The scope of catalytic enantioselective tandem carbonyl ylide formation - Intramolecular [3 + 2] cycloadditions

Article abstract of DOI:10.1021/jo0343735

Catalytic enantioselective tandem carbonyl ylide formation-intramolecular 1,3-dipolar cycloaddition reactions of 2-diazo-3,6-diketoesters show promising scope in terms of asymmetric induction as the tethered alkene/alkyne dipolarophile component is varied. Cycloadditions were found to occur in moderate to very good yields, with a difference in ee exhibited by the electronically different 2-diazo-3,6-diketoesters 1, 25 and 33, 34. Values for ee of up to 90% for alkene dipolarophiles and up to 86% for alkyne dipolarophiles were obtained.

Full text of DOI:10.1021/jo0343735

Th e Scop e of Ca ta lytic En a n tioselective Ta n d em Ca r bon yl Ylid e
F or m a tion -In tr a m olecu la r [3 + 2] Cycloa d d ition s
David M. Hodgson,* Agne`s H. Labande, Franc¸oise Y. T. M. Pierard, and
Mar´ıa AÄ . Expo´sito Castro
Dyson Perrins Laboratory, Department of Chemistry, University of Oxford, South Parks Road,
Oxford OX1 3QY, U.K.
david.hodgson@chem.ox.ac.uk
Received March 21, 2003
Catalytic enantioselective tandem carbonyl ylide formation-intramolecular 1,3-dipolar cycloaddition
reactions of 2-diazo-3,6-diketoesters show promising scope in terms of asymmetric induction as
the tethered alkene/alkyne dipolarophile component is varied. Cycloadditions were found to occur
in moderate to very good yields, with a difference in ee exhibited by the electronically different
2-diazo-3,6-diketoesters 1, 25 and 33, 34. Values for ee of up to 90% for alkene dipolarophiles and
up to 86% for alkyne dipolarophiles were obtained.
SCHEME 1. Ta n d em Ca ta lyzed Ca r bon yl Ylid e
F or m a tion -Cycloa d d ition
Catalytic asymmetric synthesis is widely recognized as
an attractive and powerful concept to access enantioen-
riched materials.1 The utility of developing this concept
in the context of tandem carbonyl ylide formation-1,3-
dipolar cycloaddition reaction sequences from diazo
substrates (Scheme 1, for example) lies in the rapid
generation of molecular complexity being coupled with
simultaneous control over the generation of absolute
stereochemistry.2 Such cascade processes, catalyzed in
the racemic sense using achiral copper or rhodium
complexes, were developed by Ibata and, particularly,
Padwa as a method to concisely construct oxapolycycles
containing an embedded reduced furan ring.3 Since the
late 1960s and with increasing intensity from approxi-
mately 1990 onward, chiral catalysts have been investi-
gated with considerable success for generating enan-
tioenriched products in many important transformations
of diazo compounds, such as cyclopropanations and, more
recently, C-H insertions.3b In contrast, catalytic enan-
tioselective rearrangements and cycloadditions involving
ylides from diazo compounds have been investigated only
sporadically and with comparatively little success until
the past few years.4 In part, this reflects the challenging
role required of the chiral catalyst in which, for good
enantioselectivity in the reaction shown in Scheme 1, for
example, it must smoothly transform the diazo substrate
1 into a catalyst-associated ylide 2 and then efficiently
mediate facial selectivity in the cycloaddition event
[making the reasonable assumption that cycloaddition via
the achiral (catalyst-free) carbonyl ylide would lead to a
racemic cycloadduct]. Following our first demonstration
in 19975 of enantioselective carbonyl ylide cycloaddition
in up to 52% ee with 2-diazo-3,6-diketoesters such as 1
utilizing Davies’ prolinate catalyst Rh2(S-DOSP)4 4,6
Hashimoto and co-workers subsequently reported high
levels of asymmetric induction (up to 92% ee) with
catalyst Rh2(S-BPTV)4 5 in intermolecular cycloadditions
using diazodiones with DMAD as the dipolarophile.7
Although application of Hashimoto’s optimized catalyst-
solvent combination (Rh2(S-BPTV)4 5, PhCF3 as solvent)
(1) Catalytic Asymmetric Synthesis, 2nd ed.; Ojima, I., Ed.; Wiley-
VCH: New York, 2000.
(2) (a) Cycloaddition Reactions in Organic Synthesis; Kobayashi, S.,
J ørgensen, K. A., Eds.; Wiley-VCH: Weinheim, 2002. (b) Synthetic
Applications of 1,3-Dipolar Cycloaddition Chemistry Toward Hetero-
cycles and Natural Products; Padwa, A., Pearson, W. H., Eds.; J ohn
Wiley & Sons: New York, 2002.
(3) (a) Padwa, A.; Weingarten, M. D. Chem. Rev. 1996, 96, 223-
269. (b) Doyle, M. P.; McKervey, M. A.; Ye, T. Modern Catalytic
Methods for Organic Synthesis with Diazo Compounds; J ohn Wiley &
Sons: New York, 1998, Chapter 7. (c) Clark, J . S. Nitrogen, Oxygen
and Sulfur Ylide Chemistry; Oxford University Press: Oxford, 2002.
(d) Mehta, G.; Muthusamy, S. Tetrahedron 2002, 58, 9477-9504.
(4) (a) Hodgson, D. M.; Pierard, F. Y. T. M.; Stupple, P. A. Chem.
Soc. Rev. 2001, 30, 65-99. (b) Hodgson, D. M.; Stupple, P. A.; Forbes,
D. C. In Rodd’s Chemistry of Carbon Compounds, Topical Volume,
Asymmetric Catalysis; Sainsbury, M., Ed.; Elsevier: Oxford, 2001; p
6599. (c) Kitagaki, S.; Hashimoto, S. J . Synth. Org. Chem. J pn. 2001,
59, 1157-1169.
