Dendritic bis(oxazoline)copper(II) catalysts. 2. Synthesis, reactivity, and substrate selectivity

Article abstract of DOI:10.1021/jo970383s

A series of dendritic bis(oxazoline) ligands 1-4 were synthesized to evaluate the effects of the degree of branching of a dendritic sector on both the reactivity and selectivity of their corresponding copper(II) complex- catalyzed Diels-Alder reaction between cyclopentadiene and a crotonyl imide. Kinetic studies unveiled a two-step mechanism of the Diels-Alder reaction, in which a reversible binding of the dienophile to the copper complex was followed by a rate-determining reaction between the resulting dienophile- catalyst complex with the diene. Furthermore, two interesting features emerged: first, the formation constant of the dienophile-catalyst complex decreased gradually on going from the lower to higher generations, and secondly, while the Diels-Alder reaction rate constant remained essentially the same from the zeroth to second generation catalysts, it dropped abruptly for the third generation one. These observations were rationalized as a consequence of a folding-back of the dendritic sectors toward the catalytic unit at the third generation, so that increase in steric size impeded both the reactivity and binding profiles of the catalytic system. This behavior was reminiscent of related phenomena observed by others from solvatomatic, photophysical, and viscosity studies. In line with this reasoning, a slight but noticeable substrate selectivity was observed for the third generation catalyst, which was absence for the lower ones, in competitive kinetic studies involving two dienophiles of different steric sizes.

Full text of DOI:10.1021/jo970383s

5116
J . Org. Chem. 1997, 62, 5116-5127
Den d r itic Bis(oxa zolin e)cop p er (II) Ca ta lysts. 2.1 Syn th esis,
Rea ctivity, a n d Su bstr a te Selectivity
Hak-Fun Chow* and Chi Ching Mak
Department of Chemistry, The Chinese University of Hong Kong, Shatin, NT, Hong Kong
Received February 28, 1997X
A series of dendritic bis(oxazoline) ligands 1-4 were synthesized to evaluate the effects of the degree
of branching of a dendritic sector on both the reactivity and selectivity of their corresponding
copper(II) complex-catalyzed Diels-Alder reaction between cyclopentadiene and a crotonyl imide.
Kinetic studies unveiled a two-step mechanism of the Diels-Alder reaction, in which a reversible
binding of the dienophile to the copper complex was followed by a rate-determining reaction between
the resulting dienophile-catalyst complex with the diene. Furthermore, two interesting features
emerged: first, the formation constant of the dienophile-catalyst complex decreased gradually on
going from the lower to higher generations, and secondly, while the Diels-Alder reaction rate
constant remained essentially the same from the zeroth to second generation catalysts, it dropped
abruptly for the third generation one. These observations were rationalized as a consequence of a
folding-back of the dendritic sectors toward the catalytic unit at the third generation, so that increase
in steric size impeded both the reactivity and binding profiles of the catalytic system. This behavior
was reminiscent of related phenomena observed by others from solvatomatic, photophysical, and
viscosity studies. In line with this reasoning, a slight but noticeable substrate selectivity was
observed for the third generation catalyst, which was absence for the lower ones, in competitive
kinetic studies involving two dienophiles of different steric sizes.
In tr od u ction
advantage of using polymer-supported catalysts is the
ready recoverability of the catalyst system, which often
simplifies product separation.
The design of catalyst systems to mimic enzyme
functions has been a topic of current interest. Enzyme
molecules, in contrast to man-made catalysts, are unique
not only for their high substrate- and enantioselectivities
but also for their efficiency in catalyzing reactions with
extremely fast turnover rates. It is well documented that
the supremacy of enzyme catalysts lies in their well-
defined three dimensional structures. The creation of an
active site within a large enzyme molecule provides the
requisite geometry which holds the reacting partners in
close proximity, thereby facilitating their union at ex-
tremely fast rates, while the size and shape of this
catalytic pocket control the substrate- and enantioselec-
tivity of enzymatic transformation.
Enzyme molecules are unique in that they are homog-
enous, monodisperse macromolecular catalysts with well-
defined three-dimensional topology in sharp contrast to
the heterogeneous, polydisperse polymer-supported cata-
lysts. Their well-defined three-dimensional structures
combined with their monodispersity allow enzyme mol-
ecules to exert powerful substrate- and enantioselective
control. While enzyme molecules are designed by nature
and synthesized in vivo with high efficiency, it is ex-
tremely difficult, if not impossible, for chemists to prepare
monodisperse and structurally defined polymer catalysts.
Recent success in reactions promoted by soluble poly-
mer-bound catalysts,5 especially in light of their improved
reactivity and selectivity, has prompted us to look into
the similar uses of soluble dendritic catalysts. An
attractive feature of dendrimers is that they can be
prepared with controllable topology and with specific
functionalities located at predetermined positions within
the dendritic matrix.6 Furthermore, dendrimers can also
be specifically engineered with certain physical and
chemical properties so as to allow possibilities in fine-
tuning of catalytic activity and selectivity. As a result,
the idea of incorporating catalytic centers into a dendritic
structure to produce catalytic dendrimers with high
Recent progress in various areas in host-guest chem-
istry have enabled chemists to design receptor or host
molecules approaching enzyme-substrate binding ef-
ficiency.2 When a catalytic center is assembled into a
receptor molecule, an artificial enzyme model is produced.
A number of successful enzyme models have been re-
ported which are invariably small molecules with con-
siderably sophisticated internal structures.3 On the other
hand, polymer chemists have long been engaging in
efforts to incorporate catalytic unit(s) into a polymer
structure, in hope that a properly fabricated polymer
chain may impart onto the catalytic sectors higher
reactivity and selectivity.4 Obviously, the additional
(4) For a review, see Blossey, E. C.; Ford, W. T. In Comprehensive
Polymer Science. The Synthesis, Characterisation, Reactions and
Applications of Polymers; Allen, G., Bevington, J . C., Ed.; Pergamon
Press: New York, 1989; Vol. 6, p 81.
(5) (a) Bergbreiter, D. E.; Weatherford, D. A. J . Org. Chem. 1989,
54, 2726-2730. (b) Han, H.; J anda, K. D. J . Am. Chem. Soc. 1996,
118, 7632-7633.
(6) For reviews, see (a) Tomalia, D. A.; Naylor, A. M.; Goddard, W.
A., III. Angew. Chem., Int. Ed. Engl. 1990, 29, 138-175. (b) Mekel-
burger, H.-B.; J aworek, W.; Vo¨gtle, F. Angew. Chem., Int. Ed. Engl.
1992, 31, 1571-1576. (c) Newkome, G. R.; Moorefield, C. N.; Baker,
G. R. Aldrichim. Acta 1992, 25, 31-38. (d) Tomalia, D. A.; Durst, H.
D. Top. Curr. Chem. 1993, 165, 193-313. (e) Tomalia, D. A. Adv. Mater.
1994, 6, 529-539. (f) Issberner, J .; Vo¨gtle, F. Angew. Chem., Int. Ed.
Engl. 1994, 33, 2413-2420.
* Author to whom correspondence should be addressed. Fax:
26035057; Tel: 26096344; e-mail: hfchow@cuhk.edu.hk.
X Abstract published in Advance ACS Abstracts, J une 15, 1997.
(1) Part 1: Chow, H.-F.; Mak, C. C. Macromolecules 1997, 30, 1228-
1230.
(2) For comprehensive reviews on discussion of host-guest com-
plexes, see Comprehensive Supramolecular Chemistry; Atwood, J . L.,
Davies, J . E., MacNicol, D. D., Vo¨gtle, F., Ed.; Pergamon Press: Oxford,
1996; Vols. 1 and 2.
(3) For
a recent review, see Feiters, M. C. In Comprehensive
Supramolecular Chemistry; Atwood, J . L., Davies, J . E., MacNicol, D.
D., Vo¨gtle, F., Ed.; Pergamon Press: Oxford, 1996; Vol. 10, p 267.
S0022-3263(97)00383-6 CCC: $14.00 © 1997 American Chemical Society

