Substituent effects on rates of rhodium-catalyzed allene cyclopropanation

Article abstract of DOI:10.1016/j.tetlet.2010.09.143

Rates of cyclopropanation for mono- and disubstituted allenes have been measured relative to standard substrates in reaction with aryldiazoacetate esters catalyzed by Rh₂(S-DOSP)₄. Phenylallene derivatives exhibited a linear correlation of rate with σ⁺ coefficients, indicating a resonance-based effect, though the magnitude of the effect for allenes is less than that reported for other cyclopropanations. Relative reaction rates for aliphatic allenes were found to be similar to those for aryl-substituted allenes, but silicon substitution was found to give a 5- to 14-fold rate increase. The rate enhancement effect for 1-silyl allenes can partially make up for loss of rate and regioselectivity, with 1-trimethylsilyl-1,2-butadiene exhibiting high levels of enantioselectivity and diastereoselectivity in reaction with the chiral catalyst.

Full text of DOI:10.1016/j.tetlet.2010.09.143

Tetrahedron Letters 51 (2010) 6429–6432
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Substituent effects on rates of rhodium-catalyzed allene cyclopropanation
⇑
Timothy M. Gregg , Russell F. Algera, John R. Frost, Furqan Hassan, Robert J. Stewart
Department of Chemistry and Biochemistry, Canisius College, Buffalo, NY 14208, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Rates of cyclopropanation for mono- and disubstituted allenes have been measured relative to standard
substrates in reaction with aryldiazoacetate esters catalyzed by Rh2(S-DOSP)4. Phenylallene derivatives
exhibited a linear correlation of rate with r+ coefficients, indicating a resonance-based effect, though
the magnitude of the effect for allenes is less than that reported for other cyclopropanations. Relative
reaction rates for aliphatic allenes were found to be similar to those for aryl-substituted allenes, but sil-
icon substitution was found to give a 5- to 14-fold rate increase. The rate enhancement effect for 1-silyl
allenes can partially make up for loss of rate and regioselectivity, with 1-trimethylsilyl-1,2-butadiene
exhibiting high levels of enantioselectivity and diastereoselectivity in reaction with the chiral catalyst.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 12 September 2010
Revised 28 September 2010
Accepted 28 September 2010
Available online 7 October 2010
Keywords:
Allene cyclopropanation
Rhodium carboxylate
Hammett equation
Beta-silicon effect
Alkylidenecyclopropane
Intermolecular alkene cyclopropanation has been shown to be
an effective enantioselective methodology for the preparation of
densely functionalized cyclopropanes in good yields and high
enantiomeric excess.1–5 The reaction of aryl-substituted alkenes
is particularly high-yielding and provides very high levels of dia-
stereoselectivity in rhodium(II) mediated reactions with donor-
substituted diazoacetates. This efficiency is attributed, in part, to
rate enhancement due to conjugation between the reacting alkene
and the aromatic ring.6
We have recently reported the highly enantioselective cyclo-
propanation of substituted allenes using aryldiazoacetate-derived
rhodium carbenoids, but inconsistent yields were suggestive of sig-
nificantly reduced cyclopropanation rates, because slower reac-
tions allow competitive dimerization and other side-reactions of
the carbenoid species.7 The rate of allene cyclopropanation may
suffer due to the lack of conjugation-derived stabilization in the
transition state as well as to the inherent lower reactivity of an
sp-hybridized nucleophile proceeding through a three-membered
ring transition state.8
In our earlier work with aryldiazoacetate-derived carbenoids,
electronic influences also seemed to play a role in substrate reactiv-
ity, with an electron-withdrawing substituent causing reduced yield
and an electron-donor substituent greatly improving the yield.7 To
get a better understanding of steric and electronic constraints in
allene cyclopropanation, we have undertaken the current study,
with the aim of rationalizing effects of specific groups and substitu-
tion patterns. We compare reaction rates for a variety of allenes
(Eq. 1), observing effects that either accelerate or decelerate their
reaction with diazoacetate 2 in the presence of a Rh(II) catalyst.