(5) Hodgson, D. M.; Stupple, P. A.; J ohnstone, C. Tetrahedron Lett.
1997, 38, 6471-6472.
(6) (a) Davies, H. M. L.; Bruzinski, P. R.; Lake, D. H.; Kong, N.;
Fall, M. J . J . Am. Chem. Soc. 1996, 118, 6897-6907. (b) Davies, H.
M. L. Eur. J . Org. Chem. 1999, 2459-2469.
(7) Kitagaki, S.; Masahiro, A.; Kataoka, O.; Matsuno, K.; Umeda,
C.; Watanabe, N.; Hashimoto, S. J . Am. Chem. Soc. 1999, 121, 1417-
1418.
10.1021/jo0343735 CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/04/2003
J . Org. Chem. 2003, 68, 6153-6159
6153

Hodgson et al.
tioselective intramolecular process, only studies with a
simple, unsubstituted terminal alkene have been re-
ported in detail thus far (Scheme 1).9 Thus, the current
study would give an idea of the scope of this methodology
for asymmetric synthesis and hopefully provide some
insight into the nature of the factors influencing asym-
metric induction. To achieve this aim, it was considered
important to study substrates that allowed examination
of variations in the length of the tether to the dipolaro-
phile, as well as its level and pattern of alkyl substitution
and degree of unsaturation (alkene, alkyne).15 A strategy
is also demonstrated that permits greater flexibility in
the nature of the dipolarophile substituent, especially for
the introduction of polar functionalities.
The original synthesis of 19 was adapted for the
construction of cycloaddition substrates 25-34 (Scheme
2). In this strategy, unsaturated 4-ketoacids 8-14 were
chemoselectively homologated by activation as the acyl
imidazolides and reaction with the magnesium salt of
mono tert-butyl malonate.16 This chemistry led to 3,6-
diketoester functionality from which the desired cycload-
dition substrates 25-34 arose following diazo transfer
at the doubly activated methylene group. The 4-ketoacids
8-14 were accessed via lithiation of 2,3-dihydrofuran and
alkylation with the appropriate alkenyl iodide or (TMS-
protected) alkyne, followed by direct oxidation of the
crude alkylated dihydrofuran using J ones’ reagent.17 For
the cases of the TMS-protected alkynes, protodesilylation
conveniently occurred concomitantly during oxidation,
leading to the terminal alkynes 11 and 13. (E)-Alkene
25 and (Z)-alkene 26 were synthesized to study if the
asymmetric cycloaddition process exhibited stereospeci-
ficity. Whereas (E)-alkene 25 originated from com-
mercially available (E)-4-hexen-1-ol (E:Z, 98:2 by GC),
Lindlar reduction of the acetylenic diketoester derived
from 4-ketoacid 12 gave, after diazo transfer, (Z)-alkene
26 (Z:E, 90:10).
F IGURE 1. Chiral rhodium catalysts.
to ester 1 at 25 °C resulted in only essentially racemic
cycloadduct 3 (90% yield, 1% ee), further catalyst studies
using ester 1 eventually led us to a hydrocarbon-soluble
variant of Pirrung’s phosphate catalyst Rh2(R-BNP)4 6:8
Rh2(R-DDBNP)4 7 was found capable of generating
intramolecular cycloadduct (+)-3 (absolute configuration
as shown in Scheme 1) in up to 90% ee.9 To probe
electronic effects of the dipole on asymmetric induction,
enantioselective cycloadditions of diazoketone-derived
aryl-substituted carbonyl ylides have been studied.10
Using Rh2(R-DDBNP)4 7 we have also recently reported
significant enantiomeric excesses (up to 92% ee) in
intermolecular cycloadditions with strained alkenes (such
as norbornene) and 2-diazo-3,6-diketoesters similar to 1,
but lacking side chain unsaturation.11 Enantioselective
carbonyl ylide type cycloadditions of (aromatic) oxidopy-
ryliums have also been studied. Although only low ee
values have been observed so far in intramolecular
cycloadditions of oxidopyryliums (up to 19% ee),12 for the
intermolecular process up to 93% ee using DMAD and
catalyst 5 has been recorded13 and up to 98% ee has been
reported using a 3-acryloyl-2-oxazolidinone dipolarophile
with a chiral Lewis acid-Rh catalyst combination.14
The principal focus of the work described herein was
to develop an appreciation of the effect on ee of changing
the dipolarophile in intramolecular carbonyl ylide cy-
cloadditions, using substrates related to 1 with catalysts
4 and 7. With respect to the dipolarophile in the enan-
To increase flexibility in the nature of the dipolarophile
in intramolecular cycloaddition substrates, we examined
a cross-metathesis strategy.18 With methyl acrylate and
the 3,6-diketoesters 49 and 17 (precursors to cycloaddi-
tion substrates 1 and 27), metathesis reactions using the
second-generation ruthenium catalyst (PCy3)(L)Rud
CHPhCl2 (L ) 1,3-dimesityl-4,5-dihydroimidazol-2-ylidene)
pleasingly produced unsaturated diketodiesters 23 and
24 in excellent yields (96% and 87%, respectively) and
with good control over alkene stereochemistry (E:Z, 93:
3) (Scheme 3). Furthermore, subsequent diazo transfer
proceeded smoothly in the presence of the R,â-unsatur-
ated ester functionality to deliver cycloaddition precur-
sors 33 and 34 in good yields (84% and 82%, respectively).
(8) Pirrung, M. C.; Zhang, J . Tetrahedron Lett. 1992, 33, 5987-5990.