Dendritic Bis(oxazoline)copper(II) Catalysts
J . Org. Chem., Vol. 62, No. 15, 1997 5117
reactivity and selectivity has been one of the long-
standing goals in dendrimer chemistry.
Our experimental results showed that the substrate
selectivity was again higher for the third generation
dendritic catalyst, whereas the first generation analog
exhibited indistinguishable substrate selectivity as that
of a nondendritic compound, an observation that is in line
with the sudden change of dendrimer conformation at
the third generation reported earlier by us.1
At the outset of the present work, only a few catalytic
dendrimers were known.7-12 They were assembled by
either one or two approaches. The first involved the
attachment of a dendritic sector(s) to a single catalytic
moiety followed by the study of its influence on catalyst
reactivity and selectivity. The second approach entailed
the deployment of multiple catalytic units onto a den-
dritic matrix and the subsequent scrutiny for signs of
cooperativity among these catalytic units. Notwithstand-
ing the dissimilarity of these two approaches, both of
them address the singular issues in catalytic chemistry,
namely, reactivity and selectivity.
Resu lts a n d Discu ssion s
1. Syn th esis. The catalytic moiety chosen for this
study is a bis(oxazoline) ligand, metal complexes12 of
which are a well established class of catalysts capable of
promoting reactions such as cyclopropanation of olefins
and the Diels-Alder reaction. The dendritic fragments
employed to modulate the reactivity and properties of the
catalytic core are the known polyether dendritic sectors
5-8.13 Two approaches were regarded to be viable for
anchoring the dendritic units onto the catalytic moiety
(Scheme 1). One involved the direct coupling of a pivotal
bis(oxazolinyl) dibromide 9 to two equiv of dendritic
phenol [Gn]-OH 5-8. Alternatively, one could first
attach the dendritic phenol to a bis(hydroxyethyl)amido
dibromide 10 and subsequently cyclize the diamide to the
desired bis(oxazoline). Demurred at the likelihood of self
N-alkylation in a molecule such as 9, we decided to take
the latter approach.
The synthetic operation entailed the initial preparation
of the central core 10 and the attachment to it of the
various generation dendritic sectors 5-8. Construction
of the core began with the readily available 4-(benzyloxy)-
benzyl alcohol 11 (Scheme 2). Treatment of the alcohol
with concentrated hydrobromic acid in glacial acetic acid
afforded the corresponding benzyl bromide 12 as a white
solid in 64% yield. Di-C-alkylation of diethyl malonate
with 2.1 mol equiv of bromide 12 (NaH, THF) furnished
the diester 13 as an oil in 74% yield. Alkaline hydrolysis
of 13 in aqueous ethanol resulted in the formation of a
separable mixture of the decarboxylated monoacid 14
(27%) and the desired diacid 15 (65%). However, pro-
longed heating of the reaction mixture led to the pre-
dominant formation of the decarboxylation product 14.
Conversion of diacid 15 to the corresponding diacyl
chloride with oxalyl chloride in DMF, followed by quench-
ing with excess 2-ethanolamine gave diamide 16 as a
white solid in 69% yield. At this point the benzyl
protective groups were dismantled by hydrogenolysis to
give bisphenol 17, which was then O-alkylated to give
dibromide 10 in a combined yield of 69% by reaction with
excess 1,3-dibromopropane in the presence of anhydrous
K2CO3 in acetone.
Some years ago Brunner7 reported the preparation of
a series of low generation chiral dendrimers containing
a catalytic diphos rhodium complex unit. Interestingly,
the resulting catalytic dendrimers not only preserve their
catalytic activity but display enhanced reactivity. How-
ever, the structure-reactivity issue was not pursued in
detail. Ford8 was first to capitalize on the “multiple
catalytic centers” concept and designed a series of poly-
ether-based catalytic dendrimers terminated with mul-
tiple quaternary ammonium ions which were capable of
promoting unimolecular decarboxylation of 6-nitrobenz-
isoxazole-3-carboxylate. Positive cooperative reactivity
was observed for the higher generation catalysts which
was not present in lower generation analogs. Shortly
afterward, DuBois9 and van Koten10 described the prop-
erties of two different series of multicenter dendritic
catalysts but were unable to detect any positive cooper-
ativity in them. On the other hand, the substrate
selectivity issue was addressed by Moore in a study of a
number of dendritic porphyrin-based catalysts.11 It was
noted that substrate selectivity was higher for the higher
generation dendritic catalysts. All these developments
suggested that dendritic catalysts rivaled polymer-sup-
port catalysts in terms of manipulability of their physical,
chemical, and topological properties and had the potential
to become an alternative type of insoluble or soluble
catalysts.
Until recently there were few studies on the effect of a
dendritic sector on the reactivity and selectivity of single-
center catalytic systems. Earlier we disclosed1 the
synthesis of a series of dendritic bis(oxazoline) ligand 1-4
and demonstrated that the reactivity and substrate
binding profile of their corresponding metal complex-
catalyzed Diels-Alder reaction were strongly dependent
upon the size of the dendritic sector in a manner that
the substrate binding ability deteriorated with increasing
size of the dendritic fragment, and that a sudden change
of catalyst conformation and reactivity occurred from the
second to the third generation. These initial findings
provided the foundations of the work described in this
paper. Reported herein are additional findings on the
substrate selectivity of this novel series of dendritic
catalysts, together with the full details of their synthesis.
The structures of the compounds in Scheme 2 were
characterized by NMR and MS techniques. Thus, for the
1
diester 13, the unique H-NMR resonance signal of the
oxygenated benzylic protons at δ 5.03 together with the
singlet at δ 3.15 ascribable to the benzylic protons
adjacent to the quaternary carbon confirmed its struc-
1
ture. The H-NMR spectrum of monoacid 14 gave a set
(7) Brunner, H. J . Organomet. Chem. 1995, 500, 39-46.
(8) Lee, J .-J .; Ford, W. T.; Moore, J . A.; Li, Y. Macromolecules 1994,
27, 4632-4634.
of complex resonance signals at δ 2.50-3.10 while that
of the symmetric diacid 15 exhibited a simple singlet at
δ 3.22 diagnosed as the benzylic protons adjacent to the
quaternary carbon. The structural identity of diamide
16 was established by its 1H-NMR spectrum which
showed two characteristic quartets at δ 3.31 and 3.56
(9) Miedaner, A.; Curtis, C. J .; Barkley, R. M.; DuBois, D. L. Inorg.
Chem. 1994, 33, 5482-5490.
(10) Knapen, J . W. J .; van der Made, A. W.; de Wilde, J . C.; van
Leeuwen, P. W. N. M.; Wijkens, P.; Grove, D. M.; van Koten, G. Nature
1994, 372, 659-663.
(11) Bhyrappa, P.; Young, J . K.; Moore, J . S.; Suslick, K. S. J . Am.
Chem. Soc. 1996, 118, 5708-5711.
(12) For a review, see Pfaltz, A. Acc. Chem. Res. 1993, 26, 339-
345.
(13) Chow, H.-F.; Chan, I. Y.-K.; Mak, C. C.; Ng, M.-K. Tetrahedron
1996, 52, 4277-4290.

5118 J . Org. Chem., Vol. 62, No. 15, 1997
Chow and Mak
Ch a r t 1
attributable to the methylene protons adjacent to the
nitrogen atom and hydroxyl group, respectively. The
presence of a triplet at δ 7.69 provided evidence for the
presence of the nitrogen-bonded hydrogen. The success-
ful deprotection of the benzyl groups from 16 was readily
confirmed by the absence of the signal for the oxygenated
benzylic proton at δ 5.00 and the signal for the corre-
of the diol 18 into the dibromide 22 using carbon
tetrabromide/triphenylphosphine followed by intra-
molecular cyclization of 22 in the presence of NaOH in
EtOH/THF afforded the bis(oxazoline) dendrimer G0 1
as a pale yellow liquid in 43% overall yield from 10. In
a similar manner, coupling of the higher generation
polyether dendritic sectors 2-4 onto 10 followed by
bromination/cyclization furnished the higher generation
dendritic ligands 2-4 in 46, 25, and 27% overall yields
from 10, respectively.
It should be noted that the bromination and cyclization
steps of the above reaction sequence take place at the
site of the catalytic core against a landscape of dendritic
sectors which expands with each upward generation.
Therefore, the appraisal of the success of these reactions
by NMR methods was not expected to be easy at high
generations, owing to the disproportionality in intensity
of the relevant signals. Fortunately, the methylene
protons at the reaction site displayed triplets at measur-
1
sponding benzylic carbon at δ 69.9 in the H- and 13C-
NMR spectra, respectively, which were present in those
of compound 15. For diamide 10, its structure was
supported by a distinct quintet at δ 2.29 for the central
methylene protons of the three-carbon linker. Mass
spectra data provided additional evidence for the molec-
ular identities of all the aforementioned compounds.
With both the catalytic core 10 and the dendritic
fragments 5-8 on hand, the stage was set for the
assemblage of the catalytic core onto the various dendritic
sectors (Scheme 3). Reaction of 4-tert-butylphenol 5 with
dibromide 10 gave the diol 18 in 60% yield. Conversion

Dendritic Bis(oxazoline)copper(II) Catalysts
J . Org. Chem., Vol. 62, No. 15, 1997 5119
Sch em e 1. P r ep a r a tion of Den d r itic Bis(oxa zolin e) Liga n d s
Sch em e 2a
a
(i) HBr, HOAc; (ii) NaH (2 eq), diethyl malonate, THF; (iii) NaOH, H2O/EtOH; (iv) DMF, (COCl)2; (v) ethanolamine; (vi) H2, 10%
Pd-C; (vii) 1,3-dibromopropane, K2CO3, acetone.
ably different chemical shift values for the series of diols
[Gn]-NHCH2CH2OH (δ ∼3.6), dibromides [Gn]-
NHCH2CH2Br (δ ∼3.3) and bis(oxazoline) Gn (δ ∼4.2,
atoms which were not present in the precursors. In
addition, the unique 13C-NMR resonance signals at about
δ 31.2 for all the dibromides 22-25 was also diagnostic
of the carbons adjacent to the bromine atoms.
1
CH2O), which were not obscured by the H-NMR signals
arising from the dendritic sectors. In addition, the diols
18-21, dibromides 22-25, and bis(oxazoline)s 1-4 all
displayed significantly different TLC mobilities so that
monitoring of the progress of the reactions could be
carried out with ease. The tert-butyl groups on the
surface sector served as markers for the dendrimer
generation of these molecules. Therefore, by focusing on
the relative intensities of the tert-butyl signals at δ ∼1.3
and the oxygenated methylene signal at δ ∼3.6 of
compounds 18-21, one could ascertain whether two
dendritic sectors had been incorporated. For example,
experimental integration ratios for these two signals for
the diols 18-21 (from the zeroth to third generation) of
4.5:1, 8.1:1, 20.7:1, and 36.0:1, respectively, indeed closely
matched the theoretical values of 4.5:1, 9:1, 18:1, and 36:
1. The successful conversions of the above dendritic diols
[Gn]-NH(CH2)2OH to the corresponding dibromides [Gn]-
NH(CH2)2Br were substantiated by the appearance of the
Upon cyclization to the bis(oxazoline) dendritic ligands
1-4, Gn (n ) 0-3), an appreciable downfield shift (from
δ 3.5 to δ 3.8) of the triplet signals due to the methylene
protons adjacent to the nitrogen atom together with the
disappearance of the triplets for the nitrogen-bonded
hydrogens were noted. A comparison of the 1H-NMR
spectra of dendritic ligands 1-4 of different generations
showed that the intensities of 1H signals due to the
catalytic core decreased gradually with respect to those
of the signals arising from the dendritic sectors toward
the higher generation ligands. Infrared spectroscopy
offered little information since the spectra were over-
whelmed by the signals arising from the polyether
dendritic sectors. Nevertheless the stretching frequen-
cies at about 1660 cm-1 in the IR spectra of G0 1 and G1
2 were consistent with the presence of oxazoline C)N
bond although this signal could not be observed for the
higher generation dendrimers. Additional evidence for
the formation of the bis(oxazoline) dendrimers was
secured from their satisfactory mass spectra. Thus,
1
distinctive H-NMR resonance triplet centered at about
δ 3.3 for the methylene protons adjacent to the bromine

5120 J . Org. Chem., Vol. 62, No. 15, 1997
Chow and Mak
Sch em e 3. Sch em a tic Dia gr a m Sh ow in g th e Syn th etic Tr a n sfor m a tion s of th e Va r iou s Den d r itic Sector s
5-8 to th e Cor r esp on d in g Bis(oxa zolin e) Den d r itic Liga n d s 1-4
dendrimers 1-4 showed ions at 747 (M + H+), 1460 (M
+ H+), 2883 (M+), and 5731 (M+), respectively, which
corresponded to their molecular ion peaks within experi-
mental error.
The educts for the Diels-Alder reaction were cyclo-
pentadiene (Cp) and crotonyl imide 26 (R). The dendritic
copper(II) catalysts were prepared by stirring a 1:1
mixture of individual dendritic ligand (Gn) and copper(II)
triflate in anhydrous dichloromethane under nitrogen
until complete dissolution of the solid inorganic copper
reagent. The reaction was monitored by UV spectroscopy
to ensure complete complexation as all the copper(II)
dendritic complexes gave UV spectra with a λmax at
approximately 720 nm. As expected, dendritic ligands
of higher generations (G2 and G3) required longer time
(3-5 h) for complete complexation than did ligands of
lower generations (1.5 h). The reaction was conducted
by mixing a preformed solution of a ligand-copper(II)
triflate complex, Cp, and R in anhydrous dichloromethane,
and the resulting solution was stirred at 25.0 ( 0.1 °C.
The kinetics were followed by taking aliquots at specified
time intervals for gas chromatographic analysis. The
starting materials and adduct 27 (P) all possessed
different retention times allowing easy assay of their
relative concentrations at any particular time interval.
The rate law of the reaction was studied by employing
pseudo-order kinetics technique on a nondendritic bis-
(oxazoline)copper complex 28‚Cu(OTf)2. The pseudo-
order conditions were satisfied by keeping concentrations
of all but one of the components constant. This condition
was automatically fulfilled by the catalyst component
because it was continuously regenerated and therefore
remained constant during the reaction. The concen-
trations of the other components were maintained es-
sentially constant by using a large excess of the reagents.
Only the initial kinetic data of the reaction with less than
15% product conversion were taken into consideration.
The kinetic results are summarized in Figure 1. It was
found that the initial rate of the reaction d[P]/dt was
directly proportional to [28‚Cu(OTf)2] and to [Cp], while
a plot of 1/[R] vs 1(d[P]/dt) gave a straight line not passing
2. Kin etics a n d Mech a n ism of th e Den d r itic-
Com p lex-Ca ta lyzed Diels-Ald er Rea ction . Attach-
ment of dendritic sectors to a catalytic center will
invariably affect the polarity and steric environment
around its vicinity. As a result, its catalytic reactivity
and selectivity can be modulated. The successful prepa-
ration of the novel series of dendritic ligands 1-4 having
a single catalytic unit embedded in a dendritic matrix of
different sizes provided us the opportunity to probe the
microenvironment around the catalytic center, to identify
the role of the dendritic sector in this type of single site
dendritic catalyst, and to examine their practical useful-
ness in catalytic chemistry.
Although the bis(oxazoline)copper(II) complex-cata-
lyzed Diels-Alder reaction had been known for some
time, little effort was devoted to elucidate the mechanism
of this reaction.14 Hence, the kinetics and mechanism
of the Diels-Alder reaction in our study were investi-
gated prior to the study of the effect of dendrimerization
on the properties of the catalytic bis(oxazoline) center.
(14) For discussions on plausible mechanisms: see (a) Evans, D. A.;
Miller, S. J .; Lectka, T. J . Am. Chem. Soc. 1993, 115, 6460-6461. (b)
Evans, D. A.; Murry, J . A.; Norcross, R. D.; Miller, S. J . Angew. Chem.,
Int. Ed. Engl. 1995, 34, 798-800.

