Molecular engineering of heterogeneous catalysts: An efficient catalyst for the production of hydrogen peroxide

Article abstract of DOI:10.1006/jcat.1996.0162

A goal of chemists for many years has been to control the structure of solid-state compounds to induce desired chemical properties. A porous metal phosphonate compound with chemically active pillaring groups, which allow for the direct production of hydrogen peroxide from hydrogen and oxygen, has been designed. The porous compound, Zr₂(PO₄)(O₃PCH₂CH ₂-bipyridinium-CH₂CH₂PO₃)X ₃ · 3H₂O (X = halide) was prepared by low-temperature hydrothermal reaction. The organic groups of this material act as pillars for the inorganic layers, leading to large pores (ca. 8 × 9 A) in the solid. Platinum and palladium colloids can be incorporated into the structure by ion-exchanging the free halide with a solution of K₂MCl₄ (M = Pd or Pt) followed by hydrogen reduction. In these materials, the Pt and Pd colloids act as microelectrodes for the reduction of viologen by hydrogen, which selectively reduces oxygen to hydrogen peroxide. Treatment of aqueous suspensions of these porous catalysts with H₂ and O₂ gases at atmospheric pressure leads to H₂O₂ solutions with concentrations as high as 0.21 M. Reactions carried out at higher pressures lead to significantly higher concentrations of H₂O₂. The yield of H₂O₂ based on H₂ consumed was estimated at 40%.

Full text of DOI:10.1006/jcat.1996.0162

JOURNAL OF CATALYSIS 161, 62–67 (1996)
ARTICLE NO. 0162
Molecular Engineering of Heterogeneous Catalysts: An Efficient Catalyst
for the Production of Hydrogen Peroxide
Kenneth P. Reis, Vijay K. Joshi,1 and Mark E. Thompson2
Received December 11, 1995; revised February 6, 1996; accepted February 12, 1996
organic molecules are used the pore sizes range from 3 to
˚
13 A (2), while the use of micelles or vesicles as templates
A goal of chemists for many years has been to control the
structure of solid-state compounds to induce desired chemical
properties. A porous metal phosphonate compound with chem-
ically active pillaring groups, which allow for the direct pro-
duction of hydrogen peroxide from hydrogen and oxygen, has
been designed. The porous compound, Zr2(PO4)(O3PCH2CH2-
bipyridinium-CH2CH2PO3)X3 ꢀ 3H2O (X = halide) was prepared by
low-temperature hydrothermal reaction. The organic groups of this
material act as pillars for the inorganic layers, leading to large pores
˚
can give materials with pore sizes up to 100 A, or larger (3).
Forallofthesesolids, theporesofthematerialarefilledwith
the template when they are isolated, giving them no avail-
able void space. The inorganic frameworks themselves are
typically quite thermally stable, however, so that the truly
porous solid can be obtained by burning the template out.
Microporous materials have many important applica-
tions. The high charge densities and open structures found
inzeolitesmakethemgoodion-exchangematerialsandhet-
erogeneous catalysts (1, 2, 4, 5). Since zeolites are selective
with respect to pore size they can function as molecular
sieves and shape selective catalysts. The catalytic or ion-
exchange sites in these materials are typically acid sites or
metal ions of the inorganic framework.
A wealth of selective chemistry has been developed for
organic compounds, but the nature of how zeolites are pre-
pared makes it very difficult to combine the selectivity of
organic catalysts with the porous and robust nature of the
inorganic frameworks of zeolites. The organic groups used
in synthesizing zeolites are burned out of the solid to give
the porous material. In this paper, we describe a system
that allows us to construct robust open frameworks com-
posed of organic and inorganic components, capable of
carrying out selective chemical transformations. The syn-
thetic methodology is mild enough to allow active organic
molecules to be incorporated into the walls of these porous
solids. Herein we have focused on a porous solid with
a cationic framework [i.e., Zr2(PO4)(O3PCH2CH2-4,40-
˚
(ca. 8 ꢁ 9 A) in the solid. Platinum and palladium colloids can be in-
corporated into the structure by ion-exchanging the free halide with
a solution of K2MCl4 (M = Pd or Pt) followed by hydrogen reduc-
tion. In these materials, the Pt and Pd colloids act as microelectrodes
for the reduction of viologen by hydrogen, which selectively reduces
oxygen to hydrogen peroxide. Treatment of aqueous suspensions of
these porous catalysts with H2 and O2 gases at atmospheric pressure
leads to H2O2 solutions with concentrations as high as 0.21 M. Re-
actions carried out at higher pressures lead to significantly higher
concentrations of H2O2. The yield of H2O2 based on H2 consumed
c
was estimated at 40%. ꢂ 1996 Academic Press, Inc.
