Reduced-symmetry Deep-cavity Cavitands

Article abstract of DOI:10.1021/ol006569m

Formula presented The syntheses of reduced-symmetry deep-cavity cavitands by two-stage stereoselective bridging with substituted benzal bromides is reported. Conditions for the optimal formation of the trisbridged derivatives were readily established. However, it was not possible to determine conditions which selectively promoted formation of either one of the two bisbridged species, or the monobridged compound, above the other products. A comparison of yields for A/B bisbridged derivatives verses A/C bisbridged derivatives may indicate that the one-pot formation of deep-cavity cavitands occurs primarily through the former species.

Full text of DOI:10.1021/ol006569m

ORGANIC
LETTERS
2000
Vol. 2, No. 24
3845-3848
Reduced-Symmetry Deep-Cavity
Cavitands
Jodie O. Green, John-Henry Baird, and Bruce C. Gibb*
Department of Chemistry, UniVersity of New Orleans, New Orleans, Louisiana 70148
bgibb@uno.edu
Received September 9, 2000
ABSTRACT
The syntheses of reduced-symmetry deep-cavity cavitands by two-stage stereoselective bridging with substituted benzal bromides is reported.
Conditions for the optimal formation of the trisbridged derivatives were readily established. However, it was not possible to determine conditions
which selectively promoted formation of either one of the two bisbridged species, or the monobridged compound, above the other products.
A comparison of yields for A/B bisbridged derivatives verses A/C bisbridged derivatives may indicate that the one-pot formation of deep-
cavity cavitands occurs primarily through the former species.
In the development of new host-guest systems, two inter-
related molecular information considerations exist, which
necessitates an increase in the scale of host molecules. In
the first instance, the host must simply be of sufficient scale
to bind or encapsulate an increasing range of ever larger
guests. Harder to attain perhaps is the need to pack
information tightly into the binding site. Thus, precisely
positioning the myriad of groups necessary for generating
highly specific binding requires a deft touch that we are still
not able to accomplish.
Investigations into the development of large host molecules
with concave cavities have been underway for some time
now and have focused on the use of calixarenes1 and
resorcinarenes.2,3 Enlarging the cavities of such molecules
can be carried out by a variety of approaches. For example,
bridging of resorcinarenes with 1,2-dihalo quinoxaline or
pyrazine groups has led to the development of the velcrands,
rigid molecules with deep, concave cavities.3-8 Taking the
host properties of these molecules further, Rebek et al.
developed preorganized variants capable of kinetically stable
guest binding in both organic9,10 and even aqueous media.11
Furthermore, corresponding dimers of these molecules, be
they held together by covalent12 or noncovalent forces,13 have
demonstrated a number of unusual binding properties such
as novel stereochemical relationships between encapsulated
guests.14
With a goal of forming large preorganized molecular
cavities, we recently demonstrated the efficient formation
of deep-cavity cavitands (DCCs) via the stereoselective
bridging of resorcinarenes with benzal bromide.15 More
recently, we noted that this process, involving the irreversible
formation of eight new covalent bonds and the generation
(7) Soncini, P.; Bonsignore, S.; Dalcanale, E.; Ugozzoli, F. J. Org. Chem.
1992, 57, 4608-4612.
(8) Cram, D. J.; Heung-Jin, C.; Bryant, J. A.; Knobler, C. B. J. Am.
Chem. Soc. 1992, 114, 7748-7756.
(9) Rudkevich, D. M.; Hilmersson, G.; Rebek, J., Jr. J. Am. Chem. Soc.
1997, 119, 9911-9912.
(1) Gutsche, C. D. Calixarenes; Royal Society of Chemistry: Cambridge,
1992.
(2) Ho¨gberg, A. G. S. J. Org. Chem. 1980, 45, 4498-4500.
(3) Cram, D. J.; Cram, J. M. Container Molecules and Their Guests;
Royal Society of Chemistry: Cambridge, 1994.
(4) Dalcanale, E.; Soncini, P.; Bacchilega, G.; Ugozzoli, F. Chem.
Commun. 1989, 500-503.
(5) Moran, J. R.; Ericson, J. L.; Dalcanale, E.; Bryant, J. A.; Knobler,
C. B.; Cram, D. J. J. Am. Chem. Soc. 1991, 113, 5707-5714.
(6) Cram, D. J.; Karbach, S.; Kim, H.-E.; Knobler, C. B.; Maverick, E.
F.; Ericson, J. L.; Helgeson, R. C. J. Am. Chem. Soc. 1988, 110, 2229-
2237.
(10) Rudkevich, D. M.; Hilmersson, G.; Rebek, J., Jr. J. Am. Chem. Soc.
1998, 120, 12216-12217.
(11) Haino, T.; Rudkevich, D. M.; Rebek, J., Jr. J. Am. Chem. Soc. 1999,
121, 11253-11254.
(12) Tucci, F. C.; Renslo, A. R.; Rudkevich, D. M.; Rebek, J., Jr. Angew.
Chem., Int. Ed. 2000, 39, 1076-1079.
(13) Renslo, A. R.; Tucci, F. C.; Rudkevich, D. M.; Rebek, J., Jr. J. Am.
Chem. Soc. 2000, 122, 4573-4582.
(14) Tucci, F. C.; Rudkevich, D. M.; Rebek, J., Jr. J. Am. Chem. Soc.
1999, 121, 4928-4929.
(15) Xi, H.; Gibb, C. L. D.; Stevens, E. D.; Gibb, B. C. Chem. Commun.
1998, 1743-1744.
10.1021/ol006569m CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/31/2000