Dendritic Bis(oxazoline)copper(II) Catalysts
J . Org. Chem., Vol. 62, No. 15, 1997 5121
F igu r e 1. Initial rate of product (P) formation vs [28‚Cu(OTf)2]0 ([R]0 ) 0.0843 M, [Cp]0 ) 1.52 M) (left). Initial rate of (P)
formation vs [Cp]0 ([R]0 ) 0.32 M, [28‚Cu(OTf)2]0 ) 5.0 mM) (middle). Plot of 1/{d[P]/dt} vs 1/[R]0 ([Cp]0 ) 2.39 M, [28‚Cu(OTf)2]0
) 1.22 mM) (right).
Sch em e 4. P r op osed Mech a n ism of th e
through the origin. Hence the rate equation of the Diels-
Alder reaction has the form:
Diels-Ald er Rea ction
k1k2[28‚Cu(OTf)2][R][Cp]
k-1 + k1[R]
d[P]
dt
)
(1)
Ta ble 1. Kc a n d k2 Va lu es of th e Diels-Ald er Rea ction s
The rate law is consistent with the mechanism shown
in Scheme 4. The copper-containing catalyst 28‚Cu(OTf)2
first forms a 1:1 complex with the dienophile (R). The
ligand-dienophile complex 28‚Cu(OTf)2-R can either
react with an incoming Cp molecule to form the Diels-
Alder adduct (P) with the liberation of a free catalyst
28‚Cu(OTf)2, or dissociate to regenerate the free catalyst
and the dienophile.
Applying steady state approximation on the catalyst-
dienophile complex 28‚Cu(OTf)2-R, eq 2 can be ob-
tained,15 where subscript 0 denotes the initial concen-
trations of the reactants.
catalyst
Kc (M-1
)
103k2 (M-1 s-1
)
G0‚Cu(OTf)2
G1‚Cu(OTf)2
G2‚Cu(OTf)2
G3‚Cu(OTf)2
10.4 ( 0.2
9.8 ( 1.1
7.8 ( 1.1
5.7 ( 0.5
3.3 ( 0.1
3.3 ( 0.3
3.2 ( 0.4
1.9 ( 0.2
readily be extrapolated from the Lineweaver-Burk plot,16
data of which was obtainable by measuring the initial
rates of the reaction at different dienophile concentra-
tions [R]0.
The effect of the various dendritic fragments on the
catalytic reactivity and binding capability was addressed
by looking at the binding and rate constants of the
various dendritic ligand-copper(II) complex-catalyzed
reactions (Table 1 and Figure 2). In comparing the
binding constants (Kc) and the rate constants (k2) of the
different dendrimer-catalyzed reactions, two important
findings were noteworthy. First, the binding constant
decreased with increasing dendrimer generation. The
destablization could be attributed to an increased distor-
tion of the catalyst-dienophile complex (Gn‚Cu(OTf)2-
R) from its optimal geometry as a result of the increasing
steric repulsion between the two large dendritic sectors
at the higher generations. Secondly, the rate constant
k2 for the Diels-Alder reaction remained essentially
constant from the zeroth to the second generation cata-
lysts, but dropped suddenly at the third generation. This
rate-determining process involved the Diels-Alder reac-
tion between the catalyst-dienophile complex and Cp,
and the rate should be controlled by the steric accessibil-
ity of the catalytic cavity. The sudden drop of k2 for the
G3 catalyst can be ascribed to a sudden change of the
steric environment around the catalytic cavity from the
second to the third generation. We postulated that the
catalytic core which was essentially open to the sur-
roundings at G0-G2 became partially buried in the
interior of the dendritic matrix at G3. For the lower
generation catalytic dendrimers, the two dendritic sectors
are capable of disposing themselves in such a way so as
to avoid steric repulsion between them and to allow the
catalytic core to be exposed to the surroundings. Con-
k1k2[28‚Cu(OTf)2]0[R][Cp]
k-1 + k1[R] + k2[Cp]
d[P]
dt
)
(2)
If the rate of the Diels-Alder reaction between
28‚Cu(OTf)2-R and Cp is slow compared to both the rates
of formation and the decomplexation of the 28‚Cu-
(OTf)2-R complex, i.e., k-1 . k2[Cp] and k1[R] . k2[Cp],
we have
k1k2[28‚Cu(OTf)2]0[R][Cp]
k-1 + k1[R] + k2[Cp]
d[P]
dt
)
=
k1k2[28‚Cu(OTf)2]0[R][Cp]
k-1 + k1[R]
(3)
Thus, the initial rate of the reaction becomes:
k1k2[28‚Cu(OTf)2]0[R][Cp]
k-1 + k1[R]
d[P]
dt
) νinit
)
f
k1k2[28‚Cu(OTf)2]0[R]0[Cp]0
k-1 + k1[R]0
(4)
Equation 4 is the modified Michaelis-Menten equation
for enzyme kinetics and is equivalent to the experimental
rate law expressed by eq 1. The binding constant (Kc )
k1/k-1) of the bis(oxazoline)copper(II)-dienophile complex
and the rate constant k2 of the Diels-Alder reaction could
(15) Gates, B. C. Catalytic Chemistry; J ohn Wiley & Sons, Inc.: New
York, 1992; p 7.
(16) Segel, I. H. Enzyme Kinetics; J ohn Wiley & Sons, Inc.: New
York, 1975; p 46.

5122 J . Org. Chem., Vol. 62, No. 15, 1997
Chow and Mak
F igu r e 2. Linewaver-Burk plot for G0- (top left), G1- (top right), G2- (bottom left), and G3- (bottom right) copper(II)-catalyzed
Diels-Alder reaction of Cp and crotonyl imide 26 (R).
avoid the steric repulsion between them, folded backward
and acted as an envelope for the hydrophilic metal center.
Similarly, inward folding of chain ends toward the center
with increasing generation number was also detected by
rotational-echo double-resonance NMR study.17 The sud-
den transition from an extended to a globular structure
from the third to the fourth generation of a series of
polyether-based dendrimers has also been inferred from
a solvatochromatic investigation.18 In another study of
a family of carboxylate-terminated starburst dendrimers,
the occurrence of a structural change between the third
and fourth generations was also demonstrated by photo-
induced electron transfer measurements.19 Fre´chet and
co-workers further noted that the intrinsic viscosity
profile of low-generation polyether dendrimers increased
with molecular weight but that of the higher generation
counterparts showed a reverse trend.20 Such behaviors
were again interpreted by a conformational change from
an extended configuration to a globular one. The kinetic
results obtained in our study therefore served to dem-
onstrate that profound effects on the reactivity of a metal
complex can be brought about by a conformational change
on the dendritic sectors of its ligands.
F igu r e 3. Transition from an exposed to a partially buried
active site from lower generation to the third generation
dendrizyme.
sequently, k2 was insensitive to the dendrimer generation
in these cases. However, for the third generation analog,
the steric repulsion between the two large dendritic
sectors might be so severe that one or both of them has
to move toward the catalytic core to relieve the steric
strain, resulting in a decrease of steric accessibility
around the reaction center (Figure 3).
Changes in the physical properties of dendrimers
across generations have been interpreted as a conse-
quence of a change of conformation in them. For ex-
amples, the encapsulation of hydrophilic metal ions by
large hydrophobic dendritic sectors was used to explain
the unexpected solubility of higher generation dendritic
rhodium phosphine complexes in nonpolar solvents such
as pentane.7 Such unusual solubility, however, was not
observed for the analogous dendritic complexes of lower
generations. It was proposed that the hydrophobic
sectors of the higher generation dendrimers, in order to
(17) Wooley, K. L.; Klug, C. A.; Tasaki, K.; Schaefer, J . J . Am. Chem.
Soc. 1997, 119, 53-58.
(18) Hawker, C. J .; Wooley, K. L.; Fre´chet, J . M. J . J . Am. Chem.
Soc. 1993, 115, 4375-4376.
(19) Moreno-Bondi, M. C.; Orellana, G.; Turro, N. J .; Tomalia, D.
A. Macromolecules 1990, 23, 910-912.
(20) Mourey, T. H.; Turner, S. R.; Rubinstein, M.; Fre´chet, J . M. J .;
Hawker, C. J .; Wooley, K. L. Macromolecules 1992, 25, 2401-
2406.

