Synthesis of 2-Ethenylcyclopropyl Aryl Ketones via Intramolecular SN2-like Displacement of an Ester

Article abstract of DOI:10.1021/acs.orglett.6b02588

The efficient synthesis of trans-2-ethenylcyclopropyl aryl ketones via an intramolecular S_N2-like displacement of an allylic ester is reported. A novel 1,5-acyl shift process is also observed that contributes to the product mixture. Theoretical calculations provide a rationale for the observed product ratio.

Full text of DOI:10.1021/acs.orglett.6b02588

Letter
pubs.acs.org/OrgLett
Synthesis of 2‑Ethenylcyclopropyl Aryl Ketones via Intramolecular
SN2‑like Displacement of an Ester
Michael E. Jung,* Daniel L. Sun, Timothy A. Dwight, Peiyuan Yu, Wei Li, and K. N. Houk*
Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095-1569, United States
S
* Supporting Information
ABSTRACT: The efficient synthesis of trans-2-ethenylcyclo-
propyl aryl ketones via an intramolecular SN2-like displace-
ment of an allylic ester is reported. A novel 1,5-acyl shift
process is also observed that contributes to the product
mixture. Theoretical calculations provide a rationale for the
observed product ratio.
or a projected total synthesis of members of the morphine
alkaloid family, we envisioned the palladium-promoted
allowed us to confidently assign the structure. In addition we
performed a NOESY analysis which also confirmed the trans
structure.2
Since cyclopropanes and specifically cyclopropyl ketones are
important components of natural products3 and intermediates
for synthesis,4 we decided to examine the generality and scope of
this process, with the hope of finding conditions to make it much
higher yielding.
We first studied the effect of the leaving group and found that
benzoates functioned very well. Changing the ligand from
triphenylphosphine to dppe also helped the reaction. Finally
changing the base to lithium hexamethyldisilazide (LiHMDS)
had a very significant effect on the yield of the process. Thus, the
allylic benzoates 7 and 8 afforded the corresponding products 6
and 9 under these conditions in quite high yields, 85% and 89%
respectively (Scheme 3). We wanted to test the amount of the
palladium catalyst needed for this process and therefore lowered
the catalyst loading. In all cases, the reaction still occurred. It was
somewhat of a surprise to find that even with no catalyst, in new
flasks that had never seen a metal, the desired products were
obtained in good yields; thus, the phenyl ketone 7 gave 6 in 84%
yield and 8 gave 9 in 80% yield (Scheme 4). Therefore, the
F
cyclization of the allylic acetate 2 on to the benzofuran unit at the
3-position to give the tricycle 3, which contains three of the five
rings present in morphine 1. However, treatment of the allylic
acetate 2 with a palladium(0) catalyst and the base DBU did not
produce any of the tricycle 3 but rather the unexpected 2-
ethenylcyclopropyl aryl ketone 4 in 24% yield as nearly a single
stereoisomer (Scheme 1). In order to prove the structure and
Scheme 1. Synthesis of trans-2-Vinylcyclopropyl Ketones
Scheme 3. Variation of Leaving Group, Ligand, and Base
stereochemistry of the product, we carried out the analogous
reaction with the phenyl analogue 5 which under similar
treatment afforded the trans-2-ethenylcyclopropyl phenyl ketone
6 in 23% yield (Scheme 2). Comparison of the proton NMR data
of this compound with those for the known trans isomer1
Scheme 2. Synthesis of Ketone 6
Received: August 29, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02588
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
d to the same symmetrical intermediate E to lead to both
observed products.
Scheme 4. Palladium-Free Conditions
We have carried out a number of cyclization studies with
various aroyloxy aryl ketones, and the results are shown in Table
1. Obviously the symmetrical compound 13a (entry 1) gives only
the sole product 14a. The benzofuryl ketone 13b (entry 2) gives
only the unrearranged ketone product 14b when starting with
13b but gives a 73:11 mixture of the ketone 15b, the rearranged
product, and the phenyl ketone 14b, the unrearranged product,
when starting with the isomeric ester 13l (corresponding to A′).