INTRODUCTION
The controlled synthesis of microporous materials has
been an elusive goal for many years. Most microporous
solids (e.g., zeolites) are constructed of silicate tetrahedra
or mixtures of aluminate, silicate, or phosphate tetrahedra
(1). The corner-sharing tetrahedra of these solids can form
a variety of arrangements, leading to a large number of
unique structures (2). Substitution on the framework can
lead to a charged structure, which is balanced by anions or
cations in the pores of the material. Most of these porous
materials are synthesized under hydrothermal conditions
using small organic templates. These organic molecules are
typically charged and act as nucleation sites around which
the inorganic framework forms. This approach leads to ma-
terials whose pore sizes can vary over a wide range. If small
bipyridinium-CH2CH2PO3)X3 ꢀ 3H2O
(
= halide), this
dialkyl-bipyridinium moiety will also be referred to as vi-
ologen]. Ion exchange of the free halide ions for anionic
metal complexes (i.e., MCl24ꢀ, M = Pt, Pd), followed by re-
duction with hydrogen, gives materials with fine metal par-
ticles trapped in the pores of the material. The metal parti-
cles act as catalysts for the direct reduction of viologen by
hydrogen. Reduced viologen is a potent reductant, capable
of reducing both organic and inorganic substrates.
1 Current address: Summit Research Laboratory, Big Pond Road, Box
F, Hugenot, NY 12746.
2 To whom correspondence should be addressed. Fax: (213) 740-0930.
E-mail: Thompson@chem1.usc.edu.
We have examined the porous viologen-containing mate-
rials described above as catalysts for the direct production
62
0021-9517/96 $18.00
c
Copyright ꢁ 1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.

MOLECULAR ENGINEERING OF HETEROGENEOUS CATALYSTS
63
of hydrogen peroxide from hydrogen and oxygen gases.
Presently, hydrogen peroxide is commercially produced by
the AO process (6), developed in the late 1940s and early
1950s (7). In this homogeneous process, the first step in-
volves the hydrogenation of 2-alkyl-9,10-anthraquinone to
the corresponding hydroquinone. Oxidation of the hydro-
quinone with oxygen yields hydrogen peroxide. Hydrogen
peroxide is then extracted with water from the working
solution. The aqueous peroxide solution is always contam-
inated with organic solvent and has to be purified, which
is costly for a homogeneous process. For hydrogen pero-
xide to be used in many industrial applications all traces of
organic solvent must be removed (6). For environmental
reasons it would be beneficial to remove organic solvents
from the process completely. For these reasons an efficient
heterogeneous catalyst for producing hydrogen peroxide
that would function in an all-aqueous system is highly desir-
able. Heterogeneous catalysts for the direct production of
hydrogen peroxide from hydrogen and oxygen have been
reported (8). These catalysts consist of group VIII metal
particlesoninertsupports. Whilethesematerialsshowgood
activity, theyrequirehighpressuresofhydrogenandoxygen
to prepare useful concentrations of hydrogen peroxide. The
system presented here is a heterogeneous catalyst that uses
an organic catalyst (viologen) as the source of reducing po-
tential to generate H2O2 from O2. The catalyst also reduces
hydrogen peroxide to water in the presence of hydrogen,
but at a much slower rate, leading to very efficient hydro-
gen peroxide production. Direct comparison of the catalyst
reported here with a supported metal catalyst shows that
our materials are significantly more active at low pressures.
