Broad spectrum catalytic system for the deep oxidation of toxic organics in aqueous medium using dioxygen as the oxidant

Article abstract of DOI:10.1021/ja990244n

In water, metallic palladium was found to catalyze the deep oxidation of a wide variety of functional organics by dioxygen at 80-90°C in the presence of carbon monoxide or dihydrogen. Several classes of organic compounds were examined: benzene, phenol and substituted phenols, nitro and halo organics, organophosphorus, and organosulfur compounds. In every case, deep oxidation to carbon monoxide, carbon dioxide, and water occurred in high yields, resulting in up to several hundred turnovers over a 24 h period. For substrates susceptible to hydrogenation, the conversions were generally higher with dihydrogen than with carbon monoxide. For organophosphorus compounds, the system presents the first examples of catalytic cleavage of phosphorus - alkyl bonds.

Full text of DOI:10.1021/ja990244n

J. Am. Chem. Soc. 1999, 121, 7485-7492
7485
Broad Spectrum Catalytic System for the Deep Oxidation of Toxic
Organics in Aqueous Medium Using Dioxygen as the Oxidant
Anne Pifer, Terrence Hogan, Benjamin Snedeker, Robert Simpson, Minren Lin,
Chengyu Shen, and Ayusman Sen*
Contribution from the Department of Chemistry, The PennsylVania State UniVersity,
UniVersity Park, PennsylVania 16802
ReceiVed January 25, 1999
Abstract: In water, metallic palladium was found to catalyze the deep oxidation of a wide variety of functional
organics by dioxygen at 80-90 °C in the presence of carbon monoxide or dihydrogen. Several classes of
organic compounds were examined: benzene, phenol and substituted phenols, nitro and halo organics,
organophosphorus, and organosulfur compounds. In every case, deep oxidation to carbon monoxide, carbon
dioxide, and water occurred in high yields, resulting in up to several hundred turnovers over a 24 h period. For
substrates susceptible to hydrogenation, the conversions were generally higher with dihydrogen than with
carbon monoxide. For organophosphorus compounds, the system presents the first examples of catalytic cleavage
of phosphorus-alkyl bonds.
4
This paper encompasses the description of a catalytic system
for the deep oxidation of toxic organics to carbon monoxide,
carbon dioxide, and water in aqueous medium using dioxygen
oxidations and the “Fenton” systems involving H2O2 and a
5
soluble transition metal catalyst. Again, although they have the
advantages of broad substrate applicability and cleanliness, these
systems are not optimal in many situations. For example,
photons are relatively expensive, and the photooxidation systems
cannot be employed where there is a dearth of sunlight or where
large volumes of contaminated water are involved. In Fenton
oxidations, a portion of the H2O2 is wasted because of a parallel
(metal-catalyzed) decomposition pathway. Additionally, soluble
catalysts that are usually used are difficult to remove following
decontamination. Very recently, metal complexes of macrocyclic
ligands have been used as catalysts in conjunction with the
1
as the oxidant. Two broad classes of organic compounds have
been examined: those that model the organic pollutants found
in water and those that model the common chemical warfare
agents.
Toxic organics in water constitute an important environmental
2
hazard. Many are byproducts of industrial production. Several
are introduced into aqueous systems through their usage, e.g.,
as biocides. Additionally, chlorinated organics often result from
the conversion of organic impurities during chlorination of
municipal water. Any procedure for the removal of toxic
organics from water must meet the following criteria. First, the
process must be economical, i.e., only inexpensive reagents and
catalysts may be used. Second, the procedure should be
applicable to a broad spectrum of toxic organics with a variety
of functional groups. Finally, the procedure should not result
in the introduction of anything to the water that needs to be
removed subsequently, i.e., simultaneous water purification and
6
persulfate ion as the oxidant. Again, the systems suffer from
one or more drawbacks: (a) incomplete oxidation of the
substrate, (b) use of expensive ligands and their eventual
oxidative degradation, and (c) the use of expensive, oxidants.
With respect to chemical warfare agents, it is estimated that
there are 30 000 metric tons of chemical weapons stored in the
continental U.S. Since the Chemical Weapons Convention
Treaty mandating the eradication of all chemical weapons by
the year 2007 took effect in April, 1997, and Public Law 102-
484 requiring the destruction of unitary chemical weapons by
7
8
contaminant destruction should be feasible.
One obvious solution to the problem is bioremediation.3
However, many of the toxic organics are xenobiotic in character.
In addition, when enzymes with low substrate specificity
encounter foreign molecules, products that are xenobiotic often
result. Among the catalytic oxidation systems, two that have
been studied most extensively are the TiO2-catalyzed photo-
(
4) Reviews: (a) Pelizzetti, E.; Minero, C.; Vincenti, M. In Technologies
for EnVironmental Cleanup: Toxic and Hazardous Waste Management;
Avogadro A., Ragaini, R. C., Eds.; Euro Courses: Environmental Manage-
ment, 1994; Vol. 2, p 101. (b) Fox M. A.; Dulay, M. T. Chem. ReV. 1993,
9
3, 341. (c) Minero, C.; Pelizzetti, E.; Pichat, P.; Sega, M.; Vincenti, M.
EnViron. Sci. Technol. 1995, 29, 2226. (d) Dong C.; Huang, C.-P. AdV.
Chem. Ser. 1995, 244, 291. (e) Dillert, R.; Fornefett, I.; Siebers, U.;
Bahnemann, D. J. Photochem. Photobiol., A: Chem. 1996, 94, 231.
(5) Recent representative examples: (a) Koyama, O.; Kamagata, Y.;
Nakamura, K. Water Res. 1994, 28, 895. (b) Kiwi, J.; Pulgarin, C.; Peringer,
P. Appl. Catal., B 1994, 3, 335. (c) Sorokin A.; Meunier, B. J. Chem. Soc.,
Chem. Commun. 1994, 1799. (d) Fortuny, A.; Ferrer, C.; Bengoa, C.; Font
J.; Fabregat, A. Catal. Today 1995, 24, 79. (e)Li, Z. M.; Comfort, S. D.;
Shea, P. J. J. EnViron. Qual. 1997, 26, 480.
(6) (a) Sorokin, A.; Seris, J.-L.; Meunier, B. Science 1995, 268, 1163.
(b) Sorokin, A.; Seris, J.-L.; Meunier, B. Chim. Ind. 1995, 151.
(7) U.S. Chemical Weapons Stockpile Information; Declassified News
Release; Office of Assistant Secretary of Defense (Public Affairs): Wash-
ington, DC; January 22, 1996.
(1) Preliminary reports: (a) Hogan, T.; Simpson, R.; Lin M.; Sen, A.
