Substituent effects in the double diastereotopic differentiation of α-diazophosphonates via intramolecular cyclopropanation

Article abstract of DOI:10.1016/S0957-4166(03)00081-8

The investigation of substituent effects in the double diastereotopic differentiation of substituted α-diazophosphonates using intramolecular cyclopropanation catalyzed by Rh₂(OAc)₄ is reported. Carbene facial selectivity in these tr

Full text of DOI:10.1016/S0957-4166(03)00081-8

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 873–880
Substituent effects in the double diastereotopic differentiation of
a-diazophosphonates via intramolecular cyclopropanation
Joel D. Moore and Paul R. Hanson*
Department of Chemistry, University of Kansas, 1251 Wescoe Hall Drive, Lawrence, KS 66045-7582, USA
Received 10 December 2002; accepted 10 January 2003
Abstract—The investigation of substituent effects in the double diastereotopic differentiation of substituted a-diazophosphonates
using intramolecular cyclopropanation catalyzed by Rh2(OAc)4 is reported. Carbene facial selectivity in these transformations is
dictated by substrate control in either of two ways: (i) exploitation of (R)-pantolactone as an auxiliary incorporated into the
carboester functionality while probing olefinic substituent effects, or (ii) utilization of chiral, non-racemic allylic alcohols
incorporated into the phosphonate moiety. © 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
this system is dictated by chelation of the carbonyl
moiety in the (R)-pantolactone to the metallo-car-
benoid center effectively ‘locking’ the molecule in a
rigid conformation and therefore blocking access to the
si face of the reacting carbene (Davies model).6 This
method affords non-racemic, P-stereogenic [3.1.0]-
bicyclic phosphonates 2 with good levels of olefin-
(10.4:1) and diastereofacial-selectivity (6.9:1). The
diastereomeric ratios in this study were conveniently
The rich biological and synthetic profile of phosphorus-
containing compounds1 has prompted the development
of new strategies for their synthesis. It is known that
stereogenicity at or around a phosphorus center has
profound effect on the biological and chemical proper-
ties.2 Current synthetic efforts are therefore aimed at
the development of asymmetric methods to access
structurally diverse P-chiral phosphorus compounds
that can have an assortment of medicinal, agricultural,
and synthetic applications.3 As part of our continuing
program aimed at the synthesis of diverse P-chiral
P-heterocycles,4 we now report a study investigating
substituent effects in the double diastereotopic differen-
tiation of appropriately substituted a-diazophosphono-
acetates employing intramolecular Rh2(OAc)4-catalyzed
cyclopropanation.
1
determined using H decoupled 31P NMR analysis.
Intramolecular cyclopropanation (ICP) mediated by
Rh2(OAc)4 continues to provide a powerful tool for the
construction of constrained systems from their diazo
precursors, with high levels of diastereoselectivity and
enantioselectivity having been achieved in numerous
instances.5 We recently reported a double diastereotopic
differentiation strategy on an a-diazophosphonoacetate
template
1
that utilized Rh2(OAc)4-catalyzed
intramolecular cyclopropanation (ICP) employing the
(R)-pantolactone auxiliary (Scheme 1).4b Selectivity in
* Corresponding author. E-mail: phanson@ku.edu
Scheme 1. Rationale for double diastereoselective ICP.
0957-4166/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0957-4166(03)00081-8

874
J. D. Moore, P. R. Hanson / Tetrahedron: Asymmetry 14 (2003) 873–880
1
In order to associate each signal in the H decoupled
31P NMR spectra to its corresponding diastereomer,
and to verify the group and facial selectivity, a correla-
tion experiment was initially undertaken. Formic acid-
mediated hydrolysis of racemic cyclopropanated,
tert-butyl ester diastereomers ( )-3 (ꢀ6:1, cis:trans),7
followed by DCC coupling with (R)-pantolactone, pro-
duced the corresponding four diastereomers 2-cis-PR/2-
cis-PS and 2-trans-PR/2-trans-PS, preserving the ꢀ6:1
cis:trans ratio (Scheme 2). Subsequent 1H decoupled 31
P
NMR analysis of this sample and comparison to the
experimental 31P spectra allowed the assignment of each
diastereomeric pair (cis versus trans) as denoted in Fig.
1. Furthermore, both cis-diastereomers were separated
using silica gel chromatography and X-ray crystallo-
graphic analysis of each, 2-cis-PR and 2-cis-PS, pro-
vided full unambiguous assignment (Fig. 2). The
production of 2-cis-PR as the major diastereomer is
consistent with an s-trans orientation of the RhꢀC and
PꢀO p-systems (opposing-dipole) incorporating a B-
type orientation of the reacting olefin and occurring
from the re-face of the rhodium carbene (Scheme 1).
Thus far, unambiguous assignment of each trans-
diastereomer (2-trans-PR versus 2-trans-PS) via X-ray
crystallographic analysis has not been possible. Tenta-
Figure 2. ORTEP representations of major and minor cis-
diastereomers.
tive assignment of the furthest downfield resonance as
2-trans-PS (Fig. 1) was made on the basis of analogy to
the cis-diastereomeric series with the major trans-
diastereomer also occurring from re-face attack on the
rhodium carbenoid occurring from a B-type transition
state.
2. Results and discussion
With these results in hand, we reasoned that investiga-
tion of the cyclopropanation of the corresponding bis-
methallyl and bis-crotyl phosphonate esters, 7a and 7b
(Scheme 3), respectively, would provide additional evi-
dence for our previously proposed models vida infra.