FIG. 1. Schematic representation of Zr2(PO4)PVX3. Viologen groups
shown in bold are in the foreground.
ing groups between adjacent inorganic layers. We have re-
cently found that cationic pillaring groups can be used to
prepare porous structures. The pillars in this compound
consist of alkyl bisphosphonic acids with dicationic 4,40-
bipyridinium (viologen) groups incorporated into the alkyl
chain, i.e., H2O3PCH2CH2 viologenCH2CH2PO3H2X2 (
=
halide), abbreviated PV. The electrostatic repulsion of ad-
jacent viologen groups prevents them from forming a dense
segregated phase or collapsing (13).
We have recently determined the crystal structure of one
of these porous materials (14). The compound has a for-
mula unit of Zr(PO4)PVX3 ꢀ 3H2O. A schematic drawing
of the structure is shown in Fig. 1, and a ball-and-stick
representation viewed down the axis is shown in Fig. 2.
RESULTS AND DISCUSSION
Metal phosphate and phosphonate compounds have The water molecules and halides that fill the pores of the
been examined by several groups as robust materials for material have been removed in Fig. 2 for clarity. Under
the organization of organic molecules in the solid state (9). our synthetic conditions the halide ions of the material are
Compounds prepared with tetravalent metals are highly principally fluoride, so the compound is referred to as a
insoluble, requiring mineralizing agents for the growth of trifluoride, even though we have not measured the exact
even microcrystalline materials. The most common formula ratio of fluoride to the other halides present. The Zr atoms
for these materials is M(O3P-R)2 (M = tetravalent metal are octahedrally coordinated by two oxygen atoms from
ion, R = OH, H alkyl, aryl), which tend to crystallize in lay- the phosphate groups, three oxygen atoms from the violo-
ered structures. In these materials, the R groups project gen phosphonate groups (in a facial geometry), and an F
into the interlamellar space. Unlike most solid-state reac- atom pointing into the organic layer. The viologen phos-
tions, metal phosphates and phosphonates are made at low phonate molecules bridge the inorganic lamellae and form
temperatures from aqueous solutions (9–11). This allows a criss-cross stack, which minimizes the electrostatic repul-
for well-ordered materials to be synthesized, well below sions of adjacent viologen groups. The closest viologen–
the temperatures that would lead to decomposition of the viologen contact across the pore (along ) is approximately
˚ ˚
8 A, with a layer-to-layer distance of 13.5 A giving rise to
organic group.
If a mixture of phosphoric and phosphonic acids is used fairly large pores. The cavities are occupied by three water
during synthetic preparation, mixed metal phosphonates molecules and one halide ion per formula unit. An alternate
can be obtained (9, 11, 12). Bisphosphonic acids have been way of listing the compound, which emphasizes the differ-
used in mixed phosphonates, i.e., Zr(O3P-OH)2 (O3P-R0- ence between the different types of halide in the structure,
PO3)lꢀ ( = 0–1), and generally act as bridging or pillar- is [(ZrF)2(PO4)PV]+ Fꢀ ꢀ 3H2O.

64
REIS, JOSHI, AND THOMPSON
FIG. 2. Structure of Zr2(PO4)PVF3 viewed along the axis (shown in perspective). The intrapore water molecules and halide atoms have been left
out for clarity.
The double positive charge on the viologen groups in
If a mixture of Pt and Pd salts is used, the Pt : Pd ratio
Zr2(PO4)PVX3 gives the compounds a high positive charge found in the bulk solid is similar to that in the solution (as
density, which is counterbalanced with halide ions (12, 13). shown by ICP analysis of the dissolved solids and solutions).