Catal. Lett. 1996, 40, 95. (b) Hogan, T.; Simpson, R.; Lin M.; Sen, A.
Catal. Lett. 1997, 49, 59.
(2) Reviews: (a) Drinking Water and Health; National Academy of
Sciences: Washington, DC, 1977; Chapter VI. (b) Organic Contaminants
in Waste Water, Sludge and Sediment; Quaghebeur, D.; Temmerman, I.,
Angeletti, G., Eds.; Elsevier: New York, 1989. (c) Organic Contaminants
in the EnVironment; Jones, K. C., Ed.; Elsevier: New York, 1991.
(3) Reviews: (a) Gottschalk, G.; Knackmuss, H.-J. Angew. Chem., Int.
Ed. Engl. 1993, 32, 1398. (b) Bioremediation of Chlorinated and Polycyclic
Aromatic Hydrocarbon Compounds; Hinchee, R. E., Leeson, A., Semprini,
L., Ong, S. K., Eds.; Lewis: Boca Raton, FL, 1994. (c) Abramowicz, D.
A.; Olson, D. R. CHEMTECH 1995, 36.
(8) Ember, L. Chem. Eng. News 1996, 74, 9.
1
0.1021/ja990244n CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/06/1999

7
486 J. Am. Chem. Soc., Vol. 121, No. 33, 1999
Pifer et al.
Scheme 1. Mechanism of Pd-Catalyzed Oxidation of
December 31, 2004, has already been enacted, there is signifi-
cant pressure to develop novel methods for the decomposition
of these chemicals.9,10 The optimal method for the destruction
of chemical weapon stockpiles requires an efficient system
which involves minimum exposure of the operators to the toxins.
As such, a closed system that converts these compounds into
benign products is required. Particularly desirable would be a
mild catalytic procedure for deep oxidation using dioxygen.
Most chemical warfare agents contain phosphorus-carbon (e.g.,
Tabun, Sarin, Soman, and VX) or sulfur-carbon (e.g., mustard
gas) bonds, and their neutralization requires the cleavage of
these bonds. Herein, we report a novel system for the efficient
catalytic oxidative cleavage of phosphorus-alkyl and sulfur-
alkyl bonds in model compounds under mild conditions, by
using dioxygen as the oxidant. Previous work has shown that
certain metal-phosphine complexes of Fe, Co, Ni, and Rh
Organic Substrates
1
2
Table 1. Oxidation for para-Substituted Phenolsa
substrate
(mmol)
conversion
(%)
turnovers/surface
Pd
substrate
1
3
undergo stoichiometric phosphorus-carbon cleavage. How-
ever, the phosphorus-carbon bonds in question were generally
phosphorus-aryl bonds and the ease of phosphorus-carbon
p-nitrophenol
p-chlorophenol
p-bromophenol
p-iodophenol
phenol
0.18
0.20
0.15
0.11
0.27
10
17
78
94
46
20
37
125
119
137
2
3
bond rupture follows the order P-Csp > P-Csp > P-Csp ,
3
14
with P-Csp bond cleavage observed very rarely. In one such
report, the activation of the P-Csp3 bond by using electron-
withdrawing CF3 groups was required to effect attack by
a
Reaction conditions: 25 mg of phenol, 15 mg of 5% Pd/C (0.9
µmol of surface Pd atoms), 3 mL of D
2
O/DCl (pH ) 3), 100 psi CO,
1
5
hydroxide. To our knowledge, the present work provides the
first detailed examples of catalytic cleavage of simple phos-
phorus-alkyl bonds.
1000 psi N , 100 psi O , 90 °C, 24 h.
2
2
The third step in the oxidation process involves the metal-
catalyzed oxidation of the substrate by the hydrogen peroxide
equivalent. It is possible to replace carbon monoxide and
dioxygen by hydrogen peroxide; however, unless the latter is
added slowly (see below) the amount of substrate oxidized
relative to the hydrogen peroxide consumed is low due to the
catalytic decomposition of hydrogen peroxide occurring in
parallel with the oxidation. It is this latter undesirable reaction
that made the combination of carbon monoxide and dioxygen
more effective than hydrogen peroxide. Starting with carbon
monoxide and dioxygen, hydrogen peroxide is formed at a low
steady rate through the first two catalytic reactions and is used
more efficiently for substrate oxidation. This was verified by
carrying out the oxidation of phenol under conditions similar
to that shown in Table 1, except that the gas mixture was
replaced by 50% hydrogen peroxide added slowly over a period
of 24 h through a syringe pump. A 40% conversion of phenol
was obtained which is only slightly lower than that observed
with a carbon monoxide-dioxygen mixture.
Several control experiments are also of relevance to the above
catalytic system. First, significantly lower oxidation rates were
observed when a soluble palladium(II) salt, such as K2PdCl4,
was employed instead of supported metallic palladium. This
suggests that the active catalyst is not a soluble palladium
species, and indeed, the rapid reduction of K2PdCl4 to metallic
palladium is observed under the reaction conditions. Presumably,
the lower reaction rate observed starting with the former
Results and Discussion
A. Catalytic System. We have previously described a
catalytic system for the direct, low temperature, oxidation of
methane and lower alkanes.16 In this system, metallic palladium
was found to catalyze the oxidation of alkanes, including
methane, by dioxygen in aqueous medium at 70-100 °C in the
presence of carbon monoxide. While carboxylic acids are the
initial products (and the reaction can be stopped at this stage
by using a large amount of the starting alkane), the ultimate
oxidation products are carbon monoxide and carbon dioxide.
Using ethane as a test substrate, over 1000 turnovers () mmol
of substrate reacted/mmol of surface Pd atoms) were observed
over a 24 h period at 90 °C. Mechanistic studies previously
reported indicate that the overall transformation encompasses
three catalytic steps in tandem (Scheme 1).16a The first is the
water-gas shift reaction (WGSR) involving the oxidation of
carbon monoxide to carbon dioxide with the simultaneous
formation of dihydrogen. The second catalytic step involves the
combination of dihydrogen with dioxygen to yield hydrogen
peroxide (or its equivalent, M-OOH). The formation of
hydrogen peroxide through palladium-catalyzed reaction be-
17
tween carbon monoxide, water, and dioxygen has also been the
subject of several patent applications.18
(9) National Research Council, ReView and EValuation of AlternatiVe
Chemical Disposal Technologies; National Academy Press: Washington,
DC, 1996.
(compared to 5% Pd/carbon) is due to different aggregation of
the metallic palladium formed in situ. A second control
experiment indicated that the carbon support is not oxidized by
the system. The essential role of carbon monoxide in achieving
(
10) National Research Council, AlternatiVe Technologies for the
Destruction of Chemical Agents and Munitions; National Academy Press:
Washington, DC, 1993.
