Dendrimeric catalysts for the activation of hydrogen peroxide. Increasing activity per catalytic phenylseleno group in successive generations

Article abstract of DOI:10.1021/ol990836a

(matrix presented) Dendrimeric polyphenylselenides are prepared in high yield using propyloxy spacers to connect the phenyseleno groups to the dendrimeric core. The selenides catalyze the oxidation of bromide with hydrogen peroxide to give positive bromine species that can be captured by cyclohexene in two-phase systems. The increase in the rate of catalysis exceeds statistical contributions for the first few generations with 3, 6, and 12 phenylseleno groups.

Full text of DOI:10.1021/ol990836a

ORGANIC
LETTERS
1999
Vol. 1, No. 7
1043-1046
Dendrimeric Catalysts for the Activation
of Hydrogen Peroxide. Increasing
Activity per Catalytic Phenylseleno
Group in Successive Generations
Charles Francavilla, Frank V. Bright, and Michael R. Detty*
Departments of Chemistry and Medicinal Chemistry,
State UniVersity of New York at Buffalo, Buffalo, New York 14260
mdetty@acsu.buffalo.edu
Received July 20, 1999
ABSTRACT
Dendrimeric polyphenylselenides are prepared in high yield using propyloxy spacers to connect the phenyseleno groups to the dendrimeric
core. The selenides catalyze the oxidation of bromide with hydrogen peroxide to give positive bromine species that can be captured by
cyclohexene in two-phase systems. The increase in the rate of catalysis exceeds statistical contributions for the first few generations with 3,
6, and 12 phenylseleno groups.
We report our initial studies toward the development of
highly efficient catalysts for peroxide activation that are based
on dendrimeric molecules terminating in phenylseleno
groups. Although the dendrimeric catalysts have minimal
water solubility with values of log P > 4, they are efficient
catalysts in two-phase systems and show large increases in
catalytic activity per phenylseleno group with each successive
generation of dendrimer.
chemical means. Traditionally, the molar activity of catalysts
has been optimized through structure-activity relationships
derived from substituent changes. In dendrimeric molecules,
catalytic functionality can be placed at the terminus of the
individual arms of the molecule and the reactivity of the
molecule on a molar basis then becomes a function of the
total number of individual reactive groups as well as the
reactivity of the individual groups. Dendrimers also impose
order within the molecules since reactive groups will be the
same distance from the central core. In certain catalytic
systems, a dendrimer effect is observed in which the catalytic
activity of individual groups increases with dendrimer
generation.1,2
Recent developments in the synthesis of dendrimeric
molecules1 suggest that materials with improved catalytic
activity can be designed through statistical rather than
(1) For reviews of dendrimeric molecules: (a) Tomalia, D. A.; Durst,
H. D. Top. Curr. Chem. 1993, 165, 193. (b) van der Made, A. W.; van
Leeuwen, P. W. N. M.; de Wilde, J. C.; Brands, R. A. C. AdV. Mater.
1993, 5, 466. (c) Service, R. F. Science 1995, 267, 458-459. (d) Dagani,
R. Chem. Eng. News 1996, 74, 20. (e) Newkome, G. R.; Moorefield, C.;
Vo¨gtle, F. Dendritic Macromolecules: Concepts, Syntheses, PerspectiVes;
VCH: Weinheim, Germany, 1996. (f) Gorman, C. AdV. Mater. 1998, 10,
295. (g) Chow, H.-F.; Mong, T. K.-K.; Nongrum, M. F.; Wan, C.-W.
Tetrahedron 1998, 54, 8543-8660.
Diorganotellurides are catalysts for oxidations with hy-
drogen peroxide.3,4 Peroxide oxidation gives the correspond-
ing dihydroxytellurane, which then acts as an oxidant
(2) For one example, see: Lee, J.-J.; Ford, W. T.; Moore, J. A.; Li, Y.
Macromolecules 1994, 27, 4632-4634.
10.1021/ol990836a CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/03/1999

