Stille coupling of an aziridinyl stannatrane

Article abstract of DOI:10.1021/jo4005052

An aziridinyl stannatrane 8 couples with aryl or alkenyl halides RX under modified Stille conditions to afford substituted aziridines. Efficient coupling at room temperature is possible in the best examples in the presence of ( ^tBu₃P

Full text of DOI:10.1021/jo4005052

Note
pubs.acs.org/joc
Stille Coupling of an Aziridinyl Stannatrane
Naresh Theddu and Edwin Vedejs*
Department of Chemistry, University of Michigan, Ann Arbor, Michigan 48109, United States
S
* Supporting Information
ABSTRACT: An aziridinyl stannatrane 8 couples with aryl or alkenyl
halides RX under modified Stille conditions to afford substituted
aziridines. Efficient coupling at room temperature is possible in the best
examples in the presence of (tBu3P)2Pd and CuOP(O)Ph2 (CuDPP).
ecent reports have described the first examples of
encouraged to see that aziridine transfer is possible from 1,
suggesting that the key transmetalation step involving the
aziridinyl C−Sn bond is not beyond reach when copper salts
are used along with the palladium source. Accordingly, we
investigated a modified Stille coupling using a more strongly
polarized stannatrane (more precisely, a trideoxastannatrane)
derivative in place of 1 to improve C−Sn reactivity.9 As
outlined below, several cases were found where this approach
allows efficient Stille cross-coupling to give substituted
aziridines under mild conditions. We have also encountered
cases where the substrates display unexpected reactivity.
Preparation of the key stannatrane 8 began with standard
tin−lithium exchange from 1 to the lithioaziridine 2,1,10
followed by reaction with the chlorodeoxastannatrane 7
(Scheme 1).9 According to characteristic NMR signals due to
the aziridine CH adjacent to Sn (1H δ 0.63 ppm, doublet, J =
7.0 Hz), this procedure gave 8 in situ. However, attempts to
purify 8 by chromatography on silica gel were not successful
due to conversion into 9 by protodestannylation. This behavior
was somewhat unexpected because analogous stannatranes
containing an exocyclic sp3 hybridized Sn−C bond survive
chromatography.9 Evidently, the increased s-character in the
Sn−C bond due to the strained ring together with the adjacent
nitrogen electron pair are responsible for the relatively greater
polarization in the Sn−C bond of 8. At the extreme, such an
effect might induce carbenoid generation by α-elimination and/
or conversion to dimeric alkenes as observed by Hodgson et al.
in the case of lithiated aziridines at −78 to 0 °C.11 However,
crude 8 was quite stable under inert conditions and survived
refluxing toluene over 2 days. None of the unusual
decomposition products previously encountered in Negishi
coupling using the chlorozinc derivative 3 were observed in the
course of the present study.
R
palladium-catalyzed cross-coupling of aziridine derivatives
using modified Negishi activation procedures.1 Since these
advances were partly the result of exploiting re-optimized
ligands for the palladium catalyst,2 we were interested to know
whether similar improvements are possible using the readily
accessible 2-tributylstannyl aziridine 1,1a thereby avoiding
conversion to 2 and 3 as required for Negishi coupling (Figure
1). We also expected that a Stille procedure would provide a
Figure 1. Metalated aziridines as potential coupling partners.
greater safety margin with respect to aziridine decomposition
compared to the Negishi method. A preliminary survey of
improved Stille conditions gave no hint of coupling between 1
and iodobenzene.3−5 However, with the more reactive halo
esters 4 and 5, we did see low yields of the coupled product 6
in the presence of Cu salts (Table 1).6−8 Attempts to improve
the above experiments were not promising, but we were
Table 1. Attempted Coupling of 1 with 4 or 5
yield
of 6
entry ArX
conditions
a
1
2
3
4
4
4
Pd2(dba)3/Sphos THF, reflux, 16 h
Pd(tBu3P)2 (5%) PhMe, 100 °C, 20 h
ND
b
trace
c
Pd(tBu3P)2 (5%) CuCN (10%)3,6 PhMe, 100 °C,
21%
20 h
Undesired protodestannylation during attempted purification
of 8 raised concerns about Stille applications because the crude
reagent contains Bu4Sn as the byproduct of tin−lithium
exchange. However, 8 was expected to have considerably
higher metal exchange reactivity compared to that of Bu4Sn due
to the effect of transannular nitrogen lone pairs,9 the same
e
d
4
5
5
5
Pd(tBu3P)2 (5%) CuTC (1 equiv)4,5 DMF, rt, 16 h trace
e
d
Pd(tBu3P)2 (5%) CuTC (1 equiv)5,7 THF, 50 °C,
10%
16 h
g
f
6
5
Pd(tBu3P)2 (5%) CuDPP (1.5 equiv)8 THF, 50 °C, ND
16 h
a
b
Coupling product not detected by NMR assay. 31% C6H5CO2Me
c
d
isolated. 54% of C6H5CO2Me isolated. 10% C6H5CO2Me isolated.
e
f
CuTC = Cu(I) thiophenecarboxylate. 90% of aziridine 1 recovered.
g
Received: March 8, 2013
CuDPP = CuOP(O)Ph2.
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo4005052 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
Scheme 1.
effect that is partly responsible for the surprisingly facile
protodestannylation of 8. Modified Stille cross-coupling
reactions were therefore conducted using unpurified 8,
prepared from 1.5 equiv of 1 for each equivalent of the
intended halide coupling substrate. For purposes of character-
ization, 8 could be separated from nearly all of the byproducts
by thin layer chromatography on neutral alumina, but
fortunately, this level of purity was not necessary for successful
Stille coupling.
