Oxygenation of hydrocarbons mediated by mixed-valent basic iron trifluoroacetate and valence-separated component species under Gif-type conditions involves carbon- and oxygen-centered radicals

Article abstract of DOI:10.1002/1521-3773(20000703)39:13<2343::AID-ANIE2343>3.0.CO;2-T

Hydrogen-atom abstraction by hydroxyl radicals takes place to generate both tert- and sec-adamantyl radicals in Gif-type oxygenation of adamantane by H₂O₂ in pyridine/trifluoroacetic acid when the reaction is mediated by [Fe(O₂CCF₃)₂(py)₄] or [Fe₂O(O₂CCF₃)₄(py)₆], which are formed by dissociation of [Fe₃O(O₂CCF₃)₆- (L)₃] in pyridine (L = H₂O, DMSO; see scheme).

Full text of DOI:10.1002/1521-3773(20000703)39:13<2343::AID-ANIE2343>3.0.CO;2-T

COMMUNICATIONS
[8] P. OꢁBrien, Angew. Chem. 1999, 111, 339; Angew. Chem. Int. Ed. 1999,
38, 326.
[9] A. E. Rubin, K. B. Sharpless, Angew. Chem. 1997, 109, 2751; Angew.
Chem. Int. Ed. Engl. 1997, 36, 2637.
[10] A. M. Baranger, P. J. Walsh, R. G. Bergman, J. Am. Chem. Soc. 1993,
115, 2753.
[11] S. W. Krska, R. L. Zuckerman, R. G. Bergman, J. Am. Chem. Soc.
1998, 120, 11828, and references therein.
Oxygenation of Hydrocarbons Mediated by
Mixed-Valent Basic Iron Trifluoroacetate and
Valence-Separated Component Species under
Gif-Type Conditions Involves Carbon- and
Oxygen-Centered Radicals**
Amy E. Tapper, Jeffrey R. Long, Richard J. Staples,
and Pericles Stavropoulos*
[12] P. J. Walsh, F. J. Hollander, R. G. Bergman, Organometallics 1993, 12,
3705.
[13] F. R. W. P. Wild, M. Wasiucionek, G. Huttner, H. H. Brintzinger, J.
Organomet. Chem. 1985, 288, 63.
A remarkable series of iron-based systems for oxidizing
hydrocarbonsÐsuch as the century-old Fenton reagent,[1] the
biologically relevant Udenfriend system,[2] and the more
recently developed Gif systems[3]Ðhave received detailed
attention, but the nature of the active oxidants involved (free
[14] Enantioresolved complexes 1 and 5 were confirmed to be enantiopure
by NMR analysis of the products formed by treating the zirconium
complexes withenantiopure ( R)-mandelic acid, as described by earlier
workers: a) A. Schafer, L. Zsolnai, G. Huttner, H. H. Brintzinger, J.
Organomet. Chem. 1987, 328, 87; b) R. B. Grossman, W. M. Davis,
S. L. Buchwald, J. Am. Chem. Soc. 1991, 113, 2321; c) B. Chin, S. L.
Buchwald, J. Org. Chem. 1997, 62, 2267.
[15] The reaction was stopped 10 min after addition of the allene.
[16] Enantioenrichments of the allenes were determined by chiral vapor-
phase chromatography, in which the resolution is calibrated using
racemic mixtures.
[17] Y. Noguchi, H. Takiyama, T. Katsuki, Synlett 1998, 543.
[18] The configuration of cyclononadiene was assigned by comparison with
material prepared in partially enantioenriched form by resolution with
[(À)-(isopinylcampheyl)2BH]2. Resolution using this reagent is
known to provide cyclononadiene enriched in the S enantiomer.
See: a) R. D. Bach, J. W. Holubka, C. L. Willis, J. Am. Chem. Soc.
1982, 104, 3980; b) W. R. Moore, H. W. Anderson, S. D. Clark, J. Am.
Chem. Soc. 1973, 95, 835.
.
.