Dendritic Bis(oxazoline)copper(II) Catalysts
J . Org. Chem., Vol. 62, No. 15, 1997 5123
Ta ble 2. In itia l Velocities of Diels-Ald er Ad d u ct
F or m a tion P r om oted by Differ en t Ca ta lysts (a ll r a tes
n or m a lized to 1 m M of ca ta lyst con cen tr a tion )
level of substrate selectivity. On the other hand, a
notable difference of the rates of formation of the two
Diels-Alder adducts was observed for the reaction
catalyzed by the third generation complex. The higher
substrate selectivity brought about by the third genera-
tion catalyst can be put into better perspective if one
compares the ratio of the initial velocities (krel) for the
formation of the two Diels-Alder adducts. For the
nondendritic 28‚Cu(OTf)2 and the G1‚Cu(OTf)2 catalysts,
the ratio of the initial velocities for the formation of 31
to 32 was 1.05 and 1.07, respectively, whereas the
corresponding value for the G3‚Cu(OTf)2 catalyst was
1.18. These figures are consistent with our notion that
the partially folded G3 dendritic sectors imposed signifi-
cant steric crowding around the catalytic site. As a
result, the Diels-Alder reaction involving the bulkier
dienophile 30 reacted with Cp at a slower rate than the
less hindered competitor 29, whereas in the absence of
shielding caused by the folding back of the dendritic
sectors in lower generation ligands, such discrimination
was negligible.
104d[31]/
dt (M/s)
104d[32]/
dt (M/s)
catalyst
krel
28‚Cu(OTf)2
(non-dendritic)
G1‚Cu(OTf)2
G3‚Cu(OTf)2
2.33 ( 0.04
2.21 ( 0.04
1.05 ( 0.03
1.93 ( 0.04
0.67 ( 0.01
1.80 ( 0.04
0.57 ( 0.01
1.07 ( 0.03
1.18 ( 0.03
The kinetic data indicated that under saturated kinetic
conditions, when nearly all catalysts are dienophile
bound (i.e. [Gn‚Cu(OTf)2-R]/[Gn‚Cu(OTf)2] g 102, or
when [R] ) [Gn‚Cu(OTf)2-R]/([Gn‚Cu(OTf)2] × Kc) ≈ 10
M), there is little reactivity difference among the zeroth
to the second generation catalysts. It should be pointed
out, however, that under ordinary nonsaturated kinetic
conditions (i.e. [R] e 100 M), because of the decreasing
binding constant, a gradual decrease of reactivity should
be observed on going from the zeroth to the second
generation dendritic catalyst.
3. Su bstr a te Selectivity. The encapsulation of a
catalytic functionality within a dendritic matrix may
provide the key for successful substrate control of a
catalytic reaction. Dendritic catalysts of different gen-
erations may therefore display differential substrate
selectivity in competitive experiments. Dienophiles 29
and 30 of slightly different steric sizes were chosen to
probe these effects in our study. Under the reaction
conditions, a 1:1 molar ratio mixture of dienophiles 29
and 30 was allowed to react with a large excess of Cp in
the presence of a catalytic amount (20 mol %) of a copper
catalyst formed by complexation of individual ligands
1-4 with Cu(OTf)2. On the basis of the kinetic study
described earlier, the third generation catalyst G3‚Cu-
(OTf)2, due to its presumably partially folded structure
around the catalytic center, would be the starting point
to exhibit discernible selectivity toward the two sub-
strates 29 and 30. For comparison purpose, the nonden-
dritic complex 28‚Cu(OTf)2 and the first generation
catalyst G1‚Cu(OTf)2 were also employed in the selectiv-
ity study. The initial velocities for the formation of the
respective Diels-Alder adducts 31 and 32 with these
three catalysts are tabulated in Table 2.
4. Con clu sion s. This paper reported the synthesis,
reactivity, and substrate selectivity of a new class of bis-
(oxazoline)-based dendritic catalyst. From both kinetic
and substrate selectivity studies, it was deduced that a
sudden change of dendrimer conformation occurred from
the second to the third generation. Furthermore, we
have shown here that enhanced substrate selectivity can
be realized with this class of dendritic catalysts, in a
manner similar to the results reported earlier by Moore
on dendritic porphyrin catalysts.11 Even though the
observed substrate selectivity enhancement for the third
generation catalyst was not significantly high to warrant
its practical usefulness in synthesis, we believe that
dendritic catalysts with much better substrate selectivity
and reactivity could be designed by carefully choosing
suitable dendritic sectors with proper disposition around
the catalytic center.
Exp er im en ta l Section
1. Syn th esis. The particulars on the analytical instru-
ments employed in the present study were previously de-
scribed.21 Unless otherwise stated, all chemicals were pur-
chased from commercial suppliers and used without further
purification. The molecular mass reported in each case is the
most abundant molecular isotopic peak for the compound.
4-(Ben zyloxy)ben zyl Br om id e (12). To a stirred solution
of 4-(benzyloxy)benzyl alcohol (11) (48.0 g, 0.22 mol) in acetic
acid (150 mL) at 0 °C was added a solution of hydrobromic
acid (100 mL, 48% in acetic acid). A white precipitate formed
immediately, and the mixture was poured into ice-water. The
white solid was suction filtered, washed with water, and
redissolved in ethyl acetate (400 mL). The organic layer was
washed with brine, dried (Na2SO4), and filtered. The filtrate
was concentrated to about 100 mL, and the bromide 12 (40.0
g, 64%) crystallized out as a white solid: mp 78-80 °C (lit.22
Examination of the kinetic profile showed that the
smaller unsaturated imide 29, as expected, reacted faster
than the bulkier analogue 30 in all three catalysts
studied. Since all competitive experiments were con-
ducted under nonsaturated kinetic conditions, the lower
generation catalyst, as anticipated, promoted a faster
turnover rate than the higher generation one. Neverthe-
less, a significantly large rate retardation was noted for
the Diels-Alder reactions promoted by the third genera-
tion dendritic catalyst. The nondendritic bis(oxazoline)
28‚Cu(OTf)2 and G1‚Cu(OTf)2 catalysts displayed similar
reactivity patterns toward the two different dienophiles,
suggesting that they possessed intrinsically the same
1
mp 81-82 °C); Rf 0.45 (hexane/ethyl acetate ) 5/1); H-NMR
(CDCl3) δ 4.49 (s, 2 H), 5.05 (s, 2 H), 6.93 (d, J ) 8.7 Hz, 2 H),
7.31 (d, J ) 8.7 Hz, 2 H), 7.34-7.50 (m, 5 H).
Dieth yl 2,2-Bis((4-ben zyloxy)ben zyl)m a lon a te (13). To
a suspension of sodium hydride (1.50 g, 62.3 mmol) in dry THF
(150 mL) at 0 °C was added dropwise a solution of diethyl
malonate (2.75 g, 17.2 mmol) in dry THF (15 mL) over 15 min.
The resulting mixture was warmed to 20 °C and stirred for a
further 15 min. (4-Benzyloxy)benzyl bromide (12) (10.0 g, 36.1
(21) Chow, H.-F.; Chan, I. Y.-K.; Chan, D. T. W.; Kwok, R. W. M.
Chem. Eur. J . 1996, 2, 1085-1091.
(22) Kashdan, D. S.; Schwartz, J . A.; Rapoport, H. J . Org, Chem.
1982, 47, 2638-2643.

