Photocatalytic degradation of 4-chlorophenol. 2. The 4-chlorocatechol pathway

Article abstract of DOI:10.1021/jo990912n

The TiO₂-mediated photocatalytic degradation of 4-chlorocatechol is studied as a branch of the degradation of 4-chlorophenol. In addition to some basic kinetic studies, the identities of many of the cyclic and acyclic intermediates, verified in most cases with authentic samples, are reported. From 4-chlorocatechol, the major product is hydroxylation to form 5-chloro- 1,2,4-benzenetriol. A small amount of 4-chloropyrogallol is also produced. Substitution to give 1,2,4-benzenetriol is observed as is oxidative cleavage of the C1-C2 bond to give the diacid. The major products of all of the triols are those of oxidative cleavages, occurring mainly between ortho hydroxy- substituted carbons to give diacids but also between one hydroxy and one unsubstituted carbon to give acid-aldehydes. Many smaller intermediates in the degradations are identified, and pathways are proposed for the larger compounds.

Full text of DOI:10.1021/jo990912n

J . Org. Chem. 1999, 64, 8525-8536
8525
P h otoca ta lytic Degr a d a tion of 4-Ch lor op h en ol. 2. Th e
4-Ch lor oca tech ol P a th w a y
Xiaojing Li, J erry W. Cubbage, and William S. J enks*
Department of Chemistry, Iowa State University, Ames, Iowa 50011-3111
Received J une 4, 1999
The TiO2-mediated photocatalytic degradation of 4-chlorocatechol is studied as a branch of the
degradation of 4-chlorophenol. In addition to some basic kinetic studies, the identities of many of
the cyclic and acyclic intermediates, verified in most cases with authentic samples, are reported.
From 4-chlorocatechol, the major product is hydroxylation to form 5-chloro-1,2,4-benzenetriol. A
small amount of 4-chloropyrogallol is also produced. Substitution to give 1,2,4-benzenetriol is
observed as is oxidative cleavage of the C1-C2 bond to give the diacid. The major products of all
of the triols are those of oxidative cleavages, occurring mainly between ortho hydroxy-substituted
carbons to give diacids but also between one hydroxy and one unsubstituted carbon to give acid-
aldehydes. Many smaller intermediates in the degradations are identified, and pathways are
proposed for the larger compounds.
In tr od u ction
Many of these publications focus on practical aspects of
photocatalysis as an advanced oxidation process for
cleanup of contaminated waters. Others are concerned
with the effect of experimental parameters on either the
initial disappearance of the 4-chlorophenol or the total
mineralization rate.
A third category of papers is concerned with the
chemical mechanisms of the degradation and the path-
ways by which chlorophenol is degraded to CO2 and Cl-.
The oxidative mineralization chemistry can involve a
number of reactive species, such as TiO2-bound hydroxyl
radicals, TiO2 valence band holes, superoxide, and other
less reactive oxidizers such as molecular oxygen and
hydrogen peroxide.
It is well established that, using TiO2 as the photo-
catalyst, two reactions compete as the first transforma-
tion, giving hydroquinone and 4-chlorocatechol (Scheme
1).2,3,21,37,38 Differences in the observed ratios of these two
reactions have been reviewed by Stafford.29,33 Despite the
plethora of information available about the first chemical
step, little is known about the subsequent chemistry that
leads to ring opening and further degradation of acyclic
compounds. For instance, the first carbon-carbon cleav-
The photocatalytic degradation of 4-chlorophenol by
TiO2 and other semiconductor catalysts has taken on an
importance in the literature that exceeds its original
significance as a model pollutant.1 It has become a
standard for evaluating experimental parameters and
examining their effects on mechanism and efficiency.2-36
(1) Barbeni, M.; Pramauro, E.; Pelizzetti, E.; Borgarello, E.; Graetzel,
M.; Serpone, N. Nouv. J . Chim. 1984, 8, 547-550.
(2) Al-Sayyed, G.; D’Oliveira, J .-C.; Pichat, P. J . Photochem. Pho-
tobiol., A 1991, 58, 99-114.
(3) Al-Ekabi, H.; Serpone, N. J . Phys. Chem. 1988, 92, 5726-5731.
(4) Al-Ekabi, H.; Serpone, N.; Pelizzetti, E.; Minero, C.; Fox, M. A.;
Draper, R. B. Langmuir 1989, 5, 250-255.
(5) Balcioglu, A.; Inel, Y. Turk. J . Chem. 1993, 17, 125-132.
(6) Bauer, R. Chemosphere 1994, 29, 1225-33.
(7) Cunningham, J .; Sedlak, P. J . Photochem. Photobiol., A 1994,
77, 255-63.
(8) Dong, C.; Huang, C.-P. Adv. Chem. Ser. 1995, 244, 291-313.
(9) Gray, K. A.; Stafford, U. Res. Chem. Intermed. 1994, 20, 835-
853.
(10) Guajardo, R.; Hernandez, C.; Leyva, E.; Lopez, A.; Moctezuma,
E. Av. Ing. Quim. 1996, 5, 242-246.
(11) Haarstrick, A.; Kut, O. M.; Heinzle, E. Environ. Sci. Technol.
1996, 30, 817-824.
(12) Hague, i. U.; Rusling, J . F. Chemosphere 1993, 26, 1301-1309.
(13) Kesselman, J . M.; Weres, O.; Lewis, N. S.; Hoffmann, M. R. J .
Phys. Chem. B 1997, 101, 2637-2643.
(14) Klausner, J . F.; Martin, A. R.; Goswami, D. Y.; Schanze, K. S.
J . Sol. Energy Eng. 1994, 116, 19-24.
(27) Serpone, N.; Sauve, G.; Koch, R.; Tahiri, H.; Pichat, P.; Piccinini,
P.; Pelizzetti, E.; Hidaka, H. J . Photochem. Photobiol., A 1996, 94, 191-
203.
(15) Leyva, E.; Moctezuma, E.; Ruiz, M. G.; Torres-Martinez, L.
Catal. Today 1998, 40, 367-376.
(28) Stafford, U.; Gray, K. A.; Kamat, P. V. J . Catal. 1997, 167, 25-
(16) Lindner, M.; Theurich, J .; Bahnemann, D. W. Water Sci.
Technol. 1997, 35, 79-86.
32.
(29) Stafford, U.; Gray, K. A.; Kamat, P. V. J . Phys. Chem. 1994,
98, 6343-6351.
(17) Martin, S. T.; Morrison, C. L.; Hoffmann, M. R. J . Phys. Chem.
1994, 98, 13695-13704.
(30) Stafford, U.; Gray, K. A.; Kamat, P. V. Heterog. Chem. Rev.
1996, 3, 77-104.
(18) Matthews, R. W. Water Res. 1990, 24, 653-660.
(19) Mills, A.; Morris, S. J . Photochem. Photobiol., A 1993, 71, 75-
83.
(31) Stafford, U.; Gray, K. A.; Kamat, P. V. J . Catal. 1997, 167, 25-
32.
(20) Mills, A.; Morris, S. J . Photochem. Photobiol., A 1993, 71, 285-
(32) Stafford, U.; Gray, K. A.; Kamat, P. V. Res. Chem. Intermed.
1997, 23, 355-388.
289.
(21) Mills, A.; Morris, S.; Davies, R. J . Photochem. Photobiol., A
1993, 70, 183-191.
(33) Stafford, U.; Gray, K. A.; Kamat, P. V. Chem. Oxid. 1997, 4,
193-204.
(22) Mills, A.; Sawunyama, P. J . Photochem. Photobiol., A 1994, 84,
305-309.
(34) Stafford, U.; Gray, K. A.; Kamat, P. V.; Varma, A. Chem. Phys.
Lett. 1993, 205, 55-61.
(23) Mills, A.; Davies, R. J . Photochem. Photobiol., A 1995, 85, 173-
(35) Wyness, P.; Klausner, J . F.; Goswami, D. Y.; Schanze, K. S. J .
Sol. Energy Eng. 1994, 116, 2-7.
178.
(24) Mills, A.; Wang, J . J . Photochem. Photobiol., A 1998, 118, 53-
63.
(25) Mylonas, A.; Papaconstantinou, E. J . Mol. Catal. 1994, 92, 261-
267.
(26) Ruppert, G.; Bauer, R.; Heisler, G. Chemosphere 1994, 28,
1447-1454.
(36) Wyness, P.; Klausner, J . F.; Goswami, D. Y.; Schanze, K. S. J .
Sol. Energy Eng. 1994, 116, 8-13.
(37) Vinodgopal, K.; Stafford, U.; Gray, K. A.; Kamat, P. V. J . Phys.
Chem. 1994, 98, 6797-6803.
(38) Theurich, J .; Lindner, M.; Bahnemann, D. W. Langmuir 1996,
12, 6368-6376.
10.1021/jo990912n CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/16/1999