(9) (a) Hodgson, D. M.; Stupple, P. A.; J ohnstone, C. J . Chem. Soc.,
Chem. Commun. 1999, 2185-2186. (b) Hodgson, D. M.; Stupple, P.
A.; Pierard, F. Y. T. M.; Labande, A. H.; J ohnstone, C. Chem. Eur. J .
2001, 7, 4465-4476.
(15) Hodgson, D. M.; Labande, A. H.; Pierard, F. Y. T. M. Synlett
2003, 59-62.
(16) (a) Brooks, D. W.; Lu, L. D.-L.; Masamune, S. Angew. Chem.,
Int. Ed. Engl. 1979, 18, 72-73. (b) Ghosh, M.; Miller, M. J . Tetrahedron
1996, 52, 4225-4238.
(10) (a) Hodgson, D. M.; Glen, R.; Redgrave, A. J . Tetrahedron Lett.
2002, 43, 3927-3930. (b) Hodgson, D. M.; Glen, R.; Grant, G. H.;
Redgrave, A. J . J . Org. Chem. 2003, 68, 581-586.
(11) Hodgson, D. M.; Labande, A. H.; Glen, R.; Redgrave, A. J .
Tetrahedron: Asymmetry 2003, 14, 921-925.
(17) Tschantz, M. A.; Burgess, L. E.; Meyers, A. I Organic Syntheses;
Wiley: New York, 1998; Collect. Vol. IX, pp 530-533.
(18) (a) Scholl, M.; Trnka, T. M.; Morgan, J . P.; Grubbs, R. H.
Tetrahedron Lett. 1999, 40, 2247-2250. (b) Scholl, M.; Ding, S.; Lee,
C. W.; Grubbs, R. H. Org. Lett. 1999, 1, 953-956. (c) Chatterjee, A.
K.; Morgan, J . P.; Scholl, M.; Grubbs, R. H. J . Am. Chem. Soc. 2000,
122, 3783-3784. (d) Chatterjee, A. K.; Choi, T.-L.; Grubbs, R. H. Synlett
2001, SI, 1034. (e) Huang, J .; Stevens, E. D.; Nolan, S. P.; Petersen,
J . L. J . Am. Chem. Soc. 1999, 121, 2674-2678. (f) J afarpour, L.; Hillier,
A. C.; Nolan, S. P. Organometallics 2002, 21, 442-444.
(12) Hodgson, D. M.; Stupple, P. A.; J ohnstone, C. Arkivoc 2003,
vii, 49-58.
(13) Kitagaki, S.; Yasugahira, M.; Anada, M.; Nakajima, M.; Hash-
imoto, S. Tetrahedron Lett. 2000, 41, 5931-5935.
(14) Suga, H.; Inoue, K.; Inoue, S.; Kakehi, A. J . Am. Chem. Soc.
2002, 124, 14836-14837.
6154 J . Org. Chem., Vol. 68, No. 16, 2003

Intramolecular Carbonyl Ylide Cycloadditions
SCHEME 2. Syn th esis of 2-Dia zo-3,6-d ik etoester Su bstr a tesa
a
(a) t-BuLi (1.2 equiv), RCHdCH2CH2(CH2)nCH2I (1 equiv), THF, -60 to 25 °C, 18 h; (b) J ones’ reagent (3 equiv), THF, 0 °C, 18 h; (c)
carbonyldiimidazole (1.2 equiv), THF, 0 °C, then Mg(O2CCH2CO2t-Bu)2 (1.1 equiv), THF, 25 °C, 18 h, then H3O+; (d) 4-(NHAc)C6H4SO2N3
(1.1 equiv), Et3N (1.1 equiv), MeCN, 0 °C, 5 h; (e) t-BuLi (1.2 equiv), RCtCHCH2(CH2)nCH2I (1 equiv), THF, -60 to 25 °C, 18 h; (f) H2
(1 atm), Lindlar’s catalyst, quinoline, petroleum ether, 25 °C, 5.5 h.
SCHEME 3. Syn th esis of 33 a n d 34 by Ca ta lyzed Cr oss-Meta th esis a n d Dia zo Tr a n sfer Rea ction s
The viability of the diazo substrates to undergo the
desired ylide formation-cycloaddition process was es-
tablished by treatment with rhodium(II) acetate in
CH2Cl2 at 25 °C (40-80% yields of cycloadducts); these
racemic cycloadducts were also used for establishing
enantiomeric purity determination assays by GC.19 Both
Rh2(S-DOSP)4 4 and Rh2(R-DDBNP)4 7 were then exam-
ined with the cycloaddition substrates in hexane. Results
in CH2Cl2 are described in Supporting Information; as
previously found with 1,9 the ee values were uniformly
lower than those in hexane. Cycloaddition substrates (E)-
and (Z)-alkenes 25 and 26, differing only in alkene
stereochemistry, successfully underwent cycloaddition to
give structurally different (epimeric) cycloadducts 35 and
36, respectively (Scheme 4). The absolute configurations
of the predominant cycloadduct enantiomers are tenta-
tively assigned as those shown, by analogy with cycload-
duct (+)-3 of known absolute configuration.9
relative proportion of (chromatographically separable)
cycloadduct that could be attributed to the minor isomer
in each case. nOe experiments with cycloadduct 36
established that the cis methyl group in the precursor
alkene 26 remained cis to the side chain that evolved into
the cyclopentyl group.20 Further evidence for the stere-
ochemistry of cycloadduct 36 is found in the fact that
hydrogenation of 40 (vide infra) leads to 36, and not 35.