(kinetically superior to hydrogen peroxide) for a variety of
substrates, including halide salts, thiols, selenols, and dye
precursors. The corresponding diorganotelluride is regener-
ated in the process to continue the catalytic cycle. Di-
organoselenides undergo the same chemistry, but they are
poorer catalysts for peroxide activation due to slower rates
of oxidation of the selenides and due to slower rates of
reaction of the corresponding dihydroxyselenanes with the
substrates. The tellurides and selenides mimic enzymes that
activate hydrogen peroxide such as horseradish peroxidase
(HRP)5 and vanadium bromoperoxidase (VBPO).6 In two-
phase systems, the most efficient catalysts have shown some
aqueous solubility as the telluride or selenide with n-octanol/
water partition coefficients (log P) of e2.3,4
functionality directly to 2 would provide the first-generation
dendritic wedge. Higher generations can be derived by the
linking of compound 2 to itself in an iterative process to
produce dendritic wedges such as 3 that can then be joined
to the central core.8
Arylselenide groups were successfully linked to phenolic
hydroxy groups through propyloxy spacers as shown in
Scheme 1. Importantly, both the introduction of the 3-bromo-
Scheme 1a
Synthesis. The oxidizing environment experienced by
peroxide catalysts limits the selection of core molecules and
stable linkages that can be employed in the synthesis of
dendrimeric arrays. The use of 1,1,1-tris(4-hydroxyphenyl)-
ethane (1; Figure 1) as a core molecule to which dendritic
a Key: (a) 18-crown-6, K2CO3, acetone, 90%; (b) CBr4, PPh3,
THF, 93%; (c) NaSePh from (PhSe)2/NaBH4 in EtOH/THF, 92%.
1-propyloxy spacer and the nucleophilic phenylselenide-
substitution reaction proceed in high yield (g90%). These
reactions should be suitable for the many iterations necessary
to produce either dendritic wedges or complete dendrimers.
In addition to demonstrating the feasibility of the chemistry,
the 3-phenoxypropyl phenylselenide 69 represents a control
monoselenide, to which the catalytic activity of the den-
drimeric molecules described below can be compared.
Attachment of three 3-bromopropanol molecules to the
core molecule 1 gave triol 7 in 90% isolated yield, as shown
in Scheme 2. The triol was converted to the tribromide 8 in
Figure 1. Building blocks for dendritic catalysts.
wedges can be attached has been elegantly developed by
Fre´chet.7 Compound 1 should be relatively unreactive toward
either hydrogen peroxide or the hypohalous acids produced
under the conditions described herein. Attachment of arms
containing a single selenide functionality directly to the
hydroxy groups of 1 would generate molecules with three
catalytic groups.
Scheme 2a
3,5-Dihydroxybenzyl alcohol (2) can be used to construct
dendritic wedges8 that are also stable to hyrogen peroxide
(Figure 1). Attachment of arms containing a single selenide
(3) (a) Detty, M. R.; Gibson, S. L. J. Am. Chem. Soc. 1990, 112, 4086-
4088. (b) Detty, M. R.; Gibson, S. L. Organometallics 1992, 11, 2147-
2155.
(4) Detty, M. R.; Zhou, F.; Friedman, A. E. J. Am. Chem. Soc. 1996,
118, 313-318.
(5) (a) Thorpe, G. H. G.; Kricka, L. J.; Mosely, S. B.; Whitehead, T. P.
Clin. Chem. 1985, 31, 1335-1341. (b) Ishikawa, E. Clin. Biochem. 1987,
20, 375-385. (c) Ikemoto, M.; Ishida, A.; Tsunekawa, S.; Ozawa, K.; Kasai,
Y.; Totani, M. Clin. Chem. 1993, 39, 794-799. (d) Larue, C.; Calzolari,
C.; Bertinchant, J.-P.; Leclercq, F.; Grolleau, R.; Pau, B. Clin. Chem. 1993,
39, 972-979.
a Key: (a) 18-crown-6, K2CO3, acetone, 90%; (b) CBr4/PPh3,
THF, 93%; (c) NaSePh from (PhSe)2/NaBH4 in EtOH/THF, 93%.
(6) (a) Butler, A.; Walker, J. V. Chem. ReV. 1993, 93, 1937-1944. (b)
Butler, A. In Bioinorganic Catalysis; Reedijk, J., Ed.; Marcel Dekker: New
York, 1992; pp 425-445. (c) Butler, A.; Carrano, C. J. Coord. Chem. ReV.
1991, 109, 61-105. (d) Wever, R.; Kreenn, M. B. E. In Vanadium in
Biological Systems; Chasteen, N. D., Ed.; Kluwer Academic Publishers:
Dordrecht, The Netherlands, 1990; pp 81-97.
(7) (a) Hawker, C.; Fre´chet, J. M. J. J. Chem. Soc., Chem. Commun.
1990, 1010-1013. (b) Hawker, C. J.; Fre´chet, J. M. J. J. Am. Chem. Soc.
1990, 112, 7638-7647.
93% isolated yield, and the tribromide 8 was then converted
to triselenide 9 in 93% isolated yield. The average yield of
each of the individual iterations of these reactions was greater
than 96%. Full characterization of all new compounds is
provided in the Supporting Information.
1044
Org. Lett., Vol. 1, No. 7, 1999