Scheme 1. Possible Products from the Stereoselective Bridging of Resorcinarenes (Octols)a
a R ) CH2CH2Ph. For series a, X ) 3,5-dibromo. For series b, X ) 4-bromo.
of four stereogenic centers, is a general process applicable
to a variety of substituted benzal bromides.16 The resulting
functionalized deep-cavity cavitands, e.g., 6 (Scheme 1),
allow access to novel molecular subunits for the study of
irreversible self-assemblies17 and offer their large concave
cavities to many aspects of the host-guest chemistry field.
Here we demonstrate that this stereoselective bridging
process can be applied in a stepwise manner to the formation
of a variety of lower symmetry deep-cavity cavitands. We
believe that such derivatives will be of considerable utility
in the development of large molecular cavities possessing
detailed and/or specific binding information within their
structures.
In theory, the treatment of octol 1 with less than an excess
of a benzal bromide bridging material can give rise to five
products: the monobridged compound 2, the A/B and A/C
bisbridged species 3 and 4, the trisbridged derivative 5, and
the fully bridged, C4V DCC 6 (Scheme 1). With an excess of
bridging material, yields of the tetrabridged DCCs are
typically ca. 55% for benzal bromides whose substituents
do not give rise to deleterious side reactions.16 These yields
corresponds to an average of 86% diastereoselectivity per
bridge18 and, assuming that this diasteroeselectivity improves
with each subsequent bridge, are indicative of a first bridging
reaction occurring with at least 55% efficiency.
Initial investigations identified the stereoselective bridging
of phenethyl octol 1 (R ) CH2CH2Ph) with 3,5-dibromoben-
zal bromide as the reaction of choice for investigating partial
bridging, primarily because the yield of DCC 6a could be
raised from the previously reported 43%16 to 67% with a
slightly modified set of conditions. With an average of over
90% diastereoselectivity per bridging process, this reaction
represented the most efficient benzal brimide bridging of an
octol that we have observed to date.
With these experimental conditions in hand, we set out to
investigate how at constant temperature (60 °C) yields of
the various derivatives 2a-6a were dependent on the
equivalents of bridging material. The results are summarized
in Table 1. Briefly, reducing the amount of bridging material
Table 1. Yields of Deep-Cavity Cavitands 2a-6aa
equiv of bridging material
DCC
8.8
4.4
3.3
2.2
1.1
2%
5%
2%
8%
trace
2
3
4
5
6
trace
trace
trace
trace
67%
trace
trace
trace
trace
68%
3%
4%
2%
33%
28%
4%
9%
4%
33%
7%
a By bridging octol 1 with 3,5-dibrombenzal bromide, DMA, 60 °C, 24
h.
(16) Xi, H.; Gibb, C. L. D.; Gibb, B. C. J. Org. Chem. 1999, 64, 9286-
9288.
(17) Gibb, C. L. D.; Stevens, E. D.; Gibb, B. C. Chem. Commun. 2000,
363-364.
(18) This value assumes no intermolecular bridging between the mono-,
bis-, and trisbridged intermediates.
steadily decreased the yield of the tetrabridged species 6a
but without a concomitant increase in the yield of the partially
3846
Org. Lett., Vol. 2, No. 24, 2000

