Investigation of functionalized α-chloroalkyllithiums for a stereospecific reagent-controlled homologation approach to the analgesic alkaloid (-)-epibatidine

Article abstract of DOI:10.1002/chem.201302511

Four putative functionalized α-chloroakyllithiums RCH ₂CHLiCl, where R=CHCH₂ (18 a), CCH (18 b), CH ₂OBn (18 c), and CH[O(CH₂)₂O] (18 d), were generated in situ by sulfoxide-lithium exchange from α-chlorosulfoxides, and investigated for the stereospecific reagent-controlled homologation (StReCH) of phenethyl and 2-chloropyrid-5-yl (17) pinacol boronic esters. Deuterium labeling experiments revealed that α-chloroalkyllithiums are quenched by proton transfer from their α-chlorosulfoxide precursors and it was established that this effect compromises the yield of StReCH reactions. Use of α-deuterated α-chlorosulfoxides was discovered to ameliorate the problem by retarding the rate of acid-base chemistry between the carbenoid and its precursor. Carbenoids 18 a and 18 b showed poor StReCH efficacy, particularly the propargyl group bearing carbenoid 18 b, the instability of which was attributed to a facile 1,2-hydride shift. By contrast, 18 d, a carbenoid that benefits from a stabilizing interaction between O and Li atoms gave good StReCH yields. Boronate 17 was chain extended by carbenoids 18 a, 18 b, and 18 d in 16, 0, and 68 % yield, respectively; α-deuterated isotopomers D-18 a and D-18 d gave yields of 33 and 79 % for the same reaction. Double StReCH of 17 was pursued to target contiguous stereodiads appropriate for the total synthesis of (-)-epibatidine (15). One-pot double StReCH of boronate 17 by two exposures to (S)-D-18 a (≤66 % ee), followed by work-up with KOOH, gave the expected stereodiad product in 16 % yield (d.r.～67:33). The comparable reaction using two exposures to (S)-D-18 d (≤90 % ee) delivered the expected bisacetal containing stereodiad (R,R)-DD-48 in 40 % yield (≥98 % ee, d.r.=85:15). Double StReCH of 17 using (S)-D-18 d (≤90 % ee) followed by (R)-D-18 d (≤90 % ee) likewise gave (R,S)-DD-48 in 49 % yield (≥97 % ee, d.r.=79:21). (R,S)-DD-48 was converted to a dideuterated isotopomer of a synthetic intermediate in Corey's synthesis of 15. Copyright

Full text of DOI:10.1002/chem.201302511

DOI: 10.1002/chem.201302511
Investigation of Functionalized a-Chloroalkyllithiums for a Stereospecific
Reagent-Controlled Homologation Approach to the Analgesic Alkaloid
(ꢀ)-Epibatidine
Christopher R. Emerson, Lev N. Zakharov, and Paul R. Blakemore*[a]
In memory of Professor Robert E. Gawley (1949–2013) a pioneer of chiral carbanion chemistry
Abstract: Four putative functionalized
a-chloroakyllithiums RCH2CHLiCl,
where R=CHCH2 (18a), CCH (18b),
CH2OBn (18c), and CH[O(CH2)2O]
its precursor. Carbenoids 18a and 18b
showed poor StReCH efficacy, particu-
larly the propargyl group bearing car-
benoid 18b, the instability of which
was attributed to a facile 1,2-hydride
shift. By contrast, 18d, a carbenoid
that benefits from a stabilizing interac-
tion between O and Li atoms gave
good StReCH yields. Boronate 17 was
chain extended by carbenoids 18a,
18b, and 18d in 16, 0, and 68% yield,
respectively; a-deuterated isotopomers
D-18a and D-18d gave yields of 33 and
79% for the same reaction. Double
StReCH of 17 was pursued to target
contiguous stereodiads appropriate for
the total synthesis of (ꢀ)-epibatidine
(15). One-pot double StReCH of boro-
nate 17 by two exposures to (S)-D-18a
(ꢁ66%ee), followed by work-up with
KOOH, gave the expected stereodiad
product in 16% yield (d.r.~67:33). The
comparable reaction using two expo-
sures to (S)-D-18d (ꢁ90%ee) deliv-
ered the expected bisacetal containing
stereodiad (R,R)-DD-48 in 40% yield
AHCTUNGTRENNUNG
(18d), were generated in situ by sulfox-
ide–lithium exchange from a-chloro-
sulfoxides, and investigated for the ste-
reospecific reagent-controlled homolo-
gation (StReCH) of phenethyl and 2-
chloropyrid-5-yl (17) pinacol boronic
esters. Deuterium labeling experiments
revealed that a-chloroalkyllithiums are
quenched by proton transfer from their
a-chlorosulfoxide precursors and it was
established that this effect compromis-
es the yield of StReCH reactions. Use
of a-deuterated a-chlorosulfoxides was
discovered to ameliorate the problem
by retarding the rate of acid-base
chemistry between the carbenoid and
(ꢂ98%ee,
d.r.=85:15).
Double
StReCH of 17 using (S)-D-18d (ꢁ
90%ee) followed by (R)-D-18d (ꢁ
90%ee) likewise gave (R,S)-DD-48 in
49% yield (ꢂ97%ee, d.r.=79:21).
(R,S)-DD-48 was converted to a dideu-
terated isotopomer of a synthetic inter-
mediate in Coreyꢀs synthesis of 15.
Keywords: alkaloid synthesis · bor-
onic esters · carbenoids · homologa-
tion · sulfoxide–lithium exchange
Introduction
alization of a process of interest, for example, bimolecular
decomposition pathways like eliminative “dimerization”
(1!2), nucleophilic substitution (1!3), and E2 elimination
(1!4), and unimolecular pathways, such as 1,2-hydride/alkyl
shifts (1!5) and related distal intramolecular s-bond inser-
tion reactions (1!6; Scheme 1).[1b,3] Although the chemical
properties of a-haloalkyllithiums are commonly exploited,
a-Haloalkyllithiums provide the archetype for the carbenoid
concept in being species that display the ambiphilic chemical
reactivity traits characteristic of true carbenes (i.e., divalent
carbon) while retaining a closed valence shell configura-
tion.[1] As first charted in the work of Closs et al.[2] and later
more thoroughly explored in the seminal contributions of
Kçbrich and co-workers,[3] it is now well appreciated that a-
haloalkyllithiums can be utilized in a variety of synthetically
useful transformations (e.g., Darzens chemistry, cyclopropa-
nation) providing that their high reactivity and propensity to
decompose is successfully managed. This last point is impor-
tant given that a wide spectrum of diverting side-reactions
are available to a-haloalkyllithiums that may thwart the re-
[a] Dr. C. R. Emerson, Dr. L. N. Zakharov, Prof. Dr. P. R. Blakemore
Department of Chemistry, Oregon State University
Corvallis, OR 97331 (USA)
E-mail: paul.blakemore@science.oregonstate.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201302511.