II/III
HO /RO radicals or metal-bound FeIV/V O/Fe
OO(H)
ˆ
À
units) and their mode of action (radical or concerted) are
topics of current debate.[4, 5] Recent advances towards eluci-
ˆ
dating the functional behavior of high-valent Fe O units,
presumed to operate in biological monooxygenases (P-450,[6]
sMMO[7]), have cast suspicion as to whether similar metal-
centered oxidants participate in oxygenated Fenton,[4, 8] Gif,[9]
and other allegedly biomimetic systems.[10] There is now
consensus[11] that at least tBuOOH-dependent versions of
.
.
these systems involve tBuO /tBuOO and substrate-centered
.
.
radicals (RO /ROO ). The recognition that tBuOOH-sup-
ported shunt pathways of P-450-type mimics[12] frequently
[19] A. H. Hoveyda, J. P. Morken, Angew. Chem. 1996, 108, 1378; Angew.
Chem. Int. Ed. Engl. 1996, 35, 1263.
[20] E. L. Eliel, S. H. Wilen, L. N. Mander, Stereochemistry of Organic
Compounds, John Wiley and Sons, Inc., New York, 1994, p. 837.
[21] B. R. Bender, D. L. Ramage, J. R. Norton, D. C. Wiser, A. K. Rappe, J.
Am. Chem. Soc. 1997, 119, 5628.
.
.
generate tBuO /tBuOO radicals limits the usefulness of these
systems in probing mechanistic distinctions. Evidence to
support a radical mechanism[13] for mainstream H2O2- or O2/
Zn-dependent Gif-type systems is currently resting on in-
sufficient experimental basis.[14] Reported in the present study
is a persuasive case of a typical Gif reagent which performs
[22] a) T. H. Upton, A. K. Rappe, J. Am. Chem. Soc. 1985, 107, 1206; b) E.
Folga, T. Ziegler, Organometallics 1993, 12, 325.
[23] We postulate a diradical structure for the intermediate because the
thermal dimerization of allenes is known to proceed via this type of
species. For leading references, see: a) D. J. Pasto, Tetrahedron, 1984,
40, 2805; b) The Chemistry of the Allenes (Ed.: S. R. Landor),
Academic Press, London, 1982. However, withour presently available
data, we cannot distinguishbetween a diradical and the corresponding
zwitterion (presumably witha positive charge on Zr and a negative
charge on the allyl moiety) or a p-allyl complex in which the C3
fragment in intermediate C is coordinated to the Zr center. Evidence
has been proposed for diradical intermediates in several organo-
metallic reactions; for examples, see: c) S. L. Buchwald, E. V. Anslyn,
R. H. Grubbs, J. Am. Chem. Soc. 1985, 107, 1766; d) L. Cavallo; H.
Jacobsen, Angew. Chem. 2000, 112, 602; Angew. Chem. Int. Ed. 2000,
39, 589, and references therein.
oxidation of substrates withH O2 in pyridine/trifluoroacetic
acid (py/TFA) by radical pathways.
2
The reaction of [Fe3O(O2CCH3)6(H2O)3] withexcess TFA
is known[15] to yield [Fe3O(O2CCF3)6(H2O)3] ´ 3.5H2O. In our
hands, samples prepared in TFA/H2O (4/1 v/v) afford red
crystals of [Fe3O(O2CCF3)6(H2O)3] ´ 2.5H2O ´ CF3COOH (1,
see Scheme 1). The structure of 1 at 133 K (see the Supporting
Information) indicates a valence-trapped state within the
II
triangular Fe3O core (av FeIII O 1.864(8), Fe O 2.034(3) Š).
À
À
In dimethyl sulfoxide (DMSO),
1
affords red
[Fe3O(O2CCF3)6(DMSO)3] (2), whose structure at 213 K
(see the Supporting Information) reveals partial valence
trapping, as there is only a 0.065 Š difference between the
[24] The assumption of initial p-complex formation is convenient but not
required to explain the enantioselective behavior of this system; the
initial interaction between imido complex 1 and the allenes could lead
directly to the transition states A= and B=.
À
longer and shorter Fe O distances.
[25] There is, of course, a higher energy path leading from intermediate C
to the more sterically congested isomer of the metallacycle. Because
we do not observe detectable amounts of this product, we have
omitted this pathway and product from the diagram for the sake of
clarity.
