Validation of diethoxyphosphonate as an effective agent for charge transfer in Anion Relay Chemistry (ARC)

Article abstract of DOI:10.1021/ol3019709

The diethoxyphosphonate group comprises an effective agent to achieve negative charge migration in Type II Anion Relay Chemistry (ARC). The process involves a [1,4]-phosphorus-Brook rearrangement that proceeds via a phosphacyclic intermediate leading to an anion that can be captured by reactive electrophiles. In the absence of an exogenous electrophile, the anion derived via phosphorus migration undergoes internal displacement of the phosphonate group to produce a diastereomeric mixture of cyclopropanes.

Full text of DOI:10.1021/ol3019709

ORGANIC
LETTERS
2012
Vol. 14, No. 17
4470–4473
Validation of Diethoxyphosphonate as an
Effective Agent for Charge Transfer in
Anion Relay Chemistry (ARC)
Alexander Sokolsky and Amos B. Smith III*
Department of Chemistry, Laboratory for Research on the Structure of Matter, and the
Monell Chemical Senses Center, University of Pennsylvania, Philadelphia, Pennsylvania
19104, United States
smithab@sas.upenn.edu
Received July 16, 2012
ABSTRACT
The diethoxyphosphonate group comprises an effective agent to achieve negative charge migration in Type II Anion Relay Chemistry (ARC).
The process involves a [1,4]-phosphorus-Brook rearrangement that proceeds via a phosphacyclic intermediate leading to an anion that can be
captured by reactive electrophiles. In the absence of an exogenous electrophile, the anion derived via phosphorus migration undergoes internal
displacement of the phosphonate group to produce a diastereomeric mixture of cyclopropanes.
Anion Relay Chemistry (ARC) (Figure 1) has emerged
as a powerful tactic for the rapid assembly of molecular
complexity.1,2 The Type II ARC protocol (Figure 1B), in
particular, holds considerable significance given the po-
tential to mimic what in polymer chemistry is known as
“living polymerization”.3 That is, the controlled addition
of individual bifunctional linchpins in an iterative fashion,
followed by termination with an electrophile, can effec-
tively provide access to complex molecular scaffolds.
To augment the potential of the Type II ARC tactic, the
design, synthesis, and validation of effective new linchpins
with diverse migrating groups is required, which possess
both orthogonal reactivity and the kinetic potential
to undergo the requisite “Brook-like” migration under
mild conditions. In this communication, we validate the
diethoxyphosphonate group as an effective transfer agent
to achieve negative charge migration in Type II ARC.
We were attracted to the phosphonate group for two
major reasons. First, phosphorus shares with silicon the
propensity to form a strong σ-bond with oxygen in
preference to carbon, with the oxygenÀphosphorus bond
favored by ca. 20 kcal/mol.4 Second, unlike silicon, nu-
cleophilic attack at the phosphonate group held the promise
of generating a neutral five-coordinate phosphorus ring
intermediate with expulsion of an ethoxy anion.5 Assuming
the presence of an anion stabilizing group (ASG) R to the
phosphonate, readdition of the displaced alkoxide under
equilibrium conditions could in turn lead to a carbanion
capable of reacting with an electrophile (Figure 2A).
A number of [1,2]-,6 [1,3]-,7,8 and [1,4]-9,10phosphorus-
Brook rearrangements are known. The most common
(4) Quin, L. D. A Guide to Organophosphorus Chemistry; Wiley-
Interscience: New York, 2000. For comparison, the siliconÀoxygen bond
is stronger by ca. 40 kcal/mol (Brook, A. G. Acc. Chem. Res. 1974, 7, 77).
(5) Izydore, R. A.; Ghirardelli, R. G. J. Org. Chem. 1973, 38, 1790.
(6) (a) Hall, L. A. R.; Stephens, C. W.; Drysdale, J. J. J. Am. Chem. Soc.
€
€
€ €
1957, 79, 1768. (b) Demir, A. S.; Reis, B.; Reis, O.; Eymur, S.; Gollu, M.;
Tural, S.; Saglam, G. J. Org. Chem. 2007, 72, 7439. (c) Demir, A. S.;
€
€ €
Esiringu, l.; Gollu, M.; Reis, O. M. J. Org. Chem. 2009, 74, 2197.
:
:
€
ꢀ
€
(d) Demir, A. S.; Reis, O.; Igdir, A. C-.; Esiringu, I.; Eymur, S. J. Org.
Chem. 2005, 70, 10584. (e) Janzen, A. F.; Smyrl, T. G. Can. J. Chem. 1972,
50, 1205. (f) Coffinier, D.; El Kaim, L.; Grimaud, L. Synlett 2008, 2008,
1133.
(1) Smith, A. B.; Wuest, W. M. Chem. Commun. 2008, 5883.
(2) Smith, A. B.; Xian, M. J. Am. Chem. Soc. 2005, 128, 66.
(3) Webster, O. Science 1991, 251, 887.
(7) Beier, P.; Pohl, R.; Alexandrova, A. V. Synthesis 2009, 2009, 957.
(8) Obayashi, M.; Ito, E.; Matsui, K.; Kondo, K. Tetrahedron Lett.
1982, 23, 2323.
r
10.1021/ol3019709
Published on Web 08/10/2012
2012 American Chemical Society

