Exploitation of ylide steric bulk to alter cyclopropanation outcome during the reaction of 1,2-dioxines and stabilised phosphorus ylides

Article abstract of DOI:10.1039/b001975p

The exploitation of ylide steric bulk to alter cyclopropanation outcome during the reaction of 1,2-dioxines and stabilised phosphorus ylides was discussed. It was found that sterically bulky ylides favoured the formation of a different distereomeric cyclopropyl series at ambient temperatures. The analysis showed that the use of sterically bulky ester ylides under concentrated conditions favoured the formation of the 'normal' trans isomer.

Full text of DOI:10.1039/b001975p

View Article Online / Journal Homepage / Table of Contents for this issue
Exploitation of ylide steric bulk to alter cyclopropanation outcome
during the reaction of 1,2-dioxines and stabilised phosphorus ylides
Thomas D. Avery, Ben W. Greatrex, Dennis K. Taylor* and Edward R. T. Tiekink
1
Department of Chemistry, The University of Adelaide, Australia, 5005.
E-mail: dennis.taylor@adelaide.edu.au
Received (in Cambridge, UK) 14th December 1999, Accepted 13th March 2000
Published on the Web 11th April 2000
A new approach for the synthesis of diastereomerically
pure cyclopropanes from 1,2-dioxines and sterically bulky
stabilised phosphorus ylides is presented.
bulkier di-substituted ylide 2b allows for the incorporation
of another stereogenic centre within the side chain of the
cyclopropane 5. However, during these investigations we also
isolated (7%) another cyclopropane 6 which did not appear
to originate from our proposed mechanism and we speculated
that the steric bulk of the ylide may be the reason behind the
formation of 6. We now report on the exploitation of the steric
bulk of the ester moiety of the stabilised ylides to completely
alter the diastereomeric course of these cyclopropanation
reactions.
We began analysing the effect of steric bulk on cyclopropan-
ation outcome with the diphenylmethyl (DPM) grouping. Thus,
reaction of 1,2-dioxine 1 (Y = Ph) with ylide 2a (R1 = Ph2CH)
under the conditions specified within Table 1 (entry 1), resulted
only in the formation of diastereomerically pure cyclo-
propane 4 (Scheme 2). Hence the DPM grouping fails to alter
Organics containing the cyclopropane motif are of wide inter-
est since many display novel biological activity, whilst others
have been utilised as key intermediates in the synthesis of
natural and non-natural products. Through a series of mech-
anistic studies we have elucidated a clear understanding of the
complex relationship between 1,2-dioxines, and their isomeric
cis and trans γ-hydroxy enones, cis and trans hemiacetals and
β-ketoepoxides, and of how these precursors can be utilised
to construct diastereomerically pure or enantiomerically pure
diversely functionalised cyclopropanes.1 We now ‘expect’
that the reaction of 1,2-dioxines 1 with stabilised phosphorus
ylides containing an ester moiety 2a, will afford diastereo-
merically pure cyclopropanes 4 in excellent yields through
collapse of the intermediate 1,2λ5-oxaphospholanes 3 (Scheme
1). Furthermore, we have demonstrated that the use of the
Scheme 2
the cyclopropanation outcome. A far more dramatic effect on
cyclopropanation outcome was found during investigations
utilising the tert-butyl ester stabilised ylide (entries 2–4,
Table 1). Thus, reaction of 1,2-dioxine 1 (Y = Ph) with ylide 2a
(R1 = t-Bu) at 25 ЊC resulted in the formation of three cyclo-
propyl isomers (entry 3). The first of these was simply the
‘expected’ trans-cyclopropane 4 resulting from collapse of the
1,2λ5-oxaphospholane 3, while the remaining two (7 and 8)
represented a uniquely different cyclopropanation outcome.
The structure and relative stereochemistry of these two new
cyclopropanes were unambiguously elucidated from a combin-
ation of crystallography and 1H, 13C, gCOSY, gHSQC, and
Scheme 1
DOI: 10.1039/b001975p
J. Chem. Soc., Perkin Trans. 1, 2000, 1319–1321
This journal is © The Royal Society of Chemistry 2000
1319