Thus, utilizing a similar protocol for the preparation of
1, both 8a and 8b were synthesized as outlined in
Scheme 3. Treatment of PCl3 with Et3N and either
methallyl alcohol 4a or trans-crotyl alcohol 4b in
diethyl ether generated the corresponding phosphites,
trimethallyl phosphite 5a and tricrotyl phosphite 5b
which were used directly without further purification.
Subsequent Arbuzov reaction with (R)-pantolactone-
iodoacetate produced the phosphonoacetates 7a and 7b
in good to moderate yields over two steps. Final diazo
transfer afforded the cyclopropanation precursors 8a
and 8b in good yields.
Scheme 2. Synthesis of racemic cyclopropane diastereomers 2
for 31P NMR correlation experiment.
Scheme 3.
Figure 1. Experimental and correlation 1H decoupled 31P
spectra for the four cyclopropane diastereomers: 2-cis-PR/2-
cis-PS and 2-trans-PR/2-trans-PS.
Cyclopropanation of the a-diazophosphonoacetates 8a
and 8b can proceed through one of eight possible
transition states involving either of two A-type or B-

J. D. Moore, P. R. Hanson / Tetrahedron: Asymmetry 14 (2003) 873–880
875
type transition states occurring from both the re- or
si-face of the rhodium carbenoid. The ability of the
(R)-pantolactone auxiliary to effectively block the si-
face dictates re-face attack as the major pathway, lead-
ing to the four diastereomeric transition states as
depicted in Scheme 4.
propanation of the crotyl systems should display a
significant amount of selectivity erosion. Indeed this
was observed experimentally as shown in Fig. 3, thus
giving additional proof to the previously described
model.
Figure 3. 1H decoupled 31P spectra for the Rh2(OAc)4-cata-
lyzed ICP of a-diazophosphonoacetates 8a. (*) 31P resonances
corresponding to trace cyclopropane diastereomers pre-
sumably derived from the Z-crotyl impurity in the starting
material 7a.
We have tentatively assigned the diastereomers as
shown in Fig. 3 by correlation to the peak order of the
31P NMR resonances of the allyl systems. Interestingly,
the formation of the trans-PR diastereomer as the
major diastereomer in the ICP of the crotyl-based
a-diazophosphonoacetate 8a indicates the preference
for an A-type transition state occurring from the re-
face of the rhodium carbenoid with an opposing dipole
orientation of the RhꢀC and PꢀO p-systems (s-trans).
Mechanistically, this implies that this opposing dipole
orientation plays a significant role in dictating the
stereochemical course of these ICP reactions. Further-
more, additional steric factors may be operative,
whereby there are unfavorable interactions between the
terminal methyl group of the crotyl-based rhodium
carbenoid species generated from 8a (R1=H, R2=Me)
and the rhodium wall in B-type transition states
(Scheme 4).
Scheme 4. Transition states leading to 9a and 9b occurring
via re-face attack on the rhodium carbenoid.
A number of steric and electronic factors operate to
govern the diastereofacial selectivity obtained from
these four pathways, including: (i) the facial orientation
between the reacting olefin and the rhodium carbene
(A-type or B-type), which is dictated by charge stabi-
lization,8 (ii) the orientation of the RhꢀC and PꢀO
p-systems (s-trans or s-cis), in which ‘opposing’ dipole
interactions between the PꢀO and the Rh-carbene moi-
eties would favor the s-trans orientation, and (iii) steric
interactions between the substituents R1 and R2 with
the rhodium wall and/or the chiral auxiliary in the
carboester functionality. In this complex analysis, a
combination of effects is undoubtedly at play.
We next focused on the cyclopropanation of the
methallyl-based a-diazophosphonoacetate 8b. In accor-
dance to the proposed transition states (Scheme 4), this
system should generate a heavily favored B-type transi-
tion state in which positive charge build-up occurs at a
tertiary carbon. Experimentally it was found that the
selectivity in the cis-series slightly decreased (8.7:1),
when compared to the original allyl case (10.4:1.0)
while the trans selectivity increased slightly to 6.1:1.0
from 3.6:1.0 (Fig. 4). It is worth noting that the rate of
For the crotyl-phosphonate system 8a, it can be seen
that during the cyclopropanation, both A-type and
B-type transition states lead to a build-up of positive
charge on a secondary carbon. This contrasts with our
previously explored allyl-based system 1, which in the-
ory progresses through B-type transition states since
they alone generate positive charge build-up on a sec-
ondary carbon.4b We therefore presumed that cyclo-

876
J. D. Moore, P. R. Hanson / Tetrahedron: Asymmetry 14 (2003) 873–880
via si-face attack on the rhodium carbenoid undergoing
B-type transitions.4b The A-type transition states are
further disfavored due to steric interactions between the
internal methyl group of the rhodium carbenoid species
generated from 8b (R1=Me, R2=H) and the rhodium
wall (Scheme 4).
