Catalytic Cage Formation via Controlled Dimerization of Norbornadienes: An Entry to Functionalized HCTDs (Heptacyclo[6.6.0.02,6.03,13.04,11.05,9.010,14]tetradecanes)

Article abstract of DOI:10.1021/acs.orglett.6b00207

A general and practical catalytic method has been developed for the rapid synthesis of HCTD (heptacyclo[6.6.0.0^2,6.0^3,13.0^4,11.0^5,9.0^10,14]tetradecanes) and various new 7,12-disubstituted HCTDs from n

Full text of DOI:10.1021/acs.orglett.6b00207

Letter
pubs.acs.org/OrgLett
Catalytic Cage Formation via Controlled Dimerization of
Norbornadienes: An Entry to Functionalized HCTDs
2
,6 3,13 4,11 5,9 10,14
(
Heptacyclo[6.6.0.0 .0 .0 .0 .0 ]tetradecanes)
Hee Nam Lim and Guangbin Dong*
Department of Chemistry, University of Texas at Austin, Austin, Texas 78712, United States
*
S Supporting Information
ABSTRACT: A general and practical catalytic method has been
developed for the rapid synthesis of HCTD (heptacyclo-
2
,6 3,13 4,11 5,9 10,14
[
6.6.0.0 .0 .0 .0 .0 ]tetradecanes) and various new 7,12-
disubstituted HCTDs from norbornadienes. Compared to the
known approaches, this new protocol avoids stoichiometric metals,
utilizes commercially available reagents as catalysts, and affords
higher yields and significantly improved selectivity. In addition,
quadracyclane was discovered for the first time to undergo a similar endo,cis,endo cycloaddition to give HCTD in a good yield.
Derivatization of HCTDs led to efficient preparation of a range of novel homo- and heterobifunctional scaffolds that hold
potentials for biological and material applications.
ith fascinating highly symmetrical structures, carbocyclic
Scheme 1. Controlled Dimerization of NBD
W
cage compounds hold great promise for use as
1
pharmaceuticals (e.g., amino-adamantanes), energetic (e.g.,
2
nitrocubanes and CL-20), and photonic/electronic materials
3
(
e.g., fullerenes). In particular, HCTD (heptacyclo-
2,6 3,13 4,11 5,9 10,14
[
6.6.0.0 .0 .0 .0 .0 ]tetradecane) represents a unique
4
class of cage molecules (Figure 1). Analogous to cubanes and
Figure 1. Carbocyclic cage compounds.
adamantanes, HCTD can be considered a distinct repository of
cyclopentanes with D2d symmetry. In addition, the rigid,
strained, and highly compact scaffold, as well as the unusual
overall shape, has made HCTD an attractive moiety for use as
2
5
energetic materials and energy transfer spacers.
Despite these intriguing features, practical synthesis of
HCTD, particularly substituted HCTDs, remains a long-
standing challenge. HCTD was first prepared (in 2−3%
tion of NBD to HCTD without forming other observable
dimers, although the yield was moderate (26% based on Mo)
with excess NBD. Clearly, to thoroughly explore the potential
of HCTD or functionalized HCTDs in biomedical or material
applications, a practical, efficient, and selective synthesis of
HCTD and its derivatives must be developed. Hence, a catalytic
approach without the need of stoichiometric metal carbonyls
would be highly desirable.
9
yield) in 1961 via iron carbonyl [Fe(CO) ]-mediated
5
endo,cis,endo cyclodimerization of norbornadiene (NBD)
6
(
Scheme 1A). Later, this transformation was also observed
7
using other transition metals, such as rhodium; however, these
reactions generally produced a complex mixture with many
other isomeric byproducts, and the overall yields of HCTD
were low due to the poor selectivity during the NBD-
dimerization process (at least 14 pathways to dimerize NBD
Our research was inspired by the seminal work of Mitsudo
8
and co-workers reported in 1999, in which a novel half-cage
8
have been reported to date). A significant improvement was
made by Chow and co-workers in 1985; they found that
Received: January 21, 2016
Mo(CO) can selectively promote the desired cyclodimeriza-
6
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b00207
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
compound (PCTD) was obtained selectively in a Ru-catalyzed
NBD-dimerization reaction (Scheme 1B). During the course of
this study, an uncommon solvent effect was observed: while
THF and DMF gave almost exclusively the PCTD product, the
use of DMSO as solvent led to forming a significant amount of
HCTD (66−70% GC yield, 45% isolated yield) along with
PCTD (23−26% yield). While the selectivity for HCTD is
moderate (2.5−3.0:1) and the [Ru(cod)(cot)] catalyst used
needs additional preparation, this earlier finding clearly
indicated that catalytic formation of HCTD is feasible.