R2
CO2Me
R2
N2
CO2Me
Ar
Ar =
p-BrC6H4
R1
2
Ar
ð1Þ
R3
R1
Rh2(S-DOSP)4
Hexane, rt
R3
1
For arylallenes (1, R1 = Ar, R2 = R3 = H), cyclopropanation occurs
at the terminal -bond, which is not conjugated with the aryl ring.
Electronic effects are not communicated from the aryl ring to the
reactive -bond via direct resonance. We sought to characterize
p
The literature of allene addition reactions paints a varied pic-
p
ture of factors that affect reactivity of the allene
p-bonds. For
the kind of effects aryl substituents have on the kinetics of
cyclopropanation.
osmylation9 and epoxidation,10 substituents activate the nearer
bond for reaction, while with palladium11–13 and other metal spe-
cies,14–16 an alkyl substituent at one end hinders addition, leading
to preferential involvement of the less substituted bond. Allene
cyclopropanation using malonate- and acetate-derived rhodium
carbenoids is also strongly directed by such steric factors, and is
well precedented.17
We compared the reaction rate of phenylallene, 3, with that of
para-substituted arylallene substrates 4a–d by direct competition.
A 1:1 mixture of two allenes (3 and one of 4a–d) with 2 (0.1 equiv
relative to allene) in the presence of Rh2(S-DOSP)4 (see Fig. 1) in
hexane at rt gave a mixture of cyclopropanation products. The ratio
of products in the crude reaction mixture, as observed by 1H NMR,
was taken to be the ratio of reaction rates, kAr/kPh. The observed
rate ratios for substrates 4a–d relative to 3 are reported in Table 1.
⇑
Corresponding author. Tel.: +1 716 888 2259; fax: +1 716 888 3112.
E-mail address: greggt@canisius.edu (T.M. Gregg).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.09.143

6430
T. M. Gregg et al. / Tetrahedron Letters 51 (2010) 6429–6432
Support for this came from an experiment wherein styrene and
Coplanar
H
R
H
R
3 reacted in direct competition under standard cyclopropanation
conditions (Table 2, entry 1). Products derived from both sub-
strates were observed by 1H NMR. However, the very small propor-
tion of product formed from 3 allowed only a rough estimate of the
reaction rate ratio, which we put as between 20 and 40 to 1 (see
Supplementary data). This rate enhancement of styrene over 3 is
roughly the same as that reported for styrene over other non-
conjugated alkenes. Davies reported a 50 to 1 ratio for styrene
versus 1-hexene.6
H
H
O Rh
δ+
Ph
E
.
N
O Rh
4
E
SO2Ar
Ph
Rh2(S-DOSP)4
Ar = p-C12H25-C6H4
Rh2L4
H
H
H
Rh2L4
TS1
Figure 1. Calculated allene cyclopropanation TS, which suggests bond formation
out of conjugation when R is an aromatic ring.
Because allenes gain no reactivity increase when substituted
with a phenyl ring, rates of cyclopropanation of aryl allenes and
alkyl allenes should be similar. Competition between 3 and
1,2-nonadiene gave products in a 1 to 2 ratio (Table 2, entry 2),
confirming the lack of rate enhancement with aromatic substitu-
ents. Moving to a consideration of steric effects, for cyclopropana-
tion of monosubstituted allenes, TS1 suggests that a single R
group, whether alkyl or aryl, would not appear to offer hindrance
to substrate–carbenoid interaction. For 1,1-disubstituted allenes,
however, one group would be oriented away, and the second
group (the smaller of the two if they are not identical) would
be projecting toward the carbenoid, potentially raising the TS
energy and slowing the reaction rate. We first addressed the
question of what might be the magnitude of this geminal substi-
tution effect.