Figure 1. (A) Type I ARC; (B) Type II ARC.
prototype, orginally reported by Wadsworth and Emmons,9
entails the reaction of stabilized phosphonate anions with
epoxides to form cyclopropanes (Figure 2A), a reaction
process that typically requires both a high temperature and
prolonged reaction time. Independent studies by Singh,11
Merschaert,12 and Ghirardelli5 suggested that the reaction
proceeds through a mechanism similar to what we now
term Type I ARC, involving a stepwise “Brook-like” re-
arrangement involving initial explusion of an ethoxide
anion. Readdition of the ethoxide and completion of
the CfO phosphorus migration furnishes a stabilized
anion that undergoes intramolecular displacement of the
diethoxyphosphate to generate the trans cyclopropane.
More recently, Krawczyk et al.13 reported that the rear-
ranged adduct 9, obtained by addition of a nucleophile to
the aldehyde of bifunctional phosphonate linchpin 8
(Figure 2B), after isolation and deprotonation, can under-
go intramolecular cyclization to form a cyclopropane ring.
To the best of our knowledge, this is the only example of
what is a formal Type II phorphorus “ARC-like” process,
albeit achieved in a stepwise fashion.
Figure 2. Phosphorus-Brook rearrangement.
To explore this scenario, we examined the reaction of
alcohol 11 possessing a β-diethoxyphosphonate group, en-
visioned to furnish two-component adduct 13 (Figure 2C).
If successful, we would turn to a three-component Type II
ARC process involving generation of the corresponding
oxyanion by nucleophilic addition to aldehyde 12.
To this end, alcohol 11 was treated with potassium
hexamethyldisilazide (KHMDS), followed by addition of
allyl bromide, initially employing tetrahydrofuran (THF),
dichloroethane(DCE), and/ordimethylformamide(DMF)
as solvent systems at À78 °C. Under these conditions, only
the phosphorus-Brook rearranged product 15 was observed
(Table 1, entries 1À3). However, upon addition of an in-
creasing amount of hexamethylphosphoramide (HMPA),
employing first THF and then DMF as the solvent system
with allyl bromide as the electrophile, the two component
adduct 14 was observed, in conjunction with diallylated
adduct 17, allylated phosphacycle 16, and the phosphorus-
Brook product 15 (Table 1, entries 4À6).
Given that ring closure has been implicated as the rate-
determing step in the WadsworthÀEmmons cyclopropa-
nation reaction,9 the possibility of performing the phos-
phorus-Brook rearrangement at lower temperature might
permit the derived anion to react with an exogenous
electrophile, thus expanding the scope of the ARC tactic
(Figure 2C).
Further reaction optimization was guided by the mech-
anistic hypothesis outlined in Figure 3, which accounts for
the formation of 14 as well as byproducts 15, 16, and 17.
Specifically, initial deprotonation of 11 was envisioned to
lead to phosphacycle 18, which in the presence of excess
base, could undergo deprotonation at the carbon bearing
the phosphonate. Alkylation with allyl bromide would
lead to 16. Alternatively, in the presence of only 1.0 equiv
of KHMDS, the liberated alkoxide and phosphacycle 18,
presumably in equilibrium with carbanion A, would lead
to alkylation with allyl bromide to furnish the Type II
(9) Wadsworth, W. S.; Emmons, W. D. J. Am. Chem. Soc. 1961, 83,
1733.
(10) Bray, C. D.; de Faveri, G. J. Org. Chem. 2010, 75, 4652 and
references cited therein.
(11) Singh, A. K.; Rao, M. N.; Simpson, J. H.; Li, W.-S.; Thornton,
J. E.; Kuehner, D. E.; Kacsur, D. J. Org. Process Res. Dev. 2002, 6, 618.
(12) Delhaye, L.; Merschaert, A.; Delbeke, P.; Brione, W. Org.
Process Res. Dev. 2007, 11, 689.
(13) Krawczyk, H.; Wasek, K.; Kedzia, J.; Wojciechowski, J.; Wolf,
W. M. Org. Biomol. Chem. 2008, 6, 308.
Org. Lett., Vol. 14, No. 17, 2012
4471