(11) Flamm, K. J.; Kwan, Q.; McNulty, W. B. Chemical Agent and
“
difficult” oxidations is shown by the following competition
Munitions Disposal: Summary of the U.S. Army’s Experience; Report
SAEPEO-CDE-IS-87005, 1987.
1
2
12
experiment between CH3 CH3 (500 psi, 0.03 M solution
(12) Chemical Warfare Agents; Somani, S. M., Ed.; Academic: San
13
12
concentration) and CH3 CH2OH (0.18 mmol, 0.06 M solution
concentration), both in the presence and absence of 100 psi of
CO. The reaction conditions were similar to those de-
scribed below. In the absence of added carbon monoxide, only
Diego, 1992.
(
(
(
13) Garrou, P. E. Chem. ReV. 1985, 85, 171 and references therein.
14) Carty, A. J. Pure Appl. Chem. 1982, 54, 113 and references therein.
15) Aksinenko, A. Y.; Pushin, A. N.; Sokolov, V. B. Phosphorus, Sulfur,
Silicon Relat. Elem. 1993, 84, 249.
16) (a) Lin M.; Sen, A. J. Am. Chem. Soc. 1992, 114, 7307. (b) Lin,
M.; Hogan, T. E.; Sen, A. J. Am. Chem. Soc. 1997, 119, 6048.
17) Gosser, L. W. U.S. Patent 4,681,751, 1987.
(
(18) (a) Rossella, B.; D′Aloisio, R.; Bianchi, D. Eur. Pat. Appl. EP 788,-
998, 1997. (b) Bianchi, D.; Rossella, B.; D’Aloisio, R.; Ricci, M.; Soattini,
S. Eur. Pat. Appl. EP 808,796 A1, 1997.
(

System for Oxidation of Toxic Organics in Water
J. Am. Chem. Soc., Vol. 121, No. 33, 1999 7487
1
3
12
CH3 CO2H (0.03 mmol, 18 turnovers) was formed by the
oxidation of 13CH3 CH2OH. When carbon monoxide was
added, 40% of the products were derived from CH3 CH3. The
products from this reaction were: CH3 CH2OH (0.008 mmol,
12
12
12
12
12
12
12
4
turnovers), CH3 CO2H (0.012 mmol, 7 turnovers), and
13
12
CH3 CO2H (0.030 mmol, 16 turnovers). Thus, the more inert
ethane was oxidized only in the presence of added carbon
monoxide. The requirement of a coreductant (carbon monoxide)
makes the overall reaction formally analogous to the monooxy-
genases in which only one of the two oxygen atoms in the
19
dioxygen molecule is used for substrate oxidation.
Finally, it was of some interest to find out whether the overall
stoichiometry in the oxidation was close to what is shown in
Scheme 1; the system would obviously be less useful if many
equivalents of carbon monoxide were oxidized for every mole
of substrate that was converted. Accordingly, the oxidation of
phenol was carried out to 54% conversion under conditions
similar to that shown in Table 1. A total of 0.14 mmol of phenol
was oxidized, and 1.58 mmol of carbon dioxide was formed.
Assuming 1:1 stoichiometry for conversion of the substrate and
carbon monoxide to carbon dioxide, the formation of 1.68 mmol
of carbon dioxide was expected. While the agreement with
experimental results is gratifying, it raises another interesting
issue given that there are six carbons and five hydrogens in
phenol that have to be oxidized. It suggests that the coreductant
is only required for the most difficult oxidations. Indeed, as
discussed above, carbon monoxide is required as a coreductant
in the oxidation of an alkane to the alcohol but is not required
for the subsequent oxidation of the alcohol.
B. Oxidation of Aromatics, Nitro and Halo Organics. 1.
Scope. Since the above catalytic system was able to effect the
deep oxidation of molecules as unreactive as methane under
unusually mild conditions, we have explored the ability of the
system to catalyze the deep oxidation of hazardous organics.
Several classes of organic compounds were examined: benzene,
phenol and substituted phenols, nitro and halo organics, orga-
nophosphorus, and organosulfur compounds. Typical reaction
conditions were as follows: substrate dissolved in 3 mL of DCl
acidified D2O, 5% Pd/carbon (60 µmol surface Pd atoms/g
catalyst, as determined by dihydrogen chemisorption studies)
as catalyst, 100 psi O2, 100 psi CO, 1000 psi N2, 80-90 °C. In
eVery case, deep oxidation to carbon monoxide, carbon dioxide,
and water occurred in high yields, resulting in up to seVeral
hundred turnoVers oVer a 24 h period. No catalysis was
observed when carbon alone was employed (although a small
loss of reactants (∼10%) was observed, presumably due to
absorption). On the other hand, 5% Pd/Al2O3 was also found
to be an active catalyst; however, the turnover rate was
somewhat lower than that for 5% Pd/carbon. Thus, under similar
reaction conditions, the conversions observed starting with 0.175
mmol of CH3CH2CH2SO3Na were 74% for 5% Pd/Al2O3 and
Figure 1. Plot of the ratio of the log of the rate of oxidation of para-
substituted phenol to the parent phenol versus electron affinity of the
substituent. Reaction conditions: 100 psi CO, 1000 psi N
, 3 mL of D O/DCl (pH ) 3), 15 mg of 5% Pd/C (0.9 µmol of
surface Pd atoms), 25 mg of C OH, 25 mg of p-X-C OH, 24 h at
0 °C.
2
, 100 psi
O
2
2
H
6 5
6 4
H
9
The deep oxidation of a number of para-substituted phenols
was achieved under typical reaction conditions (Table 1). The
products of the decomposition of p-bromophenol (25 mg, 0.145
1
mmol) were examined by H NMR spectroscopy and were found
to included p-benzoquinone (0.08 mmol) and formic acid (0.28
mmol). Trace amounts of acetic and glycolic acids were also
detected. The remainder of the p-bromophenol was converted
-
to COx, H2O, and Br . At longer reaction times, no organics
were observed in solution from any of the substrates. A series
of competition experiments were performed and, as can be seen
in Figure 1, the rate of conversion of the substrate decreased
with increasing electronegativity of the para-substituent, with
an approximately linear correlation between the electron affinity
of the substituent and the ratio of the log of the rate of oxidation
of the substituted phenol to the parent phenol. This is consistent
with an initial electrophilic attack at the ring. While the exact
nature of the electrophile remains to be elucidated, one
possibility in analogy with monoxygenases is a metal-oxo
species generated on the metal surface (see Conclusion).