[26] One can always postulate more complicated mechanisms that will also
account for a particular set of experimental results. Therefore, as
noted by two referees, we cannot rule out alternative scenarios in
which the faster reacting allene enantiomer reacts by a concerted
mechanism and the slower reacting allene reacts by a stepwise process
proceeding through intermediate C or through a less stable diastereo-
isomer of (S,S,R)-10. Experiments are under way aimed at distinguish-
ing these possibilities.
[*] Prof. P. Stavropoulos, A. E. Tapper
Department of Chemistry, Boston University
590 CommonwealthAvenue, Boston, MA 02215 (USA)
Fax : (‡1)617-353-6466
E-mail: stavro@chem.bu.edu.
Prof. J. R. Long
University of California, Berkeley, CA (USA)
Dr. R. J. Staples
Harvard University, Cambridge, MA (USA)
[**] This work was supported by the U.S. Environmental Protection
Agency and the NIH/NIEHS (superfund).
[27] H. B. Kagan, J. C. Fiaud, Top. Stereochem. 1988, 18, 249.
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
Angew. Chem. Int. Ed. 2000, 39, No. 13
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3913-2343 $ 17.50+.50/0
2343

COMMUNICATIONS
Surprisingly, solutions of 1 or 2 in pyridine afford green
[FeII(O2CCF3)2(py)4] (3) and red [FeI2IIO(O2CCF3)4(py)6] ´ 2py
(4, Scheme 1). Apparently, the stronger N-donor moiety
Fe environment withan imposed C2 axis bisecting the
symmetry-related N(1)-Fe(1)-N(1A) and N(2)-Fe(1)-N(2A)
angles. Compound 4, prepared independently from [Et4N]2-
[Fe2OCl6] and CF3CO2Na, features a nearly linear m-oxo
bridge, linking two ferric sites that differ slightly in their
metrical parameters (Figure 2).[16]
F3C
F3C
L
O
F3C
O
O
O
O
CF3
CF3
O
O
F3C
F3C
Fe
O
O
O
N
N
O
py
N
N
N
Fe
O
Fe
O
Fe
Fe
+
O
O
O
N
N
O
Fe
N
N
O
L
L
O
O
O
N
O
O
O
O
O
F3C
F3C
F3C
CF3
F3C
3
4
1, L = H2O
2, L = DMSO
Scheme 1. Dissociation of [Fe3O(O2CCF3)6(L)3] (L ˆ H2O, DMSO) in
pyridine.
II
À
weakens the trans-oriented Fe O ligation to cause complete
dissociation of the parent Fe3O core structure. Compound 3 is
also obtained from a solution of [Fe(O2CCF3)2]n in pyridine.
The structure of 3 (Figure 1)[16] reveals a distorted octahedral
Figure 2. The structure of 4. Selected bond lengths [Š] and angles [8]:
À
À
À
Fe(1) O(1) 1.7878(13), Fe(2) O(1) 1.7854(14), Fe(1) N(2) 2.304(2),
À
À
À
Fe(1) N(3) 2.160(2), Fe(1) O(2) 2.0341(15), Fe(2) N(5) 2.310(2),
À
À
Fe(2) N(4) 2.195(2), Fe(2) O(6) 2.0415(15); Fe(1)-O(1)-Fe(2) 169.48(10).
Table 1 shows profiles of products derived from oxidations
of the benchmark substrate adamantane (5 mmol) by the
system 3 (or 4)/H2O2 (0.2/2.0 mmol) in py/TFA (30.0/3.0 mL)
under a stream of Ar, O2 (4%) in N2, or pure O2. Similar
results (not shown) are obtained with 1, apparently because 1
dissociates to 3 and 4 in py/TFA. In addition to the expected
oxo products, 2- and 4-adamantylpyridines are obtained not
only for the tert-adamantyl positions (as previously recog-
nized),[3] but also for the sec-adamantyl sites, especially under
Ar. The presence of these coupled products provides direct
evidence for the generation of tert- and sec-adamantyl
radicals.[17] The reported absence of sec-alkylpyridines in the
product profile of Gif oxygenations had led Barton and
Doller[3] to propose that at least the activation of sec C H
bonds is brought about by nonradical pathways. Under O2,
the ratio of products derived due to competition between O2
À
Figure 1. The structure of 3. Selected bond lengths [Š] and angles [8]:
À
À
À
Fe(1) O(1) 2.069(2), Fe(1) N(1) 2.220(3), Fe(1) N(2) 2.205(3); O(1)-
Fe(1)-O(1A) 169.90(13), N(1)-Fe(1)-N(1A) 88.95(14).