5124 J . Org. Chem., Vol. 62, No. 15, 1997
Chow and Mak
mmol) was added and the mixture heated under reflux for 2
h. Water was added carefully into the reaction mixture and
excess THF removed under reduced pressure. The residue was
taken up in ethyl acetate, washed with brine, dried (Na2SO4),
and filtered. Concentration of the filtrate followed by flash
chromatography on silica gel (hexane/ethyl acetate ) 8/1)
provided the product 13 (7.0 g, 74%) as a pale yellow syrup:
Rf 0.35 (hexane/ethyl acetate ) 5/1); IR (film) cm-1 1730, 1244;
1H-NMR (CDCl3) δ 1.16 (t, J ) 7.1 Hz, 6 H), 3.15 (s, 4 H), 4.09
(q, J ) 7.1 Hz, 4 H), 5.03 (s, 4 H), 6.88 (d, J ) 8.7 Hz, 4 H),
7.09 (d, J ) 8.7 Hz, 4 H), 7.20-7.45 (m, 10 H); 13C-NMR
(CDCl3) δ 13.8, 38.2, 60.3, 61.0, 69.8, 114.4, 127.3, 127.8, 128.4,
131.0, 136.9, 157.6, 171.0; MS (EI, m/ z) 552 (M+, 4), 308 (18),
197 (28), 91 (100), 65 (24). Anal. Calcd for C35H36O6: C, 76.06;
H, 6.57. Found: C, 75.81; H, 6.69.
N,N′-Bis(2-h yd r oxyeth yl)-2,2-bis[4-(3-br om op r op oxy)-
b en zyl]-1,3-p r op a n ed ia m id e (10). A mixture of the di-
phenol 17 (0.20 g, 0.50 mmol), 1,3-dibromopropane (3.00 g, 15.0
mmol), and anhydrous potassium carbonate (0.20 g, 1.45
mmol) in acetone (50 mL) was heated under reflux for 3 d.
Upon cooling to room temperature, the reaction mixture was
filtered through a pad of Celite. The filtrate was concentrated
and the residue purified by flash chromatography on silica gel
(hexane/ethyl acetate ) 1/5 gradient to ethyl acetate) to afford
the dibromide 10 (0.25 g, 78%) as a white solid: mp 121-122
°C; Rf 0.45 (ethyl acetate); IR (film) cm-1 3344, 1660; 1H-NMR
(CDCl3) δ 2.29 (quin, J ) 6.1 Hz, 4 H), 2.83 (br s, 2 H), 3.26 (s,
4 H), 3.34 (q, J ) 5.0 Hz, 4 H), 3.50-3.65 (m, 8 H), 4.05 (t, J
) 5.8 Hz, 4 H), 6.79 (d, J ) 8.7 Hz, 4 H), 7.08 (d, J )8.6 Hz,
4 H), 7.73 (t, J ) 5.2 Hz, 2 H); 13C-NMR (CDCl3) δ 29.8, 32.2,
42.4, 42.6, 59.4, 61.4, 65.3, 114.2, 128.5, 130.4, 157.6, 173.3;
MS (FAB, m/ z) 645 (M + H+, 54); HRMS calcd for C27H37N2-
O679Br81Br 645.0998, found 645.0986. Anal. Calcd for
C27H36N2O6Br2: C, 50.33; H, 5.63; N, 4.35. Found: C, 50.86;
H, 5.70; N, 4.37.
Gen er a l P r oced u r e for Syn th esis of Den d r itic N,N′-
Bis(2-h yd r oxyeth yl)d ia m id es ([Gn ]-NH(CH2)2OH, n ) 0
to 3) 18-21. A mixture of dibromide 10 (1.0 mol equiv), 4-tert-
butylphenol or dendritic phenol [Gn]-OH13 (2.3 mol equiv), and
potassium carbonate (3.5 mol equiv) in acetone was heated
under reflux. The reaction time required was 3 d for n ) 0-2
and 5 d for n ) 3, respectively. The reaction mixture was
filtered through a pad of silica gel to remove the inorganic
materials. The filtrate was concentrated and the crude oily
substance purified as described in the following text.
2,2-Bis(4-(ben zyloxy)ben zyl) Ma lon ic Acid (15). A mix-
ture of the diester 13 (5.8 g, 10.5 mmol) and aqueous sodium
hydroxide (35 mL, 4.0 M) in ethanol (100 mL) was stirred at
90 °C for 1 d. The mixture was cooled to 0 °C and acidified
with concentrated HCl to pH 2. The mixture was extracted
with ethyl acetate (3 × 100 mL), and the combined organic
layers were washed with brine, dried (Na2SO4), and filtered.
Concentration of the filtrate followed by flash chromatography
on silica gel (hexane/ethyl acetate/ethanol ) 2/1/1) afforded
the diacid 15 (3.4 g, 65%) as a white solid: mp 168-170 °C;
Rf 0.44 (hexane/ethyl acetate/ethanol ) 2/1/1); IR (film) cm-1
3441, 1731; 1H-NMR (acetone-d6, COOH not observed) δ 3.22
(s, 4 H), 5.08 (s, 4 H), 6.93 (d, J ) 8.8 Hz, 4 H), 7.20 (d, J )
8.8 Hz, 4 H), 7.30-7.55 (m, 10 H); 13C-NMR (acetone-d6) δ 40.7,
61.2, 70.4, 115.4, 128.5, 129.3, 129.4, 131.8, 138.4, 158.9, 174.1;
MS (FAB, m/ z) 496 (M+, 35). Anal. Calcd for C31H28O6: C,
74.98; H, 5.68. Found: C, 75.20; H, 5.70. The monoacid 14
was obtained in 27% yield as a white solid: mp 138-140 °C;
Rf 0.78 (hexane/ethyl acetate/ethanol ) 2/1/1); IR (film) cm-1
3441, 1703; 1H-NMR (acetone-d6, COOH not observed) δ 2.50-
3.10 (m, 5 H), 5.05 (s, 4 H), 6.90 (d, J ) 8.7 Hz, 4 H), 7.13 (d,
J ) 8.6 Hz, 4 H), 7.25-7.50 (m, 10 H); MS (EI, m/ z) 452 (M+,
3), 161 (28), 131 (45), 117 (74), 91 (100). Anal. Calcd for
C30H28O4: C, 79.62; H, 6.24. Found: C, 79.47; H, 6.17.
N,N′-Bis(2-h yd r oxyeth yl)-2,2-bis(4-(ben zyloxy)ben zyl)-
1,3-p r op a n ed ia m id e (16). Oxalyl chloride (1.6 mL) was
added dropwise to a stirred solution of DMF (1.4 mL) in dry
dichloromethane (10 mL) at 0 °C over 5 min. A solution of
the diacid 15 (2.5 g, 5.0 mmol) in dry dichloromethane (20 mL)
was added in one portion. After stirring for 30 min, a solution
of 2-aminoethanol (2 mL) in acetonitrile (20 mL) was added
to the solution at 0 °C. After 30 min, the reaction mixture
was filtered and the filtrate concentrated under reduced
pressure. The residue was purified by flash chromatography
(hexane/ethyl acetate/ethanol ) 4/2/1) to give the diamide 16
(2.0 g, 69%) as a white solid: mp 134-136 °C; Rf 0.35 (hexane/
[G0]-NH(CH2)2OH 18. This was prepared from 4-tert-
butylphenol and purified by flash chromatography on silica
gel (hexane/ethyl acetate ) 1/3) to give [G0]-NH(CH2)2OH 18
(60%) as a white foam: Rf 0.15 (hexane/ethyl acetate ) 1/3);
1
IR (film) cm-1 3600-3100, 1660; H-NMR (CDCl3) δ 1.28 (s,
18 H), 2.19 (quin, J ) 6.0 Hz, 4 H), 2.84 (br s, 2 H), 3.23 (br s,
4 H), 3.30 (q, J ) 4.9 Hz, 4 H), 3.57 (t, J ) 4.6 Hz, 4 H), 4.07
(t, J ) 6.1 Hz, 4 H), 4.10 (t, J ) 6.1 Hz, 4 H), 6.76 (d, J ) 8.6
Hz, 4 H), 6.83 (d, J ) 8.8 Hz, 4 H), 7.05 (d, J ) 8.6 Hz, 4 H),
7.28 (d, J ) 8.8 Hz, 4 H), 7.77 (t, J ) 5.4 Hz, 2 H); 13C-NMR
(CDCl3) δ 29.4, 31.5, 34.0, 42.5, 42.9, 59.5, 61.8, 64.5, 64.6,
114.1, 114.4, 126.2, 128.4, 130.6, 143.5, 156.6, 158.1, 173.4;
MS (FAB, m/ z) 783 (M+, 28). Anal. Calcd for C47H62N2O8:
C, 72.10; H, 7.98; N, 3.58. Found: C, 72.29; H, 8.05; N, 3.51.
[G1]-NH(CH2)2OH 19. This was prepared from [G1]-OH
613 and purified by flash chromatography on silica gel (hexane/
ethyl acetate ) 1/2) to yield [G1]-NH(CH2)2OH 19 (65%) as a
white foam; Rf 0.15 (hexane/ethyl acetate ) 1/2); IR (film) cm-1
3600-3100, 1667; 1H-NMR (CDCl3) δ 1.29 (s, 36 H), 2.21 (quin,
J ) 6.1 Hz, 12 H), 2.52 (br s, 2 H), 3.24 (s, 4 H), 3.32 (q, J )
5.1 Hz, 4 H), 3.59 (br s, 4 H), 4.08 (t, J ) 5.8 Hz, 12 H), 4.11
(t, J ) 5.9 Hz, 12 H), 6.09 (s, 6 H), 6.79 (d, J ) 8.7 Hz, 4 H),
6.84 (d, J ) 8.9 Hz, 8 H), 7.06 (d, J ) 8.7 Hz, 4 H), 7.29 (d, J
) 8.9 Hz, 8 H), 7.63 (t, J ) 5.6 Hz, 2 H); 13C-NMR (CDCl3) δ
29.3, 31.4, 33.9, 42.4, 42.8, 59.4, 61.7, 64.4, 64.6, 94.4, 114.1,
114.4, 126.2, 128.5, 130.6, 143.5, 156.7, 158.0, 160.9, 173.5;
1
ethyl acetate/ethanol ) 4/2/1); IR (film) cm-1 3346, 1660; H-
NMR (CDCl3) δ 2.76 (t, J ) 4.3 Hz, 2 H), 3.25 (s, 4 H), 3.31 (q,
J ) 5.3 Hz, 4 H), 3.56 (q, J ) 4.3 Hz, 4 H), 5.00 (s, 4 H), 6.85
(d, J ) 8.7 Hz, 4 H), 7.07 (d, J ) 8.7 Hz, 4 H), 7.30-7.42 (m,
10 H), 7.69 (t, J ) 5.5 Hz, 2 H); 13C-NMR (CDCl3) δ 42.4, 42.6,
59.4, 61.6, 69.9, 114.6, 127.4, 127.8, 128.4, 128.5, 130.5, 136.9,
157.7, 173.4; MS (FAB, m/ z) 583 (M + H+, 100). Anal. Calcd
for C35H38N2O6: C, 72.14; H, 6.57; N, 4.81. Found: C, 72.06;
H, 6.54; N, 4.79.
MS (FAB, m/ z) 1495 (M+, 5). Anal. Calcd for C91H118N2O16
:
C, 73.06; H, 7.95; N, 1.87. Found: C, 72.76; H, 8.11; N, 1.74.
[G2]-NH(CH2)2OH 20. This was prepared from [G2]-OH
713 and purified by flash chromatography on silica gel (hexane/
ethyl acetate ) 1/2) to afford [G2]-NH(CH2)2OH 20 (50%) as a
white foam: Rf 0.29 (hexane/ethyl acetate ) 1/2); IR (film) cm-1
N,N′-Bis(2-h yd r oxyeth yl)-2,2-bis(4-h yd r oxyben zyl)-1,3-
p r op a n ed ia m id e (17). A suspension of the dibenzyl ether
16 (1.0 g, 1.71 mmol) and 10% palladium-on-charcoal (100 mg)
in absolute ethanol (30 mL) was stirred under an atmospheric
pressure of hydrogen at room temperature. After 3 h, the
suspension was filtered through a pad of Celite and the filtrate
concentrated. The residue was recrystallized from ethyl
acetate-chloroform (1/1) to give the diphenol 17 (0.55 g, 88%)
as a white solid: mp 199-200 °C; Rf 0.42 (ethyl acetate/ethanol
1
3600-3100; H-NMR (CDCl3) δ (s, 72 H), 2.23 (quin, J ) 6.0
Hz, 28 H), 3.22 (br s, 4 H), 3.27-3.33 (m, 4 H), 3.52-3.62 (m,
4 H), 4.07-4.13 (m, 56 H), 6.08 (s, 18 H), 6.78 (d, J ) 8.7 Hz,
4 H), 6.84 (d, J ) 8.8 Hz, 16 H), 7.05 (d, J ) 8.6 Hz, 4 H), 7.28
(d, J ) 8.8 Hz, 16 H), 7.61 (t, J ) 5.4 Hz, 2 H); 13C-NMR
(CDCl3) δ 29.3, 31.5, 34.0, 42.5, 43.2, 59.4, 62.0, 64.4, 64.6,
94.2, 114.0, 114.4, 126.2, 128.3, 130.5, 143.4, 156.6, 158.0,
160.7, 173.4; MS (FAB, m/ z) 2924 (M+, 12). Anal. Calcd for
C179H230N2O32: C, 73.59; H, 7.93; N, 0.96. Found: C, 73.43;
H, 7.96; N, 0.90.
1
) 5/1); IR (KBr) cm-1 3500-2800, 1661; H-NMR (methanol-
d4) δ 3.39 (s, 4 H), 3.44 (t, J ) 6.0 Hz, 4 H), 3.69 (t, J ) 6.0
Hz, 4 H), 4.80-5.05 (br s, 6 H), 6.83 (d, J ) 8.6 Hz, 4 H), 7.17
(d, J ) 8.6 Hz, 4 H); 13C-NMR (acetone-d6) δ 42.9, 43.7, 61.0,
61.4, 115.9, 128.7, 131.6, 157.1, 174.8; MS (FAB, m/ z) 403 (M
+ H+, 40). Anal. Calcd for C21H26N2O6: C, 62.67; H, 6.51; N,
6.96. Found: C, 62.56; H, 6.54; N, 6.89.
[G3]-NH(CH2)2OH 21. This was prepared from [G3]-OH
813 and purified by flash chromatography on silica gel (hexane/
ethyl acetate ) 2/1 gradient to 1/1) to give [G3]-NH(CH2)2OH
21 (46%) as a white foam; Rf 0.15 (hexane/ethyl acetate ) 3/2);