Perhalogenated aromatics were also examined. 2,4,6-Trichlo-
rophenol, a toxic byproduct from paper mills, was decomposed
under similar reaction conditions. Of note is that trichlorophenol
is poorly soluble in water and yet oxidation occurred. Starting
with 25 mg (0.13 mmol) of 2,4,6-trichlorophenol, 94% conver-
sion was noted after 1 d (100 turnovers/surface Pd atoms). The
9
5% for 5% Pd/carbon.
1
analysis was performed by examining the H NMR spectra of
Phenol was found to be the initial oxidation product of
benzene. p-Benzoquinone, glycolic, and formic acids were
formed subsequently. That these products arose from benzene
both the aqueous layer and the CDCl3 extract. Trace amounts
of acetic and formic acids were the only carbon-containing
products found. The chloride ions formed in solution were
determined gravimetrically by precipitation with AgNO3. The
results indicated that the chlorine atoms present in the substrate
were quantitatively converted to chloride ions.
(
and not by the hydrogenation of carbon monoxide) was verified
13
by starting with C6H6 which lead to the formation of the
1
3
corresponding C-labeled products. Because of the poor
solubility of benzene in water, as described later, quantitative
studies of benzene oxidation were carried out in a mixture of
perfluorobutyric acid and water.
In a similar manner 4,4′-bibromobiphenyl (25 mg, 0.08 mmol)
-
was oxidized with an 40% conversion to COx, H2O, and Br
(
35 turnovers/surface Pd atoms). Again, only traces of acetic
(
19) Recent reviews on monooxygenases and their synthetic analogues:
1
and formic acids were detected by H NMR spectroscopy, and
bromide ions were quantitatively determined by precipitation
with AgNO3.
(
a) Metalloporphyrins in Catalytic Oxidations; Sheldon, R. A., Ed.; Marcel
Dekker: New York, 1994. (b) Metalloporphyrins Catalyzed Oxidations;
Montanari, F.; Casella, L., Eds.; Kluwer: Dordrecht, 1994.

7488 J. Am. Chem. Soc., Vol. 121, No. 33, 1999
Pifer et al.
Table 2. Oxidation of Nitro, and Halo Aromaticsa
conversion
conversion with
CO/O2 (%)
side products with
H2/O2 (mmol)
side products with CO/O2
substrate
nitromethane
with H2/O2 (%)
(mmol)
94
86
+99
+99
98
ammonium (0.01)
nitrate (>0.04)
formic acid (0.06)
ammonium (0.10)
nitrate (<0.04)
formic acid (tr)
ammonium (0.06)
nitrate (<0.04)
formic acid (tr)
ammonium (0.07)
nitrate (<0.04)
formic acid (tr)
ammonium (0.02)
nitrate (>0.08)
formic acid (0.06)
ammonium (0.01)
nitrate (>0.04)
formic acid (0.02)
ammonium (0.06)
nitrate (>0.08)
p-nitrophenol
m-nitrophenol
nitrobenzene
nitrosobenzene
+99
+99
+99
+99
formic acid (0.21)
ammonium (0.06)
nitrate (>0.08)
formic acid (0.07)
ammonium (0.04)
nitrate (>0.08)
formic acid (0.08)
ammonium (0.09)
nitrate (<0.04)
+99
formic acid (tr)
aminophenol (0.04)
aniline (0.01)
hydroquinone (tr)
ammonium (0.03)
nitrate (<0.04)
azobenzene
aniline
96
85
ammonium (0.04)
nitrate (<0.04)
aniline (0.02)
formic adid (0.01)
aniline (0.02)
hydroquinone (tr)
ammopnium (0.07)
nitrate (<0.04)
62
98
87
99
88
77
86
ammonium (0.05)
nitrate (<0.04)
formic acid (tr)
ammonium (0.10)
nitrate (<0.04)
formic acid (0.08)
ammonium (0.05)
nitrate (>0.04)
1,2-dinitrobenzene
1,4-dinitrobenzene
1,3-dinitrobenzene
1-chloro-3-nitrobenzene
formic acid (0.04)
ammonium (tr)
99
ammonium (0.14)
nitrate (>0.08)
formic acid (tr)
ammonium (0.10)
nitrate (<0.04)
nitrate (<0.04)
formic acid (tr)
ammonium (0.01)
nitrate (<0.04)
73
formic acid (0.05)
ammonium (0.02)
nitrate (>0.08)
+99
ammonium (0.08)
nitrate (<0.04)
formic acid (0.04)
aniline (0.06)
ammonium (0.03)
nitrate (>0.08)
formic acid (0.06)
formic acid (0.01)
formic acid (0.03)
1-chloro-4-nitrobenzene
80
96
85
92
ammonium (0.02)
nitrate (>0.04)
formic acid (0.03)
p-chlorophenol
formic acid (0.09)
hydroquinone (0.06)
formic acid (0.09)
formic acid (0.08)
m-chlorophenol
p-bromophenol
+99
+99
89
99
formic acid (0.02)
formic acid (0.08)
hydroquinone (0.03)
p-hydroxybenzoic acid (0.04)
formic acid (0.04)
m-bromophenol
+99
+99
formic acid (0.09)
hydroquinone (tr)
m-hydroxybenzoic acid (0.08)
a
Reaction conditions: 0.25 mmol substrate, 30 mg of 5% Pd/C, 3 mL of H
C, 16 h.
2 2 2 2
O/HCl (pH ) 1), 100 psi H or CO, 1000 psi N , 100 psi O , 130
°
2 2 2
Table 3. Comparison of CO/O vs H /O
at Lower Conversionsa
conversion with
H2/O2 (%)
conversion with
CO/O2 (%)
side products with
H2/O2 (mmol)
side products with
CO/O2 (mmol)
substrate
p-nitrophenol
79
31
ammonium (0.04)
formic acid (0.06)
ammonium (tr)
formic acid (0.03)
hydroquinone (0.07)
formic acid (0.09)
phenol (0.17)
hydroquinone (0.04)
formic acid (0.03)
hydroquinone (0.10)
unidentified
ammonium (0.03)
formic acid (0.13)
p-chlorophenol
89
57
aniline
nitrobenzene
38
92
16
67
unidentified
ammonium (0.11)
aniline (0.28)
hydroquinone (0.01)
a
Reaction conditions: 1 mmol substrate, 15 mg of 5% Pd/C, 3 mL of H
2
2 2 2
O/HCl (pH ) 1), 100 psi H or CO, 1000 psi N , 100 psi O , 90 °C,
1
6 h.