However, it was surprising to note that this important discovery
has been overlooked for decades with no attention to be
advanced for HCTD synthesis. Herein, we describe our
development of an improved catalytic method for selective
synthesis of HCTD and its derivatives from NBD and 7-
substituted NBDs (Scheme 1C). The protocol utilizes
commercially available reagents and tolerates air and some
moisture; thus it is user-friendly and operationally simple.
To understand the role of each reaction parameter, a number
of control experiments were carried out (Table 1). First,
Mitsudo’s protocol with Ru(cod)(cot) was successfully
11
reproduced, which gave a 2:1 mixture of HCTD 2a and
PCTD 3a (Table 1, entry 2). In the absence of the ruthenium
complex or the reductant, neither 2a nor 3a was formed (Table
1, entries 3 and 4), suggesting that the reaction was catalyzed
by a low-valent Ru complex. The use of an electron-deficient
olefin ligand is also critical, as in the absence of
dimethylfumarate or use of N,N-dimethylacrylamide instead,
the yields were considerably lower (Table 1, entries 5 and 6).
Nearly the same results were obtained using 2.5 mol % of [Ru(p-
cymene)Cl ] (Table 1, entry 7); further reducing of the loading
2 2
to 1 mol % led to a much slower reaction albeit with the same
selectivity (Table 1, entry 8). Gratifyingly, the reaction
temperature can be lowered to 90 °C without much influence
on the yield and selectivity (Table 1, entry 9). It is worthy to
note that, when other solvents, such as toluene and THF, were
8
used, no significant conversion was observed (Table 1, entries
The prior mechanistic study suggested that forming the
8
1
0 and 11) suggesting a major difference between Mitsudo’s
undesired PCTD was initiated by a Ru−H-mediated 1,2-
10
and this catalytic system, in which the former gave more than
addition of NBD. Thus, it is logical to hypothesize that one
key to promote the desired cage formation would be to avoid
forming a ruthenium−hydride species, which indicates that
choosing an appropriate ruthenium precatalyst would be
critical. From a practical viewpoint, the study was initiated by
examining a range of commercially available Ru(II) salts in
combination with appropriate reductants, e.g., Mn or Zn. To
our delight, preliminary screening revealed that, using a
catalytic amount of Mn powder as reductant, several Ru(II)
chlorides, such as [Ru(cod)Cl ], [Ru(nbd)Cl ], and [Ru-
9
1
0% conversion to PCTD in the absence of DMSO (Scheme
B). Nevertheless, both systems affirmed the indispensable role
of DMSO, although the exact reason remained to be defined. In
addition, attempts to use a catalytic amount of DMSO were
unsuccessful (Table 1, entry 12). Zn was found less reactive
than Mn; however, a similar yield can be obtained when more
Zn was used (Table 1, entry 13). Furthermore, in contrast to
the [Ru(cod)(cot)]-based procedure that requires argon
8
atmosphere, this [Ru(p-cymene)Cl ] system tolerated air
2
2
2
2
and moisture (Table 1, entries 14 and 15). Interestingly, under
a higher concentration (3.0 M), the selectivity was remarkably
increased to 21:1 (Table 1, entry 16).
(
benzene)Cl ] , afforded HCTD 2a in good yields (ca. 70%)
2 2
and high selectivity (8:1−10:1). Ultimately, the [Ru(p-
cymene)Cl ] /Mn combination proved to be optimal, which
2
2
With the optimal conditions in hand, the scope of using
gave a 75% yield of 2a with 11:1 selectivity favoring the cage
formation (Table 1, entry 1).
12
various 7-substituted NBDs was explored (Scheme 2). First,
C −Ot-Bu and OTBS substituted NBDs (1b and 1c) smoothly
7
underwent the endo,cis,endo cyclodimerization to furnish the
oxygenated HCTDs (2b and 2c) in good yields. The structure
of 2c was unambiguously confirmed by X-ray crystallography.
In contrast, the reaction of the OBn-substituted substrate 1d
a
Table 1. Selected Reaction Optimization
13
was much less selective, wherein formation of HCTD was
likely hampered by an undesired intramolecular coordination of
the aryl group with the Ru-catalyst. We hypothesized that use
of an aromatic cosolvent should minimize chelation of the aryl
group in the NBD substrate with the catalyst; indeed, with the
use of a toluene/DMSO (1:1) mixed solvent, improved yield
for the cage formation was achieved. The same trend was also
found when using C -aryl- and benzyl-substituted NBDs (1h−
7
1l). On the other hand, a THF/DMSO (7:1) mixed solvent
was found more effective for alkyl-substituted NBDs (2e and
f) than pure DMSO. In addition to alkoxy, siloxy, alkyl,
benzyl, and aryl groups, acetals (i.e., MEM ether, 2f) and esters
g were also proved compatible. Notably, in this study, a range
2
2
of new C7,12-diarylated HCTDs were made available in
moderate but synthetically useful yields, which were not
achieved previously with other methods.