Cycloadduct 35 is therefore unambiguously assigned as
the methyl epimer of 36, and the stereochemistry of
cycloadduct 38 arising from trans alkene 28 is assigned
by analogy to that of 35. We conclude from the reactions
of alkenes 25 and 26 that the intramolecular addition to
simple nonpolarized alkenes of the putative catalyst
complexed carbonyl ylide (cf 2, Scheme 1) is stereospe-
cific. As a corollary we note that these reactions illustrate
for the first time the synthetically important ability to
control up to four stereocenters in such an enantioselec-
tive intramolecular cycloaddition process. We were also
pleased to observe that ee values (up to 80% and 83%
Although both cycloaddition substrates 25 and 26
contained minor amounts of the other alkene stereoiso-
mer (as discussed above), there was no increase in the
(20) Key nOe’s for cycloadduct 36 are shown in Supporting Informa-
tion.
(19) See Supporting Information for details.
J . Org. Chem, Vol. 68, No. 16, 2003 6155

Hodgson et al.
SCHEME 4. Cycloa d d ition Rea ction s w ith Alk en e
Dia zo Su bstr a tes 25-28
presumed slower rate of cycloaddition. A slower rate of
cycloaddition could potentially affect the level of asym-
metric induction if catalyst dissociation from the putative
catalyst-associated ylide (eg 2, Scheme 1) occurred to give
(some of) the achiral catalyst-free carbonyl ylide prior to
the cycloaddition step.4a Intriguingly, a more substituted
alkene appeared more sensitive to variation in tether
length with respect to ee, since for the (E)-alkene 28 there
was a noticeable reduction in ee compared with (E)-
alkene 25 (e.g., using Rh2(R-DDBNP)4 7 at 0 °C, 59% ee
compared to 72% ee, respectively).
Cycloadditions with the substrates bearing an ester
conjugated to the alkene dipolarophile proceeded in good
yields, whichever catalyst was used (Scheme 5). However,
the ee values obtained were much lower than those
observed with unpolarized alkene dipolarophiles. We did
not study Hashimoto’s optimized catalyst-solvent com-
bination (Rh2(S-BPTV)4 5 in PhCF3)7 with the unpolar-
ized systems above, since that combination had previ-
ously been shown with 1 to generate only racemic
cycloadduct 3.9 However, it was of interest to determine
if the presence of an electron-deficient dipolarophile
altered the outcome. In the event, although cycloadditions
proceeded efficiently (93% and 95% for 43 and 44,
respectively), ee values were low (18% and 17%, respec-
tively). Our previous studies of related carbonyl ylide
cycloadditions indicated that a complex blend of elec-
tronic effects from the dipole and dipolarophile, together
with the nature of the catalyst, contribute to the origin
of asymmetric induction.10 However, the current results
with catalysts 4 and 7 (Schemes 4 and 5) follow a trend
that when the dipole and dipolarophile have similar
electronic characteristics (both electron-deficient for 33
and 34) the ee is lower than when they oppose (electron-
deficient dipole and simple nonpolarized alkene dipo-
larophile).
for 35 and 36, respectively) were only slightly lower than
the best obtained with the unsubstituted alkene sub-
strate 1. The reaction’s ability to tolerate any structural
variation in the dipolarophile while maintaining good ee
values was by no means assured, since Hashimoto had
observed a significant erosion in ee (and yield) for one
(oxidopyrylium) substrate as the dipolarophile was even
varied from DMAD (74% ee) to the corresponding diethyl
ester (46% ee); 25% ee was observed with the di(tert-
butyl) ester.13
It is also interesting to note that for the cycloadditions
of (E)- and (Z)-alkenes 25 and 26, under a given set of
reaction conditions, there were small but significant
differences in ee values between the two geometric
isomers, but the effect was opposite for the carboxylate
and phosphate catalysts. That is, Rh2(S-DOSP)4 4 gave
slightly higher ee with (E)-25 compared to (Z)-26 (at
25 °C, 53% ee, versus 42% ee, respectively), whereas
Rh2(R-DDBNP)4 7 gave slightly better asymmetric induc-
tion with (Z)-26 (for example, at 25 °C, 75% ee using (Z)-
26, versus 64% ee using (E)-25.
Padwa originally established that the rate of intramo-
lecular cycloaddition of a carbonyl ylide containing a
hexenyl tether was slower than that for an ylide contain-
ing a tethered pentenyl dipolarophile (cf. that derived
from 1 (with CO2t-Bu ) H)), as the former (but not
the latter) dipole could be trapped intermolecularly if
the reaction was run in the presence of DMAD.21 It
was therefore pleasing to observe that cycloaddition of
diazo diketoester 27, bearing a hexenyl tether, occurred
in similar levels of ee compared to the basic pen-
tenyl-substituted substrate 1 (up to 87% ee using
Rh2(R-DDBNP)4 7 at 0 °C with 27; 88% ee was observed
with 1 under these conditions), despite the former’s
With the aim of generating enantioenriched cycload-
ducts that retained useful functionality for additional
synthetic manipulations and also to further probe the
effect on ee with respect to dipolarophile, we next
examined cycloaddition substrates bearing tethered acet-
ylenic functionality (Scheme 6).22
Given the earlier comments concerning substrate
generality, it was pleasing to find that the results with
intramolecular acetylenic dipolarophiles generally paral-
leled those previously found with tethered alkenes. Also,
hydrogenation of (+)-39 ([R]25D +293.5, c ) 1.0 in CHCl3)
and (+)-40 ([R]27 +266.2, c ) 1.0 in CHCl3) gave (+)-3
D
(95% yield, [R]27 +9.9, c ) 1.0 in CHCl3) and (-)-36
D
([R]27 -15.3, c ) 1.0 in CHCl3), respectively; this
D
establishes that the same sense of asymmetric induction
is observed with both alkene and alkyne dipolarophiles.