The halide ions of these porous solids can be exchanged On ion exchange with Pt and Pd salts the materials change
for a number of other anions, by treating the solid with from white to brown-orange. The ion exchange of the metal
an aqueous solution of the desired anion. In addition to species for the halide ions is not complete after a single ex-
other halides or pseudohalides, the chloride ions can be change. Treatment of an exchanged solid with a fresh solu-
exchangedforanionicmetalcomplexes, suchasMX24ꢀ (M = tion of the metal salt results in significantly higher levels of
Pt, Pd; X = Cl, Br), as shown in the equation
metal incorporation. The Zr : (Pt + Pd) ratio was typically
15 : 1 for a sample that had only been ion exchanged once,
compared with 4 : 1 for theoretical complete exchange. Ex-
changing for 3 days with a fresh solution each day leads to
a Zr : (Pt + Pd) ratio of 5 : 1. If these exchanged solids are
(ZrF)2(PO4)PVX + 0.5K2MCl4
→ (ZrF)2(PO4)PV(MCl4)0.5 + KX. [1]

MOLECULAR ENGINEERING OF HETEROGENEOUS CATALYSTS
65
treated with hydrogen gas, the metal salts are reduced to
colloidal metal particles:
TABLE 1
Hydrogen Peroxide Production by Zr2(PO4)PV ꢀ Pt,Pd
Compounds
1
2
(ZrF)2(PO4)PV(MCl4)0.5 + H2
After 1 h
→ (ZrF)2(PO4)PVCl ꢀ M0.5 + HCl. [2]
Pt
(mol%)
[H2O2]
[H2O2]final
(m
(m
)
Turnover
)
These reduced materials are abbreviated Zr2(PO4)PV ꢀ M
(M = Pt or Pd) for materials exchanged with only a single
metal and Zr2(PO4)PV ꢀ Pt,Pd for materials exchanged with
both PtCl24ꢀ and PdCl24ꢀ simultaneously. SEM and TEM
studies of the reduced solids do not show any sign of metal
particles on the surfaces of the Zr2(PO4)PV microcrystals
of Zr2(PO4)PV ꢀ Pt or Zr2(PO4)PV ꢀ Pt,Pd, suggesting that
the Pt and Pd metal is within the pores of the solid.
In addition to providing the necessary interlayer repul-
siontogiveanopenlattice, theviologengroupswerechosen
for their potential catalytic properties. Viologens have been
studied extensively as electron acceptors in photochemical
electron transfer systems (15). The reduced form of violo-
gen is a potent reductant, capable of selectively reducing
organic and inorganic substrates (16). We have found that
the viologen groups of porous solids are readily photore-
duced (12, 17). To extract the photochemically generated
No Pt or Pd
0
22
16
51
10
10
0
0
19
13
42
8
8
0
—
0
0
0.2
7
80
210
144
73
9
44
100
2
29
<5
139
335
—
2
100 ꢂ Pt/(Pt + Pd).
Turnover = 2 ꢂ moles of H2O2/moles of viologen.
Reaction carried out at 15 psi of H2, 100 psi of O2, and 60 psi
of N2 in a thick-walled glass vessel as described in the text.
The reaction vessel was vented and charged with a fresh H2/O2
mixture three times.
reducing equivalents from the system, colloidal platinum oxide produced increases dramatically relative to the ma-
particles were incorporated into the materials (14, 18). The terials with only platinum. Table 1 shows typical results for
metal particles in these materials act as microelectrodes, the catalytic production of hydrogen peroxide from these
with the electrochemical potential determined by the re- materials. In all of these experiments, the rate of hydrogen
duced viologen molecule (19).
peroxide formation is high initially and gradually drops off
If Zr2(PO4)PV ꢀ M (M = Pt or Pd) or Zr2(PO4)PV ꢀ Pt,Pd as a steady-state concentration is reached. At atmospheric
is treated with hydrogen gas the viologen groups of the solid pressure, the steady-state concentration is reached in 6–
are rapidly reduced and the solid takes on a blue or pur- 10 h. A mixture of platinum and palladium is essential for
ple color. Exposure of this colored material to air bleaches the most active catalyst. While Zr2(PO4)PV ꢀ Pd does cat-
the solid immediately. The product of this bleaching reac- alyze the production of H2O2, the best peroxide production
tion is hydrogen peroxide. The rate of oxygen reduction was obtained using a catalyst whose colloidal particles are
by reduced viologen is very high ( = 109 ꢀ1 sꢀ1) (20). If less than 10 mol% Pt. Pt content higher than 15–20% led
streams of hydrogen and oxygen gas are passed through to a significant decrease in hydrogen peroxide concentra-
an aqueous suspension of these porous catalysts at atmo- tion. For comparison to our catalyst, a sample of the silica-
spheric pressure, hydrogen peroxide is observed within the supported Pt/Pd catalyst with an optimal ratio of Pt : Pd as
first few minutes and reaches a steady-state concentration given in Ref. (8) (i.e., 32 wt% Pt) was treated with a H2/O2
within a few hours (Table 1).