Cyclopropanation of 3 in competition with 3-phenyl-1,2-buta-
diene, 7c, gave a mixture of products (Table 2, entry 3). The
more-substituted allene, 7c, reacted 8 times slower than 3. The
8-fold rate reduction may be rationalized as a result of steric
crowding of the allene approach trajectory. It should be noted that
this by no means rules out an electronic rate effect due to the
methyl group that may be partially offsetting the steric effect.
For 1,1-disubstituted allenes, we again compared aryl and alkyl
substituents (Table 3, entry 1), with the aryl substrate, 7c, reacting
3 times slower than the alkyl, 9. The general observation of re-
duced rate for aryl substrates compared to alkyl may be steric or
electronic in origin, and examination of rate effects for a wider ar-
ray of substrates may shed light on such effects in the future. The
value of such studies, though, may be limited because most of the
1,1-disubstituted allenes we have investigated suffer from low
yields in these cyclopropanation reactions.
Table 1
Arylallene cyclopropanation rate relative to phenylallene, 3
CO2Me
Ph
Ph
p-BrC6H4
2
3
5
+
+
Rh2(S-DOSP)4
Hexane, rt
CO2Me
Ar
Ar
p-BrC6H4
4a-d
6a-d
Allene
Ar
Product ratio (6:5)
kAr/kPh
4a
4b
4c
4d
p-iPrOC6H4
p-iPrC6H4
p-ClC6H4
1.5:1
1:1
1:1.2
1:1.5
1.5
1.0
0.83
0.67
p-CF3C6H4
Substrate 4a is activated by the electron-releasing alkoxy group,
while 4c and 4d are deactivated by electron-withdrawing groups,18
consistent with a species reacting with an electron-deficient carb-
enoid.19 A linear correlation of log (kAr/kPh) with Hammett
r
+ coef-
ficients20 indicates that substituents exhibit a resonance effect on
the aromatic ring. The observed
q
value of ꢀ0.25, however, sug-
gests that the ring’s effect, in turn, on the allene reaction is greatly
attenuated compared to that observed for rhodium carbenoid
cyclopropanation of styrenes and cyclopropenation of aryl al-
kynes.21 Results in the literature support
q
values of around ꢀ0.9
for similar reactions of substituted styrenes22 and arylacetylenes.23
While a resonance-like electronic effect gives a small increase in
reactivity to donor-substituted arylallenes, nonetheless, they are
not conjugated, and we expected arylallene reactivity to be greatly
diminished relative to that of styrenes. This has been discussed in
light of our TS model for allene cyclopropanation, TS1 (see Fig. 1).7
TS1 depicts developing d+ associated with an electron-deficient
An important rate effect that could make substituted allenes
more attractive in carbenoid reactions is the rate acceleration
due to silicon substitution.25 3-(Trimethylsilyl)-1,2-butadiene
(11) was found to react 14 times faster than 9 (Table 3, entry 2).
In this case, a strong electronic effect is likely at work.7 Again refer-
ring to TS1, where R = TMS, the C–Si bond would be coplanar with
orbital in the plane of the styryl p-bond (R = Ar). As such, develop-
ing charge is not stabilized by conjugation with the aryl group.24
Table 2
Rate comparison for other substrates relative to phenylallene, 3
2, Rh2(S-DOSP)4
Hexane, rt
Competing
Substrate
Competing
Product
8a-c
Ph
+
5 +
3
7a-c
Entry
1
Competing substrate
Product (8 Ar = p-BrC6H4)
Product ratio (5:8)
CO2Me
Ph
1:>20
Ph
Ar
7a
8a
CO2Me
Ar
8b
C6H13
7b
C6H13
2
3
1:2
8:1
CH3
CH3
CO2Me
Ar
Ph
Ph
7c
8c

T. M. Gregg et al. / Tetrahedron Letters 51 (2010) 6429–6432
6431
Table 3
Silylallene rate comparison
Entry
Competing substrates
Cyclopropanation products (Ar = p-BrC6H4)
Product ratio
CH3
CH3
CH3
CO2Me
Ar
1
8c
1:3
Ph
H3C
H3C
10
7c
9
CH3
CH3
14:1a
5:1
CO2Me
Ar
2
3
9
3
10
Me3Si
Me3Si
12
11
CO2Me
Ar
Ph
11
12
5
CO2Me
Ar
Me3Si
CH3
Me3Si
13
4
11
12
1.5:1
14
CH3
a
See Ref. 7.