8526 J . Org. Chem., Vol. 64, No. 23, 1999
Li et al.
Sch em e 1
age step is of obvious fundamental interest, reflecting the
nature of the oxidative step. We have examined the
degradation of 4-chlorophenol by TiO2 using HPLC and
GC-MS analysis of the intermediates in combination
with organic synthesis of the proposed intermediates to
take a more detailed look at this chemistry. In the first
paper of this series,39 we reported over 20 intermediates
observed in the degradation of hydroquinone, represent-
ing that branch of the degradation of chlorophenol. Here
we report a similar study of the degradation of 4-chlo-
rocatechol, the other major product of the first step of
4-chlorophenol photocatalytic degradation. Together, these
give a complete set of data for the degradation intermedi-
ates of 4-chlorophenol under TiO2-mediated photocata-
lytic degradation and provide insight particularly into the
ring-opening reaction for simple aromatic compounds. We
also characterize the degradation of 4-chlorocatechol as
a function of pH and catalyst load.
ments, intermediates formed in dark control reactions
were insignificant.
The variation of the concentrations of major intermedi-
ates as a function of time is illustrated in Figure 1. The
solution was completely mineralized in about 800 min.
Among the cyclic intermediates, the halogen-containing
compounds predominate over the others under our
standard conditions. The ratio of maximum concentration
of intermediates 3 and 2 is 2.8. This compares favorably
to the results of Mills21 and Stafford.29 The values of
[intermediates]max/[1]0 are 0.036, 0.017, 0.072, and 0.10
for 2, 5, 7, and 3, respectively, after 80 min irradiation
(near the maximum concentration for these compounds),
also similar to Mills’ results, save that they do not report
6 or 7. The rate constants of photodegradation of 1-3,
independently measured and given in Table 1, are
similar. Thus, the reason more 3 is observed in degrada-
tions of 1 is its slight predominance in formation and not
selective degradation of 2.
A series of degradations was carried out using the
standard conditions but adding 1% (by volume) 2-pro-
panol. The rate of photocatalytic decay of all compounds
was substantially lower in these experiments, but that
of hydroquinone dropped to the extent that it was no
longer competitive with other dark processes (Table 1).
Figure 2 illustrates the early development of intermedi-
ates in the degradation of 1 with 2-propanol present. The
intermediates that have the 1,2-oxygen functionality are
selectively destroyed over hydroquinone, as would be
predicted from Table 1.
Resu lts
P h otoca ta lytic Degr a d a tion of 4-Ch lor op h en ol. A
set of standard conditions was established for degrada-
tion of 4-chlorophenol 1 and other compounds. Solutions
of 100 mL buffered water (5 mM phosphate, pH 7.0
unless otherwise specified) containing 2 mM of substrate
were prepared, and 50 mg Degussa P-25 TiO2 was added.
The solutions were treated in an ultrasound bath and
then saturated with O2 before photolysis. The concentra-
tions of starting materials and various intermediates
were determined by HPLC analysis with a UV-vis diode
array detector. The concentration of miscellaneous deg-
radation products were determined or approximated as
described in the Experimental Section.
P h otoca ta lytic Degr a d a tion of 4-Ch lor oca tech ol.
Degradations were carried out using 3 as starting mate-
rial in standard conditions. Control experiments neglect-
ing one of the three components (O2, TiO2, and light) were
performed. In the absence of oxygen, no change in the
concentration of 3 was observed over 21 h despite
exposure to light in the presence of TiO2. Direct photoly-
sis can thus be neglected. Dark experiments that included
O2 and TiO2 showed a slow autoxidation at pH 7.0. Less
than 7% of 3 was degraded after 10 h. A rough estimate
of 1 × 10-4 min-1 could be made for the pseudo-first-order
rate constant. With light and oxygen but no TiO2, 20%
of 3 was degraded after 10 h. However, none of the cyclic
intermediates discussed below were among the products.
Such oxygenative cleavage of chlorocatechols has been
reported elsewhere41 but did not appear to be significant,
For degradation of 1, intermediates were detected by
HPLC with retention times of 1.2, 1.7, 2.7, 3.6, 6.6, 12.1,
and 15.0 min. As was found previously,39 the 1.2 min
peak consisted of a number of ring-opened intermediates.
The 1.7 min peak was a mixture of compounds. On the
basis of their new appearance compared to degradations
of 4, they were tentatively assigned as chlorine-contain-
ing structures. The peaks at 3.6, 6.6, 12.1, and 15.0 min
were assigned to 2, 5, 7, 3, and 1. 1,2,4-Benzentrol is very
easily oxidized to 5,40 and it was established that this
also occurs during the HPLC analysis procedure. Com-
pound 6 is also easily oxidized to 7. 4-Chlororesorcinol,
2,5,4′-trihydroxybiphenyl, phenol, and 4′-hydroxyphenyl-
benzoquinone, which have been reported by others,21,38
were not observed. On the time scale of these experi-
(39) Li, X.; Cubbage, J . W.; Tetzlaff, T. A.; J enks, W. S. J . Org. Chem.
1999, 64, 8509-8524.
(40) Kurien, K. C.; Robins, P. A. J . Chem. Soc. B 1970, 855-859.
(41) Takuzo, F.; Tamio, Y.; Fukui, A.; Tanaka, T.; Yoshida, S. Angew.
Chem., Int. Ed. Engl. 1998, 37, 513-515.