Correlation experiments for 9b again provided further
proof for the 31P assignments (Fig. 5). An analogous
synthesis of the corresponding racemic mixture of the
four diastereomers 9b-cis-PR/9b-cis-PS and 9b-trans-
PR/9b-trans-PS, began with an Arbuzov reaction
between trimethallylphosphite 5b and tert-butyliodoac-
etate (Fig. 5). Subsequent diazo transfer with KOtBu
and TsN3 followed by treatment with Rh2(OAc)4 pro-
duced the diastereomeric pair of cyclopropanes as a
ꢀ5:1 mixture of cis- and trans-diastereomers. Formic
acid-mediated hydrolysis followed by DCC coupling
with the (R)-pantolactone moiety, produced the corre-
sponding four diastereomers 9b-cis-PR/9b-cis-PS and
9b-trans-PR/9b-trans-PS, preserving the ꢀ5:1 cis:trans
ratio. The corresponding chemical shifts seen in Fig. 5
parallel those attained in the correlation experiments
for the four cyclopropane diastereomers 2 derived from
the a-diazophosphonoacetate 1 presented in Fig. 1.
Figure 4. 1H decoupled 31P spectra for the Rh2(OAc)4-cata-
lyzed ICP of a-diazophosphonoacetates 8b.
The final example we investigated involved the desym-
metrization of a novel pseudo-C2-symmetric template
12 (Scheme 5) possessing a nonstereogenic, prochiral
phosphorus atom bearing two diastereotopic olefins,
with each olefin possessing two diastereotopic faces. We
had previously shown that ring-closing metathesis
(RCM) was an effective tool in the diastereotopic dif-
ferentiation
of
pseudo-C2-symmetric
phospho-
namides.4a We felt that a similar desymmetrization
strategy employing the pseudo-C2-symmetric phospho-
noacetate 12 would provide access to the novel P-chiral
[3.1.0]-bicyclic system 13 (Fig. 6). In our previously
reported desymmetrization utilizing RCM,4a olefin sub-
stitution was utilized to direct the initial metathesis
event, a prerequisite for desymmetrization strategies
involving pseudo-C2-symmetric templates.4a,9 With the
phosphonoacetate 12, this is not an issue, since diazo-
tization guarantees regiospecific carbene formation a-
to the phosphonyl group. We therefore believed that
the equatorial preference of the substituent contained in
the allylic appendage would dictate si-facial attack on
the rhodium carbene from a B-type transition state
employing the s-trans orientation of the RhꢀC and PꢀO
p-systems (opposing dipole), producing the cis-PS
diastereomer, 13-cis-PS, as the major diastereomer.
Figure 5. Experimental and correlation 1H decoupled 31P
spectra for the four cyclopropane diastereomers: 9b-cis-PR/
9b-cis-PS and 9b-trans-PR/9b-trans-PS.
reaction of 8b was considerably faster in comparison to
its allylic counterpart 1 (1–2 h versus 3–4 h reaction
time). This is in agreement with a more favorable
positive charge build-up on a tertiary carbon in the
methallyl case as compared to the build-up of positive
charge on a secondary carbon in the allylic case. The
fact that we do not see a large increase in selectivity in
the ICP of 8b would support the notion that the minor
diastereomers, cis-PS and the trans-PR, do not arise
from A-type transition states, but instead are generated
Scheme 5. Preparation of pseudo-C2-symmetric phospho-
noacetate 12.

J. D. Moore, P. R. Hanson / Tetrahedron: Asymmetry 14 (2003) 873–880
877
Figure 6. 1H decoupled 31P spectra for the Rh2(OAc)4-catalyzed
ICP of the corresponding a-diazophosphonoacetates of 12.
The synthesis of the requisite phosphonoacetate 12 began
with TMSCl-mediated deprotection of dimethylethyl
phosphonoacetate (Scheme 5). Acid chloride formation,
followed by condensation of the readily prepared chiral,
non-racemic allylic alcohol 1110 affords 12 in modest yield
(unoptimized, 20% over three steps).
Scheme 6. Transition states leading to phosphonate 13 occur-
ring via B-type transition states.
Diazotization and subsequent cyclopropanation, cata-
lyzed by Rh2(OAc)4, generated the four diastereomeric
[3.1.0]-bicycles 13 with excellent group (olefin) selectivity
(18.7:1) and good diastereofacial (cis/trans) selectiv-
ity (5.6:1). The major diastereomer arising from ICP of
the corresponding a-diazophosphonoacetate of 12 has
been tentatively assigned as the cis-PS diastereomer,
13-cis-PS. We have based this assignment on correlation
of the 31P NMR spectra with the allyl system 1 which is
consistent with the rationale noted above (Scheme 6). In
addition, chromatography elution order is consistent with
1
the cis-PS assignment,11 and H NMR chemical shift
trends support the cis-assignment.12
Correlation of the 31P spectra with the allyl system 1,
tentatively leads us to assign the major trans-diastereomer
as 13-trans-PS. The trans-assignment was further sup-
1
ported by H NMR chemical shift data.12 As shown in
Scheme 7, this product can arise from either an A-type
transition state occurring with the favorable s-trans
orientation of the RhꢀC and PꢀO p-systems (opposing
dipole), and with equatorial placement of the benzyl ether
substituent, or via a B-type transition state occurring with
the unfavorable s-cis orientation of the RhꢀC and PꢀO
p-systems, and with axial placement of the benzyl ether
substituent. This result is contrary to the expected type-B
transition state model portrayed in Schemes 6 and 7 that
predicts 13-trans-PR as the major diastereomer (B-type,
s-cis, and equatorial benzyl ether). With all factors
Scheme 7. Transition states leading to the trans diastereomers,
13-trans-PR and 13-trans-PS.

