm NMP to near 0.03 at 60-100 ppm. For Cs₃PW₁₂O₄₀ under the same conditions, the apparent quantum yield is nearly invariant (0.005-0.009) over the concentration range 10- 1000 ppm.

Full text of DOI:10.1021/es980912p

Environ. Sci. Technol. 1999, 33, 3193-3198
of an electron from the valence band to the conduction band
of TiO , resulting in the generation of a hole-electron pair.
The Photooxidative Degradation of
N-Methylpyrrolidinone in the
2
This pair can recombine within a few nanoseconds; however,
+
the positive vacancy (h ) in the valence band is a very strong
oxidant and can undergo charge transfer with adsorbed
species on the photocatalyst surface on a picosecond time
Presence of Cs PW O and TiO
3
12 40
2
Colloid Photocatalysts
scale (2, 8-10). For example, water on the surface of the TiO
2
particle can be oxidized, resulting in an adsorbed hydroxyl
radical (11). These radicals attack the aqueous organic
contaminant, generating a series of oxidized intermediates
and ultimately resulting in total degradation of the con-
†
D U A N E A . F R I E S E N ,
†
J O H N V . H E A D L E Y, A N D
C O O P E R H . L A N G F O R D * , ‡
2
taminant to CO . There can also be contributions from direct
oxidation of adsorbed organics.
National Water Research Institute, Environment Canada,
1 Innovation Boulevard, Saskatoon, SK, Canada S7N 3H5,
and Department of Chemistry, The University of Calgary,
500 University Drive NW, Calgary, AB, Canada T2N 1N4
1
Recently, there has been an interest in molecular-based
materials that perform similar functions. Some of the most
promising are polyoxometalates, tungsten-oxygen clusters,
2
such as H
results in the formation of a ligand-to-metal charge transfer
LMCT) excited state (lifetime approximately 100-200 ps in
3
PW12O40 (12-23). UV irradiation of this species
(
The mechanisms of the heterogeneous photocatalytic
behavior of TiO2 and Cs3PW12O40 for the oxidative destruction
of N-methylpyrrolidinone (NMP) in water are discussed.
Both photocatalysts generate N-methylsuccinimide (NMS)
and succinimide as the main intermediates from oxidation
at the R-carbon and methyl positions. 2-Pyrrolidinone (2P)
and 5-hydroxy-N-methylpyrrolidinone (5-HNMP) have
also been identified as primary intermediates. These
intermediates are all broken down under further photooxi-
dation. However, differences in product distribution
indicate the possibility of several mechanistic pathways.
Hydroxyl radical-like attack predominates for TiO2-
water) (24, 25) with considerable oxidizing capacity (2.63 V
vs NHE, the normal hydrogen electrode) (26). This complex
has recently been shown to photodegrade organic contami-
nants, such as chlorophenols to CO and HCl (27-31).
2
However, a major drawback to the practical use of this system
is its high water solubility, which impedes ready recovery
and reuse of the photocatalyst (and, therefore, its cost-
effectiveness).
3 40
Cesium salts of polyoxometalates, such as Cs PW12O ,
are considered to be insoluble and are well-known for their
thermal catalytic properties for reactions in both gas and
liquid phases (32-41). We have recently found that Cs
3
-
PW12
O
40 is also an effective heterogeneous photocatalyst for
photocatalyzed degradation, whereas the polyoxotungstate
appears to oxidatively destroy NMP by both OH radical
attack and direct oxidation. For TiO2, the relative quantum
the photooxidation of 2-propanol to acetone (42). The present
work is a further examination of this substance for the
photocatalytic degradation of model contaminants in water.
Specifically, two issues are addressed: (1) what are the
degradation intermediates, and (2) what is the efficiency of
contaminant degradation for Cs PW O ? These questions
-
1
yields for NMP oxidation (0.5 g L TiO2, pH 3, 254 nm, 7
-
6
-1
×
10 ein sec ) increase from <0.01 at 10 ppm NMP to
near 0.03 at 60-100 ppm. For Cs3PW12O40 under the
same conditions, the apparent quantum yield is nearly
invariant (0.005-0.009) over the concentration range 10-
3
12 40
2
are examined in the context of a standard TiO material
(Degussa P25) and are important for evaluating the prac-
ticality of this material in schemes for contaminant degrada-
tion.
1
000 ppm.