Dendritic Bis(oxazoline)copper(II) Catalysts
J . Org. Chem., Vol. 62, No. 15, 1997 5125
IR (film) cm-1 3500-3150; 1H-NMR (CDCl3) δ 1.28 (s, 144 H),
2.22 (quin, J ) 5.9 Hz, 60 H), 2.45 (br s, 2 H), 3.20 (br s, 4 H),
3.28 (q, J ) 4.0 Hz, 4 H), 3.56 (t, J ) 4.8 Hz, 4 H), 3.98-4.18
(m, 120 H), 6.08 (s, 42 H), 6.76 (d, J ) 8.6 Hz, 4 H), 6.83 (d, J
) 8.9 Hz, 32 H), 7.04 (d, J ) 8.6 Hz, 4 H), 7.27 (d, J ) 8.8 Hz,
32 H), 7.62 (t, J ) 5.5 Hz, 2 H); 13C-NMR (CDCl3) δ 29.3, 29.4,
31.5, 34.0, 42.5, 43.3, 59.4, 61.9, 64.5, 64.6, 94.4, 114.1, 114.4,
126.2, 128.3, 130.5, 143.4, 156.6, 158.0, 160.8, 173.3. Anal.
Calcd for C355H454N2O64: C, 73.85; H, 7.93; N, 0.49. Found:
C, 73.48; H, 8.11; N, 0.47.
Gen er a l P r oced u r e for Syn th esis of Den d r itic N,N′-
Bis(2-br om oeth yl)d ia m id es ([Gn ]-NH(CH2)2Br , n ) 0 to
3) 22-25. Carbon tetrabromide (5.0 mol equiv) was added to
a solution of individual dendritic N,N′-bis(2-hydroxyethyl)-
diamide [Gn]-NHCH2CH2OH (1.0 mol equiv) and triphenylphos-
phine (5.0 mol equiv) in dry THF under nitrogen at 0 °C. The
reaction mixture was then stirred at room temperature for 3
h. The mixture was concentrated under reduced pressure, and
triphenylphosphine oxide was precipitated by addition of a
solvent mixture of hexane-ethyl acetate (5/1). The mixture
was filtered through a pad of silica gel, and the filter cake
thoroughly washed with the aforementioned solvent mixture.
The combined filtrates were concentrated, and the residue was
purified as described in the following text.
[G0]-NH (CH 2)2Br 22. This was prepared from [G0]-
NH(CH2)2OH 18 and purified by flash chromatography on
silica gel (hexane/ethyl acetate ) 3/1) to give [G0]-NH(CH2)2-
Br 22 (75%) as a white solid: mp 120-121 °C; Rf 0.37 (hexane/
ethyl acetate ) 2/1); IR (film) cm-1 3303, 1667; 1H-NMR
(CDCl3) δ 1.29 (s, 18 H), 2.22 (quin, J ) 6.0 Hz, 4 H), 3.25 (s,
4 H), 3.32 (t, J ) 6.0 Hz, 4 H), 3.58 (q, J ) 5.9 Hz, 4 H), 4.09
(t, J ) 5.9 Hz, 4 H), 4.14 (t, J ) 6.0 Hz, 4 H), 6.78 (d, J ) 8.6
Hz, 4 H), 6.85 (d, J ) 8.9 Hz, 4 H), 7.05 (d, J ) 8.6 Hz, 4 H),
7.29 (d, J ) 8.8 Hz, 4 H), 7.58 (t, J ) 5.5 Hz, 2 H); 13C-NMR
(CDCl3) δ 29.5, 31.3, 31.5, 34.1, 41.4, 43.1, 59.4, 64.5, 64.7,
114.1, 114.6, 126.2, 128.2, 130.6, 143.6, 156.7, 158.2, 172.6;
MS (FAB, m/ z) 909 (M + H+, 36). Anal. Calcd for C47H60N2O6-
Br2: C, 62.12; H, 6.65; N, 3.08. Found: C, 62.12; H, 6.70; N
3.06.
[G1]-NH (CH 2)2Br 23. This was prepared from [G1]-
NH(CH2)2OH 19 and purified by flash chromatography on
silica gel (hexane/ethyl acetate ) 2/1) to give [G1]-NH(CH2)2-
Br 23 (80%) as a white foam; Rf 0.30 (hexane/ethyl acetate )
2/1); IR (film) cm-1 3347, 1668; 1H-NMR (CDCl3) δ 1.29 (s, 36
H), 2.20 (quin, J ) 6.0 Hz, 12 H), 3.24 (s, 4 H), 3.29 (t, J ) 6.0
Hz, 4 H), 3.55 (q, J ) 5.9 Hz, 4 H), 4.08 (t, J ) 5.8 Hz, 12 H),
4.10 (t, J ) 5.9 Hz, 12 H), 6.09 (s, 6 H), 6.77 (d, J ) 8.7 Hz, 4
H), 6.83 (d, J ) 8.8 Hz, 8 H), 7.04 (d, J ) 8.7 Hz, 4 H), 7.28 (d,
J ) 8.8 Hz, 8 H), 7.67 (t, J ) 5.7 Hz, 2 H); 13C-NMR (CDCl3)
δ 29.4, 31.2, 31.5, 34.0, 41.4, 43.1, 59.3, 64.5, 64.7, 94.4, 114.1,
114.5, 126.1, 128.2, 130.5, 143.4, 156.6, 158.0, 160.8, 172.5;
MS (FAB, m/ z) 1621 (M + H+, 10). Anal. Calcd for
C91H116N2O14Br2: C, 67.40; H, 7.21; N, 1.73. Found: C, 67.44;
H, 7.17; N, 1.72.
) 8.5 Hz, 32 H), 7.03 (d, J ) 8.4 Hz, 4 H), 7.28 (d, J ) 8.5 Hz,
32 H), 7.63 (t, J ) 5.4 Hz, 2 H); 13C-NMR (CDCl3) δ 29.3, 31.2,
31.5, 34.0, 41.4, 43.3, 53.9, 64.5, 64.6, 94.3, 114.1, 114.5, 126.1,
128.2, 128.3, 130.4, 143.4, 156.6, 158.0, 160.8, 172.5. Anal.
Calcd for C355H452N2O62Br2: C, 72.28; H, 7.72; N, 0.47.
Found: C, 72.30; H, 7.81; N 0.46.
Gen er a l P r oced u r e for Syn th esis of Den d r itic Bis-
(oxa zolin e)s (Gn , n ) 0-3) 1-4. To a solution of N,N′-bis-
(2-bromoethyl)diamide [Gn]-NHCH2CH2Br (0.1 mmol) in a
solvent mixture of EtOH-THF (10 mL, 1/1) was added sodium
hydroxide solution (10.0 mol equiv, 1 M). The mixture was
stirred under reflux for 8 h and quenched with water (10 mL).
The aqueous phase was extracted with ethyl acetate (3 × 20
mL), and the combined organic layers were washed with brine,
dried (Na2SO4), and filtered. Concentration of the filtrate gave
a crude pale yellow oily substance. The crude product was
then purified as described in the following text.
Bis(oxa zolin e) 28. To a solution of the dialcohol 16 (3.00
g, 5.16 mmol) and triphenylphosphine (4.10 g, 15.7 mmol) in
dry THF (25 mL) was added carbon tetrabromide (5.20 g, 15.7
mmol) under nitrogen at 0 °C. The reaction mixture was
stirred at room temperature for 3 h. Concentration of the
mixture to near-dryness followed by addition of a solvent
mixture of hexane-ethyl acetate (5/1) effectively precipitated
most of the triphenylphosphine oxide. The mixture was
filtered through a pad of silica gel and washed with the
aforementioned solvent mixture. The combined filtrates were
evaporated under reduced pressure to give a pale yellow oil.
The crude product was then purified by flash chromatography
(hexane/ethyl acetate ) 3/1) to give the corresponding di-
bromide (3.0 g, 82%) as a white solid, mp 127-129 °C; Rf 0.30
(hexane/ethyl acetate ) 2/1); IR (film) cm-1 3346, 1667; 1H-
NMR (CDCl3) δ 3.26 (s, 4 H), 3.29 (t, J ) 4.1 Hz, 4 H), 3.57 (q,
J ) 5.9 Hz, 4 H), 5.01 (s, 4 H), 6.85 (d, J ) 8.6 Hz, 4 H), 7.07
(d, J ) 8.6 Hz, 4 H), 7.30-7.50 (m, 10 H), 7.68 (t, J ) 5.6 Hz,
2 H); 13C-NMR (CDCl3) δ 31.3, 41.4, 43.1, 59.3, 70.0, 114.8,
127.4, 127.9, 128.4, 128.5, 130.5, 137.0, 157.9, 172.5; MS (L-
SIMS, m/ z) 709 (M + H+, 100). Anal. Calcd for C35H36N2O4-
Br2: C, 59.34; H, 5.12; N, 3.95. Found: C, 59.22; H, 5.12; N,
3.95. To a solution of this dibromide (2.30 g, 3.25 mmol) in
EtOH-THF (40 mL, 3/1) was added sodium hydroxide solution
(25 mL, 0.5 M). The mixture was stirred under reflux for 12
h and poured into water (20 mL). The aqueous phase was
extracted with ethyl acetate (3 × 40 mL), and the combined
organic layers were washed with brine, dried (Na2SO4), and
filtered. Concentration of the filtrate gave a crude pale yellow
oily substance which was then purified by flash chromatog-
raphy (hexane/ethyl acetate ) 1/2 gradient to ethyl acetate)
to give the bis(oxazoline) 28 (3.0 g, 82%) as a white solid: mp
160-162 °C; Rf 0.29 (hexane/ethyl acetate ) 1/3); IR (film)
1
cm-1 1659; H-NMR (CDCl3) δ 3.23 (s, 4 H), 3.79 (t, J ) 9.5
Hz, 4 H), 4.21 (t, J ) 9.5 Hz, 4 H), 5.03 (s, 4 H), 6.88 (d, J )
8.7 Hz, 4 H), 7.14 (d, J ) 8.6 Hz, 4 H), 7.30-7.50 (m, 10 H);
13C-NMR (CDCl3) δ 38.9, 48.7, 54.2, 67.3, 69.9, 114.4, 127.4,
127.9, 128.5, 129.1, 131.2, 137.1, 157.6, 167.4; MS (EI, m/ z)
546 (M+, 0.1), 348 (50), 155 (9), 91 (70), 57 (100). Anal. Calcd
for C35H34N2O4: C, 76.90; H, 6.27; N, 5.12. Found: C, 76.68;
H, 6.26; N, 5.05.
[G2]-NH (CH 2)2Br 24. This was prepared from [G2]-
NH(CH2)2OH 20 and purified by flash chromatography on
silica gel (hexane/ethyl acetate ) 3/2) to give [G2]-NH(CH2)2-
Br 24 (85%) as a white foam; Rf 0.21 (hexane/ethyl acetate )
1
3/2); IR (film) cm-1 3379; H-NMR (CDCl3) δ 1.28 (s, 72 H),
Den d r itic Bis(oxa zolin e) G0 1. This was prepared from
[G0]-NH(CH2)2Br 22 and purified by flash chromatography on
silica gel (hexane/ethyl acetate ) 2/1 gradient to ethyl acetate)
to afford G0 1 (96%) as a colorless oil: Rf 0.30 (hexane/ethyl
2.20 (quin, J ) 5.9 Hz, 28 H), 3.23 (s, 4 H), 3.29 (t, J ) 5.7 Hz,
4 H), 3.55 (q, J ) 5.9 Hz, 4 H), 4.07-4.13 (m, 56 H), 6.08 (s,
18 H), 6.77 (d, J ) 8.7 Hz, 4 H), 6.83 (d, J ) 8.8 Hz, 16 H),
7.03 (d, J ) 8.6 Hz, 4 H), 7.28 (d, J ) 8.8 Hz, 16 H), 7.58 (t, J
) 5.3 Hz, 2 H); 13C-NMR (CDCl3) δ 29.4, 31.2, 31.5, 34.0, 41.4,
43.2, 59.3, 64.5, 64.6, 64.7, 94.4, 114.1, 114.5, 126.2, 128.2,
128.3, 130.5, 143.5, 156.6, 158.1, 160.8, 172.5; MS (FAB, m/ z)
3044 (M + H+, 4). Anal. Calcd for C179H228N2O30Br2: C, 70.55;
H, 7.54; N, 0.92. Found: C, 70.34; H, 7.81; N, 0.71.
1
acetate ) 1/3); IR (film) cm-1 1658; H-NMR (CDCl3) δ 1.29
(s, 18 H), 2.22 (quin, J ) 6.0 Hz, 4 H), 3.22 (s, 4 H), 3.78 (t, J
) 9.5 Hz, 4 H), 4.11 (t, J ) 6.0 Hz, 4 H), 4.12 (t, J ) 6.1 Hz,
4 H), 4.19 (t, J ) 9.6 Hz, 4 H), 6.80 (d, J ) 8.7 Hz, 4 H), 6.84
(d, J ) 8.9 Hz, 4 H), 7.12 (d, J ) 8.6 Hz, 4 H), 7.29 (d, J ) 8.7
Hz, 4 H); 13C-NMR (CDCl3) δ 29.4, 31.4, 32.9, 39.0, 48.7, 54.2,
64.4, 67.2, 114.0, 126.1, 128.8, 131.1, 143.3, 156.5, 157.7, 167.3;
MS (FAB, m/ z) 747 (M + H+, 95); HRMS calcd for C47H59N2O6
747.4365, found 747.4370. Anal. Calcd for C47H58N2O6: C,
75.57; H, 7.83; N, 3.75. Found: C, 75.77; H, 8.14; N, 3.28.
Den d r itic Bis(oxa zolin e) G1 2. This was prepared from
[G1]-NH(CH2)2Br 23 and purified by flash chromatography on
silica gel (hexane/ethyl acetate ) 2/1 gradient to ethyl acetate)
to afford G1 2 (89%) as a white foam: Rf 0.35 (hexane/ethyl
[G3]-NH (CH 2)2Br 25. This was prepared from [G3]-
NH(CH2)2OH 21 and purified by flash chromatography on
silica gel (hexane/ethyl acetate ) 3/1 gradient to 1/1) to provide
[G3]-NH(CH2)2Br 25 (79%) as a white foam: Rf 0.50 (hexane/
ethyl acetate ) 5/2); IR (film) cm-1 3500-3300; 1H-NMR
(CDCl3) δ 1.28 (s, 144 H), 2.15-2.25 (m, 60 H), 3.21 (s, 4 H),
3.28 (t, J ) 5.8 Hz, 4 H), 3.54 (q, J ) 5.7 Hz, 4 H), 3.90-4.20
(m, 120 H), 6.09 (s, 42 H), 6.76 (d, J ) 8.4 Hz, 4 H), 6.83 (d, J