878
J. D. Moore, P. R. Hanson / Tetrahedron: Asymmetry 14 (2003) 873–880
considered, we believe that the s-trans orientation of the
RhꢀC and PꢀO p-systems may be the most influential
factor in determining facial preferences in ICP, since in
this pivotal example, it overrides the predicted preference
for a B-type transition state.13
4.4. [Bis-(2-methyl-allyloxy)-phosphoryl]-diazo-acetic
acid tert-butyl ester
tert-Butyldimethallylphosphonoacetate (425 mg, 1.40
mmol) in 4 mL CH2Cl2 was subjected to the general
diazotization protocol using KOtBu (276 mg, 2.46 mmol)
and TsN3 (485 mg, 2.46 mmol). Flash chromatography
(2:1 hexanes/EtOAc) produced 355 mg (77%) of pure
a-diazophosphonoacetate as a clear yellow oil. FTIR
3. Conclusions
1
2128, 1698, 1295 cm−1; H NMR (400 MHz, CDCl3) l
We have demonstrated that substituents play an impor-
tant role in the double diastereotopic differentiation
strategy of a-diazophosphonoacetate templates utilizing
Rh2(OAc)4-catalyzed ICP employing the (R)-pantolac-
tone auxiliary. Furthermore, we have reported a double
diastereoselective ICP of the pseudo-C2-symmetric phos-
phonate 12 that occurs with excellent selectivity. This
study represents part of our continuing program aimed
at the synthesis of diverse P-heterocycles. Additional
efforts in this area are underway and will be reported in
due course.
5.02 (s, 2H, broad), 4.91 (s, 2H, broad), 4.47 (dd,
J
HP=8.4, J=3.8 Hz, 4H), 1.74 (s, 6H), 1.44 (s, 9H); 13
C
NMR (100 MHz, CDCl3) l 139.77 (d, JCP=7.1 Hz),
113.62, 83.05, 70.44 (d, JCP=5.5 Hz), 28.14, 18.97; 31P
NMR (162 MHz, CDCl3) l 12.49; HRMS calcd for
C14H24N2O5P (M+H+) required 331.1423, found
331.1416.
4.5. [Bis-((E)-but-2-enyloxy)-phosphoryl]-acetic acid
(3R)-4,4-dimethyl-2-oxo-tetrahydrofuran-3-yl ester, 7a
Tricrotylphosphite was generated as described for
trimethallylphosphite using crotyl alcohol (2.05 mL, 24.0
mmol), Et3N (3.35 mL, 24.0 mmol) and PCl3 (1.0 g, 7.3
mmol). The crude phosphite was subsequently added
dropwise to a heated (90°C) neat solution of pantolac-
tone-iodoacetate (6.6 g, 21.8 mmol). Once complete (GC
analysis), the crude reaction was directly subjected to
flash chromatography (2:1 hexanes/EtOAc) to generate
1.9 g (86%, two steps) of pure phosphonoacete 7a as a
clear oil. [h]2D5=−2.2 (c 2.0, CHCl3); FTIR 1799, 1748,
1674, 1268 cm−1; 1H NMR (500 MHz, CDCl3) 5.80–5.73
(m, 2H), 5.59–5.53 (m, 2H), 5.34 (s, 1H), 4.51–4.44 (m,
4H), 4.00 (d, J=9.0 Hz, 1H), 3.97 (d, J=9.0 Hz, 1H),
3.07 (dd, J=55.0, JHP=14.3 Hz, 1H), 3.02 (dd, J=54.8,
4. Experimental
4.1. Representative diazotization procedure
A solution of the phosphonoacetate and CH2Cl2 was
cooled to 0°C. Base (KOtBu or Et3N) was added, the
reaction stirred for 5 min, and TsN3 was added. The
reaction was brought to room temperature and moni-
tored by TLC. Once complete, the crude reaction was
dissolved in EtOAc, washed with H2O (2×), brine (2×),
and dried (MgSO4). The organic portion was concen-
trated under reduced pressure. Flash chromatography
afforded purified a-diazophosphonoacetates.
JHP=14.3 Hz, 1H), 1.67 (s, 3H), 1.66 (s, 3H), 1.15 (s, 3H),
1.09 (s, 3H); 13C NMR (126 MHz, CDCl3) l 171.70,
164.62 (d, JCP=6.4 Hz), 131.78, 131.76, 125.42 (d,
4.2. Representative cyclopropanation procedure
A solution of the a-diazophosphonoacetate in CH2Cl2
(0.05 M) was charged with Rh2(OAc)4 (0.05 equiv.) under
an atmosphere of argon. Once the reaction was complete,
J
CP=6.4 Hz), 125.37 (d, JCP=6.2 Hz), 76.00, 75.67, 67.14
(d, JCP=6.2 Hz), 67.05 (d, JCP=6.2 Hz), 40.32, 34.24 (d,
J
CP=133.0 Hz), 22.60, 19.52, 17.50; 31P NMR (162 MHz,
the reaction was directly analyzed by H decoupled 31P
1
CDCl3) l 20.13; HRMS calcd for C16H26O7P (M+H+)
spectroscopy to elucidate stereoselectivity.
required 361.1416, found 361.1415.
