Photochemical oxidation of phenanthrene sorbed on silica gel

Article abstract of DOI:10.1021/es950769p

There have been relatively few detailed studies of PAH photochemical degradation mechanisms and products at solid/air interfaces under controlled conditions. Results from mechanistic studies on particulate simulants are important in understanding the fates of PAH sorbed on similar materials in natural settings. In this study, the photolysis of phenanthrene (PH) on silica gel, in the presence of air, has been carefully examined. Once sorbed onto the silica surface, PH is not observed to repartition into the gas phase, even under vacuum, and dark reactions of PH are not observed at the silica/air interface. Photolysis (254 nm) of PH leads to the formation of 2,2'-biformylbiphenyl (1), 9,10-phenanthrenequinone (2), cis-9,10-dihydrodihydroxyphenanthrene (3), benzocoumarin (4), 2,2'-biphenyldicarboxylic acid (5), 2-formyl-2'-biphenylcarboxylic acid (5), 2-formylbiphenyl (7),1,2-naphthalenedicarboxylic acid (8), and phthalic acid (9). These products account for 85-90% of the reacted PH. The photoproducts are independent of excitation wavelength (254 and 350 nm), and the reaction proceeds entirely through an initial step involving the addition of singlet molecular oxygen to the ground state of phenanthrene with subsequent thermal and/or photochemical reactions of the initially formed product. Singlet molecular oxygen is produced through quenching of the lowest triplet state of PH at the silica gel/air interface. The high material balance and detailed mechanistic information provided by this study serve as a standard for comparisons with the products and mechanism of PH photochemical oxidation on environmentally derived inorganic oxide particulates.

Full text of DOI:10.1021/es950769p

Environ. Sci. Technol. 1996, 30, 1776-1780
substances are toxic and many are carcinogenic. The
Photochemical Oxidation of
Phenanthrene Sorbed on Silica
Gel
Environmental Protection Agency (EPA) has included 16
PAH in their list of priority pollutants (1). PAH are found
in coaltar and on fly ash and other atmospheric particulates,
and when washed by rain, they contaminate the soil and
water. Many PAH have been shown to have photolytic
half-lives that are dependent on the nature of the substrate
to which they are sorbed (2). It has been found that
photolytic half-lives for PAH on silica, alumina, and fly ash
are generally much shorter than the half-lives on carbon
black (2); however, information concerning the products
or the photochemical mechanisms on these surfaces are
scarce. It has also been reported that some PAH are
photolytically stable on coal fly ash, although those with
benzylic hydrogens are prone to thermal oxidation (3).
Previous work in our laboratories has focused on deter-
mining the photoproducts and mechanisms of photo-
chemical oxidation of PAH sorbed onto silicas and aluminas
where structural dependence of the photolytic half-life has
been observed (2, 4). For example, the photochemical
oxidation of anthracene at a silica gel/ air interface was
shown to occur entirely by a singlet oxygen-mediated
pathway (4a); however, naphthalene and 1-methylnaph-
thalene were shown to react byan apparent electron transfer
mechanism (4h). Photochemical oxidation of 1-methoxy-
naphthalene was shown to occur by both singlet oxygen-
and electron transfer-mediated pathways, giving two
distinct sets of products (4c). Studies of other PAH have
also shown one or both ofthese mechanisms to be operative
(4).