mixture at atmospheric pressure, in a manner identical to
If the colloidal metal particles in the material are Pt, Zr2(PO4)PV ꢀ Pt,Pd, and gave a low steady-state concentra-
the steady-state concentration of H2O2 is quite low (<5 tion of H2O2 (0.07 ). The fact that significantly less hydro-
m
). The likely reason for this is that platinum metal is gen peroxide is generated suggests that the peroxide forma-
an excellent catalyst for the reduction of hydrogen perox- tion is going on at the viologen groups and not at the colloid
ide to water. Thus, the metal particles that are leading to particles.
formation of the reduced viologen may be reducing the
After approximately 20 h the catalyst begins to degrade,
peroxide in a destructive side reaction. It has recently been due to oxidation of the colloidal particles. ICP analysis of
reported that mixed Pd/Pt colloidal particles on inert sup- the used catalyst and peroxide solution indicate that the Pd
ports give respectable yields of hydrogen peroxide from partly dissolves, leading to a higher weight percentage of
hydrogen and oxygen gases (8). When an aqueous suspen- Pt in the solid. For example, chemical analysis of a spent
sion of the supported Pd/Pt catalyst is treated with 2000 psi catalyst showed an increase in Pt from 8 to 17 mol%. A
of hydrogen and oxygen, a 12.6% hydrogen peroxide so- sample richer in Pt will give lower steady-state values for
lution is generated. When a sample of Zr2(PO4)PV ꢀ Pt,Pd the hydrogen peroxide since the metal colloid itself will act
is treated with H2 and O2 gases, the level of hydrogen per- to reduce H2O2 to water.

66
REIS, JOSHI, AND THOMPSON
The concentration of O2(g) in a saturated aqueous solu- for the construction of heterogeneous catalysts can be ex-
tion at atmospheric pressure (0.25 m ) (21) can be used as tended to many different organic compounds, which will
an upper limit to the oxygen concentration in our catalytic lead to active solid catalysts for a wide range of different
system. The concentration of hydrogen peroxide produced reactions.
in our system is roughly 1000 times greater than that of the
upper limit for the dissolved oxygen present in the system.
EXPERIMENTAL
This shows that the catalysts are very effective at selectively
Porous zirconium phosphate/
reducing oxygen to hydrogen peroxide. The amount of per-
viologen phosphonate was prepared by the methods in
oxide produced is limited by the concentration of oxygen
Ref. (9). The ion-exchange reactions were accomplished
gas in solution. If the flow rate of oxygen is reduced, the
by preparing a solution consisting of K2PtCl4 and Na2PdCl4
steady-state concentration of H2O2 decreases. If the oxygen
in 100 ml of water [e.g., 0.015 g of K2PtCl4 and 0.217 g
flow to the system is interrupted completely, the concentra-
tion of H2O2 quickly drops to zero, as the catalysts are also
Na2PdCl4, corresponding to a Pt content (Pt/(Pt + Pd)) of
4.6%] and heating a suspension of 100 mg of the zirconium
compound in this solution at 60ꢃC for 12 h. The solution
was protected from exposure to light, to limit the degra-
dation of the salts to metal colloids in solution. Repeating
the exchange with a fresh solution for an additional 12 h in-
creases the Pt : Pd content of the sample. The solids change
from white to orange on exchange. The Pt/Pd mole per-
centage in the isolated material varied from run to run and
was determined for each solid sample by ICP analysis of
the dissolved solids; these numbers are listed in Table 1 for
several catalysts. While the mole fraction in the solid varied
from run to run, the obtained material is typically close to
the mole percentage of the reacting solution.
The reduction to metal colloids was achieved by bubbling
H2 gas into a stirred suspension of the material for 1 h at
60ꢃC. The suspension had a deep dark blue color caused by
the reduction of viologen. Exposure to air leads to a gray
solid.
respectablecatalystsforthereductionofH2O2 towater. The
selectivity seen in formation of hydrogen peroxide is most
likely due to the significant difference in the rates of reduc-
tion of oxygen and hydrogen peroxide by reduced viologen.