N-Methylpyrrolidinone (NMP), 1, is a cyclic, nonaromatic
amide commonly used as an industrial solvent and a
component of cleaning solutions. It has been detected in
textile mill effluents in Canada and in landfill leachates in
Japan (43, 44). The present study evaluates the efficacy of
TiO and Cs PW12O40 for the photocatalytic destruction of
2 3
NMP. Notably, its simple structure serves as an ideal model
for the general comparison of photooxidative destruction of
Introduction
2
In recent years, TiO photocatalysis has become a widely
studied option in the advanced oxidative treatment of organic
pollutants in wastewater due to its ability to destroy a wide
range of contaminants using solar or UV irradiation (1-6).
The ultimate product of the photooxidative process can be
CO
chloride). However, in certain cases (e.g., atrazine, ref 7),
facile mineralization to CO does not occur due to the
formation of a stable intermediate that is recalcitrant to TiO
2 2
and H O (and relevant inorganic ions such as nitrate or
2 3
nonaromatic contaminants by TiO and Cs PW12O40. Fur-
2
thermore, since NMP has minimal light absorption at 254
nm, complicating effects from direct photolysis can be
minimized.
2
photooxidation. These findings suggest that the development
of other photocatalytic materials, which can access other
mechanistic pathways, is important. We present herein initial
studies on a comparison of a heterogeneous polyoxotungstate
3 2
photocatalyst (Cs PW12O40) with TiO for the destruction of
a nonaromatic contaminant (N-methylpyrrolidinone) in
water.
Titanium dioxide (TiO
2
) is a UV-absorbing semiconductor
material. Absorption of near-UV light results in the promotion
*
To whom correspondence should be addressed: phone: (403)-
Experimental Section
2
20-5465; fax: (403)289-8926; e-mail: chlangfo@ucalgary.ca.
†
‡
TiO
2 3 40
(P25 grade) was obtained from Degussa Corp. H PW12O
National Water Rsearch Institute.
University of Calgary.
was synthesized according to literature methods (45). N-
1
0.1021/es980912p CCC: $18.00

1999 Am erican Chem ical Society
VOL. 33, NO. 18, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 1 9 3
Published on Web 07/31/1999

FIGURE 1. Diffuse reflectance spectrum of powdered Cs
3 40
PW12O
and TiO in silica gel (8 mg/g silica gel) with pure silica gel as a
2
background reference.
3 40
Methylpyrrolidinone was obtained from Aldrich. Cs PW12O
was synthesized by the metathesis of concentrated solutions
of CsCl and H 40. Diffuse reflectance spectra were
3
PW12O
obtained using an Ocean Optics S1000 optical fiber spec-
trometer with a reflectance probe accessory.
The arrangement for photocatalysis consisted of an
annular quartz cell (volume 150 mL) that surrounded a 15
W lamp (Philips TUV low-pressure Hg lamp for near
monochromatic output at 254 nm). Suspensions of the
photocatalyst and aqueous contaminant solutions were kept
aerated and stirred by a steady stream of air. Irradiation
periods were kept short (less than 10 min) in order to
minimize the effects of heating. Incident light intensities were
measured using ferrioxalate actinometry (46, 47). The light
intensity from this setup as measured by actinometry was
FIGURE 2. Concentration of intermediates (for known intermediates
N-methylsuccinimide NMS and succinimide S) and peak areas (for
intermediate 1 at 1.58 min elution and intermediate 2 at 2.00 min
elution) arising from the irradiation (254 nm) of 12.2 ppm NMP and
-
6
-1
determined to be 7.2 ( 0.2 × 10 ein sec . Photocatalytic
degradations were monitored using HPLC-UV (Waters 600E
controller and 484 UV absorbance detector). The degradation
of NMP and formation of intermediates was monitored with
isocratic elution (3-10% acetonitrile/ 90-97% water, 1 mL
-
1
-1
(a) 0.5 gL Cs PW O and (b) 0.5 gL TiO2.
3
12 40
3
A diffuse reflectance spectrum for Cs PW12O40 powder
diluted with silica gel (vs a pure silica gel reference) is shown
in Figure 1. The spectrum shows an absorbance maximum
-
1
min ) on a reversed-phase C18 NovaPak column. The
detection wavelength was 212 nm. The identity of interme-
diates was determined against authentic standards (Aldrich).
TOC analysis was performed with a Technicon Autoanalyzer.