For identical reaction conditions similar ee values were
often observed for alkene and alkyne systems in which
the tether length (3 or 4 methylene groups) and substitu-
tion pattern (terminal or internal unsaturation) were the
same. Typically, the alkyne systems gave cycloadduct ee
values a little less (usually within 10%) than the analo-
gous alkene systems. One noteworthy exception is found
in comparing the 4 methylene-tethered internally unsat-
urated systems 28 and 32, where the acetylenic system
(21) Padwa, A.; Hornbuckle, S. F.; Fryxell, G. E.; Zhang, Z. J . J .
Org. Chem. 1992, 57, 5747-5757.
(22) For a recent example, see: Graening, T.; Friedrichsen, W.; Lex,
J .; Schmalz, H.-G. Angew. Chem., Int. Ed. 2002, 41, 1524-1526.
6156 J . Org. Chem., Vol. 68, No. 16, 2003

Intramolecular Carbonyl Ylide Cycloadditions
SCHEME 5. Cycloa d d ition Rea ction s w ith Alk en e Dia zo Su bstr a tes 33 a n d 34
SCHEME 6. Cycloa d d ition Rea ction s w ith Alk yn e
SCHEME 7. Cycloa d d ition Rea ction s w ith Mixed
Dia zo Su bstr a tes 29-32
Dia zom a lon a te 47
the effect of the oxygen for methylene replacement on
cycloaddition efficiency and asymmetric induction. Padwa
had earlier reported that the corresponding ethyl ester
(47, t-Bu ) Et) was a viable substrate for the cycloaddi-
tion chemistry, catalyzed by rhodium acetate.23 However,
the tert-butyl ester was studied in the present work, since
our previous results indicated that larger ester substit-
uents led to the best ee values, at least with phosphate
catalysts such as 6 and 7.9
In the event, diazomalonate 47 was found to be
somewhat less reactive in the cycloaddition chemistry
than the â-ketoesters studied earlier, with 47 requiring
temperatures above ambient to achieve complete cycload-
dition in a reasonable time scale (∼3 h). Attenuated
reactivity of diazomalonates compared with R-diazo-â-
ketoesters is often observed in diazo chemistry.3b While
reaction under rhodium acetate catalysis provided tri-
cyclic lactone cycloadduct 48a in satisfactory yields
(benzene, reflux, 15 min, 74%; CH2Cl2, reflux, 15 min,
46%), reaction under asymmetric catalysis with
Rh2(S-DOSP)4 and with the phosphate catalysts generally
provided cycloadduct 48a in only modest yields (Scheme
7). In most cases the lactone 48b, arising from simple
intramolecular C-H insertion3b was also isolated as a
32 generated cycloadduct 42 with ee values some 20%
higher than those found for the corresponding alkene
derived cycloadduct 38. In particular, using substrate 32,
Rh2(S-DOSP)4 4 reproducibly delivered the highest level
of asymmetric induction (69% ee) seen with this catalyst
in any intramolecular carbonyl ylide cycloaddition thus
far and in good yield (73%). This latter result may also
be compared both with the corresponding 3 methylene-
tethered internally substituted alkyne 34, for which with
Rh2(S-DOSP)4 4 generated the cycloadduct 40 in only
39% ee (74% yield), and with the 4 methylene-tethered
terminal alkyne 31 where the cycloadduct 41 was only
generated in 24% ee. These latter results highlight once
again that a complex interplay of factors determine
asymmetric induction in these cycloadditions.
With the aim of exploring further the scope of enan-
tioselective tandem carbonyl ylide formation-intramo-
lecular cycloaddition chemistry, the mixed diazomalonate
47 was selected for examination (Scheme 7); comparison
of this substrate with 1 would allow an assessment of
(23) Padwa, A.; Dean, D. C.; Fairfax, D. J .; Xu, S. L. J . Org. Chem.
1993, 58, 4646-4655.
J . Org. Chem, Vol. 68, No. 16, 2003 6157

Hodgson et al.
solvents were distilled before use. Ethers were distilled from
sodium/benzophenone ketyl, and (halogenated) hydrocarbons
and MeCN from CaH2. “Petrol” refers to the fraction of light
petroleum ether boiling between 30 and 40 °C. Rh2(OAc)4 and
Rh2(S-DOSP)4 were purchased commercially. Synthesized Rh2-
(S-DOSP)4 was prepared according to Davies’ procedure, using
pure 4-dodecylbenzenesulfonyl chloride.6 Hashimoto’s proce-
dure was followed for the preparation of Rh2(S-BPTV)4.7,13
Rh2(R-BNP)4 was prepared according to Pirrung’s procedure.8
Rh2(R-DDBNP)4 was prepared as we previously reported.9b
Reactions were monitored by TLC using commercially avail-
able aluminum plates precoated with silica (0.25 mm, Merck
60 F254), which were developed using standard visualizing
techniques: UV fluorescence (254 nm) and/or potassium
permanganate or vanillin solution, heating. Flash chromatog-
raphy was performed on Kieselgel 60 (40-63 µm).