Table 1. Product profiles for the oxidation[a] of adamantane by H2O2 mediated by 3 or 4 and via authentic adamantyl radicals.[a,b]
System
Substrate
Products [mmol]
Ratio[c]
OH
OH
O
N
N
N
N
3/Ar
3/O2 (4%)
3/O2
0.001
0.003
0.034
0.001
0.003
0.025
0.002
nd[d]
0.004
0.124
0.120
0.128
0.100
0.088
0.091
0.029
0.085
0.077
0.078
0.065
0.059
0.056
0.032
0.132
0.124
2.4
3.5
4.2
3.3
3.5
4.5
0.016
0.027
0.002
0.011
0.015
trace
0.117
0.143
0.036
0.095
0.098
0.027
0.018
0.001
0.057
0.010
0.001
0.002
0.018
0.001
0.057
0.011
nd[d]
4/Ar
4/O2 (4%)
4/O2
3/O2 (4%)[b]
0.003
[a] See text for conditions. [b] By photolysis of the PTOC esters of Barton et al.[19] [c] Ratio of products (tertiary/secondary) obtained via tertiary and
secondary adamantyl radicals. [d] nd ˆ not detected.
2344
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3913-2344 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 13

COMMUNICATIONS
‡
and [pyNH] in trapping adamantyl radicals shifts profoundly
attack on DMSO/EtOH in aqueous pulse radiolysis experi-
ments. The reaction in Equation (3), which is known[21] to
proceed at near diffusion controlled rates, may limit the
preciseness of the assessment, further assisted by the rever-
sibility of the addition reaction of a-hydroxyethyl radicals to
in favor of oxo species at the secondary position and to a much
lesser extent at the tertiary site. Minisci et al.[18] have traced
this behavior to the superior rate constant (by two orders of
magnitude) and inferior reversibility for the addition reaction
of tert-adamantyl versus sec-adamantyl radicals to protonated
pyridine.
Generation of tert- and sec-adamantyl radicals in the
presence of 3 by photolysis of the appropriate PTOC esters
of Barton et al.[19] (Scheme 2) in py/TFA under O2 (4%)
provides ratios of oxo- versus pyridine-trapped adamantyl
‡
[pyNH] .[17]
.
‡
FeIII ‡ CH3 CHOH À! FeII ‡ H ‡ CH3CHO
(3)
However, the total amount of iron is kept at low levels with
respect to py/TFA. Most importantly, it is found that FeIII sites
are destabilized by the electron-withdrawing TFA (or pico-
linate[22]), and are rapidly reduced to FeII in the presence of
stoichiometric amounts of H2O2, probably in conjunction with
H2O2 dismutation. This further argues in support of a central
role for the FeII/H2O2 combination in generating the active
oxidant. In a reinterpretation of the Gif mechanism, Barton
et al.[23] had accepted that the FeII/H2O2 ªmanifoldº (as
opposed to FeIII/H2O2) produces substrate-based alkyl radi-
Scheme 2. Generation of authentic tert- and sec-adamantyl radicals.
cals, but maintained that the active oxidant is FeIV O.
The present results provide compelling evidence that HO
ˆ
.
products (Table 1) which for both the tertiary (0.03) and
secondary positions (5.4) are comparable to those obtained by
the analogous Gif experiment (tert 0.02, sec 3.7). Therefore
the product profiles of adamantane oxidation are entirely
dictated by the generation of tert- and sec-adamantyl radicals.
The normalized tertiary/secondary selectivities suggest that
a fairly indiscriminate oxidant is involved under Ar, coupled
to a more selective oxidant in the presence of O2. The addition
is the key hydrogen-abstracting oxidant under Ar, coupled to
a more selective oxidant (most likely substrate-centered
.