3-Bromo-1-propanol was converted to its tert-butyldi-
methylsilyl ether 10, and 2 equiv of 10 was attached to benzyl
alcohol 2 with potassium carbonate and catalytic 18-crown-6
to give 11 in 73% overall yield from 3-bromo-1-propanol.
Benzyl alcohol 11 was converted to the corresponding benzyl
bromide in two steps (Scheme 3). The mesylate 12 was first
Scheme 4a
Scheme 3a
a Key: (a) MsCl, Et3N, CH2Cl2, 90% of 12; (b) LiBr, THF, 93%
of 13; (c) 1, 18-crown-6, K2CO3, acetone, 91% of 14; (d) Bu4HF,
THF/DMF, 91% of 15; (e) (i) MsCl, Et3N, THF/CH2Cl2, (ii) LiBr,
81% of 16; (f) NaSePh from (PhSe)2/NaBH4 in 0.2 M NaOEt in
EtOH/THF, 85% of 17.
a Key: (a) 2, 18-crown-6, K2CO3, acetone, 24 h, 91% of 18; (b)
(i) MsCl, Et3N, THF/CH2Cl2 to give 19, (ii) LiBr, THF, 74% of
23; (d) Bu4NF, THF/DMF, 91% of 22; (e) (i) MsCl, Et3N, THF/
CH2Cl2, (ii) LiBr, THF, 74% of 23; (f) NaSePh from (PhSe)2/
NaBH4 in 0.2 M NaOEt in EtOH/THF, 66% of 24.
prepared (90%) and was then treated with lithium bromide
to give benzyl bromide 13 in 98% isolated yield. Three
equivalents of dendritic wedge 13 was attached to core
molecule 1 to give dendrimer 14 with six arms terminating
in tert-butyldimethylsilyl ethers in 91% isolated yield. The
silyl protecting groups were removed to give alcohol 15 in
91% yield, and 15 was converted to bromide 16 in 81%
overall yield via the mesylate as an intermediate. The addition
of NaSePh to bromide 16 gave dendrimer 17, bearing six
phenylseleno groups in 85% isolated yield. It should be noted
that direct displacement of the mesylate with NaSePh gave
17 in less than 30% isolated yield.
Two equivalents of bromide 13 was attached to the
phenolic hydroxyls of 3,5-dihydroxybenzyl alcohol (2), as
shown in Scheme 4, to give alcohol 18 in 91% isolated yield.
The alcohol 18 was converted to the corresponding bromide
in two steps via initial formation of the mesylate 19, which
was then treated with LiBr in THF to give the dendritic
wedge 20 in 81% overall yield. Compound 20 was attached
to the core molecule 1 to give dendrimer 21 in 70% yield.
The silyl protecting groups were removed to give alcohol
22 in 91% yield, and 22 was converted to bromide 23 in
two steps via the mesylate in 74% overall yield. The addition
of NaSePh to bromide 23 gave dendrimer 24 bearing 12
phenylseleno groups in 66% isolated yield.
Catalysis. Selenide 6 and dendrimers 9, 17, and 24 are
sparingly soluble in water (log P > 4), which necessitated
the use of two-phase systems for evaluation of their catalytic
activity. The oxidation of bromide with hydrogen peroxide
was followed by the initial rate of formation of trans-1,2-
dibromocyclohexane (25) and trans-2-bromocyclohexanol
(26) in a two-phase system of 0.5 M cyclohexene in CH2Cl2
for the organic phase and an aqueous phase of H2O2 (3.0
M) and NaBr (2.0 M) in pH 6 phosphate buffer at 296.0 (
0.1 K.10 The bromination of cyclohexene was rapid and
(8) (a) Groenendaal, L.; Fre´chet, J. M. J. J. Org. Chem. 1998, 63, 5675-
5679. (b) Wooley, K. L.; Hawker, C. J.; Fre´chet, J. M. J. J. Chem. Soc.,
Perkin Trans. 1 1991, 1059-1067. (c) Leon, J. W.; Kawa, M.; Fre´chet, J.
M. J. J. Am. Chem. Soc. 1996, 118, 8847-8859.
(9) All new compounds gave satisfactory elemental analyses and/or parent
ions by mass spectroscopy. 1H and 13C NMR spectra were consistent with
g97% purity. Spectral and analytical data for the new compounds of this
study can be found in the Supporting Information.
(10) In typical experiments, 0.05 mmol of 6, 9, or 17 or 0.008 mmol of
24 was added to a two-phase system of 20 mL of a stock solution of 0.5 M
cyclohexene in CH2Cl2 (8.20 g (0.100 mol) in 200 mL of CH2Cl2 with
diphenyl ether (170 mg, 1.00 mmol, 5 × 10-3 M) as an internal standard)
for the organic phase and an aqueous phase of H2O2 (12.5 mL of a 30%
solution) and NaBr (6.63 g, 0.065 mol) in 20 mL of pH 6 phosphate buffer
Org. Lett., Vol. 1, No. 7, 1999
1045