Electrospray-MS analysis was performed with an Autospec
Q mass spectrometer (Micromass, UK) with an electrospray
interface in the positive ion mode. Instrumentation settings
have been given previously (48). For electrospray-MS/ MS,
the magnet was parked at m/ z 115 or 85 (the molecular
weights of the unknown intermediates), and the quadrupole
was scanned from m/ z 40-150.
between 265 and 270 nm, in agreement with the solution
3-
absorption spectrum for PW12
light absorption for TiO
O40 (12, 29, 52). In contrast,
2
becomes significant between 300
and 350 nm. Both materials are significant absorbers at 254
nm, the irradiation wavelength used in this study.
Photodegradation of N-Methylpyrrolidinone. N-Meth-
ylpyrrolidinone (1) is rapidly degraded upon 254 nm ir-
radiation of aqueous slurries of TiO
example, at a lamp intensity of 7 × 10 ein sec and a
photocatalyst loading of 0.5 g L , the apparent first-order
2 3
or Cs PW12O40. For
-
6
-1
-
1
-
4
Results
rate constants for the disappearance of 1.2 × 10 M NMP
-
1
-1
are 0.34 min for TiO (P25) and 0.14 min for Cs PW12O .
2
3 40
Characterization of Photocatalyst Material. The physical
In contrast, direct irradiation of NMP solutions in the absence
of either photocatalyst leads to negligible photochemistry.
The molar absorptivity of NMP was measured to be less than
properties of cesium-polyoxotungstate materials, such as
Cs
3
PW12O40, are well-known from thermal catalysis studies.
The surface area for BET (Brunauer-Emmett-Teller) ad-
-
1
-1
2
-1
1 M cm at 254 nm; thus, the contaminant does not act
as a direct absorber or inner filter for the concentrations
used in these experiments.
sorption (126 m
g ) is in agreement with previous values
and indicates a mesoporous material (32, 37). In comparison,
2
-1
TiO
The particle size, as measured by light scattering in DMSO
and in H O, is approximately 300 nm. Previous studies have
indicated that Cs 40 are made up
2
(Degussa P25) has a specific surface area of 55 m g .
It is well-known that TiO
organics proceeds through the stepwise formation of inter-
mediates, usually forming CO as the ultimate oxidation
product (1-6). For both TiO and Cs 40, major inter-
2
-photocatalyzed degradation of
2
3
PW12
O
40 and Cs2.5
H
0.5PW12
O
2
of small crystallites of 10-12 nm diameter that can form
larger aggregates (32, 37, 49). These aggregates are likely
candidates for the suspended particles in liquid phase
observed in the present study. At pH 7 or below, the white
colloid remained insoluble and suspended in aqueous media.
However, at high pH values (pH > 9) the colloidal polyoxo-
metalate salt dissolved quickly and displayed a loss in the
UV absorption spectrum above 230 nm, indicating a break-
2
3
PW12O
mediates formed by oxidation that were detected and
quantified by HPLC-UV were N-methylsuccinimide (NMS)
and succinimide (S). Two additional intermediates were
detected that eluted at 1.6 min (intermediate #1) and 2.0 min
(intermediate #2) with 5:95 acetonitrile/ water under the LC
conditions described in the Experimental Section. Extended
photolysis led to the formation of more extensively oxidized
products and led to mineralization of NMP by both pho-
tocatalysts as observed by TOC analysis. For example, a
3
-
down of the PW12O40 unit due to basic conditions (29, 50,
5
1).
3
1 9 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 18, 1999

solution of 0.2 mM NMP was mineralized in 1.5 h by 0.5 g
-
1
-1
L
2 3 40
TiO and 3.5 h by 0.5 g L Cs PW12O .
Figure 2 shows the time development (as measured by
HPLC-UV) of NMP degradation and formation of the
intermediates. The growth and decay of intermediates 1 and
2
occur early in the photolysis, and the concentration of
NMS and succinimide continues to increase as the concen-
tration of these intermediates decreases, which suggests that
these may be precursors to NMS and succinimide formation.
FIGURE 3. Relative quantum yield O versus NMP concentration at
-1
2 3
pH 3 for TiO and Cs PW12O40 slurries (0.5 gL , 254 nm irradiation).