.
.
alkoxyl radicals: R !ROO !RO ) under increasing partial
pressures of dioxygen. In conclusion, the findings of this
report lend further support to the proposition[5] of a prepon-
derant, carbon- and oxygen-centered radical pathway for
mainstream Gif systems.
.
reaction of HO to DMSO [Eq. (1)] and the competitive
hydrogen abstraction from EtOH [Eq. (2)] have been used[20]
Experimental Section
.
.
ˆ
ˆ
A typical oxidation of adamantane was conducted as follows: The iron
reagent (0.20 mmol) was dissolved under anaerobic conditions in pyridine
(30.0 mL) and TFA (3.0 mL) followed by addition of adamantane (681 mg,
5.0 mmol). Degassed H2O2 (aq. 30%, 0.28 mL, 2.0 mmol) was added slowly
(6 h) using a syringe pump under a flow of specified gas. At the end of the
reaction, oxalic acid (5 equiv per Fe) and PPh3 (2 equiv per H2O2) were
added followed by the internal standard (1,3,5-triisopropylbenzene). An
aliquot (2 mL) was withdrawn for ether extraction and GC (SPB-1 column)
or GC/MS analysis.
Me2S O ‡ HO À! MeS( O)OH ‡ Me
(1)
(2)
(92%)
.
.
.
CH3CH2OH ‡ HO À! CH2CH2OH ‡ CH3 CHOH
(13.2%) (84.3%)
.
to investigate the possible involvement of HO , by monitoring
the formation of pyridine-trapped alkyl radicals produced in
these reactions under a constant stream of Ar. Table 2 reveals
that the reagent 3/H2O2 oxidizes DMSO/EtOH (5 mmol/
3 ± 10 mmol) in py/TFA (30.0/3.0 mL) as predicted by Equa-
tions (1) and (2).
Furthermore, the average kEtOH/kDMSO value of 0.32(4),
roughly evaluated from the ratio of methylpyridines over
hydroxyethylpyridines and the initial concentrations of
DMSO and EtOH, is consistent withteh ratio of rate
constants (kEtOH/kDMSO ˆ 0.29) reported[20] by virtue of HO
Received: February 16, 2000 [Z14721]
[1] C. Walling, Acc. Chem. Res. 1975, 8, 125 ± 131.
[2] S. Udenfriend, C. T. Clark, J. Axelrod, B. B. Brodie, J. Biol. Chem.
1954, 208, 731 ± 739.
[3] D. H. R. Barton, D. Doller, Acc. Chem. Res. 1992, 25, 504 ± 512.
[4] a) P. A. MacFaul, D. D. M. Wayner, K. U. Ingold, Acc. Chem. Res.
1998, 31, 159 ± 162; b) C. Walling, Acc. Chem. Res. 1998, 31, 155 ± 157;
c) S. Goldstein, D. Meyerstein, Acc. Chem. Res. 1999, 32, 547 ± 550.
[5] M. J. Perkins, Chem. Soc. Rev. 1996, 229 ± 236.
.
Table 2. Product profiles of the oxidation of DMSO/EtOH by H2O2 mediated by 3 in py (30.0 mL)/TFA (3.0 mL) under Ar.
Products [mmol]
kEtOH/kDMSO
CH3
CH3
N
OH
OH
N
N
N
N
CH3
OH
N
OH
N
1[a]
2[b]
3[c]
4[d]
0.198
0.154
0.179
0.118
0.026
0.016
0.020
0.016
0.099
0.078
0.090
0.059
0.026
0.092
0.080
0.011
0.038
0.029
0.003
0.013
0.009
0.002
0.005
0.004
0.29
0.36
0.32
[a] DMSO (5 mmol). [b] DMSO (5 mmol)/EtOH (3 mmol). [c] DMSO (5 mmol)/EtOH (7 mmol). [d] DMSO (5 mmol)/EtOH (10 mmol).