served as a good indicator of the generation of “Br+” species.
It should be noted that 25 and 26 were the only products of
reaction. The control experiments were then compared to
catalyzed reactions at 296.0 ( 0.1 K in which 2.5 mM 6, 9,
or 17 or 0.4 mM 24 was present (Table 1). The 25/26 ratios
efficient as a catalyst than 6. The catalytic activity of 24 is
more than 50 times the reactivity expected from strictly
statistical arguments based on the reactivity of 6. The
deviation from statistical rate increases is expressed in Table
1 as the relative initial rate (νinit) per phenylseleno group.
The catalysts 17 and 24 displayed high turnover numbers
for the generation of 25 and 26 in preparative runs.11 For
catalyst 17, a turnover number of g1030 mol of 25 + 26/
mol of 17 was observed while catalyst 24 gave a turnover
number of g5200 mol of 25 + 26/mol of 24. The catalysts
were recovered from the organic phase of the reaction and
were reused without apparent loss of activity.
Table 1. Rates of Catalysis for the Oxidation of Bromide by
Hydrogen Peroxide in the Presence of Phenylseleno-Containing
Molecules As Determined by the Rate of Bromination of
Cyclohexene
We speculate that the “dendrimer effect” observed in the
catalysis may be due to the micelle-like nature of the oxidized
catalyst where polar dihydroxyselenanes are layered equi-
distant from the core. Higher generation dendrimers may be
necessary to confirm this effect, since early generation
dendrimers are still capable of folding back on the core.12
However, the redox reaction with bromide proceeds rapidly
in the dendrimeric environment.4 These studies lay the
foundation for the design and synthesis of other dendrimeric
systems with improved properties that may be generally
useful to the synthetic chemist. We are currently preparing
water-soluble dendrimeric catalysts for halogenation reactions
that are free of organic solvent, and we are preparing
dendrimeric catalysts with chiral organochalcogen groups to
explore chiral halogenations.
Acknowledgment. We thank the Augmentation Awards
for Science and Engineering Research Training (AASERT)
Program, Office of Naval Research, for partial financial
support of this work.
in the catalyzed runs were similar to those in the uncatalyzed
runs, although somewhat more 2-bromocyclohexanol (26)
was formed in the presence of catalyst (Table 1).
The rates of catalysis for selenide 6 and dendrimers 9,
17, and 24 increased as the number of phenylseleno groups
increased in each molecule, but the increase far exceeded
the statistical changes in the number of phenylseleno groups.
As shown in Table 1, the monoselenide 6 displayed little
catalytic activity with a conversion rate of 0.24 mol of 25 +
26/(s (mol of 6)) when corrected for the background reaction.
The dendrimer 9 with three phenylseleno groups was 1 order
of magnitude more efficient than 6, while dendrimer 17 with
six phenylseleno groups was more than 2 orders of magnitude
more efficient than 6. Dendrimer 24 with 12 phenylseleno
groups continues the trend and is more than 600-fold more
Supporting Information Available: Full characterization
data for compounds 6-9, 11, 13-18, and 20-24. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL990836A
(11) In preparative runs, catalyst 24 (0.008 mmol) was added to a two-
phase mixture of 0.050 mol of cyclohexene in 50 mL of CH2Cl2 and an
aqueous phase of H2O2 (12.5 mL of a 50% solution) and NaBr (10.2 g,
0.100 mol) in pH 6 phosphate buffer (50 mL) at 293 K. When the rate of
appearance of products began to plateau, a second 12.5 mL aliquot of H2O2
was added (approximately 1 h). After 3 h, the cyclohexene was consumed
and 4.05 g (16.7 mmol) of 25 and 4.51 g (25.2 mmol) of 26 were isolated,
which corresponds to 41.9 mmol (84% yield) of brominated product from
0.008 mmol of 24 (a turnover number of >5200). Under identical conditions,
catalyst 17 (0.010 mmol) gave 10.3 mmol of brominated products after 3
h for a turnover number of >1030. In the latter reaction, starting cyclohexene
had not yet been consumed. The catalysts were recovered from the organic
phase following a bisulfite wash and chromatography on silica. No loss of
catalytic activity was observed upon further use.
at 296.6 ( 0.1 K. Compounds 25 and 26 were identified by GC-mass
spectroscopy with comparison to authentic samples for retention time.
Relative response factors of 5.0 × 10-3 M solutions of 25 or 26 were
determined and compared to the response of the 5.0 × 10-3 M diphenyl
ether internal standard. The organic phase was sampled periodically by gas
chromatography, and initial rates are reported on the basis of the first 25%
of reaction.
(12) Naidoo, K. J.; Hughes, S. J.; Moss, J. R. Macromoloecules 1999,
32, 331-341.
1046
Org. Lett., Vol. 1, No. 7, 1999