Intermediate 1 is almost absent for TiO
2 3
compared to Cs -
PW12
O
40. Electrospray positive-ion MS/ MS of photocatalyti-
in degradation of the intermediates, until mineralization is
achieved. The same trend is observed for Cs 40, where
the amount of NMP, NMS, and succinimide accounted for
1 (2.5 min), 71 (5 min), 56 (7.5 min), 47 (10 min), and 68%
15 min) of the initial concentration of NMP. The low
percentages at 7.5 and 10 min are at the maximum
concentrations of the primary intermediates 1 and 2 identified
above, although overlapping peaks from minor amounts of
other intermediates may also contribute. These react further
to give NMS and succinimide, resulting in the higher
percentage at 15 min. There may also be some contribution
of overlapping peaks from secondary and tertiary intermedi-
ates that lower the measured concentration of succinimide.
cally treated solutions of NMP detected significant amounts
of an intermediate with mass 115 (m/ z of M-H ) 116),
which was at much higher concentration for Cs
3
PW12O
+
3
PW12
O
40 than
8
(
for TiO
2
, a first indication of mechanistic differences.
Collision-induced dissociation of this ion resulted in a peak
at (M-H minus 18), which would be consistent with the
+
loss of water and thus with the presence of a hydroxy group
on the intermediate. From these results, we assign inter-
mediate 1 as 5-hydroxy-N-methylpyrrolidinone (5-HNMP),
shown below. This species has also been shown to be formed
concurrently with NMS upon electrochemical oxidation of
NMP in aqueous acetonitrile (53).
2 3 40
Relative quantum yields φ for TiO - and Cs PW12O -
photocatalyzed degradation of NMP were determined in an
annular cell configuration (with the slurry completely sur-
rounding the lamp) at a wavelength of high absorbance (254
nm) for both photocatalysts. Initial experiments were per-
formed to determine the reaction yield as a function of catalyst
loading in the slurry. The yields did not increase above a
-
1
2 3
loading level of 0.5 g L of TiO or Cs PW12O40, indicating
Appreciable formation of intermediate 2 was observed
maximum photon absorption for the experimental setup (54).
This loading level was used for the quantum yield deter-
minations. The extent of reaction was kept below 15-20%
conversion to limit the effects of lowered substrate concen-
for both photocatalysts. Similarly, an intermediate of mass
+
8
5 (m/ z M-H ) 86) was detected using both GC-MS and
electrospray-MS/ MS for solutions treated by either photo-
catalyst. As suggested by the results in Figure 2b, intermediate
tration. It was found that for TiO
2
the quantum yields (based
2
may be a precursor to succinimide formation and it is,
-4
on ferrioxalate actinometry) varied from 0.008 at 1 × 10
M
therefore, assigned to 2-pyrrolidinone (2P). The time de-
velopment of 2-pyrrolidinone and succinimide are similar
for both photocatalysts. Figure 2b also shows that succinimide
formation is accelerated as the rate of NMS formation
decreases, strongly suggesting that the formation of suc-
cinimide can also result from oxidative attack at the N-methyl
position of NMS.
-4
NMP to 0.022-0.027 above 6 × 10 M NMP (Figure 3). Over
-4
-2
a range of 10 -10 M NMP, the quantum yields for Cs
3
-
PW12
O
40 were nearly invariant at 0.005-0.009 (Figure 3). The
origin of the differing behaviors with respect to contaminant
concentration is still under investigation; however, in practi-
cal terms, it appears that TiO is the more efficient photo-
2
catalyst for NMP degradation.
Discussion
Despite the contrast drawn between PW12
3
-
O
40
as a “mo-
lecular-based” material and TiO
2
as a semiconductor material,
their redox characteristics are surprisingly similar. The
4
-/ 3-
ground-state redox potential for PW12
O
40
is at -0.37 V
vs NHE. The energy gap between the ground state and the
excited LMCT state is 2.99 eV, which results in an excited-
state potential vs NHE of +2.63 V (26). In comparison, the
The total concentration of NMP, NMS, and succinimide
was monitored during the reactions represented in Figure 2.