Photocatalytic Degradation of 4-Chlorophenol. 2.
J . Org. Chem., Vol. 64, No. 23, 1999 8527
F igu r e 2. Kinetics of the photocatalytic decay of an oxygen-
saturated aqueous solution of 4-chlorophenol with 1% 2-pro-
panol added. Shown is formation of 4-chlorocatechol (circles),
5-chloro-2-hydroxybenzoquinone (triangles), hydroxybenzo-
quinone (inverted triangles), and hydroquinone (squares).
F igu r e 1. Photocatalytic decay of an oxygen-saturated aque-
ous solution of 4-chlorophenol (filled circles) and formation of
intermediates 4-chlorocatechol (circles), 5-chloro-2-hydroxy-
benzoquinone (triangles), hydroxybenzoquinone (inverted tri-
angles), hydroquinone (squares), unidentified chlorinated mix-
ture (diamonds), and acyclic acids (×). The inset is a blow up
of the early time period.
Ta ble 1. P seu d o-F ir st-Or d er Ra te Con sta n ts for Deca y
of 1, 2, 3, a n d 7 u n d er Sta n d a r d Con d ition s
rate constant (10-3 min-1
)
photocatalytic
photocatalytic
decay with
1% iPrOH
compd
pH
decay
dark decay
1
2
3
7
3
3
3
7.0
7.0
7.0
7.0
2.0
4.0
8.5
6.9
6.8
6.6
1.0
0.48
2.1
0.11
0.49
0.11
1.5
13
3.3
3.7a
4.2a
15a
a
Rate constant corrected by subtraction of the dark decay rate
constant from the total decay rate constant with the light on.
because with the full standard conditions, the decay was
much more rapid. 4-Chlorocatechol was completely de-
composed in 450 min.
F igu r e 3. Photocatalytic decay of an oxygen-saturated aque-
ous solution of 4-chlorocatechol (circles) and formation of
intermediates 5-chloro-2-hydroxybenzoquinone (triangles), hy-
droxybenzoquinone (inverted triangles), unidentified interme-
diate (diamonds), and acids (×). The inset shows only 5-chloro-
2-hydroxybenzoquinone and hydroxybenzoquinone.
Using HPLC detection, degradation of 3 gave rise to
five peaks with retention times of 1.2, 1.3, 1.6, 3.6, and
6.7 min. As usual, the first peak was a combination of
acyclic acids. The peaks at 3.6 and 6.7 min were identified
as 5 and 7, representing the total concentrations of (4
and 5) and (6 and 7), respectively. The other two peaks
represented mixtures of compounds and were not further
identified by HPLC (Figure 3).
The effect of pH on the degradation of 3 was investi-
gated. The pseudo-first-order rate constants, corrected
by subtraction of the appropriate dark decay rate con-
stants, are given in Table 1. Whereas there is some
decrease in the rate as the solution becomes acidic, a
dramatic increase is noted when the pH is raised to 8.5.
The trials could not be extended above this pH because
of base-catalyzed dark decomposition of 3 in the presence
of oxygen. At pH 10, 3 completely decomposed in 10 min,
even in the absence of light. The number and identity of
detectable intermediates was relatively constant through-
out the pH range discussed here, but their concentrations
were highest at neutral pH, so this was chosen for the
experiments below in which the identities of the inter-
mediates were determined.
As expected for photocatalytic reactions, degradations
that were done without buffer resulted in acidification
of the solution. An unbuffered solution of 3 whose initial
pH was adjusted to 7.0 with NaOH was at pH 4 with
only 5% of the compound degraded (Figure 4). This is, of
course, a result of the dramatic effect on pH of a small
amount of acid production in the higher pH ranges. The
final pH was 2.76, matching well with the final theoreti-
cal value of 2.70, neglecting buffering by TiO2 or carbonic
acid.

8528 J . Org. Chem., Vol. 64, No. 23, 1999
Li et al.
Langmuir-Hinshelwood type relationship given in eqs
1 and 2, where k is a rate constant and K is an
equilibrium adsorption constant. The values obtained for
k and K are 1.5 µM min-1 and 4.0 × 104 M-1, respectively.
The dotted line in the main figure represents the rates
calculated using the parameters obtained from the
linearization.
kKC0
r0 )
(1)
(2)
1 + KC0
1
r0
1
1
)
+
k
kKC0
The early stages of degradation of 3 under standard
conditions were dominated by formation of 5 and 7.
However, 5 degrades (k ) 0.031 min-1) more rapidly than
3 (k ) 0.013 min-1), so an exact branching ratio was not
determined. These are the facile dark oxidation products
of 4 and 6, and it is quite likely that these latter two are
actually the primary products. Solutions of 4 or 6,
analyzed by HPLC, invariably led to detection of 5 or 7,
respectively, even when solutions were prepared with
argon-flushed water. The maximum concentration of 7
(ca. 0.25 mM) was 3-fold greater than that of 5.
F igu r e 4. Unbuffered degradation of 4-chlorocatechol 3.
Relative concentration of 3 (circles) and pH of the mixture
(crossed boxes). The pH of a dark (control) solution dropped
linearly from 7.0 to 5.0 over the same time period.
For further determination of the identities of degrada-
tion intermediates of 4-chlorocatechol, GC-MS detection
was used. For these experiments, the standard initial
concentration was 200 µM. Under acidic conditions, the
acyclic compounds disappeared at about the same rate
as 3, but at neutral pH their decay was slowed and the
maximum concentrations were generally higher. Thus
the neutral pH was used. To avoid the use of buffer, the
pH was maintained at 7.0 ( 0.5 by manual addition of
NaOH solution during the photolysis. After a fixed period
of time, optimized for having the greatest concentration
and number of compounds present, the degradations
were stopped. Attempts at direct analysis of dehydrated
samples were unsuccessful. As a result, the compounds
were silylated before GC analysis. Both alcohols and acids
are functionalized. Repeat runs were made in which the
reaction mixture was reduced with NaBH4 or NaBD4
prior to silylation. The resulting peaks, their ion trap
mass spectra, and confirmed or proposed structures are
given in Table 2. All of the compounds except those
proposed for peaks 21, 26, 27, 29, 35, and 36 were
confirmed by comparison to authentic samples. Struc-
tures for these latter peaks are based on the mass spectra
and certain chemical arguments laid out below. This is
not a complete list of intermediates in the degradation;
a few very small peaks in the GC traces did not yield
sufficient mass spectral data, and a few compounds may
not have survived the analytical regimen. Additionally,
a single large peak (37) was observed that was not
identified. However, it appeared to be a compound with
more than six carbons, presumably an adduct of 3 and a
smaller intermediate. Further details are given in the
Experimental Section.
F igu r e 5. Initial rate of degradation of 3 versus initial
concentration. Experiments were performed in a thermostati-
cally controlled water-jacketed Pyrex cell at 30 ( 0.5 °C. Aside
from temperature and concentration, standard conditions were
used.
It has been reported that certain anions inhibit pho-
tocatalytic degradations.42 Phosphate reduced the rate
of oxidation of salicylic acid and aniline, for instance.43
A similar effect was not observed here for 1, 3, or 7,
though the unbuffered data were necessarily less quan-
titative because of the need to manually regulate the pH
by addition of hydroxide.
The effect of concentration of 3 on its degradation rate
was also investigated over the range of 150-4000 µM
using otherwise standard conditions and a fixed temper-
ature of 30 °C. The results are shown in Figure 5. The
inset shows a linearization of the rates following a
P h ot oca t a lyt ic Degr a d a t ion of 5-Ch lor o-2-h y-
d r oxyben zoqu in on e 7. In the degradation of 3, 7 is one
of the major early intermediates. Therefore its degrada-
tion was pursued to observe further downstream prod-
ucts. Samples of 7 are straightforwardly obtained by
bubbling solutions of 6 with O2.
(42) Bahnemann, D.; Cunningham, J .; Fox, M. A.; Pelizzetti, E.;
Pichat, P.; Serpone, N. Photocatalytic treatment of waters. In Aquatic
and Surface Photochemistry; Helz, G. R., Zepp, R. G., Crosby, D. G.,
Eds., 1994; pp 261-316.
(43) Abdulla, M.; Low, G. K.-C.; Matthews, R. W. J . Phys. Chem.
1990, 94, 6820-6825.