J O H N T . B A R B A S , †
M I C H A E L E . S I G M A N , * A N D
R E Z A D A B E S T A N I
Chemical and Analytical Sciences Division, Oak Ridge
National Laboratory, P.O. Box 2008, MS 6100,
Oak Ridge, Tennessee 37831-6100
There have been relatively few detailed studies of
PAH photochemical degradation mechanisms and
products at solid/air interfaces under controlled
conditions. Results from mechanistic studies on
particulate simulants are important in understanding
the fates of PAH sorbed on similar materials in
natural settings. In this study, the photolysis of
phenanthrene (PH) on silica gel, in the presence of
air, has been carefully examined. Once sorbed
onto the silica surface, PH is not observed to repartition
into the gas phase, even under vacuum, and dark
reactions of PH are not observed at the silica/air
interface. Photolysis (254 nm) of PH leads to the
formation of 2,2′-biformylbiphenyl (1), 9,10-phenan-
threnequinone (2), cis-9,10-dihydrodihydroxyphenan-
threne (3), benzocoumarin (4), 2,2′-biphenyldicar-
boxylic acid (5), 2-formyl-2′-biphenylcarboxylic acid
(6), 2-formylbiphenyl (7), 1,2-naphthalenedicarboxy-
lic acid (8), and phthalic acid (9). These products
account for 85-90% of the reacted PH. The photo-
products are independent of excitation wavelength
(254 and 350 nm), and the reaction proceeds entirely
through an initial step involving the addition of singlet
molecular oxygen to the ground state of phenanthrene
with subsequent thermal and/or photochemical reac-
tions of the initially formed product. Singlet
The work detailed in this paper is a continuation of our
efforts to understand PAH photochemistry under controlled
conditions, which allow for mechanistic interpretations and
give results that are relevant to environmental processes
occurring on similar materials. Considerable amounts of
water-insoluble solid inorganic particles are found in the
atmosphere (5). Tropospheric concentrations of these
particulates are estimated to be in the range of 1-300 µg
m-3 (5). As much as 20-30% of inorganic atmospheric
particulates have been estimated to be oxides, including
insulators such as SiO2, Al2O3, silicoaluminates, and CaCO3
(5). The highly polar and non-semiconducting surfaces of
commercial silicas and aluminas may simulate, to a first
approximation, inorganic oxide particulates found in the
environment. Evaluating commercial silicas and aluminas
as simulants of environmentally derived inorganic oxide
particulates requires detailed mechanistic studies on both
commerically available and naturally occurring materials.
molecular oxygen is produced through quenching of
the lowest triplet state of PH at the silica gel/air
interface. The high material balance and detailed
mechanistic information provided by this study serve
as a standard for comparisons with the products
and mechanism of PH photochemical oxidation on
environmentally derived inorganic oxide particulates.
An understanding of the PAH structural dependence
observed for photolytic degradation rates and mechanisms
on silica gel and alumina will assist in understanding and
predicting the fates of PAH at similar interfaces. Studies
of the photochemical oxidation of small PAH at solid/ air
interfaces also afford the opportunity to rigorously identify
and quantitate products derived from members of this
important class of pollutant. This information forms an
experimental basis for predicting reactivity paterns in much
larger PAH having similar structural features and physical
properties (e.g., combinations of K-, L-, and bay-regions,
oxidation potentials, reactivity toward singlet molecular
oxygen, etc.) but where independent synthesis of the
Introduction
The photochemical oxidation of polycyclic aromatic hy-
drocarbons (PAH) is a topic of current interest. These
* Author to whom correspondence should be addressed; tele-
phone: (423)576-2173; fax: (423)574-4902; e-mail address:
sigmanme@ornl.gov.
† Present address: Valdosta State University, Valdosta, GA.
9
1 7 7 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 5, 1996
0013-936X/96/0930-1776$12.00/0
1996 Am erican Chem ical Society

oxidation products is much more difficult. For these
reasons, it is important to generate a detailed mechanistic
understanding of photolytic transformations occurring in
small PAH sorbed to substrates of environmental relevance.
The topic of this paper is the products and mechanism of
the photochemicaloxidation ofphenanthrene (PH) on silica
surfaces.
PH is reported to have a low vapor pressure and a sizable
organic carbon-water partition coefficient, log KOC = 4.2
(6). PH is one of the 16 PAH on the EPA list of priority
pollutants (1). In solution, PH has been shown to undergo
dye-sensitized photooxidation, and the potentially carci-
nogenic oxirane 9,10-epoxy-9,10-dihydrophenanthrene has
been suggested as a possible intermediate (7). Sorbed on
different atmospheric particulate substrates and/ or sub-
strate simulants, PH has been determined to have photolytic
half-lives of 150 (silica gel), 45 (alumina), 49 (fly ash), and
greater than 1000 h (carbon black) (2). The differences in
the reported half-lives on various supports are unexplained
for PH, as they are for other PAH, and will remain as such
until detailed mechanistic information is available for a
range of PAH on a series of substrates.