View Article Online
Table 1 Reaction of 1,2-dioxines 1 with various bulky ylides 2a
Entry
Y
R1
Solvent
Conc./M
Temp./ЊC
Cyclopropane (yield, %)
9 (yield, %)
1
2b
3
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
H
H
H
H
H
Ph2CH
t-Bu
t-Bu
CH2Cl2
CDCl3
CH2Cl2
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
Toluene
CH2Cl2
CH2Cl2
CDCl3
CHCl3
CHCl3
0.38
0.08
0.40
0.08
0.06
0.06
0.15
0.06
0.06
0.12
0.54
0.12
0.1
25
Ϫ15
25
60
Ϫ15
25
25
60
100
Ϫ15
25
4 (97) 7 (0) 8 (0)
4 (0) 7 (91) 8 (9)
4 (19) 7 (70) 8 (8)
4 (33) 7 (52) 8 (6)
4 (12) 7 (67) 8 (20)
4 (30) 7 (61) 8 (9)
4 (58) 7 (35) 8 (5)
4 (42) 7 (45) 8 (6)
4 (75) 7 (5) 8 (0)
4 (48) 7 (49) 8 (2)
4 (77) 7 (22) 8 (1)
4 (81) 7 (14) 8 (5)
4 (19) 7 (64) 8 (17)
4 (79) 7 (18) 8 (3)
3
0
3
9
0
0
2
7
20
0
0
0
0
0
4
t-Bu
5b
6
1-Ad
1-Ad
1-Ad
1-Ad
1-Ad
t-Bu
t-Bu
t-Bu
1-Ad
1-Ad
7
8
9
10c
11
12
13
14
60
25
60
0.06
a 1-Ad refers to 1-adamantyl. Yields quoted refer to those determined from analysis of the crude mixtures by 1H NMR (600 MHz). Isolated yields
were within a few percent of the values given. General procedure for reactions performed at ambient and higher temperatures: the 1,2-dioxine 1 and
ylide (1 equiv.) were combined in the appropriate solvent at the concentration specified and the mixture stirred under nitrogen for 3 days. The solvent
was removed in vacuo and the residue subjected to flash chromatography (ethyl acetate–hexane, 1:4) to afford the observed products. b A catalytic
amount (5 mol%) of Jacobson’s catalyst was also added at Ϫ15 ЊC. The mixture was kept at this temperature for 2 days after which time it was allowed
to warm to room temperature and was analysed. c Triethylamine (1 equiv.) added at Ϫ15 ЊC. The mixture was kept at this temperature for 3 days after
which time it was allowed to warm to room temperature and was analysed.
‘normal’ trans-cyclopropane isomer 4 was detected whilst the
yield of 7 had increased to 91%.
In order to further evaluate the effect of ylide steric bulk on
cyclopropanation outcome we investigated the use of the 1-
adamantyl ester which has a slightly smaller cone angle than
the tert-butyl moiety (entries 5–9). Utilisation of this ylide 2a
(R1 = 1-Ad) resulted in a decrease in yield of (7 and 8) with a
concomitant increase in the yield of the ‘normal’ trans isomer 4
(compare entries 6 and 3). Performing the reaction at elevated
temperature once again favoured the formation of the ‘normal’
trans-cyclopropane 4 (entries 8 and 9) while lower temperatures
were found to favour the formation of 7 and 8 (entry 5). An
important observation was that simply increasing the overall
reaction concentration favours the formation of the ‘normal’
trans-cyclopropane over 7 and 8 (compare entries 6 and 7). In
addition, we have previously noticed that the sterics and elec-
tronics of the substituent Y within the 1,2-dioxines 1 have little
effect on cyclopropanation outcome and as such we anticipated
that this new route to diastereomerically pure cyclopropanes of
type 7 would be applicable to a wide range of substituents.1 This
assumption was verified with the use of 1,2-dioxine 1 (Y = H)
which afforded the di-substituted cyclopropanes 4 and 7 in
Fig. 1 Molecular structure of 7 (Y = Ph, R1 = t-Bu) showing the crys-
good yields at elevated and sub-ambient temperatures respec-
tallographic numbering scheme employed: O(11)–C(11)–C(1)–C(2)
tively (entries 10–14). Once again, the use of sterically bulky
ester ylides under concentrated conditions favours the form-
ation of the ‘normal’ trans isomer 4 while dilute conditions
favour the formation of 7 and 8.
1.1(4)Њ, O(22)–C(22)–C(21)–C(2) 1.1(3)Њ and C(32)–C(31)–C(3)–C(1)
Ϫ69.2(3)Њ.
gHMBC NMR techniques.† Fig. 1 represents the molecular
structure of 7 (Y = Ph, R1 = t-Bu).‡,2
Finally, monitoring the reaction between 1,2-dioxine (1,
Y = Ph) with the tert-butyl ester ylide (entry 3) by 31P NMR at
20 ЊC revealed the initial formation of two phosphorus contain-
ing intermediates corresponding to signals at 23.99 and 24.29
ppm.§ These two intermediates which were in a relative ratio of
ca. 9:1, respectively, then decayed over 12 hours to afford the
observed cyclopropanes 7, 8 and TPPO. These chemical shifts
are inconsistent with a neutral pentavalent oxaphospholane
ring of type 3 and are more appropriately assigned as being due
to charged species.5 The exact nature of these intermediates is
yet to be fully established.
We conclude that the steric bulk of the ester moiety of
stabilised phosphorus ylides has a dramatic effect on cyclo-
propanation outcome observed during the reaction between
1,2-dioxines and stabilised ester ylides. Whereas previously,
utilisation of sterically ‘non-bulky’ ylides (e.g. Me, Et, Bn)
afforded the ‘normal’ trans diastereomerically pure cyclopro-
panes 4 in excellent yields at ambient temperature, we now
find that sterically bulky ylides (e.g. t-Bu, 1-Ad) favour the
formation of a different diastereomeric cyclopropyl series at
these temperatures. The mechanism is currently being fully
investigated and will be reported in due course.
To rule out the possibility that the minor cyclopropane 8 was
formed first under the reaction conditions, and then isomerised
to 7, we treated isolated cyclopropane 8 with excess ylide
under identical reaction conditions (with triphenylphosphine
oxide present) and found no isomerisation. We next analysed
the effect of temperature on reaction outcome. Performing the
identical reaction at 60 ЊC resulted in an increase in yield of the
‘normal’ trans-cyclopropane 4 at the expense of cyclopropanes
(7 and 8) (entry 4). A slight increase in the yield of 1,4-diketone
6 was also noticed and is expected at elevated temperatures
through a competing Kornblum–De La Mare rearrangement.3
We have previously reported the use of Co(salen)2 (salen =
2,2Ј-[ethane-1,2-diylbis(nitrilomethylidyne)]dibenzenethiolato)
or Jacobson’s catalyst to dramatically accelerate these
cyclopropanation reactions.1 These catalysts rapidly induce
rearrangement of the 1,2-dioxines to their corresponding cis γ-
hydroxy enones which are the key intermediates required for
cyclopropanation.4 Hence we next performed the identical reac-
tion at Ϫ15 ЊC in the presence of a catalytic amount (5 mol%)
of Jacobson’s catalyst (entry 2). At this temperature none of the
1320
J. Chem. Soc., Perkin Trans. 1, 2000, 1319–1321