Photocatalytic Degradation of 4-Chlorophenol. 2.
J . Org. Chem., Vol. 64, No. 23, 1999 8529
Ta ble 2. In ter m ed ia tes Detected in Degr a d a tion s of 3, 7, a n d 10a

8530 J . Org. Chem., Vol. 64, No. 23, 1999
Li et al.
Ta ble 2. (Con tin u ed )

Photocatalytic Degradation of 4-Chlorophenol. 2.
J . Org. Chem., Vol. 64, No. 23, 1999 8531
Ta ble 2. (Con tin u ed )
As observed for benzoquinone44 and hydroxybenzo-
quinone,39 7 is reduced to 6 under photocatalytic treat-
ment. In general, HPLC and GC analyses disagreed on
the relative amounts of these two compounds in solution,
with HPLC showing more of the quinone than did GC.
It can be reasonably assumed that there is a steady-state
concentration of 6 in degradations of 7, and in fact, we
will suggest below decomposition mechanisms that go via
intermediate increases because of the lack of competition
in its formation and lack of degradation.
P h ot oca t a lyt ic Degr a d a t ion of 4-Ch lor op yr o-
ga llol 8. A small quantity of 8 is observed in the
photocatalytic degradation of 3. Degradation of 8 in the
dark is complete within 2 h, and at least 10 intermediates
can be detected by HPLC. The dark reaction is relatively
fast and is clearly competitive with photocatalytic deg-
radation. GC-MS traces are nearly the same from
ordinary or dark degradations, and the resulting com-
pounds are mostly different than those from degradation
of 3. This confirms that formation of 8 is a relatively
minor process for 3.
6 in that the major oxidizing species, HO•ads and h+
,
VB
are electrophilic reagents.
In dark control experiments, a single HPLC peak that
remains unidentified was produced. Under standard
conditions, a small amount of this same compound was
observed by HPLC, but it was quickly degraded photo-
catalytically. The photocatalytic reaction was substan-
tially faster than the dark reaction. No known cyclic
compounds were observed by HPLC-only the usual mix
of acyclic acids. They are noted in Table 2. With 2-pro-
panol added, the rate of degradation of 7 is reduced by
about a factor of 7 (Table 1), and the amount of the dark
Discu ssion
Kin etics of Degr a d a tion . The Langmuir-Hinshel-
wood (LH) behavior of the decay kinetics of 3 merits brief
discussion. It is widely recognized that the rate constant
k obtained from these plots is entirely dependent on the
experimental setup. In the early applications of the LH
equation to this sort of data, the equilibrium constant K
(44) Richard, C. New J . Chem. 1994, 18, 443-445.