Marine and lacustrine sediments have been said to be
the ultimate environmental sinks for PAH (2). In the
atmosphere, PH and other small PAH have been shown to
reside mostly in the gas phase (8). The half-life for PH in
the gas phase has been determined to be less than 1 day
(9). Failure of PH to repartition from the sorbed state into
the gas phase in our experimnets (vide infra) coupled with
sluggish photolytic degradation observed by ourselves and
others (2) suggest that particulate materials may serve as
a sink for PH in the atmosphere.
(10). Similarly, calibration standards were prepared by
esterification of organic acids by diazomethane following
the same procedure.
GC/ FID analysis of 1 was performed using 2 (Aldrich
Chemical Co., Milwaukee, WI) as a calibration standard.
Quantitative analysis of3 was done with the isomeric trans-
9,10-dihydrodihydroxyphenanthrene, which was prepared
by reduction of 2 with LiAlH4, as previously described (11).
Reduction of 2 with LiAlH4 produced both trans-9,10-
dihydrodihydroxyphenanthrene and 3 in a ratio of ap-
proximately 90:1, as determined by GC and GC/ MS. The
stereochemistry of the major product from the reduction
(the trans isomer) was assigned on the basis of previous
results (11). The formation of both isomers from the
reduction of 2 allows us to assign the stereochemistry of
the photochemically derived diol, 3, as cis. Analysis of 4
was done using methyl o-phenylbenzoate as a calibration
standard. Methyl o-phenylbenzoate was prepared from
o-phenylbenzoic acid (Aldrich ChemicalCo.). Quantitation
of 6 was done using the response factor determined for the
dimethyl ester of 5 (Aldrich Chemical Co.), which has the
same number of carbon atoms. Similarly, the response
determined for 9-fluorenone (Aldrich Chemical Co.) was
used to quantitate 7. The dimethyl esters of 2,3-naphtha-
lenedicarboxylic acid (Aldrich Chemical Co.) and phthalic
acid (Aldrich Chemical Co.) were used to quantitate 8 and
9, respectively.
For the purposes of product identification, 4 was
prepared by the thermal decomposition of tert-butylperoxy
o-phenylbenzoate, as previously described (12). The final
product was analyzed by mass spectrometry; m/z (%) )
196 (100), 168 (48), 139 (52).
EPR measurements were made at room temperature on
an instrument that has been described elsewhere (13).
Diffuse reflectance spectra were recorded on a Cary-4
spectrophotometer equipped with an integrating sphere.
Baseline correction was done with the spectrum recorded
from a sample of SiO2 that had not been loaded with
organics. Emission spectra were recorded on a Spex
Fluorolog fluorimeter and are reported fully corrected.
Singlet molecular oxygen luminescence was recorded on
an instrument that has previously been described (4a,e).
Experimental Section
The SiO2 used in this work has characteristics identical to
those previously reported (4c). The N2 BET surface area
for the SiO2 was determined to be 274 m2 g-1 with an average
pore size of 60 Å (4c). The SiO2 was loaded with PH (2.5
× 10-5 mol g-1, 5.9% monolayer; 1.0 × 10-4 mol g-1, 23.8%
monolayer; and 1.0 × 10-3 mol g-1, 238% monolayer) by
adsorption from cyclohexane following the procedure
outlined in earlier papers (4). Cyclohexane was removed
under aspirator vacuum, and final traces were removed at
a vacuum of 5-15 µm of mercury. Photolysis was done in
quartz tubes irradiating in a Rayonet photoreactor (254
nm, 3.4 × 1016 photons cm-2 s-1 or 350 nm, 1.23 × 1016
photons cm-2 s-1) using a horizontal tube orientation as
previously described (4). Photolysis was done under an
atmosphere of air introduced directly from the laboratory
or under argon (ultra-high purity, Air Products Inc.)
introduced after evacuation (5 µm of mercury for 10 h).
