Highly Selective Ring Expansion of Bicyclo[3.1.0]hexenes

Article abstract of DOI:10.1021/acs.orglett.6b02641

A Ru-carbene-promoted ring expansion of bicyclo[3.1.0]hexenes with terminal alkynes is reported. The reaction delivers seven-membered carbocycles starting from readily available starting materials and was found to be highly regioselective. The resulting seven-membered ring products contain both conjugated diene and cyclopropane substructures that could be selectively reacted in subsequent transformations.

Full text of DOI:10.1021/acs.orglett.6b02641

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Letter
pubs.acs.org/OrgLett
Highly Selective Ring Expansion of Bicyclo[3.1.0]hexenes
Synthia Gratia, Kathryn Mosesohn, and Steven T. Diver*
Department of Chemistry, University at Buffalo, The State University of New York, Amherst, New York 14260, United States
S
* Supporting Information
ABSTRACT: A Ru-carbene-promoted ring expansion of bicyclo-
[3.1.0]hexenes with terminal alkynes is reported. The reaction delivers
seven-membered carbocycles starting from readily available starting
materials and was found to be highly regioselective. The resulting
seven-membered ring products contain both conjugated diene and
cyclopropane substructures that could be selectively reacted in
subsequent transformations.
even-membered rings appear in a diverse array of natural
products and bioactive molecules, and their synthesis
groups undergo addition reactions with amines,4 aldehydes,5
nitrones,6 and imines7 to provide diverse chemical structures.
Cyclic dienes also undergo chemoselective reactions, such as
Diels−Alder cycloaddition. The products of Scheme 1b contain
both 1,3-diene and cyclopropane moieties, providing two sites
for functionalization. We were interested to see if comple-
mentary functionalization of these groups would be possible in
this compact, conjugated system.
Bicyclic substrates were prepared using a simple and scalable
cyclopropanation procedure. The cyclopropanation was based
on the asymmetric cyclopropanation of cyclopentadiene using
the chiral catalyst Rh2(DOSP)4, as developed by Davies and co-
workers.8 In our case, we did not require enantiomerically
enriched material, so Rh2(OAc)4 was substituted as the catalyst,
providing racemic products. Employing the requisite diazo-
carbonyl compound, excess 1,3-cyclopentadiene and catalytic
Rh2(OAc)4, three bicyclo[3.1.0]hexenes were prepared in
moderate isolated yields (eq 1). The bicyclic substrates
containing aromatic rings were conveniently purified by
recrystallization from ethanol. Diester 1c was obtained as a
solid after chromatography.
S
remains a challenge.1 A conventional approach is predicated on
transition-metal-mediated cycloaddition reactions.1b Less often,
ring expansion of cyclopentenes has been used to access the
seven-membered ring structure.2 Though rare, catalytic methods
for ring expansion are potentially powerful due to their atom
economy, efficiency, and the accessibility of the substrates.
Several years ago, our laboratory developed a Ru-carbene-
catalyzed ring expansion of cyclopentene.2a A similar fused ring
system gave a mixture of regioisomers depending on the degree
of alkyl substitution (Scheme 1a).2b In this report, we show that a
Scheme 1. Ring-Expanding Ene-Yne Metathesis
cyclopentene fused with a cyclopropane (hereafter bicyclo-
[3.1.0]hexene) underwent regiocontrolled ring expansion with
terminal alkynes (Scheme 1b). Products 2−4 are uniquely
substituted bicyclic dienes which have additional value due to
their potential for chemoselective functionalization.
Ring expansion of bicyclo[3.1.0]hexenes gives dienyl cyclo-
propanes, a unique cyclic structure amenable to chemoselective
modification. Their reactivity should be similar to that of vinyl
cyclopropanes. Vinyl cyclopropanes provide an excellent plat-
form for the synthesis of heterocycles and natural products.3 For
example, vinyl cyclopropanes bearing electron-withdrawing
The ring expansion was optimized for a branched terminal
alkyne (Table 1). Alkyne addition over a 2 h period gave a 45%
isolated yield, but extending the addition time to 4 h resulted in
an improved 68% yield (Table 1, entries 1 and 2). Using fewer
equiv of 1a was detrimental to the product yield (entry 3). Next,
aromatic solvents were examined. Toluene gave results
comparable to those of benzene, but the yield was found to be
Received: September 2, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02641
Org. Lett. XXXX, XXX, XXX−XXX

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Organic Letters
Letter
Table 1. Optimization Studies for the Ring Expansion of
Bicyclo[3.1.0]hexene
a
b
entry
Ru cat equiv of 1a
solvent
C6H6
addition (h) yield (%)
1
2
Ru1
Ru1
Ru1
Ru1
Ru1
Ru1
Ru1
Ru2
Ru2
4
4
2
4
4
4
4
4
4
4
2
4
4
4
4
4
4
4
4
4
45
68
28
60
81
38
49
67
77
12
C6H6
3
C6H6
4
PhCH3
PhCF3 (TFT)
CH2Cl2
THF
5
6
Figure 1. Crystal structure of product 2a.