bridged intermediates 2a-5a. In fact, even with only 2.2 or
1.1 equiv of bridging material there appeared to be a bias
toward derivative 5a. Particularly poor yields of the mono-
bridged derivative 2a and bis-A/C derivative 4a (which is
statistically half as likely to form as the bis-A/B derivative
3a) were always observed.
Scheme 2. Formation of Bridging Material 7 and 8a
Wishing to improve upon the yields of these bridged
compounds, we investigated the bridging process at reduced
temperatures. As anticipated, a general trend of increasing
yields for the partially bridged derivatives was observed as
the reaction temperature was decreased. Again however the
corresponding decrease in tetrabridged DCC 6a was not
matched by a concomitant increase in yield of the lower
species. More specifically, yields for the trisbridged species
5a was optimized at 34% (48 h at 35 °C), 3% for the bis-
A/C derivative 4a (48 h at 35 °C), 13% for the bis-A/B
derivative 3a (48 h at rt), and 13% for the monobridged
derivative 2a (48 h at rt). Interestingly, in all ten reactions
investigated that shared the same solvent and base, the ratio
of 3a to 4a was found to be 5.38 ((σ ) 2.8); a range
considerably greater than the statistically expected 2:1 ratio.19
Also worthy of note was the relative ease at which the
trisbridged derivative 5a was isolated in reasonable 20-35%
yields, but conditions for isolating each of the lower species
in similar yields could not be determined.20 In these cases
optimum conditions gave broad overall product distributions
(e.g., 2a 13%, 3a 13%, 4a 3%, 5a 15%; 48 h reaction at 25
°C) and no conditions could be determined that favored one
of these product over the others.
a Key: (a) (i) n-BuLi, then DMF then H3O+, 65%. (b) (PhO)3P/
Br2, 50%. (c) NBS, benzoyl peroxide, benzene. 41%.
observed where the stereogenicity of a benzal bridge ensures
that the aromatic ring of the bridge essentially fills the cavity
of the molecule. In contrast to bridging with 7, reaction with
bridging material 8 gave DCC 11 in 64% yield and no
indication of any “miss-bridging”.
To investigate the generality of these conclusions, we also
investigated the bridging of octols with 4-bromobenzal
bromide. Briefly, similar trends were observed for the
formation of the intermediate bridged derivatives 2b-5b.
For example, optimum yields for the formation of the
trisbridged 5b were around 35% (24 h at 47 °C), 2% and
14% for the bis-A/C 4b and bis-A/B 3b derivatives respec-
tively21 (48 h at 25 °C), and 11% for the monobridged
derivative 2b (48 h at 25 °C).
With various partially bridged intermediates now in hand,
we investigated the formation of a number of reduced
symmetry DCCs. Diol 5b can be bridged a fourth time with
a variety of molecules, but with little to go wrong in terms
of inadvertent side reactions, we opted to investigate for the
first time the suitability of pyridine compounds as bridging
compounds. The formation of the requisite bridging materials
7 and 8 was relatively straightforward and is shown in
Scheme 2. Thus, in the first instance, 2,6-dibromopyridine
was monoformylated22 and the aldehyde converted to the
R,R-dihalopicoline. We initially attempted this latter conver-
23
sion with BBR3 without success. However, treatment of
For both the formation of 9(+10) and 11 it is interesting
to note that the yields are less than the theoretical 86% for
each step in a “normal” ca. 55% 4-fold bridging process.
Put another way, translating the efficiency of the above one-
step bridging processes into theoretical four-step processes
(octol 1 to respective tetrapyridyl derivative) would result
in yields of 8.5% and 18% for reactions with 7 and 8,
respectively. Is this because the pyridyl bridging materials
are not very efficient? To investigate this we once again took
the aldehyde with a triphenylphosphite/bromine mixture24
resulted in the formation of 7. Bridging material 8 was readily
synthesized by treating 5-methyl-2-bromopyridine with N-
bromosuccinimide.
Interestingly, the reaction of 5b with bridging material 7
resulted in the isolation of two isomeric DCCs: the expected
cavitand 9 and its isomer 10 in 54% yield and 16% yield,
respectively. This latter DCC is the first example we have
Org. Lett., Vol. 2, No. 24, 2000
3847