Procedures were initially evaluated with aryl halides 4 and 5
and conventional palladium sources. However, 5 mol %
Pd(PPh3)2Cl2 or Pd(CH3CN)2Cl2 in DMF at room temper-
ature or Pd2(dba)3/RuPhos in DMF at 80 °C gave no
detectable coupling product 6. On the other hand, heating 5
and 8 in DMF with 5 mol % Pd(PPh3)2Cl2 did afford
substantial coupling product 6 (Table 2, entry 1), and
Although this copper additive was introduced by Liebeskind et
al. primarily for palladium-catalyzed cross-coupling reactions
with thioesters,8a it was also shown to be effective in palladium-
free coupling applications between allylic phosphinites and
Bu3SnR (R = sp2-hybridized carbon) in an earlier thesis from
the same group.8b Metal exchange (Sn to Cu) as well as a tin
scavenging role were invoked for the CuDPP additive.8b,c In
our aziridine system, CuDPP proved to be an excellent copper
source in modified Stille reactions of the aziridinyl stannatrane
8 with aryl halides as coupling partners in the presence of
palladium. Thus, a ca. 10−20% increase in product yields was
observed in the optimized simple examples of Table 2 (entries
7, 8, 10) versus our prior study using the Negishi coupling of 3
with the same aryl halides. Most of the data points in Table 2
were obtained using 50 mg of crude 8, but entry 7 was also
tested on a ca. 0.5 g scale, resulting in similar yields of 6 (80%
with iodide 5 as limiting reagent; 85% with 8 as limiting
reagent). Also important, the best CuDPP/DMF conditions
allowed cross-coupling at room temperature, a key feature for
synthetic applications involving potentially sensitive functional
environments.
a
Table 2. Modified Stille Coupling of 8 with ArX
entry Ar-X
reagents
T, t
product
1
5
Pd(PPh3)2Cl2 (5%) DMF
100 °C,
16 h
6 (40%)
The Pd(tBu3P)2/CuDPP conditions were applied to several
more highly functionalized substrates with surprising results
(Scheme 2). The coupling of 8 with the Z-iodoacrylate ester 11
proceeded smoothly to give 12 (85%). However, the same
conditions applied to vinyl bromide afforded a second product
in addition to the expected 13. Based on the HRMS and NMR
data, the second product was assigned the alkene structure 14,
corresponding to dimerization of the aziridine subunit. We did
not investigate the dimerization pathway at the mechanistic
level, but control experiments established that formation of 14
is independent of palladium. Thus, reaction of 8 with CuDPP
(1 equiv, DMF, 16 h) in the absence of halide substrate or
palladium source gave the same alkene 14, a finding that
implicates metal exchange between 8 and CuDPP as the
initiating event leading to 14.11 The undesired dimerization did
not interfere in the case of the iodoacrylate coupling to 12,
presumably because the Stille pathway is relatively fast. On the
other hand, 14 was detected in the crude product mixture
obtained from the Stille coupling of the iodobenzoate 5 with 8
and may also be formed as a minor byproduct in some of the
other examples of Table 2.
2
3
5
4
(allyl-PdCl)2 (5%) CH3CN
Pd(dppf)Cl2 (5%) PhMe
rt, 20 h
6 (10%)
6 (15%)
110 °C,
10 h
4
5
Pd(tBu3P)2 (5%)12 PhMe
100 °C,
16 h
6 (20%)
5
6
5
5
Pd(tBu3P)2 (5%) CuI (10%) DMF rt, 20 h
6 (46%)
6 (50%)
Pd(tBu3P)2 (5%)CuDPP (150%)8
THF
rt, 24 h
rt, 4 h
rt, 4 h
7
5
4
Pd(tBu3P)2 (5%) CuDPP (150%)8
DMF
6 (90%)
6 (85%)
8
Pd(tBu3P)2 (5%) CuDPP (150%)8
DMF
9
PhI Pd2(dba)3 (5%) (2-Furyl)3P
80 °C,
20 h
10
(20%)
(10%)13 DMF
10
PhI Pd(tBu3P)2 (5%) CuDPP (150%)8
DMF
rt, 4 h
10
(90%)
a
Exploratory experiments using 50 mg crude 8 and aryl halide as
limiting reagent.
indications of aziridine coupling at room temperature were
observed with allylpalladium chloride in acetonitrile (entry 2).
Further optimization revealed that experiments conducted in
the presence of copper additives were more promising (entry 4
vs entries 5−7), and the best results were seen with Cu(I)
diphenylphosphinite (CuDPP) in DMF (entries 7, 8, 10).8
Given the modest yield of vinyl aziridine 13 obtained using
CuDPP, we also explored an alternative procedure using CuI/
CsF4 as the copper source (conditions B, Scheme 2). This
B
dx.doi.org/10.1021/jo4005052 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
with CuDPP (1 equiv) in DMF at rt (16 h), along with 55% of
recovered 15. Thus, copper−iodide exchange appears to trigger
the formation of the bis-indole 17 from 15.14 Therefore, we
attempted to couple 15 with 8 using a copper-free procedure at
60 °C in toluene with (tBu3P)2Pd (conditions C). However,
this gave no more than traces of coupling product 16, recovery
of unreacted 8, and slow conversion to deiodo indole (15 with
H in place of I; entry 5). Significant dimerization was not
observed.
Scheme 2
In view of the unexpected dimer formation from the N-Boc-
iodoindole 15, the standard Pd(tBu3P)2/CuDPP cross-coupling
procedure was attempted with the simpler N−H substrate 18.
However, the desired 19 was not detected, and nearly all of 18
was recovered (90%), along with the protodestannylated
aziridine 9 (10%) as well as the dimeric alkene 14 (75%).
Accordingly, we tested the coupling of 8 with 18 in the absence
of the copper additive CuDPP in DMF at 80 °C, but the only
coupling product proved to be the butylated indole 20 (ca.
85%),15 resulting from Stille coupling by the Bu4Sn byproduct
present in solutions of 8 (entry 7). This experiment also
returned ca. 90% of 9 from protodestannylation of 8.