5126 J . Org. Chem., Vol. 62, No. 15, 1997
Chow and Mak
1
acetate ) 1/3); IR (film) cm-1 1660; H-NMR (CDCl3) δ 1.29
31.0, 32.3, 42.4, 61.8, 119.7, 151.3, 153.3, 165.0; MS (EI, m/ z)
211 (M+, 7), 168 (11), 155 (13), 142 (9), 124 (75), 88 (74), 81
(79), 68 (79), 55 (100). Anal. Calcd for C11H17NO3: C, 62.54;
H, 8.11; N, 6.63. Found: C, 62.15; H, 8.06; N, 6.37.
(s, 36 H), 2.21 (quin, J ) 5.9 Hz, 12 H), 3.22 (br s, 4 H), 3.78
(t, J ) 9.6 Hz, 4 H), 4.09 (t, J ) 5.5 Hz, 12 H), 4.11 (t, J ) 5.8
Hz, 12 H), 4.20 (t, J ) 9.5 Hz, 4 H), 6.10 (s, 6 H), 6.81 (d, J )
8.4 Hz, 4 H), 6.84 (d, J ) 8.8 Hz, 8 H), 7.12 (d, J ) 8.5 Hz, 4
H), 7.28 (d, J ) 8.8 Hz, 8 H); 13C-NMR (CDCl3) δ 29.4, 31.5,
34.0, 39.2, 48.9, 54.2, 64.5, 64.7, 67.3, 94.3, 114.1, 126.1, 129.0,
131.2, 143.4, 156.6, 157.8, 160.8, 167.4; MS (FAB, m/ z) 1460
(M + H+, 100). Anal. Calcd for C91H114N2O14: C, 74.87; H,
7.87; N, 1.92. Found: C, 74.87; H, 7.86; N, 1.82.
Gen er a l P r oced u r e for P r ep a r a t ion of Au t h en t ic
Sa m p les of Diels-Ald er Cycloa d d u ct. A mixture of pre-
formed G0‚Cu(II) complex (0.03-0.05 mol equiv) (see experi-
mental of kinetic section described below), N-alkenoyl-2-
oxazolidinone (1.0 mol equiv), and cyclopentadiene (10 mol
equiv) in anhydrous CH2Cl2 was stirred at room temperature
for 1 d. The mixture was filtered through a pad of silica gel
and washed with CH2Cl2. The combined filtrates were con-
centrated, and the residue was purified as described in the
following text. Cycloadduct 27 was prepared according to
literature method.24
Den d r itic Bis(oxa zolin e) G2 3. This was prepared from
[G2]-NH(CH2)2Br 24 and purified by flash chromatography on
silica gel (hexane/ethyl acetate ) 2/1 gradient to ethyl acetate)
to afford G2 3 (58%) as a white foam: Rf 0.40 (hexane/ethyl
1
acetate ) 1/3); IR (film) cm-1 1667; H-NMR (CDCl3) δ 1.28
(s, 72 H), 2.20 (quin, J ) 6.0 Hz, 28 H), 3.21 (s, 4 H), 3.77 (t,
J ) 9.5 Hz, 4 H), 3.95-4.05 (m, 56 H), 4.18 (t, J ) 8.9 Hz, 4
H), 6.08 (s, 12 H), 6.09 (s, 6 H), 6.80 (d, J ) 8.5 Hz, 4 H), 6.83
(d, J 8.9 Hz, 16 H), 7.12 (d, J ) 8.7 Hz, 4 H), 7.28 (d, J ) 8.9
Hz, 16 H); 13C-NMR (CDCl3) δ 29.4, 31.5, 34.0, 39.3, 49.0, 54.3,
64.6, 64.7, 67.3, 94.5, 114.1, 126.2, 129.1, 131.2, 143.5, 156.7,
157.8, 160.8, 167.4; MS (MALDI, m/ z) 2883 (M+, 48). Anal.
Calcd for C179H226N2O30: C, 74.50; H, 7.89; N, 0.97. Found:
C, 74.75; H, 8.04; N, 0.94.
Den d r itic Bis(oxa zolin e) G3 4. This was prepared from
[G3]-NH(CH2)2Br 25 and purified by flash chromatography on
silica gel (hexane/ethyl acetate ) 2/1 gradient to ethyl acetate)
to give G3 4 (78%) as a white foam: Rf 0.15 (hexane/ethyl
acetate ) 1/1); 1H-NMR (CDCl3) δ 1.28 (s, 144 H), 2.20 (quin,
J ) 5.3 Hz, 60 H), 3.21 (s, 4 H), 3.76 (t, J ) 9.7 Hz, 4 H),
3.95-4.12 (m, 120 H), 4.17 (t, J ) 9.3 Hz, 4 H), 6.08 (s, 42 H),
6.80 (d, J ) 8.7 Hz, 4 H), 6.83 (d, J ) 8.8 Hz, 32 H), 7.12 (d,
J ) 8.7 Hz, 4 H), 7.27 (d, J ) 8.8 Hz, 32 H); 13C-NMR (CDCl3)
δ 29.3, 31.4, 33.9, 39.2, 48.9, 54.2, 64.4, 64.5, 67.2, 94.3, 114.0,
126.1, 128.9, 131.1, 143.3, 156.6, 157.7, 158.0, 160.7, 167.3;
MS (MALDI, m/ z) 5731 (M+, 19). Anal. Calcd for
C355H450N2O62: C, 74.32; H, 7.91; N, 0.49. Found: C, 74.50;
H, 8.15; N, 0.46.
Gen er a l P r oced u r e for t h e Alk en oyla t ion of 2-Oxa -
zolid in on e. To a solution of 2-oxazolidinone in anhydrous
THF (∼0.5 M) at -78 °C was added n-butyllithium (1.0 mol
equiv in hexane solution). After 15 min, freshly distilled
alkenoyl chloride (1.1 mol equiv) was added. The mixture was
stirred at -78 °C for 30 min and at 0 °C for 15 min. The
reaction was quenched with excess saturated ammonium
chloride solution and the solvent evaporated under reduced
pressure. The resulting slurry was diluted with dichloro-
methane and washed with brine. The organic layer was dried
(MgSO4), filtered, and concentrated under reduced pressure.
The product was purified by flash chromatography on silica
gel to afford the desired N-alkenoyl-2-oxazolidinone.
3-((2E)-P en ten oyl)-2-oxa zolid in on e (29). 2-Oxazolidi-
none was acylated with (E)-2-pentenoyl chloride23 according
to the general procedure detailed above and purified by flash
chromatography on silica gel (hexane/ethyl acetate ) 4/1) to
afford the title compound 29 (22%) as an oil: Rf 0.29 (hexane/
ethyl acetate ) 2/1); IR (film) cm-1 1778, 1682, 1634; 1H-NMR
(CDCl3) δ 1.11 (t, J ) 7.4 Hz, 3 H), 2.20-2.40 (m, 2 H), 4.08
(t, J ) 8.1 Hz, 2 H), 4.44 (t, J ) 8.0 Hz, 2 H), 7.15-7.35 (m, 2
H); 13C-NMR (CDCl3) δ 11.9, 25.4, 42.4, 61.8, 118.8, 152.3,
153.3, 165.0; MS (EI, m/ z) 169 (M+, 10), 140 (7), 84 (100), 82
(100), 55 (85). HRMS calcd for C8H12NO3 170.0812, found
170.0821.
3-((3′-Eth ylbicyclo[2.2.1]h ep t-5′-en -2′-yl)ca r bon yl)-1,3-
oxa zolid in -2-on e (31). This was prepared from 3-((2E)-
pentenoyl)-2-oxazolidinone (29) and purified by flash chroma-
tography on silica gel (hexane/ethyl acetate ) 4/1) to afford
cycloadduct 31 (83%) as a white solid: mp 56-58 °C; Rf 0.30
(hexane/ethyl acetate ) 4/1); IR (film) cm-1 1777, 1697; 1H-
NMR (CDCl3) δ 0.92 (t, J ) 7.4 Hz, 3 H), 1.42-2.00 (m, 5 H),
2.66 (br s, 1 H), 3.27 (br s, 1 H), 3.59 (t, J ) 3.9 Hz, 1 H),
3.90-4.10 (m, 2 H), 4.42 (t, J ) 8.1 Hz, 2 H), 5.79 (dd, J )
2.7, 5.6 Hz, 1 H), 6.37 (dd, J ) 3.2, 5.4 Hz, 1 H); MS (EI,
m/ z) 235 (M+, 12), 170 (100), 149 (22), 128 (26), 83 (100), 66
(100). Anal. Calcd for C13H17NO3: C, 62.36; H, 7.28; N, 5.95.
Found: C, 62.19; H, 7.28; N, 5.62.
3-((3′-P en tylbicyclo[2.2.1]h ep t-5′-en -2′-yl)ca r bon yl)-1,3-
oxa zolid in -2-on e (32). This was prepared from 3-((2E)-
octenoyl)-2-oxazolidinone (30) and purified by flash chroma-
tography on silica gel (hexane/ethyl acetate ) 5/1) to afford
cycloadduct 32 (84%) as a colorless oil: Rf 0.40 (hexane/ethyl
1
acetate ) 4/1); IR (film) cm-1 1778, 1698; H-NMR (CDCl3) δ
0.80-2.05 (m, 14 H), 2.64 (br s, 1 H), 3.26 (br s, 1 H), 3.58 (dd,
J ) 3.5, 4.4 Hz, 1 H), 3.90-4.10 (m, 2 H), 4.41 (t, J ) 8.2 Hz,
2 H), 5.78 (dd, J ) 2.8, 5.7 Hz, 1 H), 6.35 (dd, J ) 3.2, 5.8 Hz,
1 H); MS (EI, m/ z) 277 (M+, 6), 211 (72), 170 (6), 125 (10), 91
(21), 66 (100). HRMS calcd for C16H24NO3 278.1751, found
278.1746.
2. Kin etic Exp er im en ts. Unless otherwise indicated,
anhydrous dichloromethane was the solvent used in the kinetic
experiments as well as in preparing the stock solutions. All
reactions were stirred at 25.0 ( 0.1 °C in a temperature control
water bath and reactions monitored by taking aliquots (2 µL)
at recorded time and analyzed by gas chromatography. For
each aliquot, the integrated area under the dienophile (R) or
Diels-Alder cycloadduct (P) was corrected according to their
sensitivity ratio and such ratio was determined by injecting a
sample of known concentration of R and P. The percent
conversion for each reaction was evaluated according to the
areas of dienophile and Diels-Alder adduct. GC conditions:
injector temperature, 220 °C; detector temperature, 250 °C,
FID detector, 180 °C isothermal oven, 50 mL/min carrier gas
flow rate; retention times, dienophile 26 (R) ) 4.3 min; Diels-
Alder cycloadduct 27 (P) ) 11.5 min.
Esta blish m en t of th e Ra te Equ a tion of th e Diels-
Ald er Rea ction Usin g Bis(oxa zolin e)cop p er (II) Com p lex
28 a s th e Mod el Ca ta lyst. (a ) Va r ia tion of Cp Con cen tr a -
t ion . A stock metal complex solution (A) was prepared by
charging a 25.00 mL volumetric flask with bis(oxazoline) 28
(0.150 g, 0.28 mmol) and Cu(OTf)2 (0.100 g, 0.27 mmol) and
diluting to the mark. The resulting mixture was stirred
vigorously until the complete dissolution of the solid Cu(OTf)2.
A 5.00 mL stock solution (B) of dienophile 26 (0.703 M) was
prepared by proper weighing. To five reaction tubes were
charged 0.50 mL of each of these two stock solutions. To
another five 10.00 mL volumetric flasks were charged 40, 80,
120, 160, and 200 mg of freshly distilled Cp, respectively, and
diluted to the mark. From each of these volumetric flasks,
0.10 mL was pipetted into the reaction mixtures inside the
five reaction tubes prepared above. The resulting mixtures
were stirred at 25.0 ( 0.1 °C and aliquots taken at recorded
3-((2E)-Octen oyl)-2-oxa zolid in on e (30). 2-Oxazolidinone
was acylated with (E)-2-octenoyl chloride23 according to the
general procedure detailed above and purified by flash chro-
matography on silica gel (hexane/ethyl acetate ) 3/1) to afford
the title compound 30 (88%) as a white solid: mp 38-39 °C;
Rf 0.23 (hexane/ethyl acetate ) 3/1); IR (film) cm-1 1789, 1770,
1682, 1634; 1H-NMR (CDCl3) δ 0.89 (t, J ) 6.7 Hz, 3 H), 1.25-
1.40 (m, 4 H), 1.49 (quin, J ) 7.1 Hz, 2 H), 2.28 (dt, J ) 5.5,
7.3 Hz, 2 H), 4.06 (t, J ) 8.2 Hz, 2 H), 4.43 (t, J ) 8.1 Hz, 2
H), 7.10-7.20 (m, 2 H); 13C-NMR (CDCl3) δ 13.7, 22.1, 27.5,
(23) Lauk, U. H.; Skrabal, P.; Zollinger, H. Helv. Chim. Acta 1985,
68, 1406-1426.
(24) Narasaka, K.; Iwasawa, N.; Inoue, M.; Yamada, T.; Nakashima,
M.; Sugimori, J . J . Am. Chem. Soc. 1989, 111, 5340-5345.