trisbridged DCC 5b but this time treated it with 4-bromoben-
zal bromide. Under identical conditions the known DCC 6b16
was isolated in 84% yield. This corresponds to a theoretical
yield of 50% for the transformation of 1 to 6b which is close
to the actual yield of 54%. Consequently, it would appear
that the nucleophilic centers of 7 and 8 do indeed interfere
with their bridging ability.
from 4-fold bridging processes. Similar results were obtained
with the A/C tetrol 4b. Thus, treatment with the same
bridging materials gave DCCs 6b and 13 in 62% and 59%
yields, respectively. Consequently, although it may be the
case that formation of the A/B-bisbridged intermediate is
favored over the formation of the A/C species, it would
appear that both tetrols 3b and 4b are equally efficient in
generating tetrabridged species.
We have demonstrated here the partial stereoselective
bridging of resorcinarenes with benzal bromides. By this
stepwise process we have noted the first example of miss-
bridged DCCs in which the cavity of the cavitand is filled
by one of the benzal bridging units. We have also noted a
statistically significant bias in favor of formation of the bis-
A/B over the bis-A/C derivative. Further investigations into
the described protocols, as well as the utilization of these
and other non-C4V DCCs, are currently underway and will
be reported in due course.
Acknowledgment. We are grateful to Richard B. Cole
of the New Orleans Center for Mass Spectrometry Research
for mass analysis of the products. Special thanks also go to
Corinne Gibb and Carl Bischoff for technical assistance. This
work was partially supported by the Louisiana Board of
Regents Support Fund (1997-00)-RD-A-23, a Research
Innovation Award from the Research Corporation, and the
donors of the Petroleum Research Fund, administered by the
America Chemical Society.
Finally, having also noted an apparent preference for the
formation of A/B-bisbridged 3b over A/C-bisbridged 4b, we
investigated the efficiency of bridging these compounds
to form tetrabridged DCCs. When treated with 4- and
3-bromobenzal bromide, the bis-A/B derivative 3b gave
DCCs 6b and 12 in 61% and 65% yields, respectively. These
values correspond to an average 80% yield for each bridging
reaction and again are in line with theoretical values derived
Supporting Information Available: Experimental pro-
1
cedures, characterization data, and H NMR spectra of all
new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL006569M
(20) The isolation of comparable yields of tris-bridged products arising
from nonstereospecific bridging processes has been attributed to a relatively
high energy barrier when introducing the fourth bridge. See ref 6.
(21) Thirteen reactions gave an average ratio of 3b to 4b of 7.55 ((σ )
3.0). For the six reactions that share the same solvent and base similar,
statistically significant results were noted (3b:4b ratio of 6.11 ( σ ) 1.3).
Both of these results are statistically significantly greater than the expected
2:1 ratio.
(22) Tsukube, H.; Uenishi, J.; Higaki, H.; Kikkawa, K.; Tanaka, T.;
Wakabayashi, S.; Oae, S. J. Org. Chem. 1993, 58, 4389-4397.
(23) Lansinger, J. M.; Ronald, R. C. Synth. Commun. 1979, 9, 341-
349.
(19) These findings may indicate that the bridging route from octol to
DCC primarily involves an A/B-bisbridged intermediate. Such an apparent
preference could be attributed to the degree of preorganization of the
resorcinol rings. Thus, the introduction of the first bridge, e.g., to form 2a,
results in the preorganization of one of the resorcinol rings if the next bridge
is to go to the B position (e.g., to form 3a) but does not preorganize the
resorcinol rings involved with the second bridge going in to the C position
(e.g., to form 4a). Alternatively, these results might indicate different rates
of decomposition of 3a and 4a in the reaction medium. However if this
were the case then it would be expected that yields for the conversion of
3b into 6b/12 and 4b into 6b/13 (see text) would differ. This was not
observed.
(24) Hoffmann, R. W.; Bovicelli, P. Synthesis 1990, 657-659.
3848
Org. Lett., Vol. 2, No. 24, 2000