For TiO
2
, the total amount represented by these three
2
corresponding “ground-state” redox potential for TiO at pH
compounds was 91 (2.5 min), 77 (5 min), 83 (7.5 min), 92 (10
min), and 26% (15 min), based on the initial concentration
of NMP. The high percentage at 10 min irradiation (where
NMP has essentially disappeared) indicates efficient reaction
of primary intermediates to form NMS and succinimide and
a low degree of product adsorption. Further reaction results
3 is - 0.2 V (55). With a band gap of 3.2 eV, the potential for
reduction of the photogenerated valence-band hole is +2.9
V. The reduction potential for hydroxyl radical-hydroxide
ion in homogeneous phase is +2.62 V vs NHE (56). Therefore,
excitation of TiO
2
in water by UV light generates OH radicals
at the surface that may be the primary oxidant in photo-
VOL. 33, NO. 18, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 1 9 5

catalytic degradations (57). It appears from the above data
that it may also be thermodynamically possible for the
excited-state tungstate to generate OH radicals although the
68); in the case of NMP, this would lead directly to the
formation of NMS:
2
driving force is not as large as for TiO . Because of the small
overvoltage, however, a rapid reaction with water is unlikely,
whereas picosecond spectra (24) indicate initial oxidation of
propanol in <20 ps. Efficient water oxidation on the surface
of the tungstate colloid would depend on the surface
environment making oxidation somewhat more favorable
than in solution. Yamase has reported the detection of OH
radicals by ESR spin trapping upon irradiation of related
polyoxotungstates (58). However, polyoxotungstate species
also possess an ability to undergo reversible multielectron
reduction (59). In comparison, the E1/ 2 value for oxidation
of NMP from cyclic voltammetry is +2.1 V vs NHE (60). Thus,
for both catalysts the potential exists for oxidation by both
direct electron transfer and OH radical attack.
This is consistent with the high yield of NMS initially produced
by TiO photooxidation relative to NMP degradation (ap-
2
proximately 75% conversion at 2.5 min), and the low yield
of 5-HNMP in Figure 2b, even at short irradiation times.
The similarity in intermediates makes it tempting to
conclude that NMP degradation follows a mechanistic
pathway derived from hydroxyl radical attack for Cs PW O
Chemical and electrochemical oxidation of NMP results
in the substitution of carbonyl or hydroxyl groups at the
R-carbon position, and it is well-known that the R-carbon
of lactams such as NMP and 2-pyrrolidinone, is susceptible
to H-atom abstraction (53, 60-63). Both OH radicals and
excited-state polyoxotungstates are potent hydrogen atom
abstractors (26, 64, 65). Thus, it is not surprising that for both
photocatalysts, there is a substantial yield of NMS as a primary
3
12 40
as well. However, in the case of Cs PW O , the primary
3
12 40
product distributions (Figure 2) indicate that there are added
complexities. After 2.5 min irradiation (Figure 2a), the yield
of NMS from NMP was approximately 14%, with the
significant presence of 5-HNMP. These results are similar to
those found in the direct anodic oxidation of NMP in aqueous
acetonitrile at a platinum electrode where the product ratio
of 5-HNMP to NMS was 55:12% (yield based on conversion
from NMP) (53). This is an indication that the hydroxy species
could be preferentially formed in a direct oxidation process,
such as on the surface of a platinum electrode. The low initial
NMS yield for Cs PW O photocatalysis would then indicate
2
intermediate. An early study involving TiO photocatalysis
of NMP at relatively high concentrations (0.1 M NMP in water)
showed nearly quantitative conversion to NMS under aerated
conditions (66). However, no reaction was observed under
deoxygenated conditions, and use of a metal-ion electron
acceptor (Cu(II)) in the absence of oxygen led to the formation
of an alkene elimination product rather than the carbonyl
3
12 40
that direct oxidation of NMP on the surface of the excited
polyoxometalate colloid is an important mechanistic path-
way. Support for this possibility comes from the work of Hill
et al. on the com parative photooxidation of tetrahy-
drothiophene by solution polyoxometalate and particulate
species. These findings underline the critical role of O
-derived species) in the overall photocatalytic process.
It is generally accepted that O promotes TiO photo-
2
(or
O
2
2
2
catalytic oxidation by acting as an electron acceptor, removing
excess electrons from the surface of the semiconductor
material and forming superoxide radical anions (which can
TiO
of TiO
2
(59). Anaerobic conditions led to the loss of reactivity
; however, the polyoxometalate continued to function
2
subsequently protonate to form HO
results strongly suggest that O or the derived superoxide
also plays a reactive role in the TiO -based degradation of
NMP. Masui et al. have shown that in oxygenated nonaqueous
acetonitrile) solution, the presence of a chemical H-atom
2
radicals). The present
as a photooxidant, in part because of its ability to undergo
reversible multielectron reduction. In the present case, two-
electron oxidation of NMP by the polyoxometalate could
lead to the formation of the alcohol by H-abstraction followed
by electron transfer and water substitution of the resulting
carbocation. Further oxidation of the hydroxy group leads
to subsequent formation of the carbonyl group (i.e., NMS)
as is well-known for the polyoxometalate photooxidation of
alcohols and would account for the increased rate of
formation of NMS after 10 min in Figure 2a.