Table 1. Two Component Study
Figure 4. Bifunctional linchpins.
KHMDS HMPA
temp
yield of
entry solvent (equiv) (mL)
(°C)
14:15:16:17a 14 (%)
of enol ethers 25.14 A variety of temperature regimes and
additives did not improve the ratio of 24 relative to 25.
1
2
3
4
5
6
7
8
THF
DCE
DMF
THF
THF
THF
DMF
DMF
1.2
1.2
1.2
1.1
1.2
1.1
1.0
1.0
0
0
0
1
2
2
1
À78
0:1:0:0
0:1:0:0
0:1:0:0
0
0
À78
À78
0
À78 to rt 2.5:0:0:1
À78 to rt 3:0:0:1
46
49
43
90
90
À20
À60
1:1.1:.2:0
1:0:0:0
0.2 À60
1:0:0:0
a Determined by 1H NMR of the crude product mixture following
workup.
Figure 5. Three-component coupling with linchpin 21.
ARC product 14. At higher temperatures, 14 can be
deprotonated by carbanion A to reversibly yield carbanion
B, which, in turn, could react with excess allyl bromide to
furnish the diallylated product 17. Accordingly, the use of
1.0 equiv of both KHMDS and allyl bromide eliminated
formation of both 16 and 17 and allowed adduct 14 to be
isolated in 90% yield (Table 1, entries 7À8).
We therefore turned to epoxide linchpin 22, employing
methyl cuprate as the initiating nucleophile and methyl
iodide as the terminating electrophile15 (Table 2). After
Table 2. Optimization of Three-Component Type II ARC with
Linchpin 22
temp
HMPA
(equiv)
yield of
26a(%)
entry
(°C)
%THF
26:27b
1
À20
À20
À40
À40
40
50
50
25
0
0
10
10
20
5
27
45
66
60
82
55
55
>10:1
>10:1
>10:1
7:1
2
3
4c
5
25
25
25
10:1
Figure 3. Proposed phorphorus-Brook reaction sequence.
6
7d
À40
2
>10:1
10:1
À40
5
Turning to the proposed three-component union, linchpins
19 and 20 (Figure 4), possessing aldehydic and epoxide ele-
ctrophilic sites, respectively, were prepared (see Supporting
Information). Unfortunately, both linchpins proved unten-
able due to rapid and exclusive deprotonation of the acidic
proton R to the phosphonate. To circumvent this issue, we
introduced an R-alkyl substituent to provide linchpins 21À23.
Initial experiments were conducted with linchpin 21
employing n-butyllithium as the initiating nucleophile
and methyl iodide as the terminating electrophile. Pleas-
ingly, when n-butyllithium was added to an ethereal solu-
tion of 21 at À78 °C, followed by rapid addition of methyl
iodide in a solution of THF and HMPA (4:1) (Figure 5),
adduct 24 was isolated in 42% yield, along with 20À30%
a Isolated yields. b Ratio determined by 1H NMR. c Reaction was run
at 0.1 M. d Reaction was run with 1.5 equiv of MeI.
significant optimization, wediscoveredthataddition of the
linchpin in diethyl ether to an ethereal solution of cuprate
at À40 °C, followed immediately by addition of a solution
of methyl iodide in THF and HMPA (5 equiv) with
rapid warming to room temperature produced an 82%
yield of the desired three-component adduct 26 after 1 h.
(14) This side product arises from enolization followed by trapping
with methyl iodide.
(15) Methyl iodide was chosen to avoid diastereomeric product
mixtures.
4472
Org. Lett., Vol. 14, No. 17, 2012