Angew. Chem. Int. Ed. 2000, 39, No. 13 ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000 1433-7851/00/3913-2345 $ 17.50+.50/0
2345

COMMUNICATIONS
synthesis or by postchemical treatment.[2] Considering the
channel size, zeolites are classified as ultralarge (>12
membered rings (MR)), large (12-MR), medium (10-MR)
or small (8-MR) pore materials depending on the number of
T atoms that limits the pore aperture of their largest channels
(T represents atoms from the aluminum and silicon families).
While zeolites with small pores have found some specific
applications in, for instance, the conversion of methanol to
olefins[3], the most successful zeolitic catalysts are those based
on zeolites withmedium and large pores. More specifically,
large-pore zeolites have unique properties for dealing with
many of the oil fractions involved in refinery processes
(cracking, hydrocracking, hydroisomerization, among others),
in petrochemistry (including benzene alkylation with olefins,
isomerization and disproportionation of alkylaromatic spe-
cies), and in fine chemical production (such as alkylation,
acylation, isomerization, and esterification).
It must be noted, for many of the processes named above, a
rapid diffusion of the reactants and products is desired and
this is better achieved with large-pore tridirectional zeolites.
Until recently, however, only two large-pore tridirectional
zeolites were synthesized, faujasite and Beta, and of these
only the Beta zeolite can be directly synthesized with a high
Si:Al ratio and therefore does not need, unlike the faujasites,
a postsynthesis dealumination. Therefore, owing to the large
catalytic interest and very limited number of large-pore
tridirectional zeolites, a considerable effort has been devoted
in the last decade to produce such structures[4]. Very
recently,[5] the pure silica form of a new large-pore tridirec-
tional zeolite has been presented, named ITQ-7 (Instituto de
Tecnología Química-7). Unfortunately, the authors were
unable to introduce acidity into ITQ-7 by direct synthesis
with the trivalent (Al and Ga family) atoms in an isomorph-
ically substituted zeolite[6]. Therefore, this large-pore tridirec-
tional zeolite have had no possibilities in catalysis since only
the purely siliceous form was available. Herein, we present
the possibility to synthesize ITQ-7 with different TIII and TIV
elements isomorphically incorporated into the framework
and, in this way, acidic, catalytically active ITQ-7 materials
have been prepared.
[6] M. Sono, M. P. Roach, E. D. Coulter, J. H. Dawson, Chem. Rev. 1996,
96, 2841 ± 2887.
[7] B. J. Wallar, J. D. Lipscomb, Chem. Rev. 1996, 96, 2625 ± 2657.
[8] D. T. Sawyer, A. Sobkowiak, T. Matsushita, Acc. Chem. Res. 1996, 29,
409 ± 416.
[9] M. Newcomb, P. A. Simakov, S.-U. Park, Tetrahedron Lett. 1996, 37,
819 ± 822.
[10] P. A. MacFaul, K. U. Ingold, D. D. M. Wayner, L. Que Jr., J. Am.
Chem. Soc. 1997, 119, 10594 ± 10598.
[11] a) F. Minisci, F. Fontana, S. Araneo, F. Recupero, L. Zhao, Synlett
1996, 119 ± 125; b) D. H. R. Barton, Synlett 1997, 229 ± 230.
[12] B. Meunier, Chem. Rev. 1992, 92, 1411 ± 1456.
[13] M. J. Perkins, Chem. Soc. Rev. 1996, 229 ± 236.
[14] M. Newcomb, P. A. Simakov, S.-U. Park, Tetrahedron Lett. 1996, 37,
819 ± 822.
[15] V. I. Ponomarev, O. S. Filipenko, L. O. Atovmyan, S. A. Bobkova,
Á
K. I. Turte, Sov. Phys. Dokl. 1982, 27, 6 ± 9.
[16] Crystal data for 3 (213 K) withMo
radiation (l ˆ 0.71073 Š):
Ka
orthorhombic, space group Pccn, a ˆ 16.698(5), b ˆ 9.074(2), c ˆ
16.587(4) Š, Vˆ 2513(1) Š3, Z ˆ 4, R1 ˆ 0.0478 for 2156 data with
I > 2s(I), GOF (on F 2) ˆ 1.108. For 4 (213 K): orthorhombic, space
group P212121, a ˆ 13.3830(1), b ˆ 16.4843(2), c ˆ 24.1315(2) Š, Vˆ
5323.60(8) Š3, Z ˆ 4, R1 ˆ 0.0327 for 9486 data with I > 2s(I), GOF
(on F 2) ˆ 1.055. Further details on the crystal structure investigations
may be obtained from the Fachinformationszentrum Karlsruhe, 76344
Eggenstein-Leopoldshafen, Germany (fax: (‡49)7247-808-666;
e-mail: crysdata@fiz-karlsruhe.de), on quoting the depository num-
bers CSD-411108 (3) and -411109 (4).