7
8
PhCF3
CH2Cl2
PhCF3
9
assigned on the basis of proton NMR spectroscopy. Interestingly,
the structure shows that the 1,3-diene is distorted from planarity:
the dihedral angle (C12−C13−C14−C15) was found to be
36.8°. Last, the structure proof also revealed the relative
stereochemistry of the major diastereomer. The high diaster-
eoselectivity observed in this reaction (10:1 for 2a) was
unexpected.10
With the optimization studies complete, we investigated the
scope of the ring expansion. Two conditions were used with
equal success: (1) precatalyst Ru1 in benzene or TFT or (2) Ru2
in CH2Cl2 or benzene. The standard reaction temperature was 45
°C. The data are presented for two classes of terminal alkyne, that
of aromatic and aliphatic alkynes. In all ring expansions, no side
products were identified. It is assumed that the major competing
reaction is polymerization of the terminal alkyne. Excess
cycloalkene is thought to limit the competing diester polymer-
ization pathway. Excess cycloalkene was recovered in all cases.
Table 2 summarizes the ring expansion of aromatic alkynes for
a variety of bicyclo[3.1.0]hexenes. Phenylacetylene gave a single
product in good yield (NMR), but separation from the excess
cycloalkene proved difficult. In this case, 4-phenyl-1,2,4-triazole-
c
10
Ru3
a
Conditions: 2−4 equiv of 1a, RuX (7.5 mol %), solvent at 45 °C. The
b
c
alkyne was added as a solution over 2 or 4 h. Isolated yield. Ru3 is
the Hoveyda−Grubbs precatalyst, (H2IMes)Cl2Ru = CH(2-i-
PrOC6H4).
slightly lower (cf. entry 2 vs 4). Suspecting that benzylic CH
insertion might explain the difference between benzene and
toluene, we investigated trifluorotoluene (TFT), a lightly
fluorous solvent that lacks benzylic CH bonds.9 With precatalyst
Ru1, the highest yield (81%) was obtained in this solvent (entry
5). Other nonaromatic solvents such as THF and CH2Cl2 proved
inferior for this catalyst, giving lower isolated yields of the diene
product (entries 6 and 7). The second-generation Grubbs
precatalyst Ru2 was found to perform well in TFT at 45 °C and
even better in CH2Cl2 (entries 8 and 9). Last, the Hoveyda−
Grubbs precatalyst Ru3 gave low yields in TFT solvent (entry
10), whereas the other precatalysts worked well under these
conditions. From the optimization experiments, two different
solvents emerged as optimal using either the Umicore M2
catalyst (Ru1) or the Grubbs catalyst (Ru2) (entries 5 and 9,
respectively). The slow addition rate of the alkyne proved to be
essential to obtain the highest product yields. In each run, a
∼10:1 mixture of diastereomers of 2a was obtained. The major
diastereomer of 2a is shown in Table 1.
Table 2. Terminal Aromatic Alkyne Bicyclo[3.1.0]hexene
a
Ring Expansion
The excess cyclopentene reactant 1a could be recovered and
reused. In Table 1, isolated yields of 1a ranged from 70 to 85%
yield. This material was recrystallized from EtOH and reused in
subsequent ring expansions without any effect on yield. Recovery
of unchanged starting material indicates that little bicycloalkene
polymerization (ROMP), if any, occurred under the reaction
conditions.
The product in Table 1 was produced as a single regioisomer.