Photoproducts were removed by washing the SiO2 first
with acetone and then with methanol. Identification of
the photoproducts was initially done by GC/ MS, and the
initial assignments were confirmed by spiking product
mixtures with authentic materials where possible. The non-
acid products were quantitated on a GC (Hewlett Packard
5890) fitted with a cross-linked methyl silicone colum (HP-
1, 12 m, 0.2 mm i.d., 0.33 µm film thickness, Hewlett Packard)
and a flame ionization detector (FID). Analysis was done
using an internal standard method and calibration with
known materials. Organic acid products were analyzed by
GC/ FID as their respective methyl esters following reaction
of the product mixture with diazomethane. The diazo-
methane was generated following a standard procedure
Results and Discussion
Spectroscopic Analysis of Phenanthrene Sorbed onto
Silica. The diffuse reflectance spectra of PH on silica gel
at surface loadings of 1 × 10-5 mol g-1 (curve A), 2.5 × 10-5
mol g-1 (curve B), and 1 × 10-4 mol g-1(curve C) are shown
in Figure 1. The spectra show no sign of ground-state pair
formation as previously observed for some silica-sorbed
PAH (4). Ground-state pairs previously observed on silica
exhibited a red-shifted absorption profile in the diffuse
reflectance spectra. Similarly, the fluorescence spectra
recorded at PH loadings of 1 × 10-5 mol g-1 (curve A) and
1 × 10-4 mol g-1(curve B), Figure 2, do not reveal
contributions from excimer or excimer-like emissions,
although the spectrum recorded at the higher loading (curve
B) does have some added intensity on the red-edge of the
spectrum.
Ground-state association of sorbed PAH on silicas has
previously been detected by the presence of excimer-like
emission (4). In the case of sorbed 1-methoxynaphthalene,
restricted diffusion of the PAH on the surface allowed the
observation of excimer-like emission, although an excimer
emission has not been reported for 1-methoxynaphthalene
9
VOL. 30, NO. 5, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 7 7 7

FIGURE 1. Diffuse reflectance spectra of phenanthrene on silica gel
at (A) 1 × 10-5 mol g-1, (B) 2.5 × 10-5 mol g-1, and (C) 1 × 10-4 mol
g-1 surface loading.
FIGURE 3. Excitation (curve A, λem ) 450 nm) and emission (curve
B, λex ) 291 nm) spectra of phenanthrene (1 × 10-3 mol g-1) on
silica.
S CHEME 1
FIGURE 2. Fluorescence of phenanthrene on silica gel at (A) 1 ×
10-5 mol g-1, (B) 1 × 10-4 mol g-1, and (C) 1 × 10-3 mol g-1 loadings.
All samples were excited at 291 nm.
in solution (4c). Excimer emission from PH has previously
been observed centered at 430 nm in an ethanol/ isopen-
tane/ diethyl ether solvent at subambient temperatures;
however, no excimer fluorescence was observed at room
temperature in the mixed solvent system (14). These results
indicate a very weak stabilization of the PH excimers in
solution. The absence ofexcimer-like emission from sorbed
PH on silica suggests the absence of ground-state pairs;
however, the existence ofground-state pairs is not rigorously
excluded on this basis.
The fluorescence spectrum of PH at a surface loading
of 1 × 10-3 mol g-1 (curve C) in Figure 2 is identical to the
fluorescence observed in our laboratory for crystalline PH.