2
2
(
abstractor (phthalimide N-oxyl) leads to NMS formation in
high yield, most likely through a peroxide intermediate
derived from O substitution at the R-carbon (60). Recently,
2
Drago et al. have synthesized and characterized 5-hydro-
peroxy-NMP (5-HPNMP) (61, 62). This species easily un-
dergoes further reaction to form NMS.
A novel aspect of the present work is the early formation
2 3
of succinimide by both TiO and Cs PW12O40. In contrast to
the formation of NMS, the yield of succinimide per amount
of NMP degraded is very similar for both photocatalysts.
Previous studies of NMP oxidation by photochemical and
electrochemical means did not note the formation of
succinimide by attack at the N-methyl group, although related
seven-member ring lactams with N-methyl substituents
showed demethylation upon direct anodic oxidation indi-
cating that the N-methyl position is also potentially reactive
toward oxidation (53, 60). N-Methylene groups with methyl
or phenyl substituents also showed conversion to corre-
sponding carbonyl forms upon chemically mediated H-atom
abstraction in acetonitrile (60). In the case of photocatalyzed
oxidation, abstraction of the methyl hydrogen followed by
An oxidation mechanism can be proposed that is con-
sistent with these results. Photooxidation of NMP by TiO is
2
initiated via H-atom abstraction by a hydroxyl radical
generated upon band gap excitation of the semiconductor.
The attack occurs preferentially (but not exclusively, vide
infra) at the R-carbon, resulting in a radical based at that
position. Further attack by OH would be expected to yield
2 2
O or HO attack would lead to the formation of a reactive
aldehyde substituent that would undergo further reaction to
form 2-pyrrolidinone and formaldehyde or carbon dioxide.
Subsequent oxidation at the R-carbon position as discussed
above yields succinimide.
The following scheme summarizes the possible degrada-
2 3 40
tion pathways for TiO and Cs PW12O :
5
-hydroxy-NMP (5-HNMP) by addition; however, attack by
superoxide generated in the reductive photocatalytic step
would yield the hydroperoxide intermediate. Upon proto-
nation, 1° and 2° alkyl hydroperoxides can undergo a
rearrangement to form carbonyls and alcohols or water (67,
3
1 9 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 18, 1999

(
(
(
(
6) Serpone, N., Pelizzetti, E., Eds. Photocatalysis - Fundamentals
and Applications; Wiley-Interscience: New York, 1989.
7) Pelizzetti, E.; Maurino, V.; Minero, C.; Carlin, V.; Pramauro, E.;
Zerbinati, O.; Tosato, M. L. Environ. Sci. Technol. 1990, 24, 1559.
8) Lepore, G. P.; Pant, B. C.; Langford, C. H. Can. J. Chem. 1993,
7
1, 2051.
9) Arbour, C.; Sharma, D. K.; Langford, C. H. J. Phys. Chem. 1990,
4, 331.
9
(
(
(
(
10) Kamat, P. V.; Dimitrijevic, N. M. Sol. Energy 1990, 44, 83.
11) Serpone, N. J. Photochem. Photobiol. A 1997, 104, 1.
12) Papaconstantinou, E. Chem. Soc. Rev. 1989, 18, 1.
13) Dimotikali, D.; Papaconstantinou, E. Inorg. Chim. Acta, 1984,
8
7, 177.
(
(
(
14) Hiskia, A.; Papaconstantinou, E. Inorg. Chem. 1992, 31, 163.
15) Ioannidis, A.; Papaconstantinou, E. Inorg. Chem. 1985, 24, 439.
16) Fox, M. A.; Cardona, R.; Gaillard, E. J. Am. Chem. Soc. 1987, 109,
6
347.
(
(
17) Hill, C. L.; Bouchard, D. A. J. Am. Chem. Soc. 1985, 107, 5148.
18) Hill, C. L.; Kozik, M.; Winkler, J.; Hou, Y.; Prosser-McCartha, C.
M. In Photosensitive Metal-Organic Systems - Mechanistic
Principles and Applications, Kutal, C., Serpone, N., Eds., ACS
Symp. Ser. 1993, 238, 243.
(
(
19) Renneke, R. F.; Kadkhodayan, M.; Pasquali, M.; Hill, C. L. J. Am.
Chem. Soc. 1991, 113, 8357.
20) Renneke, R. F.; Pasquali, M.; Hill, C. L. J. Am. Chem. Soc. 1990,
1
12, 6585.