Proton NMR spectroscopy allowed a structure assignment that
was most consistent with the regioisomer shown. The C1−H
appeared at 6.34 ppm as a doublet, coupled to a single
cyclopropane methine proton. The C3−H appeared as a doublet
at 5.86 ppm with a 10.8 Hz coupling constant, shared with the
adjacent C4−H located at 6.17 ppm, as a ddd. The coupling
pattern of these three protons would not be consistent with the
alternative regioisomer.
b
R
entry
X
yield, %
c
Ph (alkene 1a)
1
2
H
2b, 61
2c, 68
2d, 77
OMe
CO2Me
Br
3
c
4
2e, 51
c
5
Me
2f, 57
2g, 67
3a, 94
3b, 72
6
NO2
H
p-BrC6H4 (alkene 1b)
CO2Me (alkene 1c)
7
8
OMe
H
c
9
4a, 82
4b, 80
4c, 74
10
11
OMe
CO2Me
Additional structural data came from a single-crystal X-ray
structure determination (Figure 1). Diene 2a was recrystallized
from IPA/hexanes, giving cubes: mp 119−121 °C. The structure
corroborates the structure assignment and regiochemistry
a
Conditions: 0.25 mmol alkyne added to 1.0 mmol 1a−c and Ru1 and
b
Ru2 (7.5 mol %) in solvent at 45 °C, over a 4 h period. Isolated
yields. Trapped as PTAD cycloadduct; yield over two steps.
c
B
DOI: 10.1021/acs.orglett.6b02641
Org. Lett. XXXX, XXX, XXX−XXX

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Organic Letters
Letter
3,5-dione (PTAD) was added directly to the crude reaction, and
the cycloadduct was isolated in 61% yield. Para-substituted
phenyl acetylenes all gave ring expansion with similar yields, but
no electronic trend emerged from these data (entries 2−6).
Another aromatic substituted bicyclo[3.1.0]hexene 1b gave
excellent results (entries 7 and 8). The diester-substituted
cyclopropane derivative 1c gave similar yields, suggesting that
there was not a significant electronic effect exerted by the
additional electron-withdrawing ester group (entries 9−11).
Table 3 summarizes the ring expansion of aliphatic alkynes for
a variety of bicyclo[3.1.0]hexenes. In general, oxygen sub-
Scheme 2. Additional Transformations
Table 3. Terminal Aliphatic Alkyne Bicyclo[3.1.0]hexene
a
Ring Expansion
cycloaddition with PTAD was observed, affording 5a and 5b in
good to excellent yields. Next, a reaction of the cyclopropane was
desired. Vinyl cyclopropane diesters are known to react with
nucleophiles in Michael-like fashion,6a,4a thereby opening the
cyclopropane ring. 1,3-Diene stabilization was expected to
enhance this reactivity profile. Accordingly, aniline addition
occurred at room temperature (vs elevated temperature for vinyl
cyclopropanes), giving primarily the lactam 6. After nucleophilic
addition, lactamization results due to the proximity of the
secondary amine with the malonyl appendage and is Lewis acid
promoted. To our knowledge, lactamization has not been
observed previously. Though further studies are warranted, it is
likely that dienyl stabilization results in a carbocation that can be
attacked syn or anti to the alkyl substituent.12 Last, treatment of
diene 2l with diimide gave position-selective reduction of the 1,3-
diene, yielding 7 in good yield.13
b
R1
entry
R2
−CH2OBz
−(CH2)4CH3
−(CH2)4OBz
−CH(OBn)CH3
−CH(OAc)CH2CH2Ph
−CH2N(Ts) (Boc)
−CH2OBz
−CH(OBz)CH3
−CH(OBn)CH3
−CH(OAc)CH2CH2Ph
−CH2OBz
yield, %
c
Ph (alkene 1a)
1
2
2h, 65
d
2i, 65
2j, 45
2k, 83
2l, 79
2m, 55
3c, 73
3d, 48
3e, 71
3f, 77
4d, 54
4e, 49
4f, 74
3
4
5
6
p-BrC6H4 (alkene 1b)
CO2Me (alkene 1c)
7
8
9
10
11
12
13
−CH(OBz)CH3
−CH(OAc)CH2CH2Ph
a
Conditions: 0.25 mmol alkyne added to 1.0 mmol 1a−c and Ru1 and
Ru2 (7.5 mol %) in solvent at 45 °C, over a 4 h period. Isolated
yields. Average of three runs. Trapped as PTAD cycloadduct; yield
over two steps.
b
Two plausible mechanistic scenarios can explain these data.