The fluorescence and excitation spectra of PH on silica at
1 × 10-3 mol g-1 loading are shown in Figure 3. The
excitation spectrum clearly shows strong transitions at
greater than 350 nm that should be visible in the diffuse
reflectance spectra in Figure 1 if microcrystalline PH were
present on the surface at lower loadings. Both the diffuse
reflectance and fluorescence spectra indicate that under
the conditions where we have studied the photochemistry
of PH on silica, at loadings less than or equal to 1 × 10-4
mol g-1, microcrystalline PH is not present. Our interest
is in the photochemistry of monomeric PH sorbed onto the
surface; therefore, no studies of the photochemistry were
silica (vide infra). We were unable to detect signals due to
superoxide or PH cation radical formation in photolyzed
samples (by electron paramagnetic resonance spectros-
copy) as were previously observed for other PAH (4g,i).
Photochem istry of Phenanthrene Sorbed onto Silica.
Samples of PH on silica (2.5 × 10-5 mol g-1 and 1.0 × 10-4
mol g-1) photolyzed slowly at 254 nm (t1/ 2 ) 15.1 h at 2.5
× 10-5 mol g-1 loading) to yield 2,2′-biformylbiphenyl (1),
9,10-phenanthrenequinone (2), cis-9,10-dihydrodihy-
droxyphenanthrene (3), benzocoumarin (4), 2,2′-biphen-
yldicarboxylic acid (5), 2-formyl-2′-biphenylcarboxylic acid
(6), 2-formylbiphenyl (7), 1,2-naphthalenedicarboxylic acid
(8), and phthalic acid (9). A direct comparison between
the half-life observed in our work (15.1 h) and the previously
reported value of 150 h on silica (2) cannot be made due
to differences in light source photon flux and other
experimental parameters. The structural formulas for the
products are shown in Scheme 1, and the yields (mole
percent of reacted PH) obtained at the two surface loadings
(2.5 × 10-5 mol g-1 and 1 × 10-4 mol g-1) are given in Table
1. As shown in Table 1, approximately 90% of the reacted
PH is accounted for as photoproducts. The high material
balance reported here reveals tht once PH is sorbed onto
the silica surface it does not repartition to the gas phase,
even under vacuum (vide supra). The product yields from
photolysis at 254 nm in Table 1 reveal that the surface
conducted at a surface coverage of 1 × 10-3 mol g-1
.
Emission from singlet molecular oxygen, generated by
direct excitation of PH on the silica surface, was directly
observed, although the signal was weak. Singlet molecular
oxygen emission was also observed from sorbed 9,10-
phenanthrenequinone, a product from PH photolysis on
9
1 7 7 8 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 5, 1996

S CHEME 2
TABLE 1
Product Yields (in mol %) from PH Photolysis on
Silica
surface loading
product
2.5 × 10-5 mol g-1a
1 × 10-4 mol g-1b
1
2
3
4
5
6
7
8
9
36
30
37
31
10
8.2
0.7
c
c
c
c
7.4
7.6
3.8
1.1
2.1
0.6
1.4
total
90.0
86.9
a
b
Photolysis tim e of 3 h, 15% conversion. Photolysis tim e of 4 h,
c
6% conversion. Trace of m aterial present, less than 0.5 m ol %.
formation of the lowest triplet state (T1) of PH from which
singlet molecular oxygen forms by energy transfer (vide
infra). The ground-state (S0) of PH reacts with the resulting
singlet molecular oxygen to give the thermally unstable
9,10-dioxetane of PH or possibly the corresponding per-
oxirane, which may rearrange to the dioxetane (vide infra).
The 9,10-dioxetane of PH in turn undergoes thermal ring
opening and further oxidation to give the various products
(indicated by ∆ r.t. in Scheme 2). Photooxidation of 2
leads to the formation of 4 and 5 as shown by the control
experiments described above and depicted in Scheme 2.
Photooxidation of 3 has also been shown to produce 1, 2,
4, and 5 on the silica surface. Bimolecular reactions are
not important since there is no appreciable change in
product distribution upon increasing the coverage of PH
on the surface. There is, however, a decrease in the yield
ofhigher oxidation products with increasing coverage (Table
1). This may be due to the formation of these products at
the more reactive sites on the silica surface.