Eventually, it was found that the desired coupling product 19
does form under copper-free conditions at 60 °C in the
presence of 5 mol % of (tBu3P)2Pd in anhydrous toluene
(conditions C; entry 8). Although 19 was the major product,
10−20% of 9 was still formed along with other products,16
suggesting that the indole NH is partly responsible for
conversion of 8 to 9.
The reasons for lack of cross-coupling reactivity between N−
H indole 18 and 8 were initially puzzling. We therefore tested
the reactivity of 18 in a more conventional Stille reaction with
Bu3SnCHCH2 in the presence of 5 mol % Pd[PPh3]4 and
CuDPP at 70 °C. No lack of reactivity was found, and the
expected vinylation product 21 was isolated in 90% yield.
During the coupling experiment, precipitation of an inter-
mediate palladium complex was observed. To obtain larger
quantities of the complex, 18 was reacted with 1.15 equiv of
Pd[PPh3]4 at 70 °C in the absence of CuDPP or stannane. This
gave an intermediate complex identified as the novel
indolylpalladium iodide 22 (97%).17 Evidently, this species
reacts normally in the Stille coupling with Bu3SnCHCH2.
However, attempts to couple 22 with the aziridinyl stannatrane
8 resulted in destruction of 8 and recovery of 22 but gave no
products of cross-coupling. Taken together, the above experi-
ments suggest that small differences in the substrate-dependent
rates of Stille coupling versus the rates of competing side
reactions are important factors in the indole series. However,
coupling does occur under conditions C, and the dependence
of yield on substrate and procedure offers broad opportunities
for optimization.
With a better understanding of the empirical parameters and
possibilities, the one additional substrate 23 was explored to
evaluate survival of a silyl ether protecting group in various
coupling procedures. As expected, coupling of 8 with 23 under
conditions B gave desilylated product 24 (63%) along with ca.
20% of the dimeric alkene 14 (Scheme 3, eq 1; Table 3, entry
10). Conditions A did afford the desired silyl ether 25 (33%),
but the major product (60%) was 14. Apparently, the Stille
coupling with the relatively electron-rich substrate 23 is slower
compared to the copper-mediated dimerization pathway from
8. Fortunately, this complication could be minimized by using
the copper-free method (conditions C) to give 25 in 63% yield
(eq 2; Table 3, entry 11).
alternative is less desirable for applications of interest in our
laboratory where survival of silicon-based protecting groups is
essential. However, the procedure worked very well in the case
of the simple Stille coupling of 8 with vinyl bromide, resulting
in 86% of the vinylated product 13.
Complications were also encountered with the iodoindole
substrate 15, which reacted with 8 to afford the expected 16 as
the minor product (40%) along with the bis-indole 17 (50%;
Table 3, entry 4). In this case, the palladium-free control
experiment gave 40% conversion to 17 upon treatment of 15
Table 3. Scheme 2 and Scheme 3 Coupling Summary
entry
halide
conditions
products
12 (85%)
a
1
EtO2CCHCHI
A
a
2
CH2CHBr
A
13 (40%), 14 (40%)
13 (86%)
b
3
CH2CHBr
B
a
4
15
15
18
18
18
23
23
23
A
16 (40%), 17 (50%)
reductive deiodination
c
5
C
a
d
6
A
9 (10%), 14 (75%)
ae
,
7
A
20 (85%)
c
8
C
19 (30%)
a
9
A
25 (33%), 14 (60%)
24 (63%), 14 (20%)
b
10
11
B
c
f
C
24 (63%)
a
Conditions A: Pd(tBu3P)2 (5%), CuDPP (150%), rt, DMF.
b
Conditions B: Pd(tBu3P)2 (5%), CuI (10%)/CsF (200%), 45 °C,
c
d
DMF. Conditions C: Pd(tBu3P)2 (5%), 60−65 °C, PhMe. 90% of 18
recovered. Reaction at 80 °C. Unreacted 8 (35%) was also recovered.
e
f
C
dx.doi.org/10.1021/jo4005052 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
128.2, 127.9, 127.4, 126.9, 126.1, 75.1, 74.4, 73.2, 55.0, 35.2, 29.5, 23.4,
7.8.
Scheme 3
Methyl 4-[(2R,3R)-3-(benzyloxymethyl)-1-tritylaziridin-2-yl]-
benzoate (6). Aryl Halide As Limiting Reagent, Preparative Scale.
In a 100 mL, oven-dried, single-neck round bottomed flask equipped
with a magnetic stirring bar and a rubber septum were placed crude
aziridinyl stannatrane 8 (0.56 g, 0.85 mmol estimated by NMR assay),
methyl 4-iodobenzoate (5, 0.147 g, 0.563 mmol), CuDPP (0.24 g, 0.85
mmol), and Pd(tBu3P)2 (0.014 g, 0.003 mmol) inside the glovebox.
This mixture was flushed with a stream of N2 and dissolved in dry
DMF (50 mL). The resulting solution was stirred at rt with
monitoring for completion of reaction (4 h). The reaction mixture
was diluted with ethyl acetate (100 mL) and water (100 mL). The
layers were separated, and the organic layer was washed with water (3
× 20 mL), dried (MgSO4), and concentrated. Silica gel chromatog-
raphy (10% ethyl acetate in hexanes) afforded 0.24 g (80%) of 6 as
clear oil, identified by NMR comparison with the reported data.1a
[α]20D = −70.0 (c 1.1, CHCl3); HRMS-ES+ (m/z) [M + H] calcd for
C37H34NO3, 540.2539, found 540.2535; IR (neat, cm−1) 3020, 1725;
1H NMR (500 MHz, CDCl3, ppm) δ 8.01 (d, J = 8.2 Hz, 2H), 7.54 (d,
J = 8.5 Hz, 2H), 7.49 (d, J = 7.0 Hz, 6H), 7.29−7.16 (m, 13H), 6.90
(dd, J = 7.5 Hz, 2.5 Hz, 2H), 4.18 (d, J = 12.0 Hz, 1H), 4.11 (d, J = 12.