Although the catalytic system described above exhibits many
perfluorobutyric acid, and CH3CH2CH2SO3Na was used as the
test substrate (125 mg, 0.76 mmol). As shown in Figure 2, a
slight enhancement in oxidation rate was observed initially with
increasing perfluorobutyric acid concentration in the solvent
mixture. The maximum rate was achieved at 50% (v/v)
perfluorobutyric acid:water mixture with a 98% conversion of
the substrate (500 turnovers/surface Pd atoms). Beyond 60%
(v/v) perfluorobutyric acid, the reaction rate decreased sharply
attractive features, one potential problem is that many toxic
organics are insoluble in water and therefore are likely to be
oxidized very slowly under typical reaction conditions. Con-
sequently, we have briefly examined alternative solvent systems
that are superior to water in dissolving organic compounds. One
such solvent system (which is not, however, useful from a
practical standpoint) consists of a mixture of water and

System for Oxidation of Toxic Organics in Water
J. Am. Chem. Soc., Vol. 121, No. 33, 1999 7489
Table 4. Oxidation of Nitro Aromatics in CF
3
CO
2
H/H
2
Oa
noseb) and explosives (e.g., di- and trinitrotoluene). These
20
2
1
compounds persist in the soil for many years. Halo organic
contaminants stem from their use as degreasing solvents,
pesticides, herbicides, and heat transfer fluids. Again, the
bioremediation of these compounds is an extremely slow
conversion
side products with
H
substrate
with H
2
/O
2
(%)
2
/O (mmol)
2
1
1
,4-dinitrobenzene
+99
+99
ammonium (0.35)
nitrate (>0.08)
ammonium (0.32)
nitrate (<0.04)
3
,2-dinitrobenzene
process.
As can be seen from Table 2, efficient oxidative degradation
of substrates occurred when either dihydrogen or carbon
monoxide was used as the coreductant. However, where there
is a difference, lower substrate decomposition was almost
invariably observed using carbon monoxide as the coreductant.
This trend became more obvious when the reactions were carried
out at lower conversion (Table 3). Note that no reaction occurred
when the coreductant was absent. There was no reaction either
when carbon monoxide but not dioxygen was present. On the
other hand, near quantitative reduction to cyclohexane deriva-
tives occurred when dihydrogen but not dioxygen was present
in the reaction mixture. Additionally, for nitrogen-containing
substrates, the formation of ammonium ion was observed. It is
clear from these experiments that while the prior reduction of
the substrate is not a requirement for its oxidative degradation,
such a step may enhance the decomposition rate, as is evident
from the comparison of the reactions involving dihydrogen and
carbon monoxide as coreductants. Interestingly, however, aniline
was slower to react than either nitro or nitroso benzene and is,
therefore, not an intermediate in the oxidation of the latter
substrates (see Tables 2 and 3). Presumably, aniline is protonated
in the reaction medium, and the resultant electron-withdrawing
anilinium group deactivates the ring toward electrophilic attack.
formic acid (0.03)
ammonium (0.20)
nitrate (<0.04)
1
,3-dinitrobenzene
+99
formic acid (tr)
a
Reaction conditions: 0.25 mmol substrate, 30 mg of 5% Pd/C,
O, 1.5 mL of CF CO H, 100 psi H , 1000 psi N , 100 psi
, 150 °C, 16 h.
1
O
.5 mL of H
2
2
3
2
2
2
In all of the oxidations involving nitrogen-containing sub-
1
strates, the ammonium ion was detected by H NMR spectro-
2
2
scopy, and in many cases the nitrate was also detected (Table
). However, the combined amount of these two products was
2
Figure 2. Plot of the conversion of CH
perfluorobutyric acid (v/v) in D O. Reaction conditions: 100 psi CO,
000 psi N , 100 psi O , 25 mg of 5% Pd/C (1.5 µmol of surface Pd
atoms), 0.76 mmol CH CH CH SO Na, 3 mL of solvent, 90 °C, 24 h.
3
CH
2
CH
2
SO
3
Na versus
significantly lower than that of the total nitrogen in the substrate,
thereby indicating that other nitrogen-containing species were
also generated in the reactions (for example, NH4NO3 is known
to decompose thermally to N2O and water). No ammonia gas
was detected by GC analysis in the headspace of the reactor
after the reaction of p-nitrophenol under standard conditions in
the presence of 100 psi of dihydrogen. To determine if partial
decomposition of the ammonium ion was occurring following
its formation, a control reaction using NH4Cl (1 mmol) as the
substrate was conducted in 1:1 H2O/CF3CO2H. Some decom-
position (approximately 20%) of the ammonium ion was
%
1
2
2
2
3
2
2
3
presumably because water was necessary for the generation of
H2O2 (Scheme 1).
The 50% (v/v) perfluorobutyric acid:water mixture was used
as the solvent to carry out the deep oxidation of two substrates
that are poorly soluble in water alone: benzene and 1,1,2,2-
tetrachloroethane. Under typical reaction conditions, starting
with 0.5 mmol of benzene in 3 mL of solvent mixture, 435
turnovers/surface Pd atoms to COx and H2O were observed in
a 24 h period. Traces of formic and glycolic acids were the
only organic products observed in solution. In the case of
1
observed by H NMR spectroscopy. Approximately half of this
decomposition was to the nitrate ion. The use of NaNO3 (1
mmol) as the substrate under standard conditions afforded no
ammonium ion, thereby indicating that the ammonium was not
produced from an intermediate nitrate ion. Additionally, when
starting with NaNO3, neither ammonia nor the ammonium ion
1
,1,2,2-tetrachloroethane, 35 turnovers/surface Pd atoms were
observed as determined by gravimetric chloride analysis.
. Use of Carbon Monoxide-Dioxygen Mixture versus
2
1
was detected by GC analysis and H NMR spectroscopy,
Dihydrogen-Dioxygen Mixture. Since the function of the
added carbon monoxide in the above system is to generate
dihydrogen (Scheme 1), an interesting issue involves the
comparison of oxidation rates starting with dioxygen and
dihydrogen versus dioxygen and carbon monoxide. Clearly, the
rate in the first case should be equal to or higher than that
observed with the latter combination. The issue is further
complicated by the question of whether in the case of aromatic
substrates a prereduction step facilitates the subsequent deep
oxidation. Accordingly, we have examined the effect of added
dihydrogen versus that of carbon monoxide in the deep oxidation
of nitro and halo aromatics. Nitro aromatics are a major class
of groundwater and soil contaminants because of their use as
pesticides (e.g., 2-sec-butyl-4,6-dinitrophenol, trade name: di-
respectively, when an organic hydrogen source, phenol (1
mmol), was added or when dioxygen was removed from the
reaction mixture.
The pH and the composition of the solvent appears to have
a significant effect on the oxidative degradation of the above
substrates. For example, using the dihydrogen/dioxygen mixture,
substituting H2SO4 for HCl but maintaining a pH ) 1 resulted
(
20) (a) Higson, F. K. In AdVanced and Applied Microbiology; Neidle-
man, S. L., Laskin, A. I., Eds.; Academic: San Diego, 1992; Vol. 37, p 1.