To understand the kinetic profile of this transformation, the
1
a
b
reaction progress was monitored by H NMR with substrate
All reactions were run on a 0.3 mmol scale. The yield was
1
1b. As depicted in Figure 2, the reaction exhibited a short or no
induction period, indicating a fast reduction of the Ru(II)
precatalyst. In addition, a high initial rate was observed for the
cage-compound formation, as during the first hour HCTD 2b
was produced in 63% yield, and thereafter the rate was
determined by H NMR using 1,1,2,2-tetrachloroethane as the internal
standard. Isolated yield. Ru(cod)(cot) (2 mol %) was used instead
of [Ru(p-cymene)Cl ] and Mn in 1.6 M for 2 h. Common organic
c
d
e
2
2
solvents including THF, toluene, DMF and 1,4-dioxane were tested.
f
9
0 mol % Zn was used.
B
DOI: 10.1021/acs.orglett.6b00207
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
Scheme 2. Substrate Scope
any light-mediated formation of NBD from quadracyclane),
dimerization of quadracyclane provided HCTD in 77% yield,
which, to the best of our knowledge, was unknown previously
14
(
Scheme 3A). Moreover, our method proved to be readily
Scheme 3. Further Exploration
scalable; 2b was isolated in 69% yield on a gram scale, and
5a
subsequent removal of the t-butyl group with TMSI provided
HCTD-diol 4 in an excellent yield (Scheme 3B).
Preparation of further functionalized cage compounds was
exemplified using diol 4 as a platform (Scheme 4). A diverse
Scheme 4. Derivatization of the HCTD−Diol
a
All reactions were run on a 0.2 mmol scale; NMR yields are
presented due to the difficulty of fully purifying these nonpolar
compounds; the numbers in parentheses are isolated yields of the pure
b
fractions. DMSO:toluene (1:1) mixed solvent was used.
c
DMSO:THF (7:1) mixed solvent was used.
Figure 2. Kinetic profile of forming 7,12-di-Ot-BuHCTD (2b) and
range of functional moieties can be introduced through simple
derivatizations of the alcohol groups. For example, azide-and
cyano-substituted HCTDs 6 and 7 were efficiently synthesized
4,9-di-Ot-BuPCTD (3b).
15
significantly diminished. In contrast, the rate of forming PCTD
was much lower throughout the reaction.
Besides NBDs, quadracyclane was also discovered to be a
competent substrate to give HCTD, although the exact
mechanism is unclear. For example, under the standard
conditions except at 90 °C in the absence of light (to avoid
via S 2-type approaches, which offers opportunities for
N
16
subsequent “Click” couplings, or access of the corresponding
17
amino and carbonyl compounds. Diacrylate 8 was prepared
simply by acylation of diol 4 using excess acryloyl chloride,
which can potentially be used as a distinct cross-linker for
polymer synthesis. Heterodiester 9 was synthesized by a two-
C
DOI: 10.1021/acs.orglett.6b00207
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
step sequence, offering an approach to access unsymmetrical
HCTDs. In particular, two biologically relevant molecules, such
as vitamins A and E, can be installed at each end of the HCTD
T.; Chiu, N.-R.; Chen, H.-C.; Chen, C.-Y.; Yu, W.-S.; Cheng, Y.-M.;
Cheng, C.-C.; Chang, C.-P.; Chou, P.-T. Tetrahedron 2003, 59, 5719−
5
730.
(
6) (a) Lemal, D. M.; Shim, K. S. Tetrahedron Lett. 1961, 2, 368−372.
b) Bird, C. W.; Colinese, D. L.; Cookson, R. C.; Hudoc, J.; Williams,
R. O. Tetrahedron Lett. 1961, 2, 373−375.
7) Acton, N.; Roth, R. J.; Katz, T. J.; Frank, J. K.; Maier, C. A.; Paul,
I. C. J. Am. Chem. Soc. 1972, 94, 5446.
1
0, in which the rigid cage core serves as a unique linker
(
holding the two vitamins in nearly orthogonal directions.
In conclusion, a practical catalytic method to prepare HCTD
(
18
and 7,12-disubstituted HCTDs was developed. Compared to
the traditional alternatives, this method provides improved
yield and selectivity, uses commercially available reagents as
catalysts, and tolerates air/moisture. In addition, quadracyclane,
an isomer of NBD, was discovered for the first time to undergo
a similar endo,cis,endo cycloaddition in a good yield. Detailed
mechanistic investigation to advance our understanding for
further reaction improvement is ongoing. Finally, more
functionalized HCTDs can be rapidly accessed through
derivatization of a diol intermediate. Exploring the potential
of these resulting novel scaffolds in biomedical and material
applications is now underway in our laboratory.