0 Hz, 1H), 3.96 (s, 3H), 3.85 (dd, J = 10.1, 4.3 Hz, 1H), 3.36 (dd, J =
10.2, 8.0 Hz, 1H), 2.54 (d, J = 6.4 Hz, 1H), 1.95 (ddd, J = 7.7, 6.5, 4.3
Hz, 1H). 13C NMR (125.7 MHz, CDCl3, ppm) δ 167.1, 143.9, 142.8,
137.8, 129.5, 129.3, 128.2, 128.0, 127.9, 127.5, 127.3, 126.8, 75.4, 72.8,
66.9, 52.1, 39.1, 38.9.
Aziridinyl Stannatrane 8 as Limiting Reagent, Preparative Scale.
In a similar procedure, aziridinyl stannatrane 8 (0.56 g, 0.85 mmol),
methyl 4-iodobenzoate (5, 0.33 g, 1.27 mmol), CuDPP (0.24 g, 0.85
mmol), and Pd(tBu3P)2 (0.032 g, 0.063 mmol) were combined as
before and stirred at rt (4 h). The same workup and purification
afforded 0.39 g (85%) of 6 as clear oil having the same spectroscopic
characteristics.
General Procedure for Small Scale Coupling (Scheme 2 and
Table 3); Preparation of 6 Using Conditions A. In a 25 mL, oven-
dried, single-neck round bottomed flask equipped with a magnetic
stirring bar and a rubber septum was placed aziridinyl stannatrane 8
(0.05 g, 0.05 mmol), methyl 4-bromobenzoate (0.007 g, 0.033 mmol),
CuDPP (0.014 g, 0.05 mmol), and Pd(tBu3P)2 (0.001 g, 0.002 mmol)
inside the glovebox. This mixture was flushed with a stream of N2 and
dissolved in dry DMF (3 mL). The resultant solution was stirred at rt
while monitoring for completion of reaction (typically 4 h). The
reaction mixture was diluted with ethyl acetate (20 mL) and water (15
mL). The layers were separated and the organic layer was washed with
water (3 × 10 mL), dried (MgSO4), and concentrated. Silica gel
chromatography (10% ethyl acetate in hexanes) afforded 0.015 g
(85%) of 6 as clear oil, identical to the material described above.
(2R,3R)-2-(Benzyloxymethyl)-3-phenyl-1-tritylaziridine (10).
Conditions A with 8 (0.055 g, 0.054 mmol) and PhI (0.007 g, 0.036
mmol); silica gel chromatography afforded 0.015 g (90%) of 10.1a
(Z)-Ethyl 3-[(2R,3R)-3-(Benzyloxymethyl)-1-tritylaziridin-2-
yl]acrylate (12). Conditions A with 8 (0.053 g, 0.050 mmol) and
ethyl (Z)-3-iodoacrylate (0.007 g, 0.033 mmol); silica gel chromatog-
raphy afforded 0.013 g (85%) of 12.1a
Overall, the combination of aziridine 8 and the CuDPP
method (conditions A) works over a range of substrates and
holds promise for coupling applications of aziridines with
valuable halides. The combined effects of stannatrane and
copper salt activation allow coupling at room temperature.
However, the stannatrane effect alone is sufficient in cases
where the reactants and products are sensitive to Cu additives
but can tolerate heating to ca. 60−80 °C. For substrates
containing silicon protecting groups, it would be advisable to
evaluate the copper-free procedure at 60 °C (conditions C) as
well as conditions A. The alternative procedure using CuI/CsF
as the copper source was not tested extensively (conditions B)
but is available for those applications where dimerization of the
aziridine subunit interferes due to marginal halide reactivity. All
of these procedures benefit from the unique Stille coupling
reactivity of the deoxastannatrane substituent in an aziridine
environment.18
EXPERIMENTAL SECTION
■
Unless otherwise noted, all reagents were used as received.
Tetrahydrofuran (THF) was distilled over sodium/benzophenone
ketyl, and toluene was distilled over CaH2 prior to use in the reactions.
Commercial copper diphenylphosphinate (CuDPP) was used as
received. All reactions were performed under an N2 atmosphere and in
oven-dried glassware unless otherwise stated. TLC analysis was
performed on Whatman 0.25 mm K6F silica gel 60 Å plates, visualized
with the aid of UV light or an appropriate stain (KMnO4 or I2). Flash
chromatography was accomplished using Silicycle Silica Gel: SiliaFlash
P60, 40−63 μm, 60 Å. High-resolution mass spectra were recorded on
a mass spectrometer with a time-of-flight (TOF) mass analyzer.