8532 J . Org. Chem., Vol. 64, No. 23, 1999
Li et al.
was considered strictly related to the binding constant
between the organic substrate and the TiO2. It has since
become clear that this connection is tenuous at best; in
the context of arguments over whether binding is neces-
sary for reaction, it has been shown that several reason-
able kinetic schemes, some of which do not include
substrate binding to the oxide, lead to the same kinetic
form.45 Indeed, the measured K values for 4-chlorophenol
routinely exceed the dark binding coefficients.19,21,24
the 2-propanol cannot at present be perfectly delimited.
Previously, the technique has been interpreted in terms
of discriminating between direct (hole) oxidation and
reaction with HO•.54 The high concentration used (ca. 130
mM) is expected to be sufficient to scavenge virtually all
diffusing hydroxyl radicals54,55 but may also compete for
surface reactions, despite the low binding constant of
2-propanol. We have found in unpublished work that, at
this concentration level, it has a measurable but not
overwhelming effect on adsorption by compounds such
as these. In the current results, all of the degradations
are significantly inhibited by the addition of the alcohol.
The level of inhibition observed for 3 is in line with
previous workers’ observations using ZnO as the photo-
catalyst.55
Although further investigations into the use of alcohols
in photocatalytic conditions and the mechanisms of their
action are clearly justified if they are to become a truly
diagnostic technique, the most intriguing result reported
here is that some degradations are inhibited more than
others. We offer speculation as to the origin of this effect.
The photocatalytic oxidation of hydroquinone is nearly
completely inhibited, though that of 4-chlorophenol,
another poorly binding substrate, is not. However, it need
not be that hydroquinone is actually unreactive under
these circumstances, only that it is not hydroxylated to
benzenetriol. If hydroquinone is oxidized to the semi-
quinone or benzoquinone level instead, it can quickly be
reduced back by acting as an electron sink for conduction
band electrons.39,44 Thus this result is consistent with the
trapping of hydroxyl radicals (free or surface-bound) by
2-propanol, regardless of whether hydroquinone under-
goes surface-bound one-electron oxidations. 4-Chlorophe-
nol has no such “reversible” reaction, perhaps explaining
its continued reactivity in the presence of 2-propanol
despite poor binding. Stafford has pointed out33 that
photocatalytic degradations of 1 that have generated
larger amounts of hydroquinone than chlorocatechol have
been carried out under conditions where surface chem-
istry was probably favored. The current results suggest
an exacerbating effect; chlorocatechol is probably selec-
tively degraded under such conditions. (Alcohol additives
have also been used to generate reducing conditions for
nitroaromatics and certain halogenated compounds; that
chemistry does not appear as relevant to the current case.
See, for instance, references 56-58.)
The observed values of K are, of course, dependent on
conditions, but the value established here (40 mM-1
)
exceeds any of the reported values for 4-chlorophenol of
which we are aware. It is also known that 4-chlorocat-
echol binds to TiO2 much more strongly than does either
hydroquinone or 4-chlorophenol.28 (It is reasonable to
hypothesize from the adsorption data that this is a
general property of structures with ortho hydroxy
groups.46-48) In a careful kinetic study using oxygen-free
conditions and a TiO2 photoelectrode, Kesselman showed
that the rate of degradation of 3 was proportional to the
amount adsorbed on the TiO2.49 From IR experiments,
they determined a binding constant of 97 mM-1 at pH 5,
a value of the same order of magnitude as K measured
here at pH 7. Nonetheless, the data here do not directly
address the significance of the measured K.
The effect of pH on photocatalysis rate and/or efficiency
is depends on the substrate,30,38,42,49-53 but a correlation
can be made in this case between the rates and substrate
adsorption. The pH of zero charge for TiO2 is widely
quoted to be about 6.4.42,46 Above this, there is a net
negative charge, and below there is a net positive charge
on the titanium dioxide particle, which can affect adsorp-
tion. The pKa values of 4-chlorocatechol are 8.8 and 12.7.
At 0.5 mM concentration of 3, Martin and co-workers
observed a maximum in total adsorption to TiO2 over the
pH range of approximately 7.5-10.46 Below about 50 µM,
the binding is bidentate-binuclear, i.e., the O atoms are
independently adsorbed to Ti atoms, whereas at the
concentrations used here, nonspecific adsorption is also
expected. The observed dependence of the initial rate of
degradation of 3 qualitatively follows the pattern of the
adsorption measured by Martin46 and thus may reflect
better hole-trapping efficiency. The initial rates of deg-
radation on the Martin photoelectrode with applied bias
voltage also showed an increase with pH, but the effect
was more modest.49 The conditions of their experiment
(immobilized photoelectrode, 10 mM KCl, N2 purge, 50
µA constant current) are sufficiently different that pars-
ing the quantitative differences in behavior is probably
not warranted.
Degr a d a tion P a th w a ys. As shown in Scheme 1,
4-chlorophenol is converted to a mixture of hydroquinone
and 4-chlorocatechol. The degradation of hydroquinone
was the subject of Part 1 of this work.39 Its first step is
the almost quantitative formation of 1,2,4-benzenetriol
(4), and much of that paper was based on analysis of the
degradation of that compound. In the same way, the
heart of the intermediate study here is based on the
degradation of 4-chlorocatechol 3. Here there is a greater
choice of first steps. One of them is to form 4, whose
degradation is already outlined. The HPLC traces rep-
resented in Figure 3 give the sum of [4] and [5] as [5]obs
because of dark oxidation; the analogous coupling is
The issue of surface reactions is ambiguously ad-
dressed by the experiments using 2-propanol as an
additive. The ambiguity comes about because the role of
(45) Turchi, C. S.; Ollis, D. F. J . Catal. 1990, 122, 178-192.
(46) Martin, S. T.; Kesselman, J . M.; Park, D. S.; Lewis, N. S.;
Hoffmann, M. R. Environ. Sci. Technol. 1996, 30.
(47) Terzian, R.; Serpone, N.; Minero, C.; Pelizzetti, E.; Hidaka, H.
J . Photochem. Photobiol., A 1990, 55, 243-249.
(48) Vasudevan, D.; Stone, A. T. Environ. Sci. Technol. 1996, 30,
1604-1613.
(49) Kesselman, J . M.; Lewis, N. S.; Hoffmann, M. R. Environ. Sci.
Technol. 1997, 31, 2298-2302.
(50) Djebbar, K.; Sehili, T. Pestic. Sci. 1998, 54, 269-276.
(51) Wei, T.-Y.; Wan, C.-C. Ind. Eng. Chem. Res. 1991, 30, 1293-
1300.
(54) Richard, C.; Boule, P. New J . Chem. 1994, 18, 547-552.
(55) Sehili, T.; Boule, P.; Lemaire, J . J . Photochem. Photobiol., A
1989, 50, 117-127.
(56) Ferry, J . L.; Glaze, W. H. Langmuir 1998, 14, 3551-3555.
(57) Calza, P.; Minero, C.; Pelizzetti, E. J . Chem. Soc., Faraday
Trans. 1997, 93, 3765-3771.
(52) Okamoto, K.; Yamamoto, Y.; Tanaka, H.; Tanaka, M.; Itaya,
A. Bull. Chem. Soc. J pn. 1985, 58, 2015-2022.
(53) Tang, W. Z.; Huang, C. P. Water Res. 1995, 29, 745-756.
(58) Ohtani, B.; Kakimoto, M.; Miyadzu, H.; Nishimoto, S.; Kagiya,
T. J . Phys. Chem. 1988, 92, 5773-5777.

Photocatalytic Degradation of 4-Chlorophenol. 2.
J . Org. Chem., Vol. 64, No. 23, 1999 8533
Sch em e 2
Sch em e 3
bound material is diprotonated, on the basis of the pH
of the solution, but may deprotonate after loss of an
electron.
The stereochemistry of 10 is different than one would
expect from Scheme 2. However, it is known that (2E,4Z)-
3-chloromuconic acid isomerizes to 10 extremely easily
in acidic water.59,60 The workup used here briefly exposed
the compounds to acidic conditions; the isomerization
may occur either during the photolysis or as a result of
the acidic workup.
observed for 6 and 7. Because of the known efficient
photocatalytic reduction of benzoquinone to hydroquinone
and the related examples observed here, we assume for
purposes of this discussion that most of the chemistry of
these two redox couples occurs from 4 and 6. However,
the quinone forms cannot be absolutely ruled out.
Four products are observed that derive directly from
3. They are two hydroxylation products 6 and 8, the
substitution product 4, and the cleavage product 10.
Although 8 degrades quickly either in the dark or with
light and so it is harder to quantify the amount that is
made, the products from its degradation are inconsistent
with degradation of 3. As a result, its small peak size is
taken at face value, and this is assigned as a minor
pathway. The other hydroxylation product 6 is a very
large peak in the GC traces; the corresponding quinone
is the largest in the HPLC traces and is assigned to be
the major product. It is presumed that all of these but
10 are the products of reaction of 3 with HO• or its
surface-bound equivalent. By comparison to 4 and 6, the
first step of degradation of 3 leads to much more
hydroxylation (i.e., smaller fraction of cleavage). This is
understandable on the basis of their respective oxidation
potentials, suggesting that the selectivity for hydroxyla-
tion versus cleavage is driven by more than just having
good binding affinity.
Degradation of 10 was also carried out, as shown in
Table 2. Peak 36 represents dihydroxylation of the C4-
C5 π-bond. The mechanism of this is still unclear, but it
was previously observed for related compounds, including
maleic acid.39 No other six-carbon compounds are ob-
served. Peaks 26 and 29 are isomeric, but of the two, only
peak 26 is obtained from 10. Peak 21 represents the only
other five-carbon piece. Although the mass spectra do not
lend themselves to easy assignments of the position of
the chlorines in these compounds, there seems no reason
to believe anything but that the C3 attachment in 10 is
maintained, making the assignments 3-chloro-4,5-dioxo-
pent-2-enoic acid (13) for peak 21 and both isomers of
3-chloro-2-oxopent-4-enedioic acid (14) for peak 26.
The pathways to get to these compounds are not
further revealed by the current data. A reasonable first
few steps are illustrated in Scheme 3, though alternatives
clearly exist. Oxidation of 15 by hydroxyl ultimately gives
17. The regiochemistry of the first step is unimportant,
because 17 can tautomerize in aqueous solution. It, in
turn, is subject to decarboxylation and further oxidation
to give 13 and 14.
At the four-carbon level, chlorofumaric acid, ketosuc-
cinic acid, and tartaric acid are the most important peaks.
Photo-Kolbe chemistry61-64 on 14 may lead to chlorofu-
maric acid, which may in turn be the source of most of
the rest of the peaks. This chemistry is too far removed
from the starting materials here to be convincingly
argued.
Degradation of 7 was also examined. For purposes of
this discussion, it is again assumed that reduction to 6
is rapid with the lamps on and the chemistry proceeds
from there. A small amount of substitution to form 9 is
observed, but the largest peak derives from cleavage
The pathway of degradation from 1,2,4-benzenetriol
has been discussed.39 The principal products of its
degradation are ring-opened compounds, shown as peaks
20 and 32.
The degradation pathways of 6 and 10 will now be
discussed. The opening of 6 to form 10 is consistent with
one-electron oxidation followed by trapping with super-
oxide (Scheme 2), as we have argued in the accompanying
paper. Intermediate 11 could potentially close to give two
dioxetanes, but 12 is not observed. However, this is
consistent with the position of the hydroxy group that
tends to localize charge and/or spin adjacent to it, thus
favoring closure on that side. An endoperoxide formed
by 1,4-closure of 11 is also a possibility. After rupture,
this would lead to a two-carbon fragment and maleic acid
or ketosuccinic acid, which are observed as peaks 9 and
18, respectively. However, other reasonable pathways
also exist to get to these compounds.
(59) Pieken, W. A.; Kozarich, J . W. J . Org. Chem. 1989, 54, 510-
512.
(60) Pieken, W. A.; W., K. J . J . Org. Chem. 1990, 55, 3029-3035.
(61) Sun, Y.; Pignatello, J . J . Environ. Sci. Technol. 1995, 29, 2065-
2072.
Protonation is another potential variable. For instance,
the 4-chlorocatechol that is specifically bonded to the TiO2
in the manner of Martin’s work46 will be deprotonated
throughout the reaction and will produce the bound
dianion of the diacid, whereas most of the nonspecifically
(62) Tahiri, H.; Ait Ichou, Y.; Herrmann, J .-M. J . Photochem.
Photobiol., A 1998, 114, 219-226.
(63) Sakata, T.; Kawai, T.; Hashimoto, K. J . Phys. Chem. 1984, 88,
2344-2350.
(64) Dixon, W. T.; Norman, R. O. C.; Buley, A. L. J . Chem. Soc. 1964,
3625-3634.