4.3. [Bis-(2-methylallyloxy)-phosphoryl]-acetic acid tert-
4.6. [Bis-((E)-but-2-enyloxy)-phosphoryl]-diazo-acetic
butyl ester
acid (3R)-4,4-dimethyl-2-oxo-tetrahydrofuran-3-yl ester,
8a
Trimethallylphosphite was generated as described before
for 7b using methallyl alcohol (3.00 g, 41.6 mmol), Et3N
(4.55 g, 45.0 mmol) and PCl3 (1.88 g, 13.64 mmol). The
crude phosphite was added dropwise to a neat solution
Phosphonoacetate 7a (45 mg, 0.15 mmol) in 0.3 mL
CH2Cl2 was subjected to the general diazotization proto-
col using Et3N (0.041 mL, 0.27 mmol) and TsN3 (44 mg,
0.22 mmol). Flash chromatography (2:1 hexanes/EtOAc)
produced 43 mg (90%) of pure a-diazophosphonoacetate
8a as a clear yellow oil. [h]2D5=+4.0 (c 0.50, CHCl3); FTIR
2136, 1789, 1712, 1280 cm−1; 1H NMR (500 MHz,
CDCl3) l 5.89–5.81 (m, 2H), 5.68–5.58 (m, 2H), 5.40 (s,
1H), 4.61–4.51 (m, 4H), 4.07 (d, J=9.0 Hz, 1H), 4.03 (d,
J=9.0 Hz, 1H), 1.74 (s, 3H), 1.72 (s, 3H), 1.22 (s, 3H),
1.14 (s, 3H); 13C NMR (126 MHz, CDCl3) l 171.58,
132.47, 132.42, 125.16, 125.10, 76.09, 75.82, 68.31 (d,
t
of butyliodoacetate (8.25 g, 34.1 mmol) at 90°C. Once
complete (GC analysis), the reaction was directly sub-
jected to flash chromatography (2:1 hexanes/EtOAc) to
afford 2.4 g (57.8%, two steps) of pure phosphonoacetate
1
as a clear oil. FTIR 1730, 1661, 1286, 1261 cm−1; H
NMR (400 MHz, CDCl3) l 4.97 (s, 2H, broad), 4.85 (s,
2H, broad), 4.41 (d, JHP=7.7 Hz, 4H), 2.85 (d, JHP=21.6
Hz, 2H), 1.68 (s, 6H), 1.37 (s, 9H); 13C NMR (100 MHz,
CDCl3) l 164.32 (d, JCP=6.3 Hz), 139.83 (d, JCP=6.6
Hz), 112.89, 81.78, 69.12 (d, JCP=6.0 Hz), 35.37 (d,
J
CP=5.6 Hz), 68.13 (d, JCP=5.4 Hz), 40.14, 22.82, 19.74,
CP=134.4 Hz), 27.59, 18.70; 31P NMR (162 MHz,
17.72, 17.71; 31P NMR (162 MHz, CDCl3) l 10.25;
HRMS calcd for C16H24N2O7P (M+H+) required
387.1321, found 387.1320.
J
CDCl3) l 22.10; HRMS calcd for C14H26O5P (M+H+)
required 305.1518, found 305.1506.

J. D. Moore, P. R. Hanson / Tetrahedron: Asymmetry 14 (2003) 873–880
879
4.7. [Bis-(2-methyl-allyloxy)-phosphoryl]-acetic acid (3R)-
4,4-dimethyl-2-oxo-tetrahydrofuran-3-yl ester, 7b
at 100°C for 72 h. The mixture was concentrated under
reduced pressure. In a 25 mL round-bottom flask purged
with argon and equipped with a drierite-filled syringe, the
resulting bis-TMS-phosphonic acid was dissolved in
CH2Cl2 (5 mL) and cooled to 0°C. Oxalyl chloride (0.19
mL, 2.2 mmol) was added followed by a single drop of
DMF (which induced gas extrusion). Once complete
(ceasing of gas extrusion), the crude reaction was concen-
trated under reduced pressure. The crude phosphonyl
dichloridate was rediluted with CH2Cl2 (5 mL), cooled
to 0°C, and both Et3N (0.45 mL, 3.2 mmol) and allylic
alcohol 11 were added. Once addition was complete, the
reaction was refluxed under argon for 1 h. The crude
reaction was diluted with EtOAc, washed with H2O (2×),
brine (2×), dried (MgSO4) and concentrated under
reduced pressure. Flash chromatography (1:1 hexanes/
EtOAc) produced pure 12 as a colorless oil (100 mg 19%,
three steps). [h]2D5=−1.8 (c 0.40, CHCl3); FTIR 1736, 1267
cm−1; 1H NMR (500 MHz, CDCl3) l 7.34–7.26 (m, 10H),
5.85 (dddd, J=40.8, 17.0, 10.6, 6.3 Hz, 2H), 5.41 (dd,
J=26.0, 17.2 Hz, 2H), 5.24 (dd, J=38.2, 10.6 Hz, 2H),
5.21–5.15 (m, 1H), 5.14–5.06 (m, 1H), 4.57 (dd, J=11.9,
6.1 Hz, 2H), 4.52 (d, J=11.9 Hz, 2H), 4.13 (q, J=7.1 Hz,
2H), 3.62–3.52 (m, 4H), 3.06 (dd, J=18.8, 14.5 Hz, 1H),
3.02 (dd, J=18.8, 14.6 Hz, 1H), 1.23 (t, J=7.1 Hz, 3H);
13C NMR (126 MHz, CDCl3) l 165.67 (d, JCP=6.9 Hz),
137.83 (d, JCP=6.7 Hz), 133.81 (d, JCP=4.0 Hz), 133.73
(d, JCP=4.3 Hz), 128.37, 128.34, 127.75, 127.67, 118.54,
118.27, 73.17, 73.14, 72.42 (d, JCP=5.2 Hz), 72.35 (d,
Step a: in a 50 mL round-bottom flask equipped with an
exit needle fitted with a drierite-filled syringe, a solution
of Et3N (2.12 mL, 15.2 mmol) and methallyl alcohol (1.00
g, 13.9 mmol) in 15 mL dry Et2O was cooled to 0°C under
an atmosphere of argon. PCl3 (636 mg, 4.6 mmol) was
addeddropwise. Onceadditionwascomplete, thereaction
was allowed to stir at ambient temperature for 1 h. The
salts were filtered and washed with dry Et2O. The resulting
organic filtrate was concentrated under reduced pressure
generating crude trimethallylphosphite, which was used
directly in the subsequent Arbuzov reaction. Step b: neat
pantolactone-iodoacetate (4.13 g, 13.9 mmol) was heated
to 90°C in a 10 mL round-bottom flask and subsequently
charged with dropwise addition of trimethallylphosphite.