[(2R,3R)-2-(Benzyloxymethyl)-1-tritylaziridin-2-yl]deoxa-
stannatrane (8). n-BuLi (0.48 M in hexanes, 2.25 mL, 1.07 mmol)
was added dropwise to the solution of (2S,3R)-2-(benzyloxymethyl)-3-
(tributylstannyl)-1-tritylaziridine1a (0.68 g, 0.98 mmol) in dry THF
(15 mL) at −60 °C. This reaction mixture was slowly warmed to −30
°C over a period of 30 min (an initial clear solution becomes a red
solution). At this time, the reaction mixture was cooled to −78 °C, and
a solution of chlorodeoxastannatrane 79 (0.32 g, 1.04 mmol) in THF
(5 mL) was added to it via cannula. After 10 min of stirring at −78 °C
the acetone−dry ice bath was removed, and the solution warmed to
room temperature. After stirring at rt for 1 h, the reaction mixture was
concentrated, and the residue obtained was dissolved in ethyl acetate
(50 mL). This solution was washed with water, and the organic layer
was dried (MgSO4) and concentrated to give 0.95 g (96%) of viscous
material containing an equimolar mixture of tetrabutyltin and
aziridinyl stannatrane 8 according to NMR assay. For characterization
purposes, a small amount of crude product was purified by analytical
TLC on neutral alumina, 5% ethyl acetate in hexanes, Rf = 0.8 and 0.5
(2R,3R)-2-(Benzyloxymethyl)-1-trityl-3-vinylaziridine (13)
and Dimeric Byproduct 14. Conditions A with 8 (0.100 g, 0.099
mmol) and vinyl bromide (1.0 M solution in THF, 0.066 mL, 0.066
mmol); silica gel chromatography afforded 0.011 g (40%) of 13 and
0.011 g (40%) of 14 as a white solid. Analytical TLC, 5% ethyl acetate
in hexanes, Rf = 0.6 and 0.3 for compounds 131a and 14 respectively;
(2R,5S,E)-1,6-bis(benzyloxy)-N2,N5-ditritylhex-3-ene-2,5-diamine
for tetrabutyltin and aziridinyl stannatrane 8, respectively. [α]20
=
D
(14), mp 132−134 °C; [α]20 = −97.13 (c 1.1, CHCl3); HRMS-ES+
−16.4 (c 1.6, CHCl3); HRMS-ES+ (m/z) [M + H] calcd for
C38H45N2OSn, 665.2554, found 665.2550; IR (film, cm−1) 2925, 1490;
1H NMR (400 MHz, CDCl3, ppm) δ 7.53 (bd, J = 7.5 Hz, 6H), 7.38−
7.19 (m, 14H), 4.50 (s, 2H), 3.76 (dd, J = 9.5, 5.5 Hz, 1H), 3.46 (dd, J
= 9.7, 5.7 Hz, 1H), 2.49−2.36 (m, 6H), 1.75−1.63 (m, 6H), 1.45−1.41
(m, 1H), 0.86−0.82 (m, 3H), 0.71−0.67 (m, 3H), 0.63 (d, J = 7.0 Hz,
1H); 13C NMR (125.7 MHz, CDCl3, ppm) δ 145.0, 138.5, 129.7,
D
(m/z) [M + H] calcd for C58H55N2O2, 811.4264, found 811.4254; IR
1
(neat, cm−1) 3332, 3028, 2925, 1492, 1448; H NMR (500 MHz,
CDCl3, ppm) δ 7.54 (bd, J = 7.5 Hz, 10H), 7.29−7.13 (m, 30 H), 5.53
(dd, J = 3.5, 1.5 Hz, 2H), 4.17, 4.13 (ABq, J = 12.0 Hz, 4H), 3.16−3.12
(m, 2H), 2.89 (dd, J = 9.0, 5.0 Hz, 2H), 2.35 (dd, J = 9.5, 5.5 Hz, 2H),
2.19 (bs, exchangeable, 2H); 13C NMR (125.7 MHz, CDCl3, ppm) δ
D
dx.doi.org/10.1021/jo4005052 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
147.0, 138.5, 132.0, 128.9, 128.2, 127.6, 127.5, 127.4, 126.1, 73.2, 72.6,
71.1, 54.0; sample contaminated with benzene (128.3).
g, 0.054 mmol), and Pd(tBu3P)2 (0.001 g, 0.002 mmol) inside the
glovebox. This flask was connected to a reflux condenser and flushed
with a stream of N2 for 5 min. Anhydrous DMF (4.0 mL) was added
to above mixture via syringe, and the resultant clear solution was
heated at 80 °C in an oil bath for 16 h. At this time, the reaction
mixture was cooled to rt and diluted with ethyl acetate (30 mL) and
water (20 mL). The layers were separated, and the organic phase was
washed with water (3 × 10 mL), dried (MgSO4), and concentrated.
Silica gel chromatography (20% ethyl acetate in hexanes) afforded
0.007 g (85%) of 20 as a solid and 0.019 g (90%) of 9 as clear oil.
Analytical TLC, 20% ethyl acetate in hexanes, Rf = 0.8, 0.6 for
compounds 9 and 20, respectively. Product 20 was identified by NMR
comparison with reported data.20 (S)-2-[(Benzyloxy)methyl]-1-
Preparation of 13 Using Scheme 2 Conditions B. In a 25 mL,
oven-dried, single-neck round bottomed flask equipped with a
magnetic stirring bar and a rubber septum were placed aziridinyl
stannatrane 8 (0.070 g, 0.069 mmol), vinyl bromide (1.0 M solution in
THF, 0.14 mL, 0.14 mmol), CuI (0.001 g, 0.007 mmol), CsF (0.021 g,
0.14 mmol), and Pd(tBu3P)2 (0.002 g, 0.003 mmol) inside the
glovebox. This mixture was flushed with a stream of N2 and dissolved
in dry DMF (5.0 mL). The resultant solution was stirred at 45 °C in
an oil bath with monitoring for completion of reaction (ca. 16 h). The
reaction mixture was worked up as describe for conditions A, and silica
gel chromatography (10% diethyl ether in hexanes) afforded 0.026 g
(86%) of 13.1a
Methyl 1-(tert-Butoxycarbonyl)-2-[(2S,3R)-3-(benzyloxy-
methyl)-1-tritylaziridine-2-yl]-indole-3-carboxylate (16) and
Dimeric Indole 17. Conditions A were used with aziridinyl
stannatrane 8 (0.050 g, 0.050 mmol) and 2-iodoindole 151a (0.013
g, 0.033 mmol). Silica gel chromatography (10% ethyl acetate in
hexanes) afforded 0.008 g (40%) of coupled product 161a as an oil and
0.004 g (50%) of dimer 17 as a white solid, analytical TLC, 10% ethyl
acetate in hexanes, Rf = 0.5 and 0.3 for compounds 16 and 17,
respectively; 1,1′-di-tert-butyl 3,3′-dimethyl 1H,1′H-[2,2′-biindole]-
1,1′,3,3′-tetracarboxylate (17): mp 182−184 °C; HRMS-ES+ (m/z)
[M + H] calcd for C30H33N2O8, 549.2237, found 549.2226; IR (film,
cm−1) 1741, 1711; 1H NMR (400 MHz, CDCl3, ppm) δ 8.33 (bd, J =
7.5 Hz, 2H), 8.20 (bd, J = 7.6 Hz, 2H), 7.44−7.35 (m, 4H), 3.65 (s,
6H), 1.17 (s, 18H); 13C NMR (125.7 MHz, CDCl3, ppm) δ 164.2,
148.9, 136.1, 135.7, 126.7, 125.5, 123.9, 121.7, 115.4, 112.5, 84.3, 51.5,
27.4.