8534 J . Org. Chem., Vol. 64, No. 23, 1999
Li et al.
Sch em e 4
natively, it is possible that 20 derives from 9. We view
this as less likely because the trioxygenated benzenes
readily cleave, and we suggest that the peak for 9 (peak
34) is small because little of it is produced. No new
chemical ideas need to be invoked to suggest routes for
the smaller five-carbon peaks.
Su m m a r y a n d Con clu sion s
The TiO2-mediated photocatalytic degradation of 4-chlo-
rophenol has been studied from the perspective of iden-
tifying the chemical pathways involved in its mineral-
ization. The well-established first step produces both
hydroquinone and 4-chlorocatechol. The previous paper
in this series and this one examine the degradation
pathways of these two compounds, respectively. In
Scheme 5 is shown a summary of the most reliable
connections made from 4-chlorophenol to the five-carbon
stage. Simple dihydroxylations have been omitted for
clarity. Undoubtedly other connections exist (some minor
five-carbon compounds are not even included), but we
believe the pathways as written to be reliable.
Over the course of this work, all of the observed six-
carbon acyclic compounds were consistent with a single
general mechanism for ring cleavage in which the first
microscopic step is single-electron oxidation of the sub-
strate. We have proposed that the radical cation is
trapped by superoxide and that a resulting dioxetane is
responsible for C-C cleavage. Although hypothetically
possible for any of the cyclic substrates, the ring-cleavage
reaction is only truly competitive with hydroxylation
reactions when ortho hydroxy groups are present on the
ring, particularly so when the ring is thrice oxygenated.
Two major rationalizations for this exist, both of which
probably contribute: (a) the ortho dihydroxy arrangement
is superior for adsorption to the TiO2, making single-
between the ortho hydroxy groups to give peak 28 ((E,E)-
4-chloro-3-hydroxymuconic acid (18) and tautomers),
confirmed by authentic sample.
An alternate cleavage is observed as small peak 35,
which is a six-carbon compound only observed on deg-
radation of 7. In the absence of authentic samples, its
assignment must be indirect. The NaBD4 results indicate
that two reducible groups are present, and the formula
gives five oxygens and the chlorine. Analysis of the
reasonable cleavages of 6 by means of the mechanism in
Scheme 2 allows assignment of the peak as 19 (Scheme
4). Cleavage of bond a gives the compound assigned in
the table and after reduction gives the correct observed
product. Cleavage of bonds b or d gives compounds with
the same formula but two too many hydrogens. Cleavage
of c leads to the major observed cleavage product, peak
28.
Numerous five-carbon compounds are observed on
degradation of 7. The two main ones are 2,4-dioxopen-
tanedioic acid (20, peak 31) and 3-chloro-2-oxopent-3-
enedioic acid (14, peak 26). Compound 20 is likely to
derive from straightforward degradation of 18. Attribut-
ing 20 is somewhat more difficult in that it has under-
gone dechlorination. Photocatalytic dechlorination is well
known (see, for instance, references 65 and 66), but
another reasonable suggestion can be made. Compound
19 (as drawn and most tautomers) is a vinylogous acid
chloride. As a result, hydrolysis should be relatively facile
to produce the enol. At that point, straightforward
oxidation chemistry can be written to get to 20. Alter-
electron transfer to h+ more likely, and (b) the more
VB
highly oxidized structures have inherently lower oxida-
tion potentials, making electron transfer more favorable.
The reported results are all consistent with much of the
degradation chemistry occurring at the surface of the
TiO2 particle.
On the basis of previous workers’ studies in photoca-
talysis, radiolysis, and homogeneous photolysis of aque-
ous hydrogen peroxide solutions, speculative degradation
pathways from the acyclic six-carbon compounds to most
of the five- and four-carbon compounds have been pro-
posed. The number of two-, three-, and four-carbon
compounds and the multiple routes that might produce
them prevent complete descriptions of the degradations
without further work. A summary of the overall proposed
paths is given as Scheme 5.
Exp er im en ta l Section
Degr a d a tion Rea ction s; Gen er a l Con d ition s. Solutions
were prepared using water from a Millipore Milli-Q Plus
purifier. When buffered, solutions contained 5 mM sodium
phosphate buffer at pH 7.0 unless otherwise indicated. Tem-
perature was controlled to 26 ( 2 °C during photolysis by
means of a fan. Standard conditions for HPLC analysis
employed 100 mL solutions in a Pyrex vessel with 50 mg of
TiO2 and the desired organic compound at 2 mM. Each mixture
was treated in an ultrasonic bath for 5 min to disperse larger
aggregates, purged with O2, and stirred for 15 min in the dark
before the irradiation was started. The mixture was continu-
ously purged with O2 throughout the experiment. Irradiations
were carried out in Rayonet mini-photochemical reactor with
(65) Mao, Y.; Scho¨neich, C.; Asmus, K.-D. Radical mediated degra-
dation mechanisms of halogenated organic compounds as studied by
photocatalysis at TiO2 and by radiation chemistry.Iin Photocatalytic
Purification and Treatment of Water and Air; Ollis, D. F., Al-Ekabi,
H., Eds.; Elsevier: New York, 1993; Vol. 3, pp 49-66.
(66) Kenneke, J . F.; Ferry, J . L.; Glaze, W. H. The TiO2-Mediated
Photocatalytic Degradation of Chloroalkenes in Water. In Photocata-
lytic Purification and Treatment of Water and Air; Ollis, D. F., Al-
Ekabi, H., Eds.; Elsevier: New York, 1993; Vol. 3, pp 179-191.