Singlet molecular oxygen addition to olefins has been
addressed theoretically at the semi-empirical level (16). The
most reactive site for the addition of singlet molecular
oxygen to PH is across the 9,10-positions, as observed for
other Diels-Alder reactions with PH and explained in terms
of the π-localization energy (17). Our observation of
products arising primarily from oxidation of the 9,10-
positions of PH under direct and sensitized photolysis are
in agreement with a singlet molecular oxygen-mediated
reaction, as are the previously reported results for sensitized
photolysis of PH in solution (7). The stereochemistry
observed for photochemically derived diol 3 suggests that
it arises from homolytic scission of the peroxide bond in
a dioxetane intermediate with subsequent hydrogen atom
abstraction (see Scheme 2). Further, dialdehydes such as
1 can easily be derived from the same dioxetane interme-
diate. An alternative mode of singlet molecular oxygen
addition across a double bond involves the initial formation
of a peroxirane (16). The peroxirane can possibly rearrange
to the dioxetane, although this rearrangement has a high
calculated barrier in peroxiranes derived from an olefin
that was part of a six-member ring (16a). Alternatively, a
peroxirane could be reduced by a double bond in PH to
produce 2 mol of oxirane, e.g., the potential carcinogen
9,10-epoxy-9,10-dihydrophenanthrene (7, 16b). This may
account for the reported formation of this compound in
solution (7); however, we do not observe any evidence for
its formation on silica.
loading has only minimal effect on the product distribution.
Control experiments showed no oxidation of PH/ silica
samples that were not exposed to light for a period of 4
days. The photolysis of PH on silica in the presence of
cosorbed 2,5-dimethylfuran, a known singlet oxygen scav-
enger (15), resulted in the formation of products derived
only from the furan, and no oxidation products from PH
were observed. When methylene blue, a good singlet
oxygen sensitizer, was cosorbed on silica with PH and
photolyzed at 650 nm (photolysis into the absorption band
for methylene blue), major products 1, 2, and 3 and minor
products 5 and 6 were produced, as in the nonsensitized
photolysis. These results indicate that oxidation of PH by
singlet molecular oxygen is the major reaction pathway in
the photooxidation on silica surfaces.
Photolysis was also done at 350 nm (65 h), and the major
products were substantially unchanged from those pro-
duced by irradiation at 254 nm. Photolysis at 350 nm did
produce trace amounts of unidentified materials, which
probably result from secondary product photochemistry
and/ or thermal instability of the major products. The half-
life for photolysis at 350 nm was not determined; however,
photolysis of PH occurred much slower than photolysis of
anthracene with the same 350-nm light source (4a).
Similarly, photolysis (450-W medium pressure mercury
vapor lamp) of PH on silica has previously been reported
to occur 50 times more slowly than anthracene (2). The
invariance of the product set with changing excitation
wavelength indicates that changes in mechanism do not
occur upon excitation to singlet states SN (N > 1) with 254-
nm light.
When 3 was sorbed on silica, it was observed to undergo
a slow thermal oxidation, proceeding through a transient
green intermediate, to form primarily quinone 2. When 3
was adsorbed on silica and photolyzed (254 nm) under the
experimental conditions used for direct PH photolysis, 2
was observed as the major product along with lesser
amounts of 1, 4, and 5. The photolysis of 2 (254 nm),
adsorbed on silica, under similar conditions gave 4 and 5.
The presence ofcosorbed tetrahydrofuran, a good hydrogen
atom source, was observed to increase the rate of photolysis
of sorbed 2.
These results can be best explained by invoking a
mechanism (Scheme 2) wherein PH absorbs light and is
promoted to an electronic state SN (N g 1), from which it
relaxes to the S1 state. Intersystem crossing (isc) leads to
9
VOL. 30, NO. 5, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 7 7 9

Maestri, B.; Mabey, W. R.; Holt, B. R.; Gould, C. Water-Related
Environmental Fateof129 PriorityPollutants;U.S. Environmental
Protection Agency: Washington, DC, 1979; EPA-440/ 4-79-029.