Scheme 1. A selection of potentially deleterious reactivity pathways avail-
able to a-haloalkyllithiums (X=Cl, Br).
16342
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16342 – 16356

FULL PAPER
the stereogenicity of the reactive carbon-atom center in
R1R2CLiX species, where R1 ¼R2 (X=halogen), has rarely
been used to any deliberate advantage in asymmetric syn-
thesis. For this attribute to be at all consequential an effi-
cient means to access a-haloalkyllithiums in a stereodefined
form is required and one must have confidence that once
generated the reactive intermediate will possess sufficient
configurational stability for its useful deployment. It was es-
tablished some time ago that a-halocyclopropyllithiums
maintain their stereochemical integrity below ꢀ788C;[4]
however, the configurational stability of acyclic a-haloalkyl-
lithiums has only more recently begun to be probed.[5] Work
from these laboratories,[6] and those of Hammerschmidt,[7]
has established that enantioenriched a-chloroalkyllithiums,
generated through sulfoxide–lithium exchange or tin–lithium
exchange, respectively, exhibit configurational stability[8] on
a timescale that extends beyond the limited temporal
window of their chemical stability at ꢀ788C.
Scheme 3. StReCH of pinacol boronic ester 11 (pin=OACTHUNTRGNE(UNG CMe2)2O) with
putative a-chloroalkyllithiums 13 generated from syn-chlorosulfoxides 12
(Ar=p-Tol or 4-ClC6H4) by sulfoxide–lithium exchange with tBuLi.
Yield and stereofidelity, defined as (%ee StReCH product 14)/(%ee car-
benoid precursor 12), determined after conversion of 14 into the corre-
sponding carbinols by treatment with aq. NaOOH. Data taken from
ref. [6b].
Our interest in a-chloroalkyllithiums was motivated by
ꢀ
a desire to realize a unified approach to carbon carbon
vious manner. As the selected data below illustrate
(Scheme 3),[6b] StReCH products 14 were obtained in gener-
ally good yield where R=Bn or iBu, but a poor outcome
was noted for R=Me and the process failed for R=iPr. Al-
though steric factors perhaps account for the last result, the
fact that substituents of lower steric demand than either Bn
or iBu performed poorly was puzzling. To probe the limits
of this kind of chemistry and to learn more about substitu-
ent effects as they relate to the stability of stereodefined a-
chloroalkyllithiums and the ease of their generation from
sulfoxides, we elected to pursue a target directed synthesis
based on an iterative StReCH strategy. As now relayed, the
vehicle chosen for this effort was the analgesic alkaloid epi-
batidine (15), a polycyclic compound that would require the
use of functionalized a-chloroalkyllithiums to access through
the advocated chain extension approach.
bond formation based on the stereospecific reagent-con-
trolled homologation (StReCH) of boronic esters
(Scheme 2). StReCH is defined as the chain extension of an
Scheme 2. Stereospecific reagent-controlled homologation (StReCH):
M1 =metal, M2 =electofugal group (metal), X=nucleofugal group.
organometallic substrate 7 by a stereodefined chiral carbe-
noid reagent 8 via a two-step process of ate-complex forma-
tion (7!9) and 1,2-metallate rearrangement (9!10).[6a] Iter-
ation allows for an “assembly-line” style synthesis of com-
plex polysubstituted alkyl moieties wherein stereochemical
and constitutional features are directly programmed by the
carbenoid presentation sequence. Since the first demonstra-
tion of StReCH of boronic esters with a-chloroalkylmagne-
sium chlorides,[6a] the concept has been better realized with
other carbenoid types, including a-chloroalkyllithiums,[6] a-
Epibatidine and synthetic plan: (ꢀ)-Epibatidine (15) is
a uniquely structured 7-azabicycloACTHUNTGRNEUNG[2.2.1]heptane possessing
an exo-configured 2-chloropyrid-5-yl unit at C2 (Scheme 4).
This unusual alkaloid was identified as a trace component of
the defensive skin secretions of the Ecuadoran poison frog,
[(diisopropylamino)carbonyloxy]alkyllithiums,[9]
a-[(2,4,6-
triisopropylbenzoyl)oxy]alkyllithiums,[10] oxiranyllithiums,[11]
aziridinyllithiums,[12] and a-lithiated N-Boc amines.[13] Of
these findings, the fact that StReCH can be effected with
species as chemically labile as a-chloroalkyllithiums is argu-
ably the most surprising result.[14,15]
Our earlier work established that simple nonfunctional-
ized stereodefined a-chloroalkyllithiums, generated in situ
by sulfoxide–lithium exchange from scalemic syn-a-chlor-
oalkyl aryl sulfoxides, are sufficiently robust for StReCH of
neopentyl glycol and pinacol boronic esters.[6] Of some note,
efficacy was found to be highly dependent on the nature of
the carbenoid/chlorosulfoxide substituent, but not in any ob-
Scheme 4. (ꢀ)-Epibatidine and an overview of the synthetic plan devised
to prepare it.