Once complete (GC analysis), the reaction was directly
subjected to flash chromatography (3:1 then 2:1 hexanes/
EtOAc) producing 670 mg (40%, two steps) of pure
phosphonoacete 7b as a clear oil. [h]2D5=−0.5 (c 0.95,
CHCl3);FTIR1798, 1748, 1660, 1270cm−1;1HNMR(400
MHz, CDCl3) l 5.38 (s, 1H), 5.04 (s, 2H, broad), 4.94
(s, 2H, broad), 4.55–4.44 (m, 4H), 4.04 (d, J=9.0 Hz, 1H),
4.00 (d, J=9.0 Hz, 1H), 3.18 (dd, J=36.4, JHP=14.4 Hz,
1H), 3.13 (dd, J=38.3, JHP=16.6 Hz, 1H), 1.76 (s, 6H),
1.19 (s, 3H), 1.13 (s, 3H); 13C NMR (100 MHz, CDCl3)
l 171.74, 164.64 (d, JCP=6.3 Hz), 139.83 (d, JCP=6.1 Hz),
139.77 (d, JCP=6.2 Hz), 113.68, 113.66, 76.14, 75.92,
69.89 (d, JCP=6.3 Hz), 69.76 (d, JCP=6.2 Hz), 40.40,
34.02 (d, JCP=134.9 Hz), 22.80, 19.69, 18.95, 18.93; 31P
NMR (162 MHz, CDCl3) l 20.17; HRMS calcd for
C16H26O7P (M+H+) required 361.1416, found 361.1401.
J
CP=5.2 Hz), 61.39, 35.42 (d, JCP=137.7 Hz), 14.07; 31
P
NMR (162 MHz, CDCl3) l 20.77; HRMS calcd for
C26H34O7P (M+H+) required 489.2042, found 489.2050.
4.10. (4S)-4-Benzyloxymethyl-2-((1S)-1-benzyloxy-
methyl-allyloxy)-2-oxo-3-oxa-2l5-phospha-bicyclo-
[3.1.0]hexane-1-carboxylic acid ethyl ester, 13-cis-PS
(major)
4.8. [Bis-(2-methyl-allyloxy)-phosphoryl]-diazo-acetic
acid (3R)-4,4-dimethyl-2-oxo-tetrahydrofuran-3-yl ester,
8b
Phosphonoacetate 7b (349 mg, 0.97 mmol) in 2.5 mL
CH2Cl2 was subjected to the general diazotization proto-
col using KOtBu (163 mg, 1.45 mmol) and TsN3 (286 mg,
1.45 mmol). Flash chromatography (2:1 hexanes/EtOAc)
produced243mg(65%)ofpurea-diazophosphonoacetate
8b as a clear oil. [h]2D5=+1.3 (c 0.30, CHCl3); FTIR 2138,
Diazotization of phosphonoacetate 12 was carried out
using the general diazotization protocol in 62% yield.
The resulting a-diazophosphonoacetate (20 mg, 0.04
mmol) was dissolved in 1.1 mL CH2Cl2 and subjected
to the general cyclopropanation protocol using 1 mg
Rh2(OAc)4. The reaction was stirred at room tempera-
ture until complete (TLC analysis, 72 h). The crude
mixture was concentrated under reduced pressure and
directly subjected to flash chromatography (2:1 hex-
anes/EtOAc) to yield 12 mg (63%) of the single
diastereomer 13-cis-PS as a clear oil.
1
1789, 1713, 1282 cm−1; H NMR (500 MHz, CDCl3) l
5.41 (s, 1H), 5.08 (s, 2H, broad), 4.97 (d, J=5.8 Hz, 2H),
4.59–4.51 (m, 4H), 4.06 (d, J=9.0 Hz, 1H), 4.03 (d, J=9.5
Hz, 1H), 1.79 (s, 3H), 1.78 (s, 3H), 1.22 (s, 3H), 1.21 (s,
3H); 13C NMR (126 MHz, CDCl3) l 171.55, 139.61 (d,
J
CP=6.9 Hz), 139.57 (d, JCP=6.9 Hz), 114.18, 114.07,
75.91, 75.74, 70.85 (d, JCP=5.6 Hz), 70.77 (d, JCP=5.5
Hz), 39.88, 22.83, 19.75, 19.02, 19.00; 31P NMR (162
MHz, CDCl3) l 10.45; HRMS calcd for C16H24N2O7P
(M+H+) required 387.1321, found 387.1325.