[17] F. Minisci, E. Vismara, F. Fontana, Heterocycles 1989, 28, 489 ± 519.
[18] F. Recupero, A. Bravo, H.-R. Bjùrsvik, F. Fontana, F. Minisci, M.
Piredda, J. Chem. Soc. Perkin Trans. 2 1997, 2399 ± 2405.
[19] D. H. R. Barton, F. Halley, N. Ozbalik, M. Schmitt, E. Young, G.
Balavoine, J. Am. Chem. Soc. 1989, 111, 7144 ± 7149.
[20] G. V. Buxton, C. L. Greenstock, W. P. Helman, A. B. Ross, J. Phys.
Chem. Ref. Data 1988, 17, 513 ± 886.
[21] F. Minisci, A. Citterio, E. Vismara, Tetrahedron 1985, 41, 4157 ± 4170.
[22] S. Kiani, A. Tapper, R. J. Staples, P. Stavropoulos, J. Am. Chem. Soc.,
submitted.
[23] D. H. R. Barton, B. Hu, D. K. Taylor, R. U. Rojas Wahl, J. Chem. Soc.
Perkin Trans. 2 1996, 1031 ± 1041.
Synthesis, Characterization, and Catalytic
Activity of a Large-Pore Tridirectional Zeolite,
H-ITQ-7**
The synthesis of isomorphically substituted zeolites was
attempted following two strategies. The first strategy consists
of synthesizing a boron-containing ITQ-7 (B-ITQ-7) sample
which already should present some weak acidity and then, in a
further step, to exchange B with Al to yield materials, named
B/Al-ITQ-7, with a much greater acidity than the B-ITQ-7
precursor. The second strategy involves the direct synthesis of
Al-ITQ-7. The two synthesis routes will be described below.
B-ITQ-7: Boron-containing ITQ-7 was formed from a gel
withthe composition SiO 2:B2O3:C14H26NOH:HF:H2O in a
molar ratio 1.0:0.01:0.50:0.50:3.0, where C14H26NOH is 1,3,3-
trimethyl-6-azonium-tricyclo[3.2.1.46,6]dodecane hydroxide.
The gel was prepared by dissolving H3BO3 (0.08 g) in a
solution of C14H26NOH (0.99m, 31.98 g). Tetraethylorthosili-
cate (TEOS, 13.46 g) was then hydrolyzed in the solution and
the mixture was stirred gently to completely evaporated the
ethanol formed. Finally, HF (1.34 g as 48.1% in water) and
purely siliceous ITQ-7 crystals (0.20 g) were added and the
mixture was homogeneized. After 7 days crystallization at
Â
Ä
Avelino Corma,* María Jose Díaz-Cabanas, and
Vicente Fornes
Â
Zeolites are probably the most widely used solid catalysts in
refining, petrochemistry, and fine chemical production. This is
especially true for the acid zeolites, (H-zeolites). Their success
is derived from properties such as high surface area, high
adsorption capacity, molecular sieve characteristics, and the
possibility of preparation witha well defined number of
uniformly active sites;[1] these sites are introduced by direct
Â
Ä
[*] Prof. A. Corma, M. J. Díaz-Cabanas, V. Fornes
Instituto de Tecnología Química
Â
UPV-CVSIC Universidad Politecnica de Valencia
Avda. de los Naranjos s/n, 46022 Valencia (Spain)
Fax : (‡34)96-387-78-09
E-mail: acorma@itq.upv.es
[**] We thank the Spanish CICYT for financial support (Project MAT97-
1016-C02-01).
2346
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3913-2346 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 13