In addition to the experiments carried out at atmospheric
pressure, the production of hydrogen peroxide was exam-
ined at pressures up to 175 psi. In these experiments, a slurry
of the catalyst was loaded into a thick-walled glass tube
and the tube charged with the H2/O2 gas mixture. After
20–24 h the vessel was vented and the gas above the solu-
tion examined by gas chromatography. During the course
of the reaction all of the hydrogen is consumed. From the
volume of the glass vessel, the initial pressure of hydrogen,
and the final concentration of H2O2, a 40% yield of H2O2
(based on hydrogen) was determined. In this closed system
the oxygen pressure was usually 100 psi, leading to higher
concentrations of dissolved oxygen than in our previous ex-
periments at atmospheric pressure. The increase in oxygen
concentration leads to a large increase in the concentration
of hydrogen peroxide produced, as expected.
In a typi-
cal experiment, 25 mg of catalyst and 10 ml of 0.1 HCl
were added to a 50-ml canonical flask. The system was
sealed with a rubber septum and hydrogen and oxygen
gas was bubbled through the aqueous suspension. The oxy-
gen stream was saturated with water vapor which slowed
CONCLUSION
The design of efficient new heterogeneous catalysts re- the solvent loss inside the reaction vessel. The perox-
quires the control of both the structure and chemical reac- ide concentration was measured by removing a known
tivity of the new material. By using porous metal phospho- amount of solution, typically 0.200 ml, and diluting to 5
nates we are able to control both of these properties of the ml with a TiOSO4 ꢀ H2SO4 ꢀ H2O/sulfuric acid solution.
new material. The catalyst described herein is active toward This reagent reacts with hydrogen peroxide quantitatively
reductionbyhydrogengasandsubsequentelectrontransfer to form an orange complex whose absorbance was mea-
to oxygen to give hydrogen peroxide. At atmospheric pres- sured at 410 nm. The hydrogen peroxide concentration was
sure, between 0.15 and 0.2 H2O2 was produced, whereas back-calculated from a Beer’s law plot which was obtained
0.3–0.4
H2O2 was produced in the higher-pressure ex- by a redox titration of standardized KMnO4 with H2O2 so-
periment. Increasing the pressure to 2000 psi, as is typically lutions.
done for heterogeneous catalysts (8), should increase the
concentration of H2O2 dramatically.
In our experimental setup, control of the hydrogen and
oxygen flow rates is very important in the production
We are currently examining the use of these porous re- of hydrogen peroxide. When the oxygen flow rate is 75
duced viologen materials for the reduction of organic sub- ml/min and the hydrogen flow rate is varied from 5 to 75
strates. By changing the size of the bisphosphonic acid pil- ml/min, the best peroxide concentration is obtained when
lars in these porous materials it may be possible to control the H2 and O2 rates are both at 75 ml/min. Higher flow
the pore sizes on these materials, leading to shape or size rates increased peroxide formation initially but the same
selective reduction catalysts. The approach described here steady state was reached.

MOLECULAR ENGINEERING OF HETEROGENEOUS CATALYSTS
67
ACKNOWLEDGMENT
32, 1357 (1993). Clearfield, A.,
10,, 89 (1990). Vermeulen, L. A., and Thompson, M. E.,
The authors thank The American Biomimetics Corporation for their
financial support of this work.
358, 656 (1992). Clearfield, A.,
10. Stein, A., Keller, S. W., and Mallouk, T. E.,
88, 125 (1988).
259, 1558 (1993).
11. Burwell, D. A., and Thompson, M. E., “ACS Symposium Series.”
(Bein, Thomas, Eds.), No. 499, Chap. 13. Washington,
DC, 1992. Putvinski, T. M., Schilling, M. L., Katz, H. E., Chidsey,
REFERENCES
C. E. D., Mujsce, A. M., and Emerson, A. B.,
Alberti, G., Costantino, U., and Perego, G. J.,
63, 455, 1986. Dines, M. B., and Griffith, P. C.,
6, 1567 (1990).
1. Barrer, R. M., “Hydrothermal Chemistry of Zeolites.” Academic
Press, New York, 1982; Suib, S. L.,
93, 603 (1993); Smith, J.