(b) Kaake, R. H.; Roberts, D. J.; Stevens, T. O.; Crawford, R. L.; Crawford,
D. L. Appl. EnViron. Microbiol. 1992, 58, 1683.
(21) Funk, S. B.; Crawford, D. L.; Crawford, R. L.; Mead G.; Davis-
Hooper, W. Appl. Biochem. Biotechnol. 1995, 51, 625.
(22) Hutson, A. C.; Sen, A. J. Am. Chem. Soc. 1994, 116, 4527.

7
490 J. Am. Chem. Soc., Vol. 121, No. 33, 1999
Table 5. Oxidation of Phosphorus and Sulfur Compoundsa
conversion with conversion with
Pifer et al.
side products with H
(mmol)
2
/O
2
side products with CO/O
(mmol)
2
substrate
H
2
/O (%) CO/O (%)
2
2
trimethylphosphine oxide
15
40
dimethylphosphinic acid (0.14)
methylphosphonic acid (tr)
formic acid (tr)
dimethylphosphinic acid (0.34)
methylphosphonic acid (0.06)
formic acid (tr)
dimethylphosphinic acid
16
11
methylphosphonic acid (0.15)
formic acid (tr)
methylphosphonic acid (0.11)
formic acid (0.27)
methylphosphonic acid
dimethylsulfide
5
66
10
61
formic acid (tr)
dimethylsulfoxide (0.02)
formic acid (tr)
formic acid (0.15)
dimethylsulfoxide (0.22)
dimethylsulfone (0.04)
formic acid (tr)
dimethylsulfoneb
9
6
dimethylsulfoxide (0.03)
methylsulfoxide (0.05)
a
Reaction conditions: 1 mmol substrate, 15 mg of 5% Pd/C, 3 mL of H
2
2 2 2
O/HCl (pH ) 1), 100 psi H or CO, 1000 psi N , 100 psi O , 90 °C,
b
2
0 h. 0.5 mmol substrate.
Table 6. Oxidation of Trimethylphosphine Oxidea
CH PO (CH P(O)(OH)
(
3
)
3
3
)
2
(CH
3
)P(O)(OH)
(mmol)
2
H
3
PO
4
2
HCO H
(mmol)
time (h)
(mmol)
(mmol)
(mmol)
TONb
0
5.75
9.0
8.0
2.0
09.5
57.5
79.0
.00
5.43
4.07
2.06
1.11
0.51
0.25
0.07
0.04
0.00
1.25
2.06
2.51
2.09
1.67
1.00
0.72
0.00
0.19
1.31
1.58
2.26
2.57
2.68
2.57
0.00
0.00
0.00
0.22
0.56
0.93
1.68
2.10
0.00
2.21
5.00
5.16
4.48
3.27
2.35
2.22
0
543
1
3
6
9
1560
2110
2763
3200
3800
4053
1
1
1
a
Reaction conditions: 100 psi CO, 1000 psi N
2
, 100 psi O
2
, 10 mL of D
2
O/DCl (pH ) 1), 50 mg Pd/carbon (3 µmol surface Pd atoms), 500
P(O)(OH)] + 2[(CH )P(O)(OH) ] + 3[H PO ]}/[surface Pd atoms].
b
mg of (CH
3
)
3
PO (5.43 mmol), 90 °C. Turnover number defined as {[(CH
3
)
2
3
2
3
4
in a similar oxidation rate for 1,3-dinitrobenzene. On the other
hand, a significantly lower conversion rate was observed in
neutral water. The use of pure CF3CO2H resulted in reduced
oxidation and formation of cyclohexane derivatives and am-
monium ion from 1,3-dinitrobenzene by hydrogenation/hydro-
genolysis. On the other hand, the use of a 1:1 (v/v) mixture of
CF3CO2H and water resulted in efficient substrate oxidation that
was generally comparable to HCl-acidified water (Table 4). The
former solvent system was, however, significantly superior for
substrates that are poorly soluble in water, such as 1,3-
dinitrobenzene.
Finally, an alternative explanation for the increase in substrate
oxidation rate with dihydrogen/dioxygen mixture over carbon
monoxide/dioxygen mixture may be that hydrogen simply
produces a higher concentration of hydrogen peroxide (see
Scheme 1). However, a comparison of the two systems for
substrates for which hydrogenation is unlikely appears to rule
out this possibility. Thus, for a series of organic phosphorus
and sulfur compounds similar (or lower) oxidation rates were
observed for the dihydrogen/dioxygen mixture (Table 5).
The most likely explanation for the superiority of the
dihydrogen/dioxygen mixture over carbon monoxide/dioxygen
mixture for the deep oxidation of aromatics is that the observed
Figure 3. Plot of the products formed from (CH
Reaction conditions: 100 psi CO, 1000 psi N , 100 psi O
O/DCl (pH ) 1), 50 mg of 5% Pd/carbon (3 µmol surface Pd atoms),
00 mg of (CH PO (5.43 mmol), 90 °C. (CH PO, (O); (CH P-
O)(OH), (4); CH P(O)(OH) , (0); H PO , (+); HCO H, (b).
3
)
3
PO versus time.
2
2
, 10 mL of
D
5
(
2
)
3 3
3
)
3
3 2
)
2
prereduction by dihydrogen converts the stronger sp C-H
3
2
3
4
2
3
bonds to weaker sp C-H bonds (e.g., the C-H bond energy
is ∼15 kcal lower in cyclohexane compared to that in benzene),
thereby facilitating their subsequent activation and oxidation.
C. Oxidation of Organophosphorus and Sulfur Com-
pounds. Oxidation of trimethylphosphine oxide produced three
phosphorus-containing products: dimethylphosphinic acid, me-
thylphosphonic acid, and phosphoric acid (Table 6). These
resulted from the cleavage of one, two, and three phosphorus-
carbon bonds, respectively. The methyl groups were oxidized
to form formic acid and ultimately CO2. From Figure 3, it is
evident that dimethylphosphinic acid and methylphosphonic acid
are intermediates in the formation of phosphoric acid. After 179
h, only a trace of (CH3)3PO remained. In 179 h, oVer 4000
phosphorus-carbon bonds were cleaVed by each Pd! A 31
P
NMR stack plot of the reaction is shown in Figure 4. Again, it
is apparent that only traces of starting material remain. A study
of the kinetics of both dimethylphosphinic acid and methylphos-
phonic acid oxidation indicated a zero-order relationship
between reaction rate and substrate concentration. Such a
relationship is ideal in the context of efficient oxidation of
substrates down to low concentrations. More significantly, both
substrates appear to be oxidized at similar rates thereby
indicating that other substituents present on the phosphorus had

System for Oxidation of Toxic Organics in Water
J. Am. Chem. Soc., Vol. 121, No. 33, 1999 7491
Figure 4. Stack plot of 31P NMR spectra obtained after the following reaction conditions: 100 psi CO, 1000 psi N
D2O/DCl (pH ) 1), 50 mg of 5% Pd/carbon (3 µmol surface Pd atoms), 500 mg of (CH PO (5.43 mmol), 90 °C.