The 7-methoxybenzofuryl ketone 13c (entry 3) gave only the
unrearranged ketone 14c. The furyl ketone 13d (entry 4) also
greatly favors, by 70% to 13%, the unrearranged ketone product
14d as does the 2-trifluoromethylphenyl ketone 13i (entry 9)
which gives a 70% to 5% ratio of 14i to 15i. The 4-
trifluoromethylphenyl ketone 13g (entry 7) gave a 60:20 ratio
of unrearranged/rearranged products 14g:15g, but this ratio
changes to 49% (unrearranged) to 43% (rearranged) when
starting with the isomeric ester 13n (entry 14). The 3-
trifluoromethylphenyl ketone 13h (entry 8) also preferred the
unrearranged ketone but by a smaller ratio (60% to 23%). Finally
as mentioned earlier the 4-methoxyphenyl ketone 13j gave a 20%
to 58% mixture of the rearranged ketone 15j to the unrearranged
ketone 14j and a somewhat similar ratio (31% to 49%) of the
same two products when starting from the isomeric ester 13m
(entries 9 and 12).
reaction must proceed by a simple SN2-like displacement of the
allylic ester by the enolate of the ketone. It is somewhat curious
that no cyclopentanone products are obtained nor are there any
4,5-dihydrofuran products. When the benzoyloxy 4-methox-
yphenyl ketone 10 was treated under these conditions, the
expected product 11 was formed but in only 20% yield. The
major product was the phenyl ketone 6, formed in 58% yield
(Scheme 5). Thus, a rearrangement was occurring to exchange
Scheme 5. Rearrangement and Cyclization
We attempted two other cyclizations of this type, in which we
replaced the vinyl group with either a methyl group or a
hydrogen atom. Treatment of the secondary benzoate 16a
afforded the cyclopropyl ketone 17a in 49% yield (Scheme 7)
while the primary benzoate 16b furnished the simple cyclopropyl
ketone 17b in 38% yield. The trans-stereochemistry of 17a was
determined by comparison of its proton NMR data with those
reported in the literature.1a,4c Several similar substrates with
different acyl groups did not afford good yields of cyclization; e.g.,
the corresponding methyl or tert-butyl ketones, the ethyl ester, or
the Weinreb amide all failed. The closest analogy in the literature
is the work of Yates5 who showed that 4-aroyloxy cycloalkanones
give cyclopropanes via SN2-like opening of an intermediate
lactone. Several syntheses of bi- and polycyclic cyclopropanes
with an ester leaving group are also known.6
the acyl groups. We showed that the opposite starting material,
namely the 4-methoxybenzoyl phenyl ketone 12, also gave a
similar ratio of the two products 11 (31%) and 6 (49%). Thus,
the two acyl groups are exchanging via a common intermediate.
We propose the mechanism shown in Scheme 6 for this novel
Scheme 6. Mechanism of Rearrangement
In order to understand the energetics of this rearrangement−
cyclization manifold, we performed density functional theory
(DFT) calculations at the SMDTHF/B3LYP-D3/6-31+G(d)
level of theory using the Gaussian09 program.7 The free energy
diagram for the reaction of benzofuryl ketone 13b is shown in
Figure 1. Deprotonation of 13b generates the corresponding
enolate 18, which can undergo the proposed intramolecular SN2-
like reaction to form cyclopropane 14b. The barrier for this
reaction is 14.1 kcal/mol, and the reaction is exergonic by 18.9
kcal/mol. Enolate 18 can also intramolecularly attack the ester
carbonyl group via TS-2 to form a five-membered ring
intermediate 19. Ring opening of 19 generates a high-energy
intermediate alkoxide 20, which can reclose to form another five-
membered ring intermediate 21. The formation of enolate 22