Chem. Eur. J. 2013, 19, 16342 – 16356
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16343

P. R. Blakemore et al.
Epipedobates tricolor.[16] So minute was the quantity of 15
initially isolated (<1 mg from 750 frogs) that the structural
elucidation studies of Daly et al. were delayed for 16 years
from the date of the sample collection (in 1976) to await the
emergence of more sensitive NMR spectrometers.[17] Epiba-
tidine elicits powerful analgesia in mammals, an effect first
suggested by a positive mouse “Straub-tail” response that
occurred at concentrations 500-times lower than the re-
quired threshold for morphine.[16a] The analgetic action in-
duced by 15 is not blocked by naloxone, an opiod receptor
antagonist, and it was discovered that this natural product
exerts its effects through activation of nicotinic acetylcholine
receptors (nAChRs).[18] The profound bioactivity of epibati-
dine, which it was once hoped might lead to a new class of
nonaddictive analgesics to supplant opiate-based pain man-
agement regimens, coupled with its accessible structure,
stimulated a flood of synthetic efforts aimed at total synthe-
sis and the generation of analogues for SAR studies. The
first total syntheses appeared a year after the structure was
revealed and epibatidine has now been prepared well over
forty times.[19,20] Although the clinical promise of epibatidine
was ultimately not fulfilled because of acute toxicity linked
to nonselective agonism of nAChR subtypes, this compound
and its analogues have become useful probes for dissecting
the complexities of nAChR biology.[21]
Results and Discussion
Synthesis of a-chlorosulfoxide precursors to functionalized
a-chloroalkyllithium reagents: As in our earlier work,[6] pu-
tative a-chloroalkyllithium reagents of interest were to be
generated through the versatile sulfoxide–lithium exchange
reaction[23,24] from appropriate a-chlorosulfoxide precursors.
These sulfoxides are typically shelf-stable, crystalline materi-
als that are readily accessed from thioethers by a two-step
sequence of sulfoxidation followed by chlorination. For the
purposes of optimization and mechanistic studies, a variety
of stereochemical and isotopomeric forms of a-chlorosulfox-
ides 24 were targeted: racemic, scalemic, syn and anti diaste-
reomers, and a-deuterated analogues. Requisite thioethers
21a,b were obtained from p-thiocresol by treatment with
homoallyl- or homopropargyl bromide in the presence of
K2CO3 in acetone solvent (Scheme 5). The more elaborate
thioethers 21c,d were prepared by conjugate addition of p-
thiocresol to acrolein to yield aldehyde 22 followed by its
reduction and O-alkylation to give 21c or acetalization to
give 21d.
Epibatidine was selected for this study because of its bio-
logical relevance and notoriety and our desire to demon-
strate that a chain extension approach is not limited to the
synthesis of linear acyclic compounds. The alkaloid was en-
visioned to emerge from an acyclic precursor (16) containing
preset stereochemistry at C1 and C2 and functionality at C4
and C5 that would enable the 7-azabicyclo-[2.2.1]heptane
core to be forged through some straightforward annulation
chemistry. The precursor 16 would itself be obtained in
short-order via a pair of sequential StReCH reactions from
pyridyl boronate 17 using functionalized carbenoids 18 of
appropriate configurations to install the correct stereochem-
istry at C1 and C2. The nitrogen atom at C1 would logically
derive from the final site of the boron atom in adduct 19;
here, the choice of a stereoinvertive or a stereoretentive
boronate amination tactic would dictate whether (S)-18 or
(R)-18 should be deployed in the second StReCH cycle. The
successful realization of this plan using stereodefined 1-
chloro-2-(1,3-dioxolan-2-yl)ethyllithium (18d) was commu-
nicated in an earlier report.[22] Herein, we disclose full de-
tails for this effort and describe for the first time our at-
tempts to achieve the same overall aim using three other a-
chloroalkyllithium reagents bearing allyl (18a), propargyl
(18b), and benzyloxyethyl (18c) groups as alternate func-
tionalized substituents.
Scheme 5. Preparation of thioethers 21a,b,c,d from p-thiocresol (20).
Thioethers 21 were advanced to racemic syn-chlorosulfox-
ides (ꢃ)-syn-H-24 by mild chemoselective sulfoxidation with
aq. H2O2 in MeOH,[25] followed by a-chlorination with N-
chlorosuccinimide (Table 1).[26] Two of these compounds
(syn-H-24a,d) were next prepared in scalemic form in both
SS and RS configurations from thioethers by Ellman–Bolm
enantioselective oxidation[27] using the appropriate enantio-
mer of Jacksonꢀs tert-leucinol derived ligand 25
(Scheme 6).[28] Excellent enantiomeric excess was obtained
from this process by extending the reaction time to allow
for a degree of over-oxidation to the sulfone; under this sce-
nario, enantiopurity of the desired product is improved
(albeit at the expense of yield) by a ligand-controlled kinetic
resolution, which selectively consumes the minor sulfoxide
enantiomer by its advancement to the sulfone. The absolute
configuration of sulfoxide (RS)-23d prepared using ligand
(R)-25 was determined by X-ray diffraction (XRD) analysis
16344
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16342 – 16356

Functionalized a-Chloroalkyllithiums
FULL PAPER
Table 1. Preparation of racemic syn-H-chlorosulfoxides 24a,b,c,d.[a]
Entry Thioether
R
Yield 23
Yield 24
syn/anti
[%]
[%]
24
1
2
3
4
21a
21b
21c
21d
CH=CH2
94
98
65
89
92
63
92
84:16
94:06
81:19
87:13
ꢄ
C CH
Scheme 7. A mechanistic hypothesis to account for the stereochemical
outcome of sulfoxide chlorination with N-chlorosuccinimide (X-Cl, X=
CH2OBn
ꢀ
CH[OACTHNUTGRENUNG(CH2)2O] 81
succinimidyl). Limited rotation about the a-C S bond in ylide 28 ac-
counts for the generation of the minor anti diastereoisomer product,
which shares the same absolute configuration on sulfur as the major syn
diastereosiomer product syn-30. Solvent separation of contact-ion pair 29
(or the intervention of a radical pathway for 1,2-electrophilic rearrange-
ment of 28) may account for partial or complete racemization depending
on exact reaction conditions. See refs. [31b,d] for elaboration on these
[a] Treatment of sulfoxides 23 with N-chlorosuccinimide (NCS) typically
also resulted in the formation of minor byproducts p-TolSO2CH2R and p-
TolSOCCl2R.
and related arguments. The usual formal charges by convention indicated
ꢀ
on sulfoxides (R2S+ O ) are omitted for clarity.