2 2
, 100 psi O , 10 mL of
3 3
)
Table 7. Oxidation of Dimethylphosphinic Acida
was 5.6 on a per bond basis. However, this number is subject
to error since phosphonoacetic acid itself was efficiently
oxidized to phosphoric acid. For example, when phosphonoace-
tic acid (0.71 mmol) was oxidized under standard conditions
for 16 h, a 46% conversion to phosphoric acid resulted.
time (CH P(O)OH (CH )P(O)(OH) PO
3
)
2
3
2
H
3
4
HCO
2
H
(h)
(mmol)
(mmol)
(mmol) (mmol) TONb
0
8
8
9.5
1.00
0.89
0.81
0.73
0.00
0.11
0.18
0.25
0.00
0.00
0.01
0.02
0.00
0.27
0.21
0.66
0
122
222
322
1
3
5
By simple filtration, the catalyst could be retrieved and reused
with no significant loss in activity. For example, 85.3% of
(CH3)3PO (0.54 mmol) was oxidized in 20 h. Removal of the
catalyst by filtration and subsequent reuse resulted in the
oxidation of 78.6% of (CH3)3PO (0.54 mmol) in 20 h. The small
loss in activity was most likely due to loss of catalyst during
filtration.
a
Reaction conditions: 100 psi CO, 1000 psi N
2
, 100 psi O
2
, 3 mL
of D O/DCl (pH ) 1), 15 mg of Pd/carbon (0.9 µmol surface Pd atoms),
2
b
9
4 mg of (CH
defined as {[CH
3
)
3
2
P(O)OH (1.00 mmol), 90 °C. Turnover number
P(O)(OH) ] + 2[H PO ]}/[surface Pd atoms].
2
3
4
Table 8. Oxidation of Methylphosphonic Acida
H (mmol) TONb
3 4 2
H PO (mmol) HCO
The deep oxidation of sulfur-containing analogues to mustard
gas, (ClCH2CH2)2S, occurred readily with the bulk of the
substrate being converted to carbon monoxide, carbon dioxide,
and water (Table 10). A 1:1 (v/v) mixture of water and
perfluorobutyric acid was used to enhance the solubility of the
substrate. Starting with dimethylsulfide (1.00 mmol), the species
observed in solution after 17 h of reaction were dimethylsul-
foxide (0.22 mmol), methanesulfonic acid (0.04 mmol), and
formic acid (0.26 mmol). Dimethylsulfoxide (0.5 mmol) itself
was converted to methanesulfonic acid (0.14 mmol), and
dimethylsulfone (0.30 mmol) after 18 h. To determine if
dimethylsulfone was an intermediate in the oxidation of
dimethylsulfoxide to methanesulfonic acid, the products from
oxidation of dimethylsulfoxide (2 mmol) and the oxidation of
a mixture of dimethylsulfoxide (1 mmol) and dimethylsulfone
time (h) (CH
3
)P(O)(OH)
2
0
8
8
9.5
1.00
0.90
0.80
0.71
0.00
0.10
0.20
0.29
0.00
0
111
222
322
1
3
5
0.15
0.28
0.23
a
Reaction conditions: 100 psi CO, 1000 psi N
of D
6 mg of CH
defined as [H PO
2
, 100 psi O
2
, 3 mL
2
O/DCl (pH ) 1), 15 m of Pd/carbon (0.9 µmol surface Pd atoms),
b
9
3
P(O)(OH)
2
(1.00 mmol), 90 °C. Turnover number
3
4
]/[surface Pd atoms].
little effect on the rate (Tables 7 and 8). To ascertain whether
the final product, orthophosphate, inhibited the oxidation, the
oxidation of 1 mmol of trimethylphosphine oxide was carried
out under identical conditions with and without the addition of
mmol of NaH2PO4. A 25% reduction in rate was observed in
the presence of the phosphate.
1
(1 mmol) was analyzed. Since the yield of methanesulfonic acid
The oxidation of triethylphosphine oxide led to 10 phosphorus-
containing products. Partial oxidation of the ethyl groups
competed with phosphorus-carbon cleavage. The phosphorus-
containing products derived from phosphorus-carbon cleavage
that were identified were diethylphosphinic acid, ethylphos-
phonic acid, and phosphoric acid. In addition phosphonoacetic
acid was identified as a product derived from the partial
oxidation of an ethyl group. Other organic products derived from
the oxidation of the ethyl group that were present in solution
were acetic and formic acids. After 128.5 h, no starting material
remained, and the major phosphorus-containing products were
phosphonoacetic acid, ethylphosphonic acid, and phosphoric
acid. A separate study of ethylphosphonic acid oxidation showed
that phosphorus-carbon cleavage was significantly favored over
carbon-hydrogen cleavage and oxidation (Table 9). From the
ratio of H3PO4 to HO2CCH2P(O)(OH)2, it can be estimated that
the relative phosphorus-carbon to carbon-hydrogen cleavage
was similar in both reactions (0.59, 0.53 mmols, respectively),
it is evident that dimethylsulfone was not an intermediate in
the oxidation of dimethylsulfoxide to methanesulfonic acid.
Unlike dimethylsulfide and dimethylsulfoxide, dimethylsulfone
and methanesulfonic acid were particularly resistant to oxidation.
Starting with 0.5 mmol of substrate, over 90% of the substrate
remained unreacted in each case after 18 h. The ease of oxidation
decreased in the order: (CH3)2S > (CH3)2SO > (CH3)2SO2 and
further supports the conclusion that the system acts as an
electrophilic oxidant.
The ethyl analogues showed a similar reactivity pattern.
Starting with diethyl sulfide (1.00 mmol), the species observed
in solution after 28 h of reaction were: diethylsulfoxide (0.13
mmol) and unreacted diethyl sulfide (0.42 mmol). Diethyl
sulfone (0.82 mmol) was much more reactive than its methyl
analogue, and after 16.5 h of reaction the species present in

7
492 J. Am. Chem. Soc., Vol. 121, No. 33, 1999
Table 9. Oxidation of Ethylphosphonic Acida
)P(O)(OH)
Pifer et al.