(
8) Mitsudo, T.; et al. J. Am. Chem. Soc. 1999, 121, 1839−1850 and
references therein..
(9) (a) Chow, T. J.; Wu, M.-Y.; Liu, L.-K. J. Organomet. Chem. 1985,
281, C33−C37. (b) Chow, T. J.; Chao, Y.-S. J. Organomet. Chem.
1
985, 296, C23. (c) Chow, T. J.; Liu, L.-K.; Chao, Y.-S. J. Chem. Soc.,
Chem. Commun. 1985, 700−701. (d) Chow, T. J.; Chao, Y.-S.; Liu, L.-
K. J. Am. Chem. Soc. 1987, 109, 797−804.
(10) Itoh, K.; Oshima, N.; Jameson, G. B.; Lewis, H. C.; Ibers, J. A. J.
Am. Chem. Soc. 1981, 103, 3014−3023.
(
11) In our hand, use of commercially available Ru (CO) provided
3
12
considerably lower yield (26%) for 2a than the reported one (66%) in
ref 8; however, the possibility that this reaction is highly sensitive to
the purity of Ru (CO) cannot be excluded.
3
12
ASSOCIATED CONTENT
Supporting Information
(12) Attempts to use C1 or C2-substituted NBDs remain
unsuccessful, which is likely due to the steric hindrance of the
substrates.
■
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b00207.
(
13) In addition to the HCTD product, several inseparable isomers
1
including endo and exo-PCTDs were observed in the crude H NMR
spectra.
(
14) For a stepwise conversion of quadracyclane to NBD and then to
HCTD, see: Marchand, A. P.; Dave, P. R. J. Org. Chem. 1989, 54,
775−2777.
15) Kirmse, W.; Streu, J. J. Org. Chem. 1985, 50, 4187−4194.
(16) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
001, 40, 2004−2021.
17) Larock, R. C. Comprehensive Organic Transformations: A Guide to
Functional Group Preparations, 2nd ed., Wiley-VCH: New York, 1999.
18) The mechanistic pathway of forming HCTD remains unclear.
Efforts on investigating the catalytic cycle are ongoing.
Full experimental and characterization data for all
compounds (PDF)
X-ray data for compound 2c (CIF)
2
(
X-ray data for compound 2k (CIF)
2
(
AUTHOR INFORMATION
Corresponding Author
■
*
(
E-mail: gbdong@cm.utexas.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank CPRIT and the Welch Foundation (F-1781) for
research grants. G.D. is a Searle Scholar and Sloan Fellow. Drs.
V. Lynch and M. C. Young (University of Texas at Austin) are
acknowledged for X-ray crystallography. Dr. M. C. Young is
also thanked for proofreading the manuscript. We are grateful
to Ms. M. McCallum (Baylor College) and Prof. J. L. Wood
(Baylor College) for the kind donation of quadracyclane.
REFERENCES
■
(
1) (a) Tilley, J. W.; Kramer, M. Prog. Med. Chem. 1981, 18, 1−44.
b) Wang, Y.; Eu, J.; Washburn, M.; Gong, T.; Chen, H.-S. V.; James,
(
W. L.; Lipton, S. A.; Stamler, J. S.; Went, G. T.; Porter, S. Curr.
Alzheimer Res. 2006, 3, 201−204.
(
2) (a) Krause, H. H. Energetic Materials. Teipel, U., Ed.; Wiley-VCH
Verlag GmbH & Co. KGaA: Weinheim, 2005; pp 1−26. (b) Klapotke,
T. M. Chemistry of High-Energy Materials; Wlater de Gruyter GmbH &
Co. KG: Berlin, New York, 2011. (c) Marchand, A. P. Tetrahedron
1
(
(
988, 44, 2377−2395.
3) (a) Giacalone, F.; Martín, N. Chem. Rev. 2006, 106, 5136−5190.
b) Lebedeva, M. A.; Chamberlain, T. W.; Khlobystov, A. N. Chem.
Rev. 2015, 115, 11301−11351.
4) Osawa, E.; Yonemitsu, O. Carbocyclic Cage Compounds: Chemistry
and Applications; VCH: New York, N.Y., 1992.
5) (a) Kumar, K.; Tepper, R. J.; Zeng, Y.; Zimmit, M. J. Org. Chem.
995, 60, 4051−4066. (b) Chiou, N.-R.; Chow, T. J.; Chen, C.-Y.;
Hsu, M.-A.; Chen, H.-C. Tetrahedron Lett. 2001, 42, 29−31. (c) Chow,
(
(
1
D
DOI: 10.1021/acs.orglett.6b00207
Org. Lett. XXXX, XXX, XXX−XXX