CH3
CO2Me
Ar
Me
The relative stereochemistry of 14 was assigned based on the
downfield chemical shift of the cyclopropane methine proton trans
to Ar (see Fig. 2). A shift about 1 ppm further upfield would be ex-
pected26 for the proton cis to Ar, as has been seen consistently for
products such as 12.
CO2Me
Ar
Me3Si
Me3Si
Htrans
Hcis δ=1.68 ppm
Htrans
δ=2.50 ppm
12
14
δ=2.76 ppm
In conclusion, we have demonstrated that electronic effects can
influence reactivity of an allene toward electrophilic rhodium carb-
enoids, though only a small effect can be attributed to resonance,
owing to the cumulene structure of the allene. Silyl groups at-
Figure 2. NMR shift correlation puts the methine H of 14 trans to Ar.
the evolving cyclopropane bond, ideally oriented to stabilize
positive charge developing at the b-position through hyper-conju-
gation. In direct competition, compound 11 reacted 5 times faster
than 3 (Table 3, entry 3). The rate acceleration attributed to the si-
lyl group in compound 11 was more than sufficient to make up for
the added steric hindrance attributed to the methyl group.
Considering the rate acceleration provided by a silyl group, we
investigated the possibility of performing enantioselective cyclo-
propanation on 1,3-disubstituted allenes. Such reactions would
give alkylidenecyclopropane products with an additional chiral
center. Without a silyl substituent, however, rate and regioselectiv-
ity for such allenes appear to be compromised. Indeed, we found
that 6-phenyl-2,3-hexadiene gave an inseparable mixture of cyclo-
propanation products, 16 and 17, indicating limited differentiation
of the two ends of the allene in the reaction TS (Eq. 2).
With an eye to boosting the steric and electronic differentia-
tion at work in a 1,3-disubstituted allene, we next investigated
1-(trimethylsilyl)-1,2-butadiene, 13. A 2-fold excess of racemic
13 reacted with 2, in the presence of Rh2(S-DOSP)4, and provided
a single cyclopropanation product, 14, with no regio- or diastero-
isomeric cyclopropane products discernable by NMR. The product
exhibited 91% ee, consistent with the enantioselectivity typically
achieved using Rh2(S-DOSP)4.7
tached to the non-reacting
b-silicon effect, accelerating reaction at the other
p
-bond, on the other hand have a strong
-bond. Of prime
p
importance for the use of allenes in alkylidenecyclopropane meth-
odology will be overcoming steric encumbrance in the substrate
caused by substituents, which can lead to reduced yields.
The feasibility of kinetic resolution of chiral allenes such as 13
and their use in practical methodology for enantioselective con-
struction of densely functionalized alkylidenecyclopropanes are
continuing and will be reported in due course.
Acknowledgments
We are grateful for a gift of Rh2(S-DOSP)4 from Prof. Huw M.L.
Davies, of Emory University. This work was supported by Canisius
College and by fellowship support from the Howard Hughes Med-
ical Research Institute for F.H., R.F.A. and R.J.S.
Supplementary data
Supplementary data associated (compound and competition
reaction characterization) with this article can be found, in the
online version, at doi:10.1016/j.tetlet.2010.09.143.
CO2Me
Ar
References and notes
Ph
2
CH3
1. Davies, H. M. L.; Bruzinski, P.; Lake, D. H.; Kong, N.; Fall, M. J. J. Am. Chem. Soc.
1996, 118, 6897–6907.