13-cis-PS (major): [h]2D5=+4.6 (c 1.00, CHCl3); FTIR
1
1727, 1279 cm−1; H NMR (400 MHz, CDCl3) l 7.36–
7.26 (m, 10H), 5.94 (ddd, J=16.8, 10.7, 6.0 Hz, 1H),
5.45 (dd, J=17.2, 1.2 Hz, 1H), 5.30–5.25 (m, 1H), 5.29
(dd, J=10.6, 1.2 Hz, 1H), 4.64–4.56 (m, 3H), 4.34 (ddd,
J=19.4, 7.2, 5.7 Hz, 1H), 4.27–4.15 (m, 2H), 3.82 (dd,
J=9.9, 5.6 Hz, 1H), 3.73 (dd, J=9.9, 7.2 Hz, 1H),
3.67–3.57 (m, 3H), 2.67 (ddd, J=8.1, 8.1, 5.9 Hz, 1H),
1.69 (ddd, J=10.1, 8.4, 4.8 Hz, 1H), 1.50–1.44 (m, 1H),
1.28 (t, J=7.1 Hz, 3H); 13C NMR (100 MHz, CDCl3)
l 167.96 (d, JCP=10.1 Hz), 137.87, 137.60, 133.64 (d,
4.9. [Bis-((1S)-1-benzyloxymethyl-allyloxy)-phosphoryl]-
acetic acid ethyl ester, 12
In an argon-flushed, screw-cap pressure tube,
dimethylethylphosphonoacetate (210 mg, 1.1 mmol) was
taken up in neat TMSCl (0.68 mL, 5.4 mmol) and stirred
J
CP=4.8 Hz), 128.44, 128.31, 127.83, 127.70, 127.61,

880
J. D. Moore, P. R. Hanson / Tetrahedron: Asymmetry 14 (2003) 873–880
118.23, 75.34 (d, JCP=4.6 Hz), 73.69, 73.10, 72.09 (d,
CP=3.9 Hz), 71.54, 62.02, 29.84 (d, JCP=8.3 Hz), 29.69,
22.35 (d, JCP=176.5 Hz), 18.49 (d, JCP=2.6 Hz), 14.04;
31P NMR (162 MHz, CDCl3) l 34.71; HRMS calcd for
C26H32O7P (M+H+) required 487.1886, found 487.1871.
synthesis and use of P-chiral compounds, see: (c) Han,
L.-B.; Zhao, C.-Q.; Onozawa, S.-y.; Goto, M.; Tanaka, M.
J. Am. Chem. Soc. 2002, 124, 3842–3843; (d) Al-Masum,
M.; Kumaraswamy, G.; Livinghouse, T. J. Org. Chem.
2000, 65, 4776–4778; (e) Denmark, S. E.; Stavenger, R. A.;
Wong, K.-T.; Su, X. J. Am. Chem. Soc. 1999, 121,
4982–4991; (f) Gilbertson, S. R.; Genov, D. G.; Rheingold,
A. L. Org. Lett. 2000, 2, 2885–2888; (g) Buono, G.; Chiodi,
O.; Wills, M. Synlett 1999, 377–388; (h) Denmark, S. E.;
Chen, C.-T. J. Am. Chem. Soc. 1995, 117, 11879–11897.
4. (a) Stoianova, D. S.; Hanson, P. R. Org. Lett. 2000, 2,
1769–1772; (b) Moore, J. D.; Sprott, K. T.; Wrobleski, A.
D.; Hanson, P. R. Org. Lett. 2002, 4, 2357–2360. For
examples of double diastereoselective processes, see: (c)
Kurth, M. J.; Brown, E. G. J. Am. Chem. Soc. 1987, 109,
6844–6845; (d) McKew, J. C.; Olmstead, M. M.; Kurth,
M. J. J. Org. Chem. 1994, 59, 3389; (e) Martin, S. F.; Spaller,
M. R.; Liras, S.; Hartmann, B. J. Am. Chem. Soc. 1994,
116, 4493–4494.
J
13-trans-PS (minor): [h]2D5=+8.2 (c 0.50, CHCl3); FTIR
1725, 1273 cm−1; 1H NMR (500 MHz, CDCl3) l 7.39–7.26
(m, 10H), 5.96 (ddd, J=17.1, 10.6, 6.5 Hz, 1H), 5.45 (d,
J=17.2 Hz, 1H), 5.29 (d, J=10.6 Hz, 1H), 5.24–5.19 (m,
1H), 4.61–4.48 (m, 3H), 4.35 (ddd, J=19.9, 6.6, 6.6 Hz,
1H), 4.16–4.10 (m, 1H), 4.08–4.02 (m, 1H), 3.76–3.55 (m,
5H), 2.59–2.53 (m, 1H), 1.72 (ddd, J=11.0, 8.4, 5.2 Hz,
1H), 1.62–1.56 (m, 1H), 1.21 (t, J=7.1 Hz, 3H); 13C NMR
(125 MHz, CDCl3) l 167.87 (d, JCP=8.8 Hz), 137.93,
137.52, 134.52, 128.45, 128.37, 127.87, 127.66, 127.59,
118.68, 77.69 (d, JCP=6.7 Hz), 73.52, 73.25, 71.99 (d,
J
CP=8.4 Hz), 61.99, 29.07, 29.02, 22.11 (d, JCP=179.8
Hz), 17.71, 14.00; 31P NMR (162 MHz, CDCl3) l 36.19.