First, the initiator RuX may insert the alkyne, giving a vinyl
carbene intermediate which can ring open the cycloalkene
(Scheme 3, path a). Regiochemistry is controlled in the ring-
c
d
stituents were well-tolerated and gave the resulting bicyclo-
[5.1.0]octadienes in good yields (entries 1−5). In all cases, a
single regioisomer was obtained. The product from the
hydrocarbon 1-octyne was difficult to separate from unreacted
alkene 1a, so the 1,3-diene was trapped as the PTAD cycloadduct
(entry 2). Additional substitution in the R2 group was well-
tolerated (entries 4 and 5). With a benzyloxy group, potential
chelation did not affect the success of the ring expansion (entry
4). A protected nitrogen-based functional group was also
acceptable (entry 6). Similar results were seen with the p-
bromophenyl-substituted cyclopropanes derived from alkene 1b
(entries 7−10). With bicycloalkenes 1b and 1c, lower yields were
found with a branched alkyne, which was opposite to that
observed with bicycloalkene 1a (entries 8 and 12 vs Table 1).
The case in Table 3, entry 1, deserves further comment. The
alkyne BzOCH2CCH gave variable yields due to alkyne-
promoted catalyst decomposition. The 65% yield in entry 1 is
an average of three yields. Three separate trials gave 74, 56, and
69% isolated yields (Grubbs Ru2, CH2Cl2, 45 °C). Unbranched
alkynes can undergo polymerization, and this process can lead to
catalyst decomposition.11 This variability can be overcome by use
of higher catalyst loadings.
Scheme 3. Plausible Mechanisms To Explain Regioselectivity
opening step, where the bulky ligands on the metal are furthest
from the substituted allylic carbon. Only the major pathway is
shown. Second, the initiator (e.g., Ru2) may open the
cycloalkene with the bulky ligands averting the allylic cyclo-
propane ring (path b). The resulting Ru-carbene can either revert
to the reactants by RCM or give alkyne insertion. Subsequent
RCM of the Z-vinyl carbene could form the seven-membered
ring product. Each mechanism requires a ring-opening step of the
bicyclo[3.1.0]hexene, which rationalizes why excess cycloalkene
is needed for a successful ring expansion. For path a, the fleeting
The diene products were further transformed by three
different chemoselective reactions (Scheme 2). First, rapid
C
DOI: 10.1021/acs.orglett.6b02641
Org. Lett. XXXX, XXX, XXX−XXX

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Organic Letters
Letter
(4) (a) Lifchits, O.; Charette, A. B. Org. Lett. 2008, 10, 2809−2812.
(b) Lebold, T. P.; Leduc, A. B.; Kerr, M. A. Org. Lett. 2009, 11, 3770−
3772. (c) Emmett, M. R.; Grover, H. K.; Kerr, M. A. J. Org. Chem. 2012,
77, 6634−6637. (d) Mackay, W. D.; Fistikci, M.; Carris, R. M.; Johnson,
J. S. Org. Lett. 2014, 16, 1626−1629.
(5) (a) Pohlhaus, P. D.; Johnson, J. S. J. Am. Chem. Soc. 2005, 127,
16014−16015. (b) Parsons, A. T.; Campbell, M. J.; Johnson, J. S. Org.
Lett. 2008, 10, 2541−2544. (c) Pohlhaus, P. D.; Sanders, S. D.; Parsons,
A. T.; Li, W.; Johnson, J. S. J. Am. Chem. Soc. 2008, 130, 8642−8650.
(d) Smith, A. G.; Slade, M. C.; Johnson, J. S. Org. Lett. 2011, 13, 1996−
1999.
(6) (a) Young, I. S.; Kerr, M. A. Angew. Chem., Int. Ed. 2003, 42, 3023−
3026. (b) Young, I. S.; Kerr, M. A. Org. Lett. 2004, 6, 139−141. (c) Sibi,
M. P.; Ma, Z.; Jasperse, C. P. J. Am. Chem. Soc. 2005, 127, 5764−5765.
(d) Young, I. S.; Williams, J. L.; Kerr, M. A. Org. Lett. 2005, 7, 953−955.
(e) Young, I. S.; Kerr, M. A. J. Am. Chem. Soc. 2007, 129, 1465−1469.
(7) (a) Carson, C. A.; Kerr, M. A. J. Org. Chem. 2005, 70, 8242−8244.
(b) Parsons, A. T.; Smith, A. G.; Neel, A. J.; Johnson, J. S. J. Am. Chem.