Rep r esen ta tive Syn th esis of a 2-Dia zo-3,6-d ik etoester
(25). (E)-6-Iod o-2-h exen e.24 A solution of (E)-4-hexen-1-ol
(3.79 g, 37.8 mmol, E:Z, 98:2 by GLC) and Et3N (7.88 mL, 56.7
mmol) in CH2Cl2 (100 mL) was cooled to 0 °C under Ar. MsCl
(4.76 g, 41.6 mmol) was added dropwise, and the solution was
stirred for 15 min at 0 °C and for 15 min at 25 °C and then
hydrolyzed with ice water. The resultant solution was diluted
with CH2Cl2 (100 mL), washed successively with HCl 10% w/w
(75 mL), saturated aqueous NaHCO3, (75 mL) and H2O (2 ×
75 mL), dried (MgSO4), filtered, and concentrated in vacuo to
give the crude mesylate as an oil. The mesylate was added to
a solution of NaI (14.2 g, 94.5 mmol) in acetone (80 mL) and
heated to reflux for 14 h. After cooling to room temperature,
the reaction mixture was diluted with petrol, washed with
water, dried (MgSO4), filtered, and concentrated in vacuo to
give pure (E)-6-iodo-2-hexene (6.40 g, 81%): 1H NMR (200
MHz, CDCl3) δ 5.61-5.27 (2H, m), 3.20 (2H, t, J ) 5.0 Hz),
2.11 (2H, q, J ) 7.0 Hz), 1.92-1.79 (2H, m), 1.76 (3H, d, J )
5.5 Hz).
byproduct. As related byproducts were not observed in
our earlier studies, this suggests that carbonyl ylide
formation (and/or cycloaddition) is slower from the di-
azomalonate 47 compared with the â-ketoester sub-
strates. Moreover, the ee values for tricyclic lactone cyclo-
adduct 48 were significantly lower than those obtained
with the original cycloaddition substrate 1, with only the
Rh2(R-DDBNP)4 catalyst providing any encouraging signs
of asymmetric induction, albeit in very low chemical
yield.
Clearly, the oxygen for methylene replacement resulted
in severe erosion of both cycloaddition efficiency and
asymmetric induction. The lower cycloaddition efficiency
may be due to the propensity for esters to exist in the
s-trans conformation, together with the inductive effect
of the newly introduced ester group reducing the pro-
pensity of the ketone to participate in carbonyl ylide
formation. Although both of these factors are also opera-
tive in Padwa’s successful corresponding ethyl ester
system,21 in the current case a competing C-H insertion
(with the tert-butyl ester) serves to reduce yields. If one
of the major factors determining asymmetric induction
in these cycloadditions is insertion selectivity of the
tethered ketone into the re or si face of the putative
intermediate chiral metallocarbene,4a then the origin of
the reduced asymmetric induction when using 47 could
be that this insertion is (more) reversible in the case of
47, since the derived metallocarbene is less electrophilic
and the ketonic oxygen lone pair less nucleophilic than
that from 1. Thus, reversibility in the formation of the
catalyst-complexed ylide (cf. 2) could compromise asym-
metric induction.
It is important to note that the results described with
Rh2(S-DOSP)4 4 above related to material prepared
according to the procedure of Davies6 but using isomeri-
cally pure dodecyl substituents. The commercially avail-
able Rh2(S-DOSP)4 catalyst (which contains a mixture
of undecyl to tridecyl substituents) gave essentially
identical yields for all the alkyne substrates 29-32, and
ee values were also unaffected for the terminal alkynes
29 and 31 (all reactions with Rh2(S-DOSP)4 4 have been
carried out at 25 °C). Intriguingly however, the methyl-
substituted alkynes 30 and 32 reproduceably delivered
lower ee values when using the commercial catalyst: 38-
39% ee (2 runs) for 40 compared with 50% ee, and 57-
59% ee (2 runs) for 42 compared with 68-71% ee (3 runs).
In summary, enantioselective intramolecular 1,3-
dipolar cycloadditions of unsaturated 2-diazo-3,6-dike-
toester-derived carbonyl ylides show promising scope in
terms of asymmetric induction as the tethered dipolaro-
phile component is varied; a structurally related di-
azomalonate 47 is less successful. The stereospecificity
of the process, the creation of up to four stereocenters, a
potentially versatile cross-metathesis strategy for sub-
strate preparation, and enantioselective intramolecular
cycloadditions with alkynes (in up to 86% ee) have all
been demonstrated for the first time.
(E)-4-Oxod ec-8-en oic Acid (8). A stirred solution of 2,3-
dihydrofuran (2.99 g, 42.7 mmol) in THF (25 mL) was cooled
to -78 °C before dropwise addition of t-BuLi (21.5 mL of a 1.7
M solution in pentane, 36.1 mmol) via cannula. The solution
was warmed to -5 °C over 1 h before being recooled to -78
°C. (E)-6-Iodo-2-hexene (6.32 g, 30.1 mmol) was added drop-
wise via syringe, and the resulting solution was warmed to
25 °C and then stirred for 16 h. The reaction mixture was
recooled to 0 °C and quenched by careful addition of saturated
aq NH4Cl solution (25 mL). The aqueous phase was extracted
with Et2O (3 × 25 mL), and the combined organic layers were
dried (MgSO4) before concentration under reduced pressure.
The resultant crude alkylated dihydrofuran was dissolved in
THF (80 mL) and J ones’ reagent17 (32.1 mL of a 2.7 M aq
solution) was added dropwise with vigorous stirring. After 18
h the reaction mixture was diluted with Et2O (75 mL) and H2O
(75 mL) and stirred vigorously for 30 min. The aqueous phase
was separated and extracted with Et2O (4 × 50 mL), and the
combined organic components were washed with H2O (3 × 75
mL) and extracted with 10% aq NaOH solution (4 × 50 mL).
The combined basic portions were cooled to 0 °C and acidified
to pH 1 with concentrated HCl. The cloudy aqueous component
was extracted with CH2Cl2 (4 × 50 mL) and the combined
organic components dried (MgSO4). Concentration under
reduced pressure gave the keto acid 8 as a cream solid (2.27
g, 43% over 2 steps). Mp 45 °C; IR (KBr disk) 3375, 2934, 1699,
1
1380 cm-1; H NMR (200 MHz, CDCl3) δ 5.53-5.27 (2H, m),
2.77-2.60 (4H, m), 2.44 (2H, t, J ) 7.4 Hz), 1.98 (2H, q, J )
6.7 Hz), 1.72-1.59 (5H, m); 13C NMR (50 MHz, CDCl3) δ 209.0,
178.7, 130.5, 126.1, 41.8, 36.7, 31.8, 27.6, 23.3, 17.8; MS (CI+)-
m/z 202 (M + NH4+, 100), 185 (M + H+, 29), 184 (M, 11), 158
(82); HRMS (ES, [M + NH4]+) calcd 202.1443, measured
202.1443.