5. (a)Doyle, M. P.;McKervey, M. A.;Ye, T. ModernCatalytic
Methods for Organic Synthesis with Diazo Compounds; John
Wiley & Sons, Inc: New York, 1998; (b) Doyle, M. P.;
Forbes, D. C. Chem. Rev. 1998, 98, 911–936 and references
cited therein.
Acknowledgements
This investigation was supported by funds provided by
the National Institutes of Health (National Institute of
General Medical Sciences, RO1-GM58103) and the Her-
man Frasch Foundation (442-HF97, administered by the
American Chemical Society). The authors thank Huw M.
L. Davies for many thoughtful discussions, and Dr. David
Van der Velde for assistance with NMR measurements,
Dr. Todd Williams for HRMS analysis, and Dr. Douglas
R. Powell for X-ray analysis. The authors also thank
Daiso Co., Ltd. Fine Chemical Department for kindly
donating the (S)-benzyl-glycidyl ether used in this study.
6. (a) Davies, H. M. L.; Huby, N. J. S.; Cantrell, W. R., Jr.;
Olive, J. L. J. Am. Chem. Soc. 1993, 115, 9468–9479; (b)
Davies, H. M. L.; Cantrell, W. R. Tetrahedron Lett. 1991,
32, 6509–6512.
7. Moore, J. D.; Sprott, K. T.; Hanson, P. R. J. Org. Chem.
2002, 67, 8123–8129.
8. Davies, H. M. L.; Doan, B. D. J. Org. Chem. 1999, 64,
8501–8508.
9. For desymmetrization strategies involving pseudo-C2-sym-
metric compounds, see: (a) Magnuson, S. R. Tetrahedron
1995, 51, 2167–2213; (b) Schreiber, S. L.; Poss, C. S. Acc.
Chem. Res. 1994, 27, 9–17. See also: (c) Hoye, T. R.; Peck,
D. R.; Trumper, P. K. J. Am. Chem. Soc. 1981, 103, 5618;
(d) Hoye, T. R.; Peck, D. R.; Swanson, T. A. J. Am. Chem.
Soc. 1984, 106, 2738; (e) Schreiber, S. L.; Wang, Z. J. Am.
Chem. Soc. 1985, 107, 5303; (f) Ward, D. Chem. Soc. Rev.
1990, 19, 1–19; (g) Shepherd, J. N.; Na, J.; Myles, D. C.
J. Org. Chem. 1997, 62, 4558–4559.
References
1. (a) Kafarski, P.; Lejczak, B. Curr. Med. Chem.: Anti-Cancer
Agents 2001, 1, 301–312; (b) Quin, L. D. A Guide to
Organophosphorus Chemistry; Wiley-Interscience: New
York, 2000; (c) Kukhar, V. P.; Hudson, H. R. Aminophos-
phonic and Aminophosphinic Acids; John Wiley & Sons, Inc:
New York, 2000; (d) Seto, H.; Kuzuyama, T. Nat. Prod.
Rep. 1999, 16, 580–596; (e) Mader, M. M.; Bartlett, P. A.
Chem. Rev. 1997, 97, 1281–1301; (f) Franz, J. E.; Mao, M.
K.; Sikorski, J. A. Glyphosate: A Unique Global Herbicide;
American Chemical Society: Washington, DC, 1997; (g)
Engel, R. Handbook of Organo-phosphorus Chemistry;
Marcel Dekker, Inc: New York, 1992; (h) Kafarski, P.;
Lejczak, B. Phosphorus, Sulfur and Silicon 1991, 63, 193–
215.
2. (a) Dhawan, B.; Redmore, D. Phosphorus and Sulfur 1987,
32, 119; (b) Kukhar, V. P.; Svistunova, N. Y.; Solodenko,
V. A.; Soloshonok, V. A. Russ. Chem. Rev. 1993, 62, 261;
(c) Natchev, I. A. Bull. Chem. Soc. Jpn. 1988, 61, 3699–3704;
(d) Kovach, I. M.; Mckay, L.; Vander Velde, D. Chirality
1993, 5, 143–149.
10. 11 was prepared from addition of trimethylsulfonium ylide
to (S)-benzyl-glycidyl ether, see: Davoille, R. J.; Ruther-
ford, D. T.; Christie, S. D. R. Tetrahedron Lett. 2000, 41,
1255–1259.
11. After chromatographic separation, we found that the
relative position of the major diastereomer 13 in the elution
order of all four diastereomers was identical to the position
of 2-cis-PS, therefore supporting the assignment of this
major diastereomer as 13-cis-PS.
12. The 1H chemical shift of the protons on the bicyclic system
which are endo to the P=O group are further downfield
than their exo counterparts.
3. For reviews on the asymmetric synthesis of phosphorus
compounds, including P-chiral phosphorus compounds,
see: (a) Kolodiazhnyi, O. I. Tetrahedron: Asymmetry 1998,
9, 1279–1332; (b) Pietrusiewicz, K. M.; Zablocka, M. Chem.
Rev. 1994, 94, 1375–1411. For recent examples of the
13. It is worth noting that the olefin selectivity within the
trans-diastereomers of 13 (13-trans-PS versus 13-trans-PR)
also substantially increased to a ratio of 7.4:1 (from 3.6:1
for the allyl-based system 1).