from 21 is calculated to be fast via TS-3, but TS-4 is 3 kcal/mol
above TS-1; thus, only 14b is formed from 18. This is in
agreement with the experimental observation that only 14b is
formed. Starting from ketone 13l, enolate 22 is formed after
deprotonation and converts into 15b via TS-4 which is only 1.3
kcal/mol above 20, leading to 14b. The formation of a small
amount of 15b is predicted. We have also computed the free
energy profiles for the reactions of 4-methoxyphenyl ketone 13j
rearrangement. Formation of the ketone enolate B from the
aroyloxy ketone A and ejection of the ester leaving group would
give C (path a). But this enolate B could also attack the carbonyl
of the ester (path b) to give the alkoxide D. Ejection of the
alkoxide from D would give the symmetrical 1,3-dicarbonyl
compound E. Attack of the alkoxide on the opposite ester would
give D′ which would proceed via B′ to give C′ (path c), the other
observed product. And the pathway could also be entered by
beginning with the opposite aroyloxy ketone A′ leading via path
B
DOI: 10.1021/acs.orglett.6b02588
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Cyclization of Aroyloxy Aryl Ketones
entry
compd
Ar1
Ar2
14 yield (%)
15 yield (%)
time (h)
1
a
b
c
d
e
f
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
84
80
75
70
49
47
60
60
70
20
43
11
49
49
0
0
1.5
2.5
5.0
2.0
3.0
1.0
1.75
1.5
1.5
2.5
3.5
3.0
2.5
2.5
2
benzofuryl
7-(OMe)benzofuryl
furyl
3
0
4
13
46
43
20
23
5
5
4-CH3C6H4
4-FC6H4
4-CF3C6H4
3-CF3C6H4
2-CF3C6H4
4-MeOC6H4
4-Me2NC6H4
Ph
6
7
g
h
i
8
9
10
11
12
13
14
j
58
42
73
31
43
k
l
benzofuryl
4-MeOC6H4
4-CF3C6H4
m
n
Ph
Ph
Scheme 7. Cyclization of Other Esters
Table 2. Predicted and Experimental Product Ratios
entry
Ar1
Ar2
pred. ratio
exper. ratio
1
2
3
4
5
6
benzofuryl
Ph
Ph
159:1
1:9
>20:1
1:7
benzofuryl
Ph
4-MeOC6H4
Ph
1:2
1:3
4-MeOC6H4
Ph
2:1
2:1
4-CF3C6H4
Ph
5:1
3:1
4-CF3C6H4
2:1
1:1
and 4-trifluoromethylphenyl ketone 13g. The predicted ratios of
unrearranged to rearranged product (C:C′) are comparable to
the experimental ratios. The results are summarized in Table 2.
For 4-methoxyphenyl ketone 13j, the corresponding TS-1 is
0.5 kcal/mol higher in energy than that of TS-4. This
corresponds to a 1:2 ratio of C to C′ (Table 2, entry 3). The
intermediate alkoxide is lower in energy than both TSs. Starting
from the isomeric ester 13j, the ratio of C to C′ is predicted to be
2:1 (Table 2, entry 4). For 4-trifluoromethylphenyl ketone 13g,
Figure 1. Free energy profile for the reaction of benzofuryl ketone 13b. Energies are in kcal/mol.
C
DOI: 10.1021/acs.orglett.6b02588
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(e) Liu, L.; Montgomery, J. Org. Lett. 2007, 9, 3885−3887. (f) Ogoshi,
S.; Nagata, M.; Kurosawa, H. J. Am. Chem. Soc. 2006, 128, 5350−5351.
(g) Liu, L.; Montgomery, J. J. Am. Chem. Soc. 2006, 128, 5348−5349.
(h) Attah-Poku, S. K.; Alward, S. J.; Fallis, A. G. Tetrahedron Lett. 1983,
24, 681−684. (j) Mohrbacher, R. J.; Cromwell, N. N. J. Am. Chem. Soc.
1957, 79, 401−408. (k) Danishefsky, S. Acc. Chem. Res. 1979, 12, 66−72.
(l) Doyle, M. P.; McKervey, M. A.; Ye, T. In Modern Catalytic Methods
for Organic Synthesis with Diazo Compounds: From Cyclopropanes to
Ylides; Wiley: New York, 1998; pp 163−220. (m) Negretti, S.; Cohen, C.
M.; Chang, J. J.; Guptill, D. M.; Davies, H. M. L. Tetrahedron 2015, 71,
7415−7420.
the corresponding alkoxide intermediate is slightly higher in
energy than either TS-1 or TS-4. Therefore, the unrearranged
product C is predicted to be predominant (Table 2, entries 5 and
6).