(
(
(
(
(
21) Renneke, R. F.; Hill, C. L. J. Am. Chem. Soc. 1988, 110, 5461.
22) Renneke, R. F.; Hill, C. L. J. Am. Chem. Soc. 1986, 108, 3528.
23) Yamase, T.; Usami, T. J. Chem. Soc., Dalton Trans. 1988, 183.
24) Friesen, D. A.; Langford, C. H., unpublished observations.
25) Hill, C. L.; Bouchard, D. A.; Kadkhodayan, M.; Williamson, M.
M.; Schmidt, J. A.; Hilinski, E. F. J. Am. Chem. Soc. 1988, 110,
5
471.
(
26) Hill, C. L.; Prosser-McCartha, C. M. In Photosensitization and
Photocatalysis Using Inorganic and Organometallic Compounds,
Kalyanasundaram, K., Gratzel, M., Eds.; Kluwer: Dordrecht,
1
993; pp 307-330.
(
27) Mylonas, A.; Papaconstantinou, E. J. Photochem. Photobiol. A.
1996, 94, 77.
The oxidation proceeds through a combination of path-
ways to varying degrees for both photocatalysts. Direct
oxidation of the contaminant as well as OH radical attack
appears to be an important contributor to the degradation
by the polyoxometalate. Further oxidation of succinimide in
(28) Mylonas, A.; Hiskia, A.; Papaconstantinou, E. J. Mol. Catal. A.
996, 114, 191.
1
(
(
(
(
29) Einaga, H.; Misono, M. Bull. Chem. Soc. Jpn. 1996, 69, 3435.
30) Mylonas, A.; Papaconstantinou, E. J. Mol. Catal. 1994, 92, 261.
31) Einaga, H.; Misono, M. Bull. Chem. Soc. Jpn. 1997, 70, 1551.
32) Izumi, Y.; Ono, M.; Kitagawa, M.; Yoshida, M.; Urabe, K.
Microporous Mater. 1995, 5, 255.
2
the presence of UV-light and TiO leads to ring opening and
+ -
mineralization to NH
4 3
and NO (69). Thus, photocatalyzed
degradation appears to be an effective method for oxidative
destruction of NMP contaminants (and in general, nonaro-
matic organic substances) in wastewater, with TiO being
2
the more efficient photocatalyst as also illustrated by our
previous work with 2-propanol (42). However, the utilization
of an additional significant degradation mechanism (i.e.,
direct oxidation) by the tungstate may make it an attractive
(33) Izumi, Y.; Ono, M.; Ogawa, M.; Urabe, K. Chem. Lett. 1993, 825.
(
(
34) Izumi, Y.; Ogawa, M.; Urabe, K. Appl. Catal. A 1995, 132, 127.
35) Izumi, Y.; Ogawa, M.; Nohara, W.; Urabe, K. Chem. Lett. 1992,
1
987.
36) Tatematsu, S.; Hibi, T.; Okuhara, T.; Misono, M. Chem. Lett.
984, 865.
(
1
(37) Okuhara, T.; Mizuno, N.; Misono, M. Adv. Catal. 1996, 41, 113.
(38) Okuhara, T.; Nishimura, T.; Watanabe, H.; Na, K.; Misono, M.
Stud. Surf. Sci. Catal. 1994, 90, 419.
alternative treatment in cases where TiO
2
photocatalysis
(
(
39) Misono, M. Stud. Surf. Sci. Catal. 1993, 75, 69.
40) Lee, K. Y.; Arai, T.; Nakata, S.; Asaoka, S.; Okuhara, T.; Misono,
M. J. Am. Chem. Soc. 1992, 114, 2836.
cannot provide full mineralization. The most significant point
may prove to be the reduced dependence on O
electron acceptor.
2
as the initial
(
(
41) Misono, M. Catal. Rev.-Sci. Eng. 1987, 29, 269.
42) Friesen, D. A.; Gibson, D. B.; Langford, C. H. J. Chem. Soc., Chem.
Commun. 1998, 543.
Acknowledgments
The authors wish to thank the Natural Sciences and
Engineering Research Council of Canada for financial sup-
(
43) Rutherford, L. A.; Hennigar, P. A.; Doe, K. G.; Nicol, M. L.; Holmes,
M. M. E.; MacDonald, B. C.; Horne, W. H. Surveill. Rep. EPS
(
Environ. Can.) 1992, EPS-5-AR-93-1, 144 pp.
port, Dr. Alexander Starosud for N
2
BET measurements on
(44) Yasuhara, A. Toxicol. Environ. Chem. 1995, 51, 113.