(C
H
2 5
2
HO
2
CCH
(mmol)
2
P(O)(OH)
2
H
3
PO
4
CH
(mmol)
3
CO
2
H
2
HCO H
(mmol)
time (h)
(mmol)
(mmol)
TONb
0
8
8
9.5
1.00
0.71
0.43
0.14
0.00
0.10
0.20
0.30
0.00
0.19
0.36
0.56
0.00
0.05
0.09
0.13
0.00
0.24
0.46
0.45
0
322
622
956
1
3
5
a
Reaction conditions: 100 psi CO, 1000 psi N
2
, 100 psi O
2
, 3 mL of D
2
O/DCl (pH ) 1), 15 mg of Pd/carbon (0.9 µmol surface Pd atoms), 110
CCH P(O)(OH) ] + [H PO ]}/[surface Pd atoms].
b
mg of (C
H
2 5
)P(O)(OH)
2
(1.00 mmol), 90 °C. Turnover number defined as {[HO
2
2
2
3
4
Table 10. Oxidation of Methyl-Sulfur Compoundsa
Experimental Section
(
CH3)2S (CH3)2SO (CH3)2SO2 CH3SO3H
Appropriate precautions should be taken while working with gases
under high pressures. Particular attention should be paid to flam-
mability limits of gas mixtures. In our experiments, a 1000 psi of
dinitrogen was routinely used as a diluent.
substrate (mmol) time (h) (mmol) (mmol)
(mmol)
(mmol)
(
(
(
CH3)2Sb,c (1.00)
CH3)2SO (0.50) 18.5
CH3)2SO2 (0.50) 18.5
17.0
0.39
0.22
0.04
0.30
0.03
b
0.14
0.47
A. Materials and Equipment. Reagent grade chemicals were used
as received. Reactions under pressure were carried out in Parr general
purpose bombs using glass liners. Reaction products were identified
a
Reaction conditions: 100 psi CO, 1000 psi N
O/DCl (pH ) 1), 15 mg of Pd/carbon (0.9 µmol surface Pd atoms),
00 °C. Deep oxidation to HCO
2 2
, 100 psi O , 3 mL
of D
1
2
b
c
1
31
H and CO
2
occurred. Solvent: 1.5
by their H NMR and/or P NMR spectra recorded on a Br u¨ ker AM
300 FT-NMR spectrometer using solvent resonance at appropriate
frequency or an external standard consisting of a capillary tube
2
mL of CF CF CF CO H + 1.5 mL of D
3
2
2
2
2
O/DCl (pH ) 1).
solution were: ethanesulfonic acid (0.15 mmol), acetic acid
0.08 mmol), formic acid (0.13 mmol), along with unreacted
2
containing 1 µL of DMSO in 60 µL of D O for lock, reference, and
(
1
31
3 4
integration standard (for H NMR spectra) or H PO (for P NMR
sulfone (0.60 mmol). After 20.5 h of reaction time the following
species were observed in solution starting with ethane sulfonic
acid (0.2 mmol): methylsulfate (0.03 mmol), acetic acid (0.06
mmol), formic acid (0.12 mmol), along with unreacted starting
material (0.03 mmol).
spectra). The identity of specific NMR resonances was confirmed by
comparison to standard reference spectra and/or co-injection of
1
standards. The ammonium ion was quantified by H NMR spectros-
copy.22 The procedure for detection of nitrate ion in solution was
adapted from a colorimetric method used for low concentrations of
nitrate in groundwater samples.24 A brucine-sulfanilic acid reagent
2 4
was prepared, and when added, along with NaCl(aq) and H SO , to a
reaction solution containing nitrate, the solution turned bright pink. The
gaseous products in several reactions were identified via gas-phase
Conclusion
In conclusion, we have discovered an unusually versatile
catalytic system for the deep oxidation of toxic organics in water.
This system possesses several attractive features not found
simultaneously in other reported systems. These are (a) the
ability to directly utilize dioxygen as the oxidant, (b) the ability
to carry out the deep oxidation of a particularly wide range of
functional organics, and (c) the ease of recovery of the catalyst
by simple filtration.
injection GC conducted on a HP 5890 GC equipped with a Supelco 15
1
ft ×
8
/ in. column with 60/80 Caroxen 1000 support. Reaction
chromatograms were compared with standards for positive identification
of products.
B. Catalyst. Commercial (Aldrich or Alfa Aesar) 5% palladium on
carbon (60 µmol surface Pd atoms/g catalyst, as determined by
dihydrogen chemisorption studies) was employed. A scanning electron
micrograph of the catalyst revealed the presence of particles ranging
from 10 to 140 Å in size. An X-ray dot map analysis showed the
palladium to be evenly dispersed throughout the carbon support with
domain sizes smaller than detection limits.
While our understanding of the mechanistic steps involved
in the overall oxidation is far from complete, it is clear that a
powerful electrophilic oxidant is generated in situ. One pos-
sibility is a metal-oxo species generated on the metal surface
through the sequence of steps outlined in eq 1. An initial metal-
hydride is expected to form through the water-gas shift reaction
or by direct reaction with dihydrogen. Such a species is known
C. General Procedure. Specific reaction conditions are given as
footnotes to the tables. In a typical reaction, the substrate was added
to 3 mL of DCl acidified D O (pH ) 1) and placed in a glass liner
2
containing an appropriate amount of 5% palladium on carbon. The liner
was placed in a high-pressure Parr reactor that was subsequently charged
to 100 psi with dihydrogen or carbon monoxide, to 1100 psi with
dinitrogen, and finally to 1200 psi with dioxygen, and heated for a
specific time period. At the end of this period, the gases were vented
and the contents of the liner were analyzed. To test the reproducibility
of the reactions, the key experiments were repeated and the results were
found to be within 10% of each other.
23
to react with dioxygen to form a metal-hydroperoxide, which
can be subsequently protonated to release water and form the
metal-oxo species (hence the requirement of an acid). The
sequence of
(1)
Acknowledgment. This research was funded by the National
Science Foundation and the U.S. Department of Energy.
steps is precedented and bears resemblance to that occurring in
monoxygenases.
19
JA990244N
(
23) (a) Specific example: Hosokawa, T.; Nakahira, T.; Takano, M.;
(24) Skougstad, M. W.; Fishman, M. J.; Friedman, L. C.; Erdmann, D.
E.; Duncan, S. S. Techniques of Water-Resources InVestigations of the
United States Geological SurVey; US Department of Interior: Washington,
DC; Book 5, Chapter A1, p 429.
Murahashi, S. J Mol. Catal. 1992, 74, 489. (b) Review: Catalytic Oxidations
with Hydrogen Peroxide as Oxidant; Strukul, G., Ed.; Kluwer Academic:
Dordrecht, 1992.