ꢀ
nation can be loosely understood according to the frame-
work advanced by Montanari and Klein (Scheme 7).[31b,d] Ini-
tial chlorination at sulfur is followed by deprotonation lead-
ing to an ylide (28) that suffers 1,2-electrophilic rearrange-
ment by a fragmentation/recombination process (possibly
via the illustrated contact ion pair 29) leading to the ob-
served syn-chlorosulfoxide 30 with the net inversion of con-
figuration at sulfur a consequence of rehybridization during
addition of chloride anion to thioxocarbenium ion 29. Signif-
icant primary kinetic isotope effects observed during sulfox-
ide chlorination with PhICl2 indicate that a-CH bond break-
ing can be rate limiting.[32] We observed a practical implica-
tion of this fact during the synthesis of (ꢃ)-syn-D-24b from
a,a-dideuterosulfoxide DD-23b [Eq. (1)]. In this case, treat-
ment of the racemic sulfoxide with NCS in CH2Cl2 at RT re-
sulted in less than 5% conversion to chlorosulfoxide after
20 h and heating was required to obtain (ꢃ)-syn-D-24b. By
contrast, the non-deuterated starting material 23b gave
a 92% yield of (ꢃ)-syn-H-24b in just 4 h under otherwise
identical reaction conditions at RT (Table 1, entry 2).
Scheme 6. Preparation of scalemic (SS)-syn-H-chlorosulfoxides 24ad by
enantioselective sulfoxidation using Jacksonꢀs ligand (S)-25 followed by
non-racemizing a-chlorination with N-chlorosuccinimide and K2CO3.
Enantiomeric products (RS)-syn-H-24a,d were prepared in an identical
fashion from (RS)-23a,d obtained using ligand (R)-25 in the oxidation
step (not illustrated). Chlorosulfoxide products were recrystallized to im-
prove isomeric purity (d.r. after recrystallization indicated in parenthe-
ses).
using the resonant scattering technique;[29] the result was in
accord with Jacksonꢀs precedent[28] for the stereochemical
outcome of this highly efficient process. Further conversion
to scalemic chlorosulfoxides syn-H-24 was achieved by the
nonracemizing chlorination protocol popularized by Yama-
kawa.[30,31] This transformation is notable from two stand-
points: 1) the presence of (insoluble) powdered K2CO3 in
the CH2Cl2 reaction medium dramatically slows down
chlorination (5–10 days for full conversion vs. 1–2 h without
K2CO3) but serves the useful purpose of suppressing racemi-
zation and, 2) chlorination proceeds with stereochemical in-
version at sulfur and preferentially generates the syn diaste-
reoisomer. Again, stereochemical outcome was confirmed to
be in agreement with literature precedent by XRD analy-
sis;[30] crystal structures were obtained for (RS)-syn-H-24a
and (SS)-syn-H-24d derived from (RS)-23a and (SS)-23d, re-
spectively.[29] The stereochemical course of sulfoxide chlori-
In StReCH reactions involving the generation of a-
chloroACHTUNGTRNEUNG
alkyllithiums from syn-a-chlorosulfoxides,[6] it is typi-
cally observed that any unconsumed chlorosulfoxide carbe-
noid precursor is returned in an epimerized form, with the
anti diastereoisomer predominating. The cause of this epi-
merization is undesired a-metallation followed by protonol-
ysis on work-up. Given that anti-a-chlorosulfoxides had not
hitherto been evaluated as precursors to chloroalkyllithiums,
we studied base-mediated epimerization of chlorosulfoxides
Chem. Eur. J. 2013, 19, 16342 – 16356
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16345

P. R. Blakemore et al.
Table 2. Synthesis of anti-D-chlorosulfoxides 24a,c,d by a-metallation of
syn-H-chlorosulfoxides 24a,c,d followed by low temperature deutero-
ACHTUNGTRENNUNG
lysis.[a]
Entry
Chlorosulfoxide
anti-D-24
syn-H-24
R
base
yield [%] anti/syn
1
2
3
4
5
6
7
8
9
24a
24a
24a
24b
24c
24d
24d
24d
24d
CH=CH2
CH=CH2
CH=CH2
LDA
NaHMDS 80
KHMDS
LDA
LDA
80
89:11
81:19
50:50
n.a.
77:23
95:05
93:07
86:14
67:33
90
C CH
0[b]
ꢄ
CH2OBn
98
CH[O
CH[O
CH[O
CH[O
N
87[c]
83[c]
82[d]
Figure 1. Capped stick representations of single-crystal X-ray diffraction
[a] Products were obtained with incomplete deuterium incorporation
(83–90%D). [b] Enyne (E)-H-42 generated in >95% yield. [c] Product
ee 13–18% lower than starting material (as determined by HPLC analy-
sis). [d] Product ee only 2–3% lower than starting material.
ꢀ
structures for four a-chlorosulfoxides viewed down the S C(a) bond. For
ꢀ
every compound, a gauche relationship is evident between C(a) Cl and
[29]
ꢀ
ꢀ
ꢀ
S O bonds and C(a) H is anti to S O.
portance in controlling ground-state conformation for these
syn-H-24 for the deliberate production of compounds anti-
H-24 and anti-D-24 (Table 2). Substrates syn-H-24a,c,d were
converted to anti deuterated congeners with good-to-excel-
lent diastereoselectivity by exposure to LDA followed short-
ly thereafter by low temperature deuterolysis with
[D4]methanol. By contrast, identical treatment of homopro-
pargyl sulfoxide syn-H-24b resulted exclusively in elimina-
tion and gave (E)-enyne H-42 in >95% yield (E/Z=96:4).
Upon further study, it was discovered that partial racemiza-
tion (ꢁ15%) can accompany the LDA (or LiHMDS)-medi-
ated epimerization of syn-H-24d.[33] This unwelcome phe-
nomenon was avoided by using either NaHMDS or
KHMDS instead of a lithium amide base. Of these two
bases, the sodium amide provided superior diastereoselectiv-
ity and it should be regarded as optimal for the synthesis of
nonracemic anti chlorosulfoxides. Complete deuterium in-
corporation could not be realized (typical %D was ca.
88%), and nondeuterated isotopomers anti-H-24 (not illus-
trated) were obtained similarly by substituting CH3OH for
CD3OD in the low temperature protonolysis step. XRD
analysis confirmed the absolute configuration of (SS)-anti-D-
24d generated using the NaHMDS/CD3OD method from
(SS)-syn-H-24d.[29]
a-chlorosulfoxides.