2. Bykowski, D.; Wu, K.-H.; Doyle, M. P. J. Am. Chem. Soc. 2006, 128, 16038–16039.
3. Lebel, H.; Marcoux, J.-F.; Molinaro, C.; Charette, A. B. Chem. Rev. 2003, 103, 977–
1050.
16
CH3
Rh2(S-DOSP)4
Hexane, rt
Ph
+
ð2Þ
ð3Þ
15
Ph
CO2Me
Ar
4. Reissig, H. U.; Zimmer, R. Chem. Rev. 2003, 103, 1151–1196.
5. Doyle, M. Angew. Chem., Int. Ed. 2009, 48, 850–852.
17
6. Davies, H. M. L.; Panaro, S. A. Tetrahedron 2000, 56, 4871–4880.
7. Gregg, T. M.; Farrugia, M. K.; Frost, J. R. Org. Lett. 2009, 11, 4434–4436.
8. Melloni, G.; Modena, G.; Tonellato, U. Acc. Chem. Res. 1981, 14, 227–233.
9. Fleming, S. A.; Liu, R. M.; Redd, J. T. Tetrahedron Lett. 2005, 46, 8095–8098.
10. Ghosh, P.; Cusick, J. R.; Inghrim, J.; Williams, L. J. Org. Lett. 2009, 11, 4672–4675.
11. Abe, Y.; Kuramoto, K.; Ehara, M.; Nakatsuji, H.; Suginome, M.; Murakami, M.;
Ito, Y. Organometallics 2008, 27, 1736–1742.
2
CO2Me
Ar
Me3Si
Me3Si
CH3
Rh2(S-DOSP)4
Hexane, rt
CH3
13
14
12. Burks, H. E.; Liu, S.; Morken, J. P. J. Am. Chem. Soc. 2007, 129, 8766–8773.

6432
T. M. Gregg et al. / Tetrahedron Letters 51 (2010) 6429–6432
13. Kumareswaran, R.; Shin, S.; Gallou, I.; RajanBabu, T. V. J. Org. Chem. 2004, 69,
7157–7170.
20. Exner, O. In Correlation Analysis in Chemistry; Chapman, N. B., Shorter, J., Eds.;
Plenum Press: New York, 1978; pp 439–489.
14. Bai, T.; Zhu, J.; Xue, P.; Sung, H. H.-Y.; Williams, I. D.; Ma, S.; Lin, Z.; Jia, G.
Organometallics 2007, 26, 5581–5589.
21. See the Supplementary data for a comparison of allene, alkene and alkyne
Hammett plots.
15. Yoshida, Y.; Murakami, K.; Yorimitsu, H.; Oshima, K. J. Am. Chem. Soc. 2010, 132,
8878–8879.
16. Wipf, P.; Pierce, J. G. Org. Lett. 2005, 7, 3537–3540.
17. Ma, S. Chem. Rev. 2005, 105, 2829–2872.
18. Small differences in rate ratios are probably not significant considering the
level of uncertainty in NMR integrations.
19. Nowlan, D. T.; Gregg, T. M.; Davies, H. M. L.; Singleton, D. A. J. Am. Chem. Soc.
2003, 125, 15902–15911.
22. Hansen, J.; Autschbach, J.; Davies, H. M. L. J. Org. Chem. 2009, 74, 6555–6563.
23. Davies, H. M. L.; Lee, G. H. Org. Lett. 2004, 6, 1233–1236.
24. The possibility of a bridging interaction between the aryl group and the
developing cyclopropane bond is intriguing, but would require that the ring be
turned out of conjugation with the intervening olefin.
25. Huval, C. C.; Singleton, D. A. J. Org. Chem. 1994, 59, 2020–2024.
26. Closs, G. L.; Moss, R. A. J. Am. Chem. Soc. 1964, 86, 4042–4053.