The interaction of oxygen with electronically excited
states can produce a theoretical optimum quantum yield
for singlet oxygen production (æ∆) of 2, a value that is rarely
observed (18). Achieving the optimum æ∆ requires an S1-
T1 energy gap . 7882 cm-1 (19). The reported value of æ∆
for PH is only 0.5 in methanol (18), and the source of singlet
oxygen, in this case, is limited to oxygen quenching of the
T1 state of PH. This is due to the small S1-T1 splitting in
PH (7300 cm-1) (19), which is less than the 0,0 excitation
energy of O2(1∆g) (7882 cm-1). Singlet oxygen production
is not depicted as accompanying intersystem crossing (isc)
in Scheme 2. It has previously been shown that the rate
of oxygen quenching of pyrene triplets on silica (1.7 × 106
m3 mol-1 s-1) is approximately twice as fast as on alumina
(9.4 × 105 m3 mol-1 s-1) (20). The relative rate of PH(T1)
quenching on silica versus alumina is likely to resemble
the behavior reported for pyrene quenching. This assertion
is supported by the fact that the two molecules are very
similar in size, thereby allowing them to occupy similar
surface sites and experience similar effects of shielding from
oxygen bombardment at the surface. The lifetime of the
triplet state of PH, relative to that of pyrene, is irrelevant
to this argument since our concern is with the rate of oxygen
quenching of the T1 state of PH on silica versus alumina.
On the basis of these arguments, the previously reported
enhancement in photolysis rate on alumina (t1/ 2 ) 45 h) (2)
relative to the rate on silica (t1/ 2 ) 150 h) (2) cannot be
explained by the differential quenching of the triplet state
of PH by oxygen on these two substrates. The enhanced
rate ofphotolysis on alumina is more likely due to additional
mechanistic pathways involving Lewis acid sites on the
alumina (20). The previously observed PH photodegra-
dation half-lives on fly ash (49 h) (2) and on carbon black
(>1000 h) (2) are more difficult to address. The large
variation in fly ash color and composition offer the
possibility of multiple mechanistic pathways involving
oxidation by singlet oxygen and/ or electron transfer. Both
processes can be initiated by energy transfer or direct
photolysis and possibly inhibited by competitive absorption
of the incident light. The long life time observed on carbon
black is likely due to severalprocesses including competitive
absorption of the incident light, energy transfer from PH
to the large constituent PAH in carbon black, and more
facile reactions of singlet oxygen with the substrate.
(2) Behymer, T. D.; Hites, R. A. Environ. Sci. Technol. 1985, 19, 1004.
(3) (a) Korfmacher, W. A.; Natusch, D. F. S.; Taylor, D. R.; Mamantov,
G.; Wehry, E. L. Science 1980, 207, 764. (b) Korfmacher, W. A.;
Wehry, W. L.; Mamantov, G.; Natusch, D. F. S. Environ. Sci.
Technol. 1980, 14, 1094.
(4) (a) Dabestani, R.;Ellis, K. J.;Sigman, M. E.J. Photochem. Photobiol.
A: Chem. 1995, 86, 231. (b) Sigman, M. E.; Chevis, E. A.; Brown,
A.; Barbas, J. T.; Dabestani, R.; Burch, E. L. J. Photochem.
Photobiol. A: Chem., in press. (c) Sigman, M. E.; Barbas, J. T.;
Chevis, E. A.; Dabestani, R. New J. Chem., in press. (d) Dabestani,
R.; Nelson, M.; Sigman, M. J. Photochem. Photobiol. A: Chem.
1995, 92, 201. (e) Barbas, J. T.; Dabestani, R.; Sigman, M. E. J.
Photochem. Photobiol. A: Chem. 1994, 80, 103. (f) Sigman, M.
E.; Zingg, S. P. In Environmental Aspects of Surface and Aquatic
Photochemistry; Helz, D., Zepp, R. G., Eds.; Lewis Publishers:
Boca Raton, FL, 1994, p 197. (g) Zingg, S. P.; Sigman, M. E.