Studies with 1-chloro-3-butenyllithium (18a): Use of 1-
chloro-3-butenyllithium (18a) as a reagent for StReCH was
approached with some trepidation given that the allylic CH
bonds in this carbenoid invite the possibility of facile de-
composition through a 1,2-hydride shift mechanism (i.e.,
1 to 5, Scheme 1). Nonetheless, the earlier finding that 1-
chloro-2-phenylethyllithium (13, R=Bn), a carbenoid with
benzylic CH bonds adjacent to the halogen, is a competent
reagent for chain extension (Scheme 3)[6] suggested that ex-
ploration of 18a could be a worthwhile exercise. Elabora-
tion of (ꢀ)-epibatidine was envisioned to proceed by double
StReCH of commercially available boronic ester 17 with
(S)-18a to yield diene 31. A short sequence incorporating
stereoretentive amination of the C B bond and ring-clos-
ing olefin metathesis from 31 would converge on cyclohex-
ene 32, an intermediate previously converted to the alkaloid
of interest by Corey and co-workers in four steps
(Scheme 8).[19b]
[35]
ꢀ
X-ray crystal structures were obtained for four distinct a-
chlorosulfoxides during the course of the above studies:
(RS)-syn-H-24a, (SS)-syn-H-24d, (SS)-anti-D-24d, and (ꢃ)-
anti-H-24d (Figure 1).[29] Of note, a gauche relationship is
ꢀ
ꢀ
evident between C(a) Cl and S O bonds in all of these
structures. Such a conformational preference is likely sup-
ported by a sCH!s*SO hyperconjugative interaction that
compensates for the gauche disposition of aryl and alkyl
moieties in anti diastereoisomers.[34] In syn diastereoisomers,
ꢀ
ꢀ
the C(a) Cl/S O gauche preference is reinforced by a nS!
s*CCl interaction.[34d] Given its varied orientation in relation
to proximal donor orbitals, s*SC appears to be of minor im-
Scheme 8. Envisioned approach to Coreyꢀs late stage (ꢀ)-epibatidine in-
termediate 32 by iterative StReCH using 1-chloro-3-butenyl-lithium
(18a).
16346
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16342 – 16356

Functionalized a-Chloroalkyllithiums
FULL PAPER
StReCH of pyridyl boronate 17 was initially evaluated
with the putative organolithium 18a generated from chloro-
sulfoxide syn-H-24a under the Barbier-type reaction condi-
tions previously regarded as optimal, that is, tBuLi as initia-
tor and PhMe as solvent (Table 3, entries 1–3). The desired
effect (entry 7), but introduction of ZnCl2 following ate-
complex formation at ꢀ788C gave an unexpected result
(entry 8). In this case, the yield of 33 was no higher than
before but it was now easily obtained in a pure form be-
cause boronate 17 was totally consumed by some new pro-
cess. Examination of the crude
product mixture revealed the
Table 3. Chain extension of boronic ester 17 with 1-chloro-3-butenyllithium (18a) generated via sulfoxide–lith-
ium exchange from chlorosulfoxides 24a.
presence of 2-chloro-5-(chloro-
methyl)pyridine, compound
a
not encountered in our earlier
experiments and which ac-
counted for a majority of the
missing mass balance of 17 in
this example. An independent
synthesis of 2-chloro-5-(chloro-
methyl)pyridine from methyl 2-
chloronicotinate confirmed the
nature of the unlikely byprod-
uct. The mechanistic origin of
this material remains unclear
but it was established that its
formation from 17 requires all
three of 24a, tBuLi, and ZnCl2.
Cognizant of the hazards as-
sociated with the use of pyro-
phoric materials, such as tBuLi,
an alternative organolithium in-
itiator for StReCH chemistry
was sought. Fortunately, it was
discovered that a nonpyrophoric
Entry Form of
Stoichiometry
Temp. range
Yield [%]
chlorosulf. R M
solvent 24a [equiv] R M [equiv] additive T1 [8C] T2
33[a]
34
ꢀ
ꢀ
1
2
3
4
5
6
7
8
(RS)-syn-H tBu-Li
(RS)-syn-H tBu-Li
(RS)-syn-H tBu-Li
(RS)-syn-H tBu-Li
(RS)-syn-H tBu-Li
(RS)-syn-H tBu-Li
(RS)-syn-H tBu-Li
(RS)-syn-H tBu-Li
(SS)-syn-H Ph-Li
(SS)-anti-D Ph-Li
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
THF
1.0
2.0
3.0
1.6
2.0
1.0
2.0
2.0
1.2
1.2
2.0
1.0
2.0
3.0
1.6
2.0
1.0
2.0
2.0
1.2
1.2
2.0
none
none
none
none
none
HMPA
LiCl
ZnCl2
none
none
none
ꢀ78
ꢀ78
ꢀ78
ꢀ78
ꢀ100
ꢀ78
ꢀ78
ꢀ78
ꢀ78
ꢀ78
ꢀ78
RT
RT
RT
858C
RT
RT
RT
RT
RT
RT
RT
10
16
<5
11
66
67
64
67
83
<2
0[b] 69
10
64
8[c] 69
9
11
67
10
THF
33[d] 75
11[e]
(RS)-syn-H Et-MgCl PhMe
0
86
[a] Product boronate 33 obtained as a mixture with unreacted 17 (typically ꢂ50%) following SiO2 chromatog-
raphy. [b] Elimination adduct (E)-buta-1,3-dienyl p-tolyl sulfoxide generated in 28% yield. [c] 2-Chloro-5-
(chloromethyl)pyridine isolated in 25% yield and all of 17 was consumed (33 obtained in pure form). [d] Total
yield of homologation adducts comprised of D-33 (21%) and its derived protodeboronated adduct D-38
(12%). [e] EtMgCl added to (RS)-syn-H-24a before the introduction of boronate 17.
homologated product 33 was obtained in low yield (ꢁ16%)
as an inseparable mixture with its unconverted precursor 17.