(45) Wu, H. J. Biol. Chem. 1920, 43, 189.
the tungstate material, Mr. Kerry Peru for the electrospray-
MS/ MS and GC-MSD analyses, and Ms. Vivian Sun for
assistance with the photooxidation experiments.
(
(
(
46) Hatchard, C. G.; Parker, C. A. Proc. R. Soc. A 1956, 235, 518.
47) Kirk, A. D.; Namasivayam, C. Anal. Chem. 1983, 55, 2428.
48) Headley, J. V.; Peru, K.; Dickson, L. To be published in the
Proceedings of the 46th ASMS Conference on Mass Spectrometry
and Allied Topics, May 31-June 4, 1998.
Literature Cited
(
49) Mizuno, N.; Misono, M. Chem. Lett. 1987, 967.
(
(
(
(
(
1) Goslich, R.; Dillert, R.; Bahnemann, D. Wat. Sci. Technol. 1997,
5, 137.
2) Hoffman, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W.
Chem. Rev. 1995, 95, 69.
3) Helz, G. R., Zepp, R. G., Crosby, D. G., Eds. Aquatic and Surface
Photochemistry; Elsevier: Amsterdam, 1994.
4) Ollis, D. F., Al-Ekabi, H., Eds. Photocatalytic Purification and
Treatment of Water and Air; Elsevier: Amsterdam, 1993.
5) Zeltner, W. A.; Hill, C. G., Jr.; Anderson, M. A. Chemtech 1993,
(50) Friesen, D., unpublished observations.
(51) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry, 4th
ed.; Wiley: New York, 1980; p 859.
(52) Varga, G. M., Jr.; Papaconstantinou, E.; Pope, M. T. Inorg. Chem.
1970, 9, 662.
(53) Okita, M.; Wakamatsu, T.; Ban, Y. J. Chem. Soc., Chem. Commun.
1979, 749.
(54) Valladares, J. E.; Bolton, J. R. In Photocatalytic Purification and
Treatment of Water and Air; Ollis, D. F., Al-Ekabi, H., Eds.;
Elsevier: Amsterdam, 1993; pp 111-120.
3
2
3, 21.
VOL. 33, NO. 18, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 1 9 7

(
(
55) Lobedank, J.; Bellmann, E.; Bendig, J. J. Photochem. Photobiol.
A 1997, 108, 89.
56) Serpone, N.; Terzian, R.; Minero, C.; Pelizzetti, E. In Photosensi-
tive Metal-Organic Systems - Mechanistic Principles and Ap-
plications, Kutal, C., Serpone, N., Eds., ACS Symp. Ser. 1993,
(64) Sun, L.; Bolton, J. R. J. Phys. Chem. 1996, 100, 4127.
(65) Yao, C. C. D.; Mill, T. In Aquatic and Surface Photochemistry;
Helz, G. R., Zepp, R. G., Crosby, D. G., Eds.; Elsevier: Amsterdam,
1994; pp 499-515.
(66) Pavlik, J. W.; Tantayanon, S. J. Am. Chem. Soc. 1981, 103, 6755.
(67) Yablokov, V. A. Russ. Chem. Rev. 1980, 49, 833.
2
38, 281.
(
(
(
(
(
57) Turchi, C. S.; Ollis, D. F. J. Catal. 1990, 122, 178.
58) Yamase, T. Inorg. Chim. Acta 1983, 76, L25.
59) Chambers, R. C.; Hill, C. L. Inorg. Chem. 1991, 30, 2776.
60) Masui, M.; Hara, S.; Ozaki, S. Chem. Pharm. Bull. 1986, 34, 975.
61) Patton, D. E.; Drago, R. S. J. Chem. Soc., Perkin Trans. 1 1993,
(
68) March, J. Advanced Organic Chemistry, 4th ed.; Wiley: New
York, 1991; p 1099.
(
69) Nohara, K.; Hidaka, H.; Pelizzetti, E.; Serpone, N. J. Photochem.
Photobiol. A 1997, 102, 265.
1
611.
Received for review September 7, 1998. Revised manuscript
received April 22, 1999. Accepted June 25, 1999.
(
(
62) Drago, R. S.; Riley, R. J. Am. Chem. Soc. 1990, 112, 215.
63) Gramain, J. C.; Remuson, R.; Troin, Y. J. Chem. Soc., Chem.
Commun. 1976, 194.
ES980912P
3
1 9 8
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 18, 1999