Exp er im en ta l Section
Gen er a l Met h od s. All reactions requiring anhydrous
conditions were conducted in flame- or oven-dried apparatus
under an atmosphere of argon. Syringes and needles for the
transfer of reagents were oven-dried and allowed to cool in a
desiccator over self-indicating silica gel prior to use. All
ter t-Bu tyl (E)-3,6-dioxododec-10-en oate (15). To a stirred
solution of (E)-4-oxodec-8-enoic acid (2.25 g, 12.2 mmol) in THF
(24) Krief, A.; Kenda, B.; Barbeaux, P.; Guillet, E. Tetrahedron 1994,
50, 7177-7192.
6158 J . Org. Chem., Vol. 68, No. 16, 2003

Intramolecular Carbonyl Ylide Cycloadditions
(60 mL) at 0 °C was added carbonyldiimidazole (2.38 g, 14.7
mmol). After 15 min at 0 °C, the ice bath was removed and
the reaction mixture was allowed to warm to 20 °C for 1 h. In
a separate flask, mono-tert-butyl malonate (4.30 g, 26.9 mmol)
was dissolved in THF (60 mL), cooled to -78 °C, and to this
was added Bu2Mg (13.4 mL of a 1.0 M solution in heptane,
13.4 mmol) via syringe. The mixture was stirred for 15 min
at -78 °C and then for 1 h at 20 °C. The solvent was removed
and the acyl imidazolide was added via cannula to the
magnesium salt, rinsing the flask with an additional portion
of THF (10 mL). After 18 h the reaction was quenched by the
addition of 10% aq citric acid solution (30 mL), the layers were
separated, and the aqueous component was extracted with
Et2O (2 × 60 mL). The combined organic components were
washed with saturated aq NaHCO3 solution (30 mL) and brine
(30 mL), dried (MgSO4), and concentrated under reduced
pressure. The crude product mixture was purified by flash
chromatography (SiO2, petrol/Et2O 8:2; Rf ) 0.17) to afford tert-
butyl (E)-3,6-dioxododec-10-enoate 15 as a clear yellow oil (2.06
g, 60%): IR (neat) 2979, 2934, 1738, 1715, 1368 cm-1; 1H NMR
(200 MHz, CDCl3) δ 5.41-5.33 (2H, m), 3.38 (2H, s), 2.76-
2.68 (4H, m), 2.41 (2H, t, J ) 7.3 Hz), 1.93 (2H, q, J ) 6.4
Hz), 1.68-1.57 (5H, m), 1.44 (9H, s); 13C NMR (50 MHz, CDCl3)
δ 209.0, 202.0, 166.3, 130.3, 125.7, 81.9, 50.6, 41.9, 36.2, 36.0,
31.8, 27.9, 23.4, 17.8; MS (CI+) m/z 300 (M + NH4+, 54), 283
(M + H+, 6), 244 (100); HRMS (ES, [M + H+]) calcd 283.1909,
measured 283.1907.
ter t-Bu tyl (E)-2-Dia zo-3,6-d ioxod od ec-10-en oa te (25).
To a stirred solution of 3,6-diketoester 15 (2.00 g, 7.08 mmol)
and 4-acetamidobenzenesulfonyl azide (1.87 g, 7.79 mmol) in
MeCN (35 mL) at 0 °C was added Et3N (1.09 mL, 9.69 mmol).
After 5 h the reaction was quenched by addition of saturated
aq NH4Cl (20 mL). The mixture was extracted with CH2Cl2
(20 mL), and the organic components were washed with brine
(20 mL), dried (MgSO4), and concentrated in vacuo. The crude
yellow oil was purified by flash chromatography (SiO2, petrol/
Et2O 9:1, Rf ) 0.2) to afford tert-butyl (E)-2-diazo-3,6-diox-
ododec-10-enoate 25 (1.95 g, 89%) as a yellow oil: IR (neat)
2978, 2934, 2132, 1715, 1655, 1369 cm-1; 1H NMR (200 MHz,
CDCl3) δ 5.41-5.37 (2H, m), 3.08 (2H, t, J ) 6.0 Hz), 2.71 (2H,
t, J ) 6.2 Hz), 2.45 (2H, t, J ) 7.4 Hz), 2.01-1.93 (2H, m),
1.67-1.60 (5H, m), 1.51 (9H, s); 13C NMR (50 MHz, CDCl3) δ
209.3, 191.7, 160.5, 130.5, 125.7, 83.1, 42.0, 36.1, 34.2, 31.9,
28.3, 23.5, 17.9; MS (CI+) m/z 326 (M + NH4+, 12), 309 (M +
H+, 27), 300 (48), 270 (88), 244 (100), 200 (85); HRMS (ES, [M
+ H]+) calcd 309.1814, measured 309.1812.
) 15.7 Hz, J ) 7.0 Hz), 5.80 (1H, dt, J trans ) 15.7 Hz, J ) 1.6
Hz), 3.69 (3H, s, (E)-isomer), 3.66 (3H, s, (Z)-isomer), 3.36 (2H,
s), 2.80-2.77 (2H, m), 2.67-2.63 (2H, m), 2.46 (2H, t, J ) 7.3
Hz), 2.7 (2H, app. qd, J ) 7.2 and 1.3 Hz), 1.72 (2H, quint, J
) 7.3 Hz), 1.43 (9H, s); 13C NMR (100 MHz, CDCl3) δ 208.2,
201.9, 166.9, 166.3, 148.3, 121.5, 81.9, 51.4, 50.5, 41.5, 36.3,
36.0, 31.3, 27.9, 21.8.