The isolation of significant quantities of the expected sulfox-
ide–ligand exchange byproduct 34 from these experiments
indicated that ample carbenoid was generated but that basic
StReCH efficacy was poor. A twofold excess of syn-H-24a
and tBuLi gave the best result, but a further increase in re-
agent stoichiometry was detrimental (entries 2 and 3). The
chloropyridine moiety is anticipated to have a poor migrato-
ry aptitude[36] and to exclude the possibility that a slow 1,2-
metallate rearrangement was responsible for low conversion,
the StReCH reaction was heated to 858C for 3.5 h prior to
quench (entry 4). This action did not improve the outcome
and suggested that ate-complex rearrangement was not nec-
essarily a significant problem—a conclusion borne out by
the subsequent high yielding homologation of 17 with
18d.[37] Conducting the initiation phase of the process at
ꢀ1008C was evaluated in an attempt to limit carbenoid de-
composition prior to its association with the boronic ester;
however, although this change resulted in a higher yield of
34, only a trace of the desired StReCH adduct 33 was gener-
ated (entry 5). Various additive effects were next examined
including the experiments shown in entries 6–8. Use of
HMPA co-solvent resulted in no chain extension whatsoever
and any chlorosulfoxide syn-H-24a not converted to 34 ex-
perienced elimination and was identified as (E)-buta-1,3-
dienyl p-tolyl sulfoxide. Addition of LiCl had no obvious
formulation of PhLi in Bu2O gives very satisfactory results
for StReCH reactions conducted in THF solvent. Applica-
tion of these safer conditions to the chain extension of 17
using syn-H-24a, produced 33 with similar efficacy to tBuLi
in PhMe (entry 9 vs. entry 1). The difficult chain extension
of 17 with 18a was next further improved by utilizing chlor-
osulfoxide anti-D-24a (entry 10). It was established during
the course of studies involving 18d that a-deuterated a-
chlorosulfoxides are superior precursors to a-chloroalkyl-
lithiums because of the reduced rate of proton transfer be-
tween the carbenoid and its progenitor (vide infra). Here,
conducting the StReCH reaction in the presence of an addi-
tional neutron resulted in a threefold improvement in the
yield of chain extension (albeit now to an isotopomer of the
original target). Finally, the synthesis of 33 employing an a-
chloroalkylmagnesium chloride reagent generated by adding
EtMgCl to (SS)-syn-H-24a, was briefly evaluated (entry 11).
Sulfoxide–magnesium exchange from a-chlorosulfoxides is
a highly efficient reaction and a process not plagued by del-
eterious proton exchange;[38] however, the resulting
ꢀ
Grignard species have a limited ability to insert into the C
B bonds of boronic esters. In line with expectation,[6a] we
found that the Mg-based carbenoid could neither homolo-
gate pinacol boronate 17 nor the analogous neopentyl glycol
boronate (not illustrated).
Throughout the above study, products 33 and D-33 gener-
ated using scalemic chlorosulfoxides syn-H-24a and anti-D-
Chem. Eur. J. 2013, 19, 16342 – 16356
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16347

P. R. Blakemore et al.
24a, were found to be optically active. Indirect determina-
tion of enantiomeric excess by the usual practice of oxida-
tion with NaOOH followed by chiral stationary phase
HPLC analysis of the resulting 28 alcohols was precluded
when it was discovered that the derivatization step in this
case resulted in apparent racemization. StReCH with 24a
was, however, verified to proceed in a stereocontrolled fash-
ion by study of the chain extension of model phenethyl bor-
onate 11 [Eq. (2)]. Carbenoid enantiomeric ratio (e.r.)
tribution following oxidative work-up revealed formation of
the alcohol from the desired second generation StReCH
adduct DD-31 (DD-36) and carbinol remnants of boronates
from interrupted chain extension stages (D-37 and 39). The
dominant component of the mixture was butenyl pyridine
D-38, the product of protodeboronation from D-33. Given
the low yield of this process, further advancement of (R,R)-
DD-36 (or its precursor DD-31) to epibatidine was not pur-
sued. As discussed in detail later, the diastereoselectivity of
a double StReCH reaction (or indeed, any process involving
the stereospecific fusion of two chiral molecules of submaxi-
mal ee) is highly sensitive to the stereochemical purity of
precursors. In this example, the low d.r. observed for (R,R)-
DD-36 was anticipated and is symptomatic of the modest
d.r. of the chlorosulfoxide used as carbenoid precursor for
each homologation stage; the predicted d.r. for (R,R)-DD-
36 was 72:28, which lies in reasonable agreement with the
recorded ratio of about 67:33.
cannot under normal circumstances exceed the d.r. of its
chlorosulfoxide precursor;[39] thus, in this example, the
61%ee observed for product (S)-35, assessed after NaOOH
oxidation to the alcohol, represents reasonable stereochemi-
cal fidelity given that the d.r. of the batch of (RS)-syn-H-24a
employed was only 85:15 (i.e., a max. product ee of 70%).
A second StReCH cycle from pyridyl benzylic boronic
ester 33 using all protio chlorosulfoxide (RS)-syn-H-24a re-
sulted in no more than a trace of the desired contiguous
stereodiad 31. In parallel to the findings of the more suc-
cessful study involving 24d (vide infra), we discovered that
further chain extension was realizable with an a-deuterated
a-chlorosulfoxide reagent, but that compounds such as 33
(and 47) are highly susceptible to deboronation. In this case,
the dideuterated boronate DD-31 could only be accessed by
executing a one-pot double StReCH reaction directly from
starting material 17 (Scheme 9). Analysis of the product dis-
Parenthetically, it is noted that boronate 33 was also tar-
geted by Aggarwalꢀs formidable “lithiation/borylation”
StReCH chemistry using the lithiated carbamate sparteine
complex (40) derived from homoallyl alcohol [Eq. (3)]. This
chemistry performed well in our hands employing other
lithiated carbamates, but the reaction of 17 and 40 gave no
trace of the desired product 33 under standard reaction con-
ditions,[9b] and the same protocol as applied to phenethyl
boronate 11 proceeded in only a low yield. Aggarwal et al.
recently also reported difficulties using homoallylic lithiated
carbamates and attributed the problem to a Li-cation/alkene
p-interaction that may prevent the ate-complex from achiev-
ing optimal alignment for 1,2-rearrangement.[40] These work-
ers found that rearrangement could be promoted by the use
of various additives (a mixture of 12-crown-4, H2O, and
TMSCl) or alternatively by exchanging the Et2O solvent for
CH2Cl2 after ate-complex formation. Conceivably, these
newer protocols might be effective for the synthesis of 33 if
this approach were to be reevaluated.