However, when the addition of the nucleophile and
electrophile was not carried out in rapid succession, a
significant amount of cyclopropane ring formation was
observed. Equally important, as illustrated in Table 3,
when a terminating electrophile was not employed, cyclo-
propane derivatives were rapidly formed in good yield
at low temperature (À78 °C).16 Both methyl linchpin 22
and benzyl linchpin 23 were competent substrates for
alkyl and phenyl cuprates. Importantly, these conditions
are quite mild compared to typical Type I ARC cyclo-
propanations.
ethoxide to 31 in the presence of MeI employing the
previously developed three-component reaction condi-
tions, 33 was obtained in 69% yield, along with 32 and
cyclopropane 34, both in 15% yield. Alternatively, without
an exogenous electrophile, cylopropane 34 resulted in 65%
yield as a single diastereomer.16 Thus, the phosphorus-
Brook reaction manifold can be accessed via stable phos-
phacycles, in a similar fashion as siloxanes in the silicone
ARC tactic.17,18
Table 3. Cyclopropane Synthesis
Figure 6. Phosphorus-Brook from a cyclic intermediate.
In summary, the diethoxyphosphonate group has been
validated as an effective transfer agent in the phorphorus-
Brook rearrangement leading to Type II Anion Relay
Chemistry (ARC). In addition, we have demonstrated that
a neutral five-coordinate phorphorus ring intermediate
can provide access to the phosphorus-Brook/ARC reac-
tion coordinate. Taken together, these results significantly
augment the Type II ARC tactic. Studies to increase the
scope and synthetic utility of the phorphorus-Brook re-
arrangement continue in our laboratory.
Finally, we were intrigued by the similarity of the phos-
phacycle 29 (Figure 6), presumably formed by nucleophilic
addition to 22 followed by cyclization with elimination of
ethoxide, and siloxane 30, which we recently reported to be
an alternative entry point to the silicon-ARC manifold.17,18
In particular, reaction of 30 with an organolithium species
generates a hypervalent silicate, which is implicated to be
an intermediate or transition state in the silicon migration.
We wonderedif similar activation of phosphacycle 29, here
by the addition of lithium ethoxide, would lead to a similar
hypervalent phosphorus species in the phosphorus-Brook
manifold.
Acknowledgment. Financial support was provided by the
National Institutes of Health (Institute of General Medical
Sciences) through Grant GM-29028. We also thank
Dr. Adam T. Hoye for providing helpful suggestions as well
as Drs. George Furst and Jun Gu, and Dr. Rakesh Kohli at
the University of Pennsylvania for assistance in obtaining
NMR and high-resolution mass spectra, respectively.
Accordingly, we prepared the related phosphacycle 31
(see Supporting Information). Upon addition of lithium
Supporting Information Available. Experimental pro-
cedures and spectroscopic and analytical data for all new
compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
(16) Cyclopropane stereochemistry assigned by NOESY and chemi-
cal shift correlation (see Supporting Information).
(17) Smith, A. B.; Tong, R.; Kim, W.-S.; Maio, W. A. Angew. Chem.,
Int. Ed. 2011, 50, 8904.
(18) Smith, A. B.; Hoye, A. T.; Martinez-Solorio, D.; Kim, W.-S.;
Tong, R. J. Am. Chem. Soc. 2012, 134, 4533.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 17, 2012
4473