Soc. 2010, 132, 9688−9692.
intermediate must engage the cycloalkene over alkyne to avoid
alkyne polymerization. In path b, the first ring-opening step is
reversible and unfavorable; mass action would help form the
reactive intermediate. At present, there is not enough supporting
mechanistic work that favors one pathway over the other;
additional mechanistic studies are ongoing.
In conclusion, a Ru-carbene-promoted ring expansion of
bicyclo[3.1.0]hexenes with terminal alkynes has been developed.
Distinct optimized conditions were identified using two different
precatalysts. The starting materials are easy to prepare, and
excess bicyclo[3.1.0]hexene can be recovered in good yield and
reused. The reaction can give two different regioisomers, but
only a single regioisomeric product is formed. Two plausible
mechanisms were proposed based on either an alkyne-first or a
bicycloalkene-first pathway involving different Ru-carbene
intermediates. The products contain 1,3-diene and cyclopropane
moieties: each of these can be selectively reacted in subsequent
transformations, illustrating the reactivity and versatility of these
products. Further efforts to expand the scope of this reaction and
mechanistic studies are currently underway.
(8) Pelphrey, P.; Hansen, J.; Davies, H. M. L. Chem. Sci. 2010, 1, 254−
257.
(9) (a) Improved initiation rates for Ru1 were found by Grela in
fluorous solvents: (b) Samojłowicz, C.; Bieniek, M.; Pazio, A.; Makal, A.;
Wozniak, K.; Poater, A.; Cavallo, L.; Woj
Chem. - Eur. J. 2011, 17, 12981−12993.
(10) Additional studies regarding the stereochemistry of this reaction
will be addressed in a full paper.
(11) Diver, S. T.; Kulkarni, A. A.; Clark, D. A.; Peppers, B. P. J. Am.
Chem. Soc. 2007, 129, 5832−5833.
(12) A product arising from anti-addition could not be identified from
this reaction.
́
cik, J.; Zdanowski, K.; Grela, K.
́
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.6b02641.
Experimental procedures and characterization data for
new compounds (PDF)
(13) Endoma-Arias, M. A. A.; Hudlicky, J. R.; Simionescu, R.;
Hudlicky, T. Adv. Synth. Catal. 2014, 356, 333−339.
X-ray data for JMC_031601 (CIF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: diver@buffalo.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Daniel Clark and William Karnofel for preliminary
studies, and Anibal R. Davalos for helpful comments. We
acknowledge Prof. Jason Benedict and Jordan Cox (UB
Chemistry) for solving the crystal structure of 2a. We also
thank Umicore Precious Metals Chemistry and Dr. Richard
Pederson (Materia) for generous gifts of Ru precatalysts used in
this work. This study was supported by the National Science
Foundation through Grant CHE-1300702.
REFERENCES
■
(1) (a) Nguyen, T. V.; Hartmann, J. M.; Enders, D. Synthesis 2013, 45,
845−873. (b) Ylijoki, K. E. O.; Stryker, J. M. Chem. Rev. 2013, 113,
2244−2266. (c) Clavier, H.; Pellissier, H. Methods Appl. Cycloaddit.
React. Org. Synth. 2014, 631−654. (d) de Oliveira, K. T.; Servilha, B. M.;
de C. Alves, L.; Desidera, A. L.; Brocksom, T. J. Stud. Nat. Prod. Chem.
2014, 42, 421−463. (e) Mascarenas, J. L.; Gulias, M.; Lopez, F.
Comprehensive Organic Synthesis, 2nd ed.; Elsevier: Amsterdam, 2014;
Vol. 5, pp 595−655.
(2) (a) Kulkarni, A. A.; Diver, S. T. Org. Lett. 2003, 5, 3463−3466.
(b) Diver, S. T.; Clark, D. A.; Kulkarni, A. A. Tetrahedron 2008, 64,
6909−6919.
(3) (a) Carson, C. A.; Kerr, M. A. Chem. Soc. Rev. 2009, 38, 3051−
3060. (b) Lebold, T. P.; Kerr, M. A. Pure Appl. Chem. 2010, 82, 1797−
1812. (c) Grover, H. K.; Emmett, M. R.; Kerr, M. A. Org. Biomol. Chem.
2015, 13, 655−671.
D
DOI: 10.1021/acs.orglett.6b02641
Org. Lett. XXXX, XXX, XXX−XXX