Studies with 1-chloro-3-butynyllithium (18b): It is evident
that a pair of sequential StReCH reactions from chloropyr-
idyl boronate 17 using the propargyl-group bearing carbe-
noid (S)-18b could provide a stereodefined 4,5-disubstituted
1,7-octadiyne analogous to diene 31 and also convertible to
epibatidine. To test the viability of such a plan, we first eval-
uated the efficacy of chloroalkyllithium generation from syn
chlorosulfoxides H-24b and D-24b (Scheme 10). Here, the
amount of sulfoxide–ligand exchange byproduct (41) pro-
duced by the action of phenyllithium on 24 was used as an
indirect metric to gauge the extent of carbenoid generation.
Scheme 9. Synthesis of contiguous stereodiad containing diene (R,R)-
DD-36 by a one-pot double StReCH reaction from B-(2-chloropyrid-5-
yl) pinacol boronate (17).
16348
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16342 – 16356

Functionalized a-Chloroalkyllithiums
FULL PAPER
sessing propargyl CH bonds vicinal to the carbenoid center,
will be an extremely short-lived intermediate. It transpired
that attempted homologation of pyridyl boronate 17 using
propargyl group bearing carbenoid 18b, generated in situ
from either syn-H-24b or syn-D-24b, failed to deliver the
desired chain extended adduct. Low yielding insertion of
18b into model boronate 11 was possible; again, it was
found that the a-deuterated chlorosulfoxide provided the
superior result and in each case significant quantities of
enynes 42 were produced [Eq. (5)].
Scheme 10. Investigation of sulfoxide–lithium exchange from racemic ho-
mopropargyl chlorosulfoxides syn-H-24b and syn-D-24b; Ar=p-Tolyl.
The degree of sulfoxide–ligand exchange from syn-H-24b
(51%) was lower than that found for other chlorosulfoxides
(ca. 65–75%) and the mass balance of the starting material
was identified in elimination adduct 42. The ready formation
of enyne 42 by treatment of syn-H-24b with LDA was noted
previously (Table 2, entry 4). Deuterated chlorosulfoxide
syn-D-24b gave a slightly higher yield of 41 (55%) and in
this case the presence of the isotopic label shed some light
on the mechanism of the competing elimination process.
Both labeled and nonlabeled enynes 42 were produced, in
a respective ratio of 47:53; the former (D-42) must have its
origin in a b-elimination, whereas the latter (H-42) is likely
mainly generated by a-elimination (i.e., a-metallation fol-
lowed by 1,2-hydride shift with loss of LiCl, cf. 1 to 5,
Scheme 1) with perhaps a small portion coming from an
“E2cB” process (i.e., a-metallation followed by attack of
a base on the carbenoid, cf. 1 to 4, Scheme 1).[3,41] The b-
elimination pathway from syn-D-24b showed a slight bias
for the (Z)-enyne[42] and the a-elimination pathway was
modestly E selective. Since the sample of H-42 derived from
syn-H-24b was predominantly E configured, and exhibited
an isomeric signature similar to that found in the sample of
H-42 produced from syn-D-24b, it may be concluded that
enyne formation from the nonlabeled chlorosulfoxide occurs
via the a-elimination pathway.
Studies with 3-alkoxy-1-chloropropylithiums (18c, 18d):
With the knowledge that neither allyl (18a) nor propargyl
group (18b) substituted chloromethyllithiums could effect
the challenging chain extension of pyridyl boronate 17 in
a reasonable yield, we moved on to consider alternate re-
agents wherein the functionalized side-chain moiety might
offer stabilization to the carbenoid. The electrophilic nature
of a-chloroalkyllithiums, a root cause of decomposition
(Scheme 1), is muted in the presence of Lewis base addi-
tives, which compete with the Cl atom for interaction with
the Lewis acidic Li atom. Ordinarily, it is the pull of the Li
ꢀ
atom on the geminal Cl atom that weakens the C Cl bond
and gives cationic character to the C atom, a phenomenon
recognized by Walborsky as “metal-assisted ionization”.[4c]
Use of exogenous Lewis bases is not a solution to help diffi-
cult chloroalkyllithium based StReCH reactions, however,
because these additives may associate with the boronic ester
and prevent its rapid engagement with the short-lived carbe-
noid. To avoid such a pitfall and yet still benefit from Lewis-
base-mediated carbenoid stabilization, we next targeted 3-
alkoxy-1-chloropropyllithiums 18c and 18d. Such reagents
are expected to be intrinsically more stable than other chlor-
oalkyllithiums because of the feasibility of chelate formation
from an internal Lewis base (as depicted above).
The above findings indicate that a-lithiated chlorosulfox-
ide 43b experiences facile a-elimination at ꢀ788C, a surpris-
ing fact when one considers that related metallates with al-
ternate substituents are often quite stable, in some cases per-
sisting even at room temperature [Eq. (4)]. Given that 43b
The generation of 3-benzyloxy-1-chloropropyllithium
(18c) through sulfoxide–lithium exchange from syn-H-24c
was investigated first (Table 4, entries 1 and 2). Byproduct
sulfoxides 34 were isolated in modest yield, indicating that
the formation of 18c was inefficient, and it was clear that a-
metallation of the chlorosulfoxide competed against sulfox-
ide–lithium interchange because 24c was recovered in an
epimerized and heavily deuterated form after addition of
CD3OD. By contrast, sulfoxide–magnesium exchange from
syn-H-24c was characteristically free from complications
and gave a high yield of a persistent a-chloroalkyl Grignard
reagent that led to the expected alkylchloride HD-46c upon
deuterolysis (entry 3). The stability of Mg-based carbenoids
is ripe for decomposition, it is to be expected that 18b, lack-
ing as it does a stabilizing sulfinyl moiety and yet still pos-
Chem. Eur. J. 2013, 19, 16342 – 16356
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16349