View Article Online
in CDCl3. No such signals are observable for the reaction of the same
1,2-dioxine with non-bulky ylides, ref. 1.
Acknowledgements
Financial support from the Australian Research Council
(ARC) is greatly acknowledged.
1 T. D. Avery, T. D. Haselgrove, T. J. Rathbone, D. K. Taylor and
E. R. T. Tiekink, Chem. Commun., 1998, 333; T. D. Avery, D. K.
Taylor and E. R. T. Tiekink, submitted for publication in J. Org.
Chem., 2000; F. N. Palmer and D. K. Taylor, J. Chem. Soc., Perkin
Trans. 1, 2000, preceding paper DOI: 10.1039/b001973i.
2 The crystal structure of cyclopropane 4 (Y = Ph, R1 = t-Bu) has been
submitted for publication elsewhere; T. D. Avery, G. Fallon, D. K.
Taylor and E. R. T. Tiekink, Z. Kristallogr., 2000, in the press. Cyclo-
propane 8 (Y = Ph, R1 = t-Bu) has been submitted for publication
elsewhere; T. D. Avery, G. Fallon, D. K. Taylor and E. R. T. Tiekink,
submitted for publication in Z. Kristallogr., 2000.
3 N. Kornblum and H. E. De La Mare, J. Am. Chem. Soc., 1951, 73,
881; M. E. Sengül, Z. Ceylan and M. Balci, Tetrahedron, 1997, 53,
10401 and references cited therein. Also see ref. 1.
4 K. E. O’Shea and C. S. Foote, J. Org. Chem., 1989, 54, 3475. Also see
ref. 1.
Notes and references
† All new compounds have been fully characterised by elemental anal-
ysis, spectroscopy and mass spectrometry.
‡ Crystal data: C22H24O3, 7 (Y = Ph, R1 = t-Bu): monoclinic P21/n,
a = 5.5486(1), b = 17.8452(8), c = 19.0424(8) Å, β = 91.930(3)Њ,
U = 1884.4(1) Å3, Z = 4, Dc = 1.186 g cmϪ3 and µ = 0.77 cmϪ1. Data
were collected at 173 K on a Nonius Kappa CCD employing Mo-Kα
radiation in the ranges 1.0Њ < θ < 30.1Њ and 1.4Њ < θ < 27.5Њ, respec-
tively. The structure was solved by SAPI91 and refined with the
TEXSAN Structure Analysis Package (Molecular Structure Corpor-
ation, 1998) of crystallographic programs. A total of 2484 reflections
with I ≥ 3.0σ(I) were used in the refinement which converged with
R = 0.073 and Rw = 0.047. CCDC reference number 207/411. See http://
www.rsc.org/suppdata/p1/b0/b001975p/ for crystallographic files in .cif
format.
5 See for example: J. C. Hanekamp, R. B. Rookhuizen, H. J. T. Bos and
L. Brandsma, Tetrahedron, 1992, 48, 5151; W. Adam, H. M. Harrer
and A. Treiber, J. Am. Chem. Soc., 1994, 116, 7581 and references
cited therein.
§
31P NMR chemical shifts referenced to 85% aqueous H3PO4 (external)
J. Chem. Soc., Perkin Trans. 1, 2000, 1319–1321
1321