Photochem. Photobiol. 1993, 57, 453. (h) Barbas, J. T.; Sigman,
M. E.; Buchanan, A. C., III; Chevis, E. A. Photochem. Photobiol.
1993, 58, 155. (i) Zingg, S. P.;Pagni, R. M.;Burns, J. H. Tetrahedron
Lett. 1991, 32, 5737.
(5) Guillard, C; Delprat, H.; Hoang-van, C.; Pichat, P. J. Atmos. Chem.
1993, 126, 47, and references cited therein.
(6) Mackay, D.; Shiu, W. Y.; Ma, K. C. Illustrated Handbook of
Physical-Chemical Propertiesand Environmental Fatefor Organic
Chemicals, Vol. II; Lewis Publishers: Chelsea, MI, 1992; pp 132-
140.
(7) Dowty, B. J.; Brightwell, N. E.; Laseter, J. L.; Griffin, G. W. Biochem.
Biophys. Res. Commun. 1974, 57, 452.
(8) Yamasaki, H.; Kuwata, K.; Miyamoto, H. Environ. Sci. Technol.
1982, 16, 189.
(9) Kwo, E. S. C.; Harger, W. P.; Arey, J.; Atkinson, R. Environ. Sci.
Technol. 1994, 28, 521.
(10) DeBoer, T. J.; Backer, H. J. Organic Synthesis, Collective Volume
IV; Rabjohn, N., Ed.; John Wiley: New York, 1963; pp 250-253.
(11) Moriconi, E. J.; Wallenberger, F. T.; O’Connor, W. F. J. Org. Chem.
1959, 24, 86.
(12) Minato, H.; Matsuzakim H. Miwa, K. Bull. Chem. Soc. Jpn. 1968,
41, 249.
(13) (a) Livingston, R.; Zeldes, H. J. Chem. Phys. 1966, 44, 1245-1259.
(b) Livingston, R.; Zeldes, H. J. Chem. Phys. 1967, 47, 4173-
4180.
(14) Azumi, T.; McGlynn, S. P. J. Chem. Phys. 1964, 41, 3131.
(15) Gollnick, K.; Griesbeck, A. Photochem. Photobiol. 1985, 41, 2057.
(16) (a) Dewar, M. J. S.; Thiel, W. J. Am. Chem. Soc. 1975, 97, 3978.
(b) Dewar, M. J. S.; Griffin, A. C.; Thiel, W.; Turchi, I. J. J. Am.
Chem. Soc. 1975, 97, 4439. (c) Inagaki, S; Fukui, K. J. Am. Chem.
Soc. 1975, 97, 7480.
(17) Dewar, M. J. S. J. Am. Chem. Soc. 1952, 74, 3357.
(18) Gorman, A. A.; Rodgers, M. A. Handbook of Organic Photo-
chemistry, Vol. II; Scaiano, J. C., Ed.; CRC Press: Boca Raton, FL,
1989; pp 229-247.
(19) Birks, J. B. Photophysics of Aromatic Molecules; Wiley-Inter-
sience: New York, 1970; pp 142-192.
(20) Thomas, J. K. Chem. Rev. 1993, 93, 301 and references cited
Acknowledgments
Research sponsored by the Division of Chemical Sciences,
Office of Basic Energy Sciences, U.S. Department of Energy
under Contract DE-AC05-84OR21400 with Lockheed-Martin
Inc.
therein.
Received for review October 18, 1995. Revised manuscript
received January 8, 1996. Accepted January 30, 1996.X
ES950769P
Literature Cited
(1) Callahan, M. A.; Slimak, M. W.; Gabelc, N. W.; May, I. P.; Fowler,
C. F.; Freed, J. R.; Jennings, P.; Durfee, R. L.; Whitmore, F. C.;
X Abstract published in Advance ACS Abstracts, April 1, 1996.
9
1 7 8 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 5, 1996