Dendritic Bis(oxazoline)copper(II) Catalysts
J . Org. Chem., Vol. 62, No. 15, 1997 5127
time and analyzed by gas chromatography as described in the
general conditions.
described in section a except the stock metal complex solution
was prepared from G2 dendrimer 3 (0.00294 M). The reaction
mixtures were stirred at 25.0 ( 0.1 °C and aliquots taken at
recorded time and analyzed by gas chromatography as detailed
in the general conditions.
(d ) Va r ia tion of Dien op h ile Con cen tr a tion for G3-
Cu (II) Com p lex. The kinetic experiment was set up as
described in section (a) except the stock metal complex solution
(G) prepared from G3 dendrimer 4 (0.0030 M). The reaction
mixtures were stirred at 25.0 ( 0.1 °C and aliquots taken at
recorded time and analyzed by gas chromatography as detailed
in the general conditions.
Com p etitive Exp er im en ts w ith Tw o Dien op h iles. (a )
Su bstr a te Selectivity for Bis(oxa zolin e)-Cu (II) Com p lex
28‚Cu (OTf)2. A metal complex solution (0.0018 M) was
prepared by diluting one-sixth of the stock metal complex
solution (A) (0.011 M) prepared above. A 1.00 mL stock Cp
solution (7.73 M) and a 10.00 mL stock solution containing
both dienophile 29 and 30 (0.03 M each) were prepared by
proper weighing. To a reaction tube was charged 0.30 mL of
the metal complex solution and 0.10 mL of each of the stock
Cp and dienophile solutions. The reaction mixture was stirred
at 25.0 ( 0.1 °C, and aliquots were taken (0.001 mL) at
recorded time and analyzed by gas chromatography. For each
aliquot, the integrated area under any dienophiles (R) or
Diels-Alder adducts (P) were corrected according to their
respective sensitivity ratios and these two ratios determined
by injecting samples with known concentration of R and P.
GC conditions: injector temperature, 220 °C; detector tem-
perature, 250 °C, FID detector, 180 °C isothermal oven, 35
mL/min carrier gas flow rate; retention times, dienophile 29
) 3.1 min; dienophile 30 ) 7.5 min; Diels-Alder cycloadduct
31 ) 8.3 min; Diels-Alder cycloadduct 32 ) 20.4 min.
(b) Su bstr a te Selectivity for G1-Cu (II) Com p lex. The
kinetic experiment was set up as described above except the
G1 metal complex solution (0.00193 M) was prepared by
diluting two-fifths of the stock metal complex (F) (0.0048 M)
prepared above. The reaction mixture was stirred at 25.0 (
0.1 °C, and aliquots were taken (0.001 mL) at recorded time
and analyzed by gas chromatography as described above.
(c) Su bstr a te Selectivity for G3-Cu (II) Com p lex. The
kinetic experiment was set up as described above except the
G3 metal complex solution (0.002 M) prepared by diluting two-
third of the stock metal complex (G) (0.003 M) prepared above.
The reaction mixture was stirred at 25.0 ( 0.1 °C, and aliquots
were taken (0.001 mL) at recorded time and analyzed by gas
chromatography as described above.
(b) Va r ia tion of Meta l Com p lex Con cen tr a tion .
A
stock solution (C) containing dienophile 26 and Cp was
prepared by charging a 2.00 mL volumetric flask with 26
(0.157 g, 1.01 mmol) and Cp (1.20 g, 18.2 mmol) and diluting
to the mark. To five reaction tubes were charged 1.00, 0.80,
0.60, 0.40 and 0.20 mL, respectively, of the stock metal complex
solution (A) and the last four tubes topped to 1.00 mL with
CH2Cl2. To each of these reaction tubes were then charged
0.20 mL of the stock solution (C) prepared above. The
resulting mixtures were stirred at 25.0 ( 0.1 °C and aliquots
taken at recorded time and analyzed by gas chromatography
as described in the general conditions.
(c) Va r ia tion of Dien op h ile Con cen tr a tion . A 3.00 mL
stock solution (D) of Cp (7.17 M) and a 2.00 mL stock solution
(E) of dienophile 26 (0.30 M) were prepared by proper
weighing. By proper dilution from the stock metal complex
solution (A), a 3.00 mL diluted stock metal complex solution
(0.00367 M) was prepared and 0.50 mL of it pipetted into five
reaction tubes followed by charging with 0.50, 0.40, 0.30, 0.20,
and 0.10 mL, respectively, of the stock dienophile solution (E)
and the last 4 tubes all topped to 1.00 mL with CH2Cl2. Then
0.50 mL of the stock Cp solution (D) was added into each of
the above reaction mixtures. The resulting mixtures were
stirred at 25.0 ( 0.1 °C and aliquots taken at recorded time
and analyzed by gas chromatography as described in the
general conditions.
Kin etic Exp er im en ts for Va r iou s Gen er a tion Den -
d r itic-Cu (II) Com p lexes. (a ) Va r ia tion of Dien op h ile
Con cen tr a tion for G0-Cu (II) Com p lex. A stock metal
complex solution was prepared by charging a 10.00 mL
volumetric flask with G0 dendrimer 1 (41.0 mg, 0.055 mmol),
Cu(OTf)2 (17.0 mg, 0.047 mmol) and diluting to the mark. The
resulting mixture was stirred vigorously until the complete
dissolution of the solid Cu(OTf)2. A 3.00 mL stock solution of
Cp (7.27 M) and a 2.00 mL stock solution of dienophile 103
(0.30 M) were prepared by proper weighing. To each of five
reaction tubes were charged 0.50 mL of the stock metal
complex solution followed by charging with 0.50, 0.40, 0.30,
0.20, and 0.10 mL, respectively, of the stock dienophile solution
and the last 4 tubes all topped to 1.00 mL. Then 0.50 mL of
the stock Cp solution was added into each of the above reaction
mixtures. The resulting mixtures were stirred at 25.0 ( 0.1
°C and aliquots taken at recorded time and analyzed by gas
chromatography as described in the general conditions.
(b) Va r ia tion of Dien op h ile Con cen tr a tion for G1-
Cu (II) Com p lex. The kinetic experiment was set up as
described above except the stock metal complex solution (F)
was prepared from G1 dendrimer 2 (0.0048 M). The reaction
mixtures were stirred at 25.0 ( 0.1 °C and aliquots taken at
recorded time and analyzed by gas chromatography as detailed
in the general conditions.
Ack n ow led gm en t. The authors wish to thank Dr.
T. L. Chan for helpful discussions, and the Research
Grants Council, Hong Kong, for the financial support
(Ref No.: CUHK466/95P).
(c) Va r ia tion of Dien op h ile Con cen tr a tion for G2-
Cu (II) Com p lex. The kinetic experiment was set up as
J O970383S