tritylaziridine (9): [α]20 = −31.95 (c 1.1, CHCl3); HRMS-ES+ (m/
D
z) [M + H] calcd for C29H28NO, 406.2171, found 406.2167; IR (film,
1
cm−1) 3020, 1580, 1495; H NMR (500 MHz, CDCl3, ppm) δ 7.36
(bd, J = 7.5 Hz, 5H), 7.34−7.20 (m, 15H), 4.53 (s, 2H), 3.86 (dd, J =
10.0, 5.0 Hz, 1H), 3.54 (dd, J = 10.0, 6.0 Hz, 1H), 1.74 (bd, J = 3.0 Hz,
1H), 1.58−1.54 (m, 1H), 1.20 (d, J = 6.0 Hz, 1H); 13C NMR (125.7
MHz, CDCl3, ppm) δ 144.5, 138.3, 129.5, 128.3, 127.7, 127.5, 127.4,
126.6, 73.7, 73.1, 72.9, 31.9, 25.7.
Methyl 2-Vinyl-1H-indole-3-carboxylate (21); Conventional
Stille Coupling from 18. In a 25 mL, oven-dried, single-neck round
bottomed flask equipped with a magnetic stirring bar and a rubber
septum were placed 2-iodoindole 1819 (0.015 g, 0.050 mmol),
tributyl(vinyl)tin (0.032 g, 0.100 mmol), tetrakis(triphenylphosphine)
palladium (0.004 g, 0.003 mmol), and CuDPP (0.014 g, 0.050 mmol)
inside the glovebox. This flask was connected to a reflux condenser
and flushed with a stream of N2 for 5 min. Anhydrous acetonitrile (5.0
mL) was added to above mixture via syringe, and the resultant green
solution was heated at 70 °C in an oil bath for 6 h. The reaction
mixture was cooled to rt and filtered over a small pad of Celite, and the
filtrate was concentrated to give a green residue. Silica gel
chromatography (20% ethyl acetate in hexanes) afforded 0.009 g
(90%) of 21 as clear oil, identified by NMR comparison with the
reported data.21
Control Experiment; Indole 15 Dimerization. In a 25 mL,
oven-dried, single-neck round bottomed flask equipped with a
magnetic stirring bar and a rubber septum were placed 2-iodoindole
15 (0.057 g, 0.142 mmol) and CuDPP (0.040 g, 0.142 mmol) in dry
DMF (4.0 mL) under nitrogen, and the mixture was stirred at rt for 16
h. Workup and chromatography as usual afforded 0.016 g (41%) of 17
as white solid and 0.031 g (55%) of unreacted 2-iodoindole 15;
analytical TLC, 10% ethyl acetate in hexanes, Rf = 0.5 and 0.4 for 15
and 17, respectively.
Bis(triphenylphosphine)(methyl 1-H indole-3-carboxylate-2-
yl)palladium(II)iodide (22). In a 25 mL, oven-dried, single-neck
round bottomed flask equipped with a magnetic stirring bar and a
rubber septum were placed 2-iodoindole 18 (0.010 g, 0.033 mmol)
and tetrakis(triphenylphosphine)palladium (0.044 g, 0.038 mmol)
inside the glovebox. This flask was connected to a reflux condenser
and flushed with a stream of N2 for 5 min. Anhydrous acetonitrile (5.0
mL) was added to above mixture via syringe, and the resultant green
solution was heated at 70 °C in an oil bath for 6 h. The suspension was
allowed to cool to rt and removal (aspirator) of solvent gave a green
solid. Chromatography (25% ethyl acetate in hexanes) gave 0.030 g
(97%) of 22 as a bright yellow solid; analytical TLC, 25% ethyl acetate
in hexanes, Rf = 0.4; mp >200 °C (decomposition); HRMS-ES+ (m/z)
Methyl 2-((2S,3R)-3-((Benzyloxy)methyl)-1-tritylaziridin-2-
yl)-1H-indole-3-carboxylate (19) [Conditions C]. In a 25 mL,
oven-dried, single-neck round bottomed flask equipped with a
magnetic stirring bar and a rubber septum were placed 2-iodoindole
1819 (0.012 g, 0.040 mmol), aziridinyl stannatrane 8 (0.100 g, 0.100
mmol), and Pd(tBu3P)2 (0.001 g, 0.002 mmol) inside the glovebox.
This flask was connected to a reflux condenser and flushed with a
stream of N2 for 5 min. Anhydrous toluene (5.0 mL) was added to the
above mixture via syringe, and the resultant brown solution was heated
at 60 °C in an oil bath for 16 h. The reaction mixture was cooled to rt
and filtered over a small pad of Celite, and the filtrate was
concentrated to give a dark residue. Silica gel chromatography (20%
ethyl acetate in hexanes) afforded 0.007 g (30%) of 19 as clear oil.
1
[M − I] calcd for C46H38NO2P2Pd, 804.1413, found 804.1419; H
NMR (500 MHz, CDCl3, ppm) δ 7.58−7.53 (m, 12H), 7.47 (bs, 1H),
7.44 (bd, J = 8.0 Hz, 1H), 7.28−7.21 (m, 6H), 7.17−7.13 (m, 12H),
6.80 (ddd, J = 8.0, 8.4, 0.8 Hz, 1H), 6.70 (ddd, J = 7.5, 8.0, 0.8 Hz,
1H), 6.60 (bd, J = 7.7 Hz, 1H), 3.72 (s, 3H); 13C NMR (125.7 MHz,
CDCl3, ppm) δ 166.5, 139.5, 134.6 (t, 2JC−P = 6.5 Hz), 131.4 (t, 2JC−P
= 23.9 Hz), 130.2, 130.1, 129.6, 127.7 (t, 2JC−P = 5.3 Hz), 119.1, 118.7,
111.1, 108.3, 50.1; 31P NMR (161.7 MHz, CDCl3, ppm) δ 20.3.