Photocatalytic Degradation of 4-Chlorophenol. 2.
J . Org. Chem., Vol. 64, No. 23, 1999 8535
Sch em e 5
magnetic stirring and fan cooling to 26 ( 2 °C. Light was
supplied by 8 × 4 W “black light” lamps whose broad emission
is centered at 360 nm. For HPLC analysis, samples (0.5 mL)
were withdrawn from the reaction vessel at regular time
intervals during the irradiation. They were centrifuged using
an Eppendorf 5415C and filtered using 0.2 µm Whatman
cellulose nitrate syringe filters to remove TiO2 before HPLC
analysis. When temperature regulation was needed, a Pyrex
vessel with an outer jacket was used and the temperature was
regulated to 30.0 ( 0.5 °C by circulation of thermostatically
controlled water.
HP LC An a lyses. The concentrations of cyclic compounds
1-7 were measured by HPLC (HP 1050 with diode array UV-
vis detector) using an ODS Hypersil C18 reverse-phase column
(5µm loop, 200 mm × 2.1 mm). All substances were simulta-
neously detected by absorbance at 220, 255, and 285 nm, and
response factors were obtained at the most favorable wave-
length for each compound. The eluent was a mixture of
methanol and water containing 0.2% acetic acid. The eluent
was 10% methanol for the first 6 min. Then the concentration
of methanol was increased to 40% methanol within 2 min, and
this proportion is held until the analysis was finished. The
flow rate was 0.5 mL min-1. The identification of the inter-
mediates by HPLC was performed by comparing the retention
times and the UV spectra with those of authentic samples.
The concentration of compounds that were not identified was
roughly approximated using the equation derived from cali-
bration measurements for the starting material from which
the intermediate was produced. The concentration of acyclic
intermediates was similarly estimated by comparing with the
absorbance of oxalic acid.
other than the absence of light. Multiple runs of certain decays
showed that rate constants were reproducible to better than
10%.
GC-MS An a lyses. The standard conditions for solutions
were modified when GC was to be used. Initial concentrations
of the organics were 200 µM. Unbuffered solutions were used,
and the pH was adjusted to the desired value ( 0.5 throughout
the photolysis by periodic addition of NaOH. After the pho-
tolysis was completed, the entire mixture was acidified to pH
2.5 by addition of HCl. Solutions were freeze-dried and
silylated as previously described.39 Some solutions were treated
with NaBH4 or NaBD4 prior to silylation as previously
described.39 GC analysis was done with a 25 m DB-5 column
coupled to either a Finnigan Magnum ion trap mass spec-
trometer for MS analysis or an FID detector for routine work.
It should be noted that ion trap mass spectra show intensity
patterns different than those of EI-MS. The temperature
program was 50 °C for 4 min and then ramp to 280 °C at 15
°C min-1
.
Ma ter ia ls. The following compounds were obtained from
Aldrich at the highest purity available and used without
further purification: 4-chlorophenol, ethylene glycol (peak 1),
lactic acid (peak 2), glycolic acid (peak 3), oxalic acid (peak 4),
malonic acid (peak 6), glycerol (peak 8), maleic acid (peak 9),
succinic acid (peak 10), fumaric acid (peak 12), chlorosuccinic
acid (peak 15), malic acid (peak 18), tartaric acid (peak 23),
1,2,4-benzenetriol (peak 24), 1,1,1,3,3,3-hexamethyldisilazane,
chlorotrimethylsilane, pyridine, and sodium borohydride. Tar-
tronic acid (peak 11) was provided by Sigma. The water
employed was purified by a Milli-Q UV plus system (Millipore)
resulting in a resistivity more than 18 MΩ cm-1. TiO2 is
Degussa P-25 which consists of 75% anatase and 25% rutile
with a specific BET surface area of 50 m2 g-1 and a primary
size of 20 nm.67 NMR spectra were obtained at 300 (1H) and
75 MHz (13C) on a Varian VXR instrument.
Decay rate constants in Table 1 were determined by plotting
the ln(conc/conc0) vs. time for the first two to three half-lives,
over which the decays fit well to this first-order treatment.
No particular significance is attached to the absolute values.
Dark decays were obtained using identical conditions as usual,