22, 567
V., 88, 149 (1988); Ocelli, M. L., and Robson, H. E. (Eds.),
(1983). Alberti, G., Costantino, U., Vivani, R., and Zappelli, P.,
233, 95 (1991). Wang, J. D., and Clearfield, A.,
35, 208 (1993). Clearfield, A., “Design of New
Materials” (A. Clearifield and D. A. Cocke, Eds.), p. 121. Plenum,
New York, 1986.
“Zeolite Synthesis,” ACS Symposium Series 398. Am. Chem. Soc.,
Washington, DC, 1989; Mortier, W. J., and Schoonheydt, R. A.,
16, 1 (1985).
2. (a) Breck, D. W., “Zeolite Molecular Sieves.” Kreiger, Malabar, FL,
1974. (b) Meier, W. M., and Olson, D. H., “Atlas of Zeolite Structure 12. Vermuleun, L. A., and Thompson, M. E.,
6, 77 (1994).
13. Poojary, D. M., Vermeulen, L. A., Vicenzi, E., Clearfield, A., and
Thompson, M. E., 6, 1845 (1994).
14. Byrd, H., Reis, K. P., Clearfield, A., Poojary, D., and Thompson, M. E.,
submitted for publication.
Types.” Butterworths, London, 1987.
3. Beck, J. S., Vartuli, J. C., and Roth, W. J.,
(1992).
4. Bond, G. C., “Heterogeneous Catalyses: Principles and Applications.”
Oxford University Press, New York, 1987.
114, 10834
15. Meyer, T. J.,
22, 163 (1989). Harriman, A., and
5. Lago, R. M., U.S. Patent 4,025,572, Mobil (1977), CA 87 67838u.
Flanigen, E. M., Bennett, J. W., Grose, R. W., Cohen, J. D., Patton,
West, M. A. (Eds.), “Photogeneration of Hydrogen.” Academic Press,
New York, 1982. Fox, M. A., and Chanon, M. (Eds.), “Photoin-
duced Electron Transfer.” Elsevier Press, Amsterdam, 1988, and refer-
ences therein. Ramamurthy, V. (Ed.), “Photochemistry in Constrained
Media.” VCH, New York, 1991, and references therein.
R. L., Krichner, R. M., and Smith, J. V.,
271, 512 (1978).
6. Goor, G., “Catalytic Oxidations with H2O2 as Oxidant” (G. Strukel,
Ed.), Chap. 2. Kluwer Academic, Norwell, MA, 1992.
7. Pfleiderer, G., and Riedl, H. J., U.S. Patent 2,369,912 (1945), CA 4729. 16. Evans, A. G., Dodson, N. K., and Rees, N. H.,
E.I. du Pont De Nemours & Co., British Patent 686,567 (1953), CA
8392. Sprauer, J. W., U.S. Patent 2,673,140 (1954), CA 10309.
8. Gosser, L. W., and Schwartz, J. T., U.S. Patent 4,772, 458 (1988), CA 109
859 (1976).
17. Vermeulen, L. A., Snover, J., Sapochak, L., and Thompson, M. E.,
115, 11767 (1993).
193175m. Gosser, L. W., U.S. Patent 4,889,705 (1989), CA 112 39260r. 18. Snover, J. L., and Thompson, M. E.,
Gosser, L. W., and Paolli, M. A., U.S. Patent 5,135,731 (1992), CA 117 19. Kiwi, J., and Gra¨tzel, M.,
116, 765 (1994).
101, 7214 (1979). Miller,
215646k.
D. S., Bard, A. J., McLendon, G., and Ferguson, J.,
103, 5336 (1981).
9. Alberti, G., and Costantino, U., “Intercalation Chemistry” (M. S.
Whittingham and A. J. Jacobson, Eds.), Chap. 5. Academic Press, San 20. Farrington, J. A., Ebert, M., Land, E. J., and Fletcher, K.,
Diego, 1982. Cao, G., Hong, H.-G., and Mallouk, T. E., 372, 314 (1973).
25, 420 (1992). Thompson, M. E., 6, 1168 (1994). Alberti, 21. Manaham, S. E., “Environmental Chemistry.” 6th ed., p. 50. CRC
G., Costantino, U., Marmottini, F., Vivani, R., and Zappelli, P., Angew. Press, Boca Raton, FL, 1994.