3-[(2R,3R)-3-((Benzyloxy)methyl)-1-tritylaziridin-2-yl]phenol
(24). Following conditions B, aziridinyl stannatrane 8 (0.068 g, 0.067
mmol), tert-butyl(3-iodophenoxy)dimethylsilane 2322 (0.015 g, 0.045
mmol), CuI (0.001 g, 0.004 mmol), CsF (0.014 g, 0.089 mmol), and
Pd(tBu3P)2 (0.001 g, 0.002 mmol) were combined and dissolved in
dry DMF (4 mL). After 16 h at 45 °C, the standard workup and
chromatography (12% ethyl acetate in hexanes) afforded 0.005 g
(20%) of 14 as a white solid and 0.014 g (63%) of 24 as clear oil;
analytical TLC, 10% ethyl acetate in hexanes, Rf = 0.5, 0.2 for 14 and
Analytical TLC, 20% ethyl acetate in hexanes, Rf = 0.6; [α]20
=
D
−89.43 (c 0.6, CHCl3); HRMS-ES+ (m/z) [M + H] calcd for
C39H35N2O3, 579.2648, found 579.2639; IR (neat, cm−1) 3332, 3035,
1
2918, 1693, 1669, 1450; H NMR (400 MHz, CDCl3, ppm) δ 9.46
(bs, exchangeable, 1H), 8.17−8.12 (m, 1H), 7.50−7.47 (m, 6H),
7.29−7.16 (m, 15H), 7.12−7.06 (m, 2H), 4.37, 4.31 (ABq, JAB = 11.9
Hz, 2H), 3.77 (dd, J = 10.5, 3.8 Hz, 1H), 3.76 (s, 3H), 3.62 (d, J = 6.5
Hz, 1H), 3.56 (dd, J = 10.9, 5.9 Hz, 1H), 2.13 (ddd, J = 8.0, 6.2, 4.1
Hz, 1H); 13C NMR (125.7 MHz, CDCl3, ppm) δ 165.8, 143.6, 143.2,
137.5, 134.3, 129.5, 128.4, 128.3, 127.7, 127.5, 127.1, 122.7, 121.7,
121.3, 111.1, 106.5, 75.4, 72.9, 67.7, 50.7, 40.1, 33.5; one carbon signal
is not resolved between 130 −125 ppm due to overlapping chemical
shifts.
Attempted Coupling of 18 with 8; Methyl 2-n-Butyl-1H-
indole-3-carboxylate (20) and Protodestannylated Aziridine 9.
In a 25 mL, oven-dried, single-neck round bottomed flask equipped
with a magnetic stirring bar and a rubber septum were placed 2-
iodoindole 1819 (0.011 g, 0.037 mmol), aziridinyl stannatrane 8 (0.054
24, respectively. Data for 24: [α]20 = −53.09 (c 0.61, CHCl3);
D
HRMS-ES+ (m/z) [M + H] calcd for C35H32NO2, 498.2428, found
1
498.2416; IR (neat, cm−1) 3410, 2950, 1610, 1445; H NMR (400
MHz, CDCl3, ppm) δ 7.52−7.47 (m, 6H), 7.28−7.19 (m, 14H), 7.02
E
dx.doi.org/10.1021/jo4005052 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
(bd, J = 7.9 Hz, 1H), 6.97−6.91 (m, 2H), 6.74 (ddd, J = 8.2, 2.7, 0.9
Hz, 1H), 4.80 (bs, exchangeable, 1H), 4.20, 4.15 (ABq, J = 12.1 Hz,
2H), 3.84 (dd, J = 10.3, 4.3 Hz, 1H), 3.41 (dd, J = 10.2, 7.5 Hz, 1H),
2.43 (d, J = 6.4 Hz, 1H), 1.87 (ddd, J = 7.8, 6.5, 4.8 Hz, 1H); 13C
NMR (125.7 MHz, CDCl3, ppm) δ 155.4, 144.1, 139.2, 138.1, 129.6,
129.3, 128.2, 127.5, 127.4, 127.3, 126.7, 120.6, 114.6, 113.7, 75.4, 72.7,
67.2, 38.8, 38.5.
Brierley, C. A. J.; Ward, J. G. J. Org. Chem. 2007, 72, 10009. (c) For
reviews covering aziridine metallation, see: Florio, S.; Luisi, R. Chem.
Rev. 2010, 110, 5128. Florio, S. Synthesis 2012, 44, 2872.
(12) Littke, A. F.; Schwarz, L.; Fu, G. C. J. Am. Chem. Soc. 2002, 124,
6343.
(13) Farina, V.; Krishnan, B. J. Am. Chem. Soc. 1991, 113, 9585.
(14) For the formation of dimeric indoles via organocopper
intermediates, see: Bergman, J.; Eklund, L. Tetrahedron 1980, 36, 1439.
(15) Gabriele, B.; Veltri, L.; Mancuso, R.; Salerno, G.; Costa, M. Eur.
J. Org. Chem. 2012, 13, 2549.
(16) Using 1.5 equiv of 8 with 18 gave a 2:1 mixture of 19:20, while a
larger excess (2.5 equiv) of 8 afforded an improved ratio of 6.7:1 along
with recovered 8 and deiodo indole (18 with I replaced by H). Both
experiments also produced 9, but the relative amount could not be
established due to overlapping NMR signals. This result suggests that
the NH group of 18 is involved in the destruction of an equivalent
amount of 8.