(E)-11-Dia zo-7,10-d ioxo-d od ec-2-en ed ioic Acid 12-ter t-
Bu tyl Ester 1-Meth yl Ester (33). General procedure as for
25. The residue, obtained from 23 (1.0 g, 3.1 mmol), was
purified by column chromatography (SiO2, petrol/ethyl acetate
8:2, Rf ) 0.16) to give a colorless liquid (0.91 g, 84%, E:Z 97:3
1
by H NMR): 1H NMR (400 MHz, C6D6) δ 7.00 (1H, dt, J trans
) 15.7 Hz, J ) 7.0 Hz), 5.92 (1H, dt, J trans ) 15.7 Hz, J ) 1.5
Hz), 3.54 (3H, s, (E)-isomer), 3.47 (3H, s, (Z)-isomer), 3.18 (2H,
t, J ) 6.1 Hz), 2.39 (2H, t, J ) 6.1 Hz), 2.04 (2H, t, J ) 7.3
Hz), 1.86 (2H, m), 1.54 (2H, quint, J ) 7.3 Hz), 1.36 (9H, s);
13C NMR (100 MHz, C6D6) δ 207.1, 191.2, 166.7, 160.7, 148.8,
122.0, 82.5, 76.2, 51.0, 41.5, 36.2, 34.9, 31.6, 28.2, 22.3; MS
(CI+) m/z 370 (M + NH4+, 6), 344 (100), 314 (18), 288 (66),
270 (12), 244 (27); HRMS (ES, [M + NH4]+) calcd 370.1978,
measured 370.1976.
Typ ica l P r oced u r e for
a Cycloa d d ition Rea ction .
7-Carbo tert-Butoxy-endo-6-methyl-11-oxatricyclo[5.3.1.01,5]-
u n d eca n -8-on e (35). To a stirred, degassed solution of
(E)-tert-butyl 2-diazo-3,6-dioxododec-10-enoate 25 (100 mg,
0.32 mmol) in dry hexane (5 mL) at 25 °C was added
Rh2(R-DDBNP)4 (9.5 mg, 0.003 mmol). After the reaction was
complete (50 min, TLC monitoring), the reaction mixture was
concentrated in vacuo. The residue was purified by flash
chromatography (SiO2, petrol/Et2O 8:2, Rf ) 0.17) to afford
cycloadduct 35 as a colorless oil (71.6 mg, 63%; Scheme 4):
[R]24 +34.9 (c ) 1.0 in CHCl3); IR (neat) 2942, 1740, 1709,
D
1448, 1368 cm-1; 1H NMR (200 MHz, CDCl3) δ 2.66-2.59 (1H,
m), 2.45-2.34 (2H, m), 2.20 (1H, quint, J ) 7.1 Hz), 2.09-
2.03 (2H, m), 1.91-1.78 (3H, m), 1.68-1.65 (1H, m), 1.60-
1.56 (1H, m), 1.49 (9H, s), 1.46-1.38 (1H, m), 1.15 (3H, d, J )
7.0 Hz); 13C NMR (50 MHz, CDCl3) δ 204.1, 166.7, 93.5, 91.7,
82.4, 53.6, 49.4, 37.0, 36.1, 33.2, 32.8, 27.9, 24.7, 14.5; MS (CI+)
m/z 298 (M + NH4+, 12), 281 (M + H+, 4), 242 (100); HRMS
(ES, [M + H]+) calcd 281.1753, measured 281.1750.
Ack n ow led gm en t. We thank the European Union
for Marie Curie Fellowships (HPMF-CT-2000-00559 to
A.H.L. and HPMF-CT-2001-01396 to M.AÄ .E.C.) and the
EPSRC for a research grant (GR/L98022). We also
thank the EPSRC National Mass Spectrometry Service
Centre for mass spectra and Dr. B. Odell for assistance
with NMR analysis.
Syn th esis of 2-Dia zo-3,6-d ik etod iester (33) by Cr oss-
Meta th esis. (E)-7,10-Dioxod od ec-2-en ed ioic Acid 12-ter t-
Bu tyl Ester 1-Meth yl Ester (23). Grubbs’ catalyst (see
Scheme 3)18 (145 mg, 0.17 mmol) was placed in a dry flask,
fitted with a condenser, and dry CH2Cl2 (20 mL) was added.
A solution of tert-butyl 3,6-dioxoundec-10-enoate 4919 (915 mg,
34.1 mmol) in CH2Cl2 (5 mL) and methyl acrylate (615 µL,
68.2 mmol) were added simultaneously via syringe, and the
mixture was heated to reflux for 14 h under argon. After
cooling to 20 °C, the mixture was concentrated in vacuo and
purified by column chromatography (SiO2, petrol/ethyl acetate
8:2, Rf ) 0.15) to give a colorless liquid (1.07 g, 96%, E:Z 97:3
Su p p or tin g In for m a tion Ava ila ble: Characterization
data of cycloaddition substrates and cycloadducts, chiral GC
conditions for all cycloadducts, and copies of 1H and 13C NMR
spectra for all novel cycloaddition substrates and cycloadducts.
This material is available free of charge via the Internet at
http://pubs.acs.org.
1
by H NMR): 1H NMR (400 MHz, CDCl3) δ 6.88 (1H, dt, J trans
J O0343735
J . Org. Chem, Vol. 68, No. 16, 2003 6159