In general the various pathways in Scheme 6 can all occur, and
there is a rather delicate balance between which products are
favored. As indicated in Table 2, computations and experiment
are in very good agreement. Were 18 and 22 (Figure 1) in rapid
equilibrium (Curtin−Hammett conditions), the ratio of
products would only depend on the relative energies of TS-1
and TS-4. Because 20 may be above either TS-1 or TS-4, as it is
in Figure 1, the product ratio is not that expected with Curtin−
Hammett conditions and sometimes is influenced by the reactant
identity.
In summary, we have developed an efficient synthesis of trans-
2-ethenylcyclopropyl aryl ketones through an intramolecular
SN2-like displacement of allylic esters. A novel 1,5-acyl shift
process is observed that contributes to the product mixture.
Theoretical calculations show that the SN2-like reactions and 1,5-
acyl shifts may have similar barriers, and whether or not Curtin−
Hammett conditions prevail depends upon the relative energies
of these processes. B3LYP/6-31+G(d)/SMD calculations
reproduce experimental results and provide insights into the
origins of the observed product ratios.
(5) Yates, P.; Anderson, C. D. J. Am. Chem. Soc. 1963, 85, 2937−2943.
(6) (a) Kawamura, S.; Chu, H.; Felding, J.; Baran, P. S. Nature 2016,
532, 90−93. (b) Hicklin, R. W.; Lopez Silva, T. L.; Hergenrother, P. J.
́
Angew. Chem., Int. Ed. 2014, 53, 9880−9883. (c) Trudeau, S.;
Deslongchamps, P. J. Org. Chem. 2004, 69, 832−838. (d) Brooks, G.;
Burgess, W.; Colthurst, D.; Hinks, J. D.; Hunt, E.; Pearson, M. J.; Shea,
B.; Takle, A. K.; Wilson, J. M.; Woodnutt, G. Bioorg. Med. Chem. 2001, 9,
1221−1231.
(7) Frisch, M. J.; et al. Gaussian 09; Gaussian, Inc.: Wallingford, CT,
2013. For the full reference, see Supporting Information.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.6b02588.
Experimental procedures and spectral characterization of
all new compounds along with proton and carbon NMR
spectra (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: jung@chem.ucla.edu.
*E-mail: houk@chem.ucla.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Prof. Craig Merlic, UCLA, for helpful discussions. We
are grateful to the National Science Foundation for financial
support of this research (CHE-1361104).
REFERENCES
■
(1) (a) Wessig, P.; Muehling, O. Helv. Chim. Acta 2003, 86, 865.
(b) Bahurel, Y.; Collonges, F.; Menet, A.; Pautet, F.; Poncet, A.;
Descotes, G. Bull. Soc. Chim. Fr. 2006, 5, 123.
(2) The NOESY experiment was performed on compound 6 and
showed a correlation between the proton α to the carbonyl and the
methine vinyl proton, thereby confirming that they were cis.
(3) (a) Chen, D. Y.-K.; Pouwer, R. H.; Richard, J.-A. Chem. Soc. Rev.
2012, 41, 4631−4642. (b) Taylor, R. E.; Engelhardt, F. C.; Schmitt, M. J.
Tetrahedron 2003, 59, 5623−5634. (c) Donaldson, W. A. Tetrahedron
2001, 57, 8589−8627. (d) Jung, M. E.; Chang, J. J. Org. Lett. 2010, 12,
2962−2965.
(4) (a) Lu, Z.; Shen, M.; Yoon, T. P. J. Am. Chem. Soc. 2011, 133,
1162−1164. (b) Tamaki, T.; Nagata, M.; Ohashi, M.; Ogoshi, S. Chem. -
Eur. J. 2009, 15, 10083−10091. (c) Clarke, C.; Foussat, S.; Fox, D. J.;
Pedersen, D. S.; Warren, S. Org. Biomol. Chem. 2009, 7, 1323−1328.
(d) Yadav, V. K.; Kumar, N. V. Chem. Commun. 2008, 3774−3776.
D
DOI: 10.1021/acs.orglett.6b02588
Org. Lett. XXXX, XXX, XXX−XXX