(2R,3R)-2-((Benzyloxy)methyl)-3-(3-((tert-butyldimethylsilyl)-
oxy)phenyl)-1-tritylaziridine (25). According to conditions C,
aziridinyl stannatrane 8 (0.068 g, 0.067 mmol), tert-butyl(3-
iodophenoxy)dimethylsilane 2322 (0.015 g, 0.045 mmol), and
Pd(tBu3P)2 (0.001 g, 0.002 mmol) were combined in anhydrous
toluene (4 mL). The resultant solution was stirred at 65 °C in an oil
bath for 16 h, and the standard workup gave a dark residue.
Chromatography (neutral alumina plates, 5% ethyl acetate in hexanes)
afforded 0.017 g (63%) of 25 as clear oil and 0.016 g (35%) of
unreacted aziridinyl stannatrane 8. Analytical TLC (neutral alumina
plate, 10% diethyl ether in hexanes, Rf = 0.7 and 0.4 for 8 and 25. Data
(17) The structure of 22 was established by X-ray crystallography
(see Supporting Information).
for 25, [α]20 = −56.81 (c 0.7, CHCl3); HRMS-ES+ (m/z) [M + H]
D
calcd for C41H46NO2Si, 612.3292, found 612.3294; IR (neat, cm−1)
(18) (a) For an interesting application in a β-lactam environment,
see: Jensen, M. S.; Yang, C.; Hsiao, Y.; Rivera, N.; Wells, K. M.;
Chung, J. Y. L.; Yasuda, N.; Hughes, D. L.; Reider, P. J. Org. Lett. 2000,
2, 1081. (b) For a systematic study of palladium-catalyzed coupling of
sec-alkyl substituted stannatranes, see: Li, L.; Wang, C.-Y.; Huang, R.;
Biscoe, M. R.; Nature Chem. 2013, 5, in press.
(19) Beccalli, E. M.; Broggini, G.; Marchesini, A.; Rossi, E.
Tetrahedron 2002, 58, 6673.
(20) Gabriele, B.; Veltri, L.; Mancuso, R.; Salerno, G.; Costa, M. E. J.
Org. Chem. 2012, 13, 2549.
(21) Laronze, M.; Laronze, J.-Y.; Nemes, C.; Sapi, J. Eur. J. Org.
Chem. 1999, 2285.
1
3050, 2920, 1615, 1455; H NMR (400 MHz, CDCl3, ppm) δ 7.53−
7.50 (m, 6H), 7.27−7.17 (m, 14H), 7.10−7.00 (m, 2H), 6.94 (dd, J =
7.1, 2.1 Hz, 1H), 6.76 (ddd, J = 8.1, 2.4, 1.1 Hz, 1H), 4.17, 4.13 (ABq,
J = 12.5 Hz, 2H), 3.85 (dd, J = 10.3, 4.3 Hz, 1H), 3.41 (dd, J = 10.2,
7.7 Hz, 1H), 2.43 (d, J = 6.4 Hz, 1H), 1.86 (ddd, J = 7.6, 6.4, 4.6 Hz,
1H), 1.00 (s, 9H), 0.21 (s, 6H); 13C NMR (125.7 MHz, CDCl3, ppm)
δ 155.5, 144.2, 138.8, 138.1, 129.6, 129.0, 128.2, 127.5, 127.3, 127.2,
126.7, 121.3, 119.3, 118.6, 75.4, 72.7, 67.3, 38.9, 38.4, 25.7, 18.2, −4.3.
ASSOCIATED CONTENT
■
S
* Supporting Information
(22) Pelter, A.; Hussain, A.; Smith, G.; Ward, R. S. Tetrahedron 1997,
53, 3879.
Copies of 1H NMR and 13C NMR spectra for all new
substances and X-ray crystallographic data tables for 22 in CIF
format. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: edved@umich.edu.
■
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) (a) Nelson, J. M.; Vedejs, E. Org. Lett. 2010, 12, 5085.
(b) Hughes, M.; Boultwood, T.; Zepetelli, G.; Bull, J. A. J. Org. Chem.
2013, 78, 844.
(2) Fu, G. C. Acc. Chem. Res. 2008, 41, 1555.
(3) (a) Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed. 1999, 38, 2411.
(b) Grasa, G. A.; Nolan, S. P. Org. Lett. 2001, 3, 119.
(4) Mee, S. P. H.; Lee, V.; Baldwin, J. E. Angew. Chem., Int. Ed. 2004,
43, 1132.
(5) (a) Catalyst generation: Fors, B. P.; Krattiger, P.; Strieter, E.;
Buchwald, S. L. Org. Lett. 2008, 10, 3505. (b) Li, H.; He, A.; Falck, J.
R.; Liebeskind, L. S. Org. Lett. 2011, 13, 3682.
(6) (a) Ye, J.; Bhatt, R. K.; Falck, J. R. J. Am. Chem. Soc. 1994, 116, 1.
(b) Kells, K. W.; Chong, J. M. J. Am. Chem. Soc. 2004, 126, 15666.
(7) Allred, G. D.; Liebeskind, L. S. J. Am. Chem. Soc. 1996, 118, 2748.
(8) (a) Wittenberg, R.; Srogl, J.; Egi, M.; Liebeskind, L. S. Org. Lett.
2003, 5, 3033. (b) Allred, G. D., Ph. D. Dissertation, Emory
University, 1997. (c) For the use of CuDPP as a tin scavenger, see:
Srogl, J.; Allred, G. D.; Liebeskind, L. S. J. Am. Chem. Soc. 1997, 119,
12376.
(9) Vedejs, E.; Haight, A. R.; Moss, W. O. J. Am. Chem. Soc. 1992,
114, 6556.
(10) Vedejs, E.; Moss, W. O. J. Am. Chem. Soc. 1993, 115, 1607.
(11) (a) Hodgson, D. M.; Humphreys, P. G.; Ward, J. G. Org. Lett.
2005, 7, 1153. (b) Hodgson, D. M.; Humphreys, P. G.; Miles, S. M.;
F
dx.doi.org/10.1021/jo4005052 | J. Org. Chem. XXXX, XXX, XXX−XXX