8536 J . Org. Chem., Vol. 64, No. 23, 1999
Li et al.
Compounds were identified in GC-MS runs by comparison
of the silylated sample retention times and fragmentation
patterns with those of silylated authentic compounds. The
compounds that correspond to peaks 5, 11, 14, 19, 20, 22, 31,
32, and 34 were prepared as previously reported.39
P ea k 16. The identity of 3,4-dihydroxybutyric acid was
confirmed by comparison to the reported mass spectrum of the
silylated derivative.74
P ea k 21. This compound has three acids and/or alcohols.
A formula of C5H7ClO4 is obtained. The parent has two
aldehydes and/or ketone and four unsaturations; the derivative
has two unsaturations. The assignment is made partly in
analogy to peak 19 of Table 1 in our previous work.39 On the
basis of its appearance on degradation of 7 and particularly
10, 3-chloro-4,5-dihydroxy-2-pentenoic acid is the most likely
structure.
P ea k 25. This compound is identical to peak 23 in Table 1
of our previous work.39 It was assigned on the basis of having
a mass spectrum identical to silylated cis-4-hydroxy-2-pen-
tenoic acid but a different retention time.
P ea k 26. This compound has three acids and/or alcohols.
A formula of C5H5ClO5 is obtained. The parent has one
aldehyde or ketone and four unsaturations, and the derivative
has three unsaturations. Both 2- and 3-chloro-2-pentenedioic
acid are possible, but the 3-chloro isomer is chosen because of
the appearance from degradation of 10. The two peaks are
presumed to be due to E/Z isomerization.
P ea k 29. This compound is an isomer of peak 26. 2-Chloro-
4-hydroxy-2-pentenedioic acid is proposed on the basis of the
fact that it does not appear on degradation of 10.
P ea k 35. This compound has four acids and/or alcohols and
a formula of C6H9ClO5. The parent has two aldehydes and/or
ketones and four unsaturations. The derivative has two
unsaturations. The proposed structure is a straightforward
degradation product of 6 (Scheme 4).39 The two peaks are
consistent with the two diastereomers of the alcohols but are
also consistent with alkene stereoisomers.
P ea k 36. This compound has four acids and/or alcohols. It
is not reduced by NaBH4 and has a formula of C6H7ClO6. It
has three unsaturations and is a simple derivative of 3-chlo-
romuconic acid. Dihydroxylation of fumaric acid and related
compounds was observed previously.39 The two diastereomers
are again consistent with the hydroxyl groups.
P ea k 37. This compound was only observed on photocata-
lytic degradation of 3. The mass of the silylated compound is
402. The large peak at 285 (i.e., loss of 117) is usually
correlated with a CO2TMS group. Both 402 and 285 contain a
single Cl atom, as indicated by the isotope pattern. This
compound is not reduced by NaBH4. Several possible formulas
can be written; perhaps the most reasonable is C10H5O6Cl-
(TMS)2, corresponding to C10H7O6Cl in the parent. If the
reaction mixture from degradation of 3 is worked up by
reduction of the aqueous volume and ether extraction, followed
by direct analysis, only two GC-MS peaks are observed. They
correspond to 3 and a compound with the following MS: 220-
(30), 205(100), and 177(25). This ether extract can be silylated
to produce a mixture whose principal components are silylated
3 and peak 37. To the extent that the correlation can be made
between peak 37 and the 220-mass (unsilylated) peak, there
is an upper limit of two TMS groups on peak 37. With this
assumption, peak 37 must be a compound with more than six
carbons and thus an adduct between 3 and some other
compound. This is consistent with peak 37 not being observed
in any of the other degradations analyzed by GC-MS reported
in this paper.
The following compounds were prepared by literature
methods: 4-chlorocatechol, 3;68 2-chloro-3-hydroxypropionic
1
acid (peak 7);69 chloromaleic acid (peak 13),70 mp 107 °C, H
NMR (acetone-d6): δ 6.52, 13C NMR: δ 164.4, 163.7, 137.8,
and 124.8; chlorofumaric acid (peak 17),70 mp 192-193 °C, 1H
NMR (acetone-d6): δ 7.27, 13C NMR: δ 164.5, 162.8, 134.2,
and 128.7; (E,E)-3-chloromuconic acid (peak 27);59,60,71 4-chlo-
ropyrogallol (peak 30),72 mp 163-164 °C, 1H NMR (acetone-
d6): δ 6.68 (d, J ) 8.7 Hz, 1 H), 6.42 (d, J ) 8.7 Hz, 1 H), 13C
NMR δ 145.7, 143.2, 135.1, 120.1, 112.4, 108.3.
P ea k 28. (E)-3-Ch lor o-4-oxoh ex-2-en ed ioic Acid . A so-
lution of HOAc (5 mL), HOOAc (32%, 2.52 mL, 12 mmol), and
ferric acetate (1 mg) was stirred at room temperature. To this
was added dropwise a solution of 5-chloro-1,2,4-benzenetriol
(0.60 g, 3.7 mmol) dissolved in HOAc (10 mL) at approximately
1 drop per minute. After an additional 1 h, the reaction was
concentrated under vacuum to give a dark residue (0.5 g).
NMR analysis of the concentrated oil showed that (E)-3-chloro-
4-oxohex-2-enedioic acid was the major component (>60%) of
a mixture. Attempts to purify the compound by chromatogra-
phy were not fruitful. Tituration of the mixture with acetone
or chloroform confirmed the assignment of the 1H NMR by
changing identities and ratios of impurities but also did not
result in pure material. In CDCl3, the closed form, i.e.,
(3-chloro-2-hydroxy-5-oxo-2,5-dihydrofuran-2-yl)-acetic acid,
predominates, whereas in acetone, the equilibrium favors the
open form. 1H NMR (CDCl3): δ 6.24 (s, 1 H), 3.21 (d, J ) 16.5
1
Hz, 1 H), 2.83 (d, J ) 16.5 Hz, 1 H). H NMR (acetone-d6): δ
6.41 (s, 1 H), 3.12 (s, 2 H). 13C NMR (75 MHz): δ 169.3, 167.5,
158.9, 120.2, 104.8 and 39.8.
P ea k 33. 5-Ch lor o-1,2,4-ben zen etr iol (6).73 1,2,4-Ben-
zenetriol (1.0 g, 7.9 mmol) was dissolved in dry ether (50 mL)
in an ice bath. Sulfuryl chloride (1.1 g, 8.1 mmol) was added
dropwise. The reaction was stirred in an ice bath for 24 h and
then refluxed for 3 h. A black residue (1.2 g) was obtained on
removal of solvent. The crude material was sublimed at 80-
90 °C at 0.2 Torr, and the sublimate was recrystalized in
benzene to give light brown crystals (0.82 g, yield 65%), mp
1
117 °C (d). H NMR (300 MHz, CDCl3): δ 6.87 ppm (s, 1 H),
6.61 (s, 1 H), 5.12 (two close singlets, 2 H, exchanges with D2O),
and 4.70 (s, 1 H, exchanges with D2O). 13C NMR (75 MHz): δ
147.0, 145.9, 139.6, 116.8, 110.1, and 105.0.
Com p ou n d s Ten ta tively Id en tified on th e Ba sis of MS
P a tter n . The mass spectra, along with reasonable constraints
based on the starting materials, generally give unique molec-
ular formulas. The number of TMS groups is readily deter-
mined from the molecular weight and certain fragments. It
corresponds to the total number of alcohol and acid sites in
the unsilylated compound. If the compound is a reduction
product, the number of aldehyde and ketone groups in the
parent is given by the additional molecular weight when
NaBD4 is used. The names used below refer to the unsilylated
compounds, just as in Table 2.
Ack n ow led gm en t. The NSF is gratefully acknowl-
edged for support of this work in the form of a CAREER
Award.
(67) DeGussa. DeGussa Technol. Bull. 1984, 56, 8.
(68) Willstatter, R.; Mu¨ller, H. E. Chem. Ber. 1911, 44, 2182-2191.
(69) Roush, W. R.; Brown, B. B. J . Org. Chem. 1992, 57, 3380-3387.
(70) Perkin, W. H. J . Chem. Soc. 1888, 53, 695-713.
(71) Pandell, A. J . J . Org. Chem. 1983, 48, 3908-3912.
(72) Horner, L.; Gowecke, S. Chem. Ber. 1961, 94, 1267-1276.
(73) J oshi, D. K.; Gold, M. H. Appl. Environ. Microbiol. 1993, 59,
1779-1785.
Su p p or tin g In for m a tion Ava ila ble: Copies of proton
NMR spectra from the preparation of authentics samples for
peak 28 in acetone and chloroform. This material is available
free of charge via the Internet at http://pubs.acs.org.
(74) Esterbauer, H.; Schubert, J .; Sanders, E. B.; Sweeley, C. C. Z.
Naturforsch. B. 1977, 32, 315-320.
J O990912N