P. R. Blakemore et al.
Table 4. Sulfoxide–metal exchange studies from a-chlorosulfoxides 24c and 24d using isotopic labels as mechanistic probes.[a]
Entry
Sulfox. 34
yield [%]
Alkylchloride 46
yield [%] HH/HD/DD
Recovered chlorosulf. 24
yield [%]
2
syn(%d)/anti(%d)
R M
quench
syn(%d)/anti(%d)
ꢀ
chlorosulfoxide
1[b]
2
3
4
5
(ꢃ)-syn-H-24c
(ꢃ)-syn-H-24c
(ꢃ)-syn-H-24c
(RS)-syn-H-24d
(RS)-anti-H-24d
(RS)-anti-D-24d
(RS)-anti-D-24d
83(0):17(0)
83(0):17(0)
tBu-Li
nBu-Li
Et-MgCl
Ph-Li
Ph-Li
Ph-Li
CD3OD
CD3OD
CD3OD
CD3OD
CD3OD
CH3OH
CD3OD
30
45
80
47
51
58
63
n.d.
n.d.
80
n.d.
n.d.
70
48
20
20
20
16
16
65(29):35(66)
14(91):86(87)
83(0):17(0)
83(0):17(0)
13:87:00
>95:<5:00
>95:<5:00
22:50:28
14:59:27
>95(0):<5(0)
7(0):93(0)
28
33
35
26
22(99):78(98)
20(98):80(99)
23(<2):77(<2)
27(89):73(96)
6
7
8(ꢂ80):92(87)
8(ꢂ80):92(87)
Ph-Li
1
[a] Isotopic and diastereoisomeric distributions determined by H NMR spectral analysis. [b] Reaction conducted in PhMe solvent.
as compared to their Li counterparts betrays a lack of reac-
tivity and the chloroalkylmagnesium chloride generated
from 24c proved incapable of chain extending pinacol boro-
nate 11. Subsequent studies with the acetal containing chlor-
osulfoxide 24d were far more promising and 24c was not
further investigated.
a doubly deuterated isotopomer. The fact that DD-46d was
not the major isotopomer in this experiment is a likely arti-
fact of the use of incompletely deuterated starting material
D-24d and the possibility of proton transfer between 18d
molecules (cf., 1 to 4, Scheme 1). The termination of an oth-
erwise identical experiment using CD3OD barely affected
the isotopic ratio observed for 46d (entry 7) indicating that
carbenoid 18d was not extant at the moment of the quench.
With the discovery of a proton transfer between a-chloro-
Utilizing the by now preferred nonpyrophoric initiation
system (PhLi dispensed in Bu2O), the generation of 1-
chloro-2-(1,3-dioxolan-2-yl)ethyllithium (18d) from various
forms of chlorosulfoxide 24d was pursued next (Table 4, en-
tries 4–7). Comparison of syn and anti isomers of H-24d
showed that neither form offered a real advantage over the
other for carbenoid formation as gauged by the yield of 34
(entries 4 and 5). Any chlorosulfoxide recovered after
quench with CD3OD was fully deuterated and exhibited
a syn/anti ratio of about 1:4 regardless of whether syn-H-
24d or anti-H-24d was used as starting material. In addition
to sulfoxides 34 and 24d, a significant quantity of the alkyl-
chloride derived from putative carbenoid 18d (46d) was ob-
tained from both experiments;[43] significantly, in spite of the
fact that the reaction mixture was quenched with CD3OD,
this material was not enriched in deuterium. In situ proton
transfer between the basic chloroalkyllithium 18d and its
weakly acidic chlorosulfoxide precursor H-24d would ac-
count for these observations [Eq. (6)], and a reversed iso-
topic labeling study using anti-D-24d established the validity
of such an argument (entry 6). Thus, sulfoxide–lithium ex-
AHCTUNGTREGaNNUN lkyllithums and their a-chlorosulfoxide precursors we had
identified yet another carbenoid decomposition pathway
that could be deleterious for StReCH of boronates. In this
case, close scrutiny of the data in Table 4 revealed one po-
tential tactic to ameliorate the problem since it was ob-
served that anti-D-24d gave higher yields of 34 than either
diastereoisomer of H-24d (cf., entries 6/7 vs. 4/5). The data
imply that deuterated chlorosulfoxides are superior carbe-
noid precursors to their nonlabeled congeners, a fact under-
standable on the basis that carbenoids will be quenched at
a slower rate by ArS(O)CDClR than ArS(O)CHClR species
due to a primary kinetic isotope effect. As documented in
a number of aforementioned examples, the use of a-deuter-
ated a-chlorosulfoxides was indeed later found to improve
the outcome of difficult StReCH reactions. In the case of
18d, a chloroalkyllithium that potentially benefits from a sta-
bilizing chelate effect, increase in StReCH yield upon ex-
ploitation of the “deuterium effect” was worthwhile but not
too dramatic. For the process of interest here, use of the all
protio chlorosulfoxide (RS)-syn-H-24d resulted in the high-
est conversion yet charted for any StReCH reaction of pyr-
idyl boronate 17 (Scheme 11). Pleasingly, a further gain in
efficacy was realized by using deuteride (SS)-anti-D-24d,
which afforded chain extended boronate (R)-D-47 in 79%
conversion and with essentially perfect stereochemical fideli-
ty (note: use of (RS)-anti-H-24d instead resulted in 62%
conversion). Purification of boronate (R)-D-47 was compli-
cated by its propensity to protodeboronate during silica gel
chromatography and this fact limited isolated yield to
a modest 53%.
change as before, but followed this time by the addition of
CH3OH instead of CD3OD, gave the expected mixture of
34, nondeuterated chlorosulfoxide H-24d, and a sample of
chloride 46d that contained a significant proportion of
With an efficient means to effect the chain extension of
boronate 17 finally identified, advancement of 47 to a stereo-
diad containing material appropriate for the elaboration of
16350
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16342 – 16356

Functionalized a-Chloroalkyllithiums
FULL PAPER
Scheme 11. StReCH of B-(2-chloro-pyrid-5-yl) pinacol boronate (17)
with 1-chloro-2-(1,3-dioxan-2-yl)ethyllithium (18d) generated by sulfox-
ide–lithium exchange from chlorosulfoxides (RS)-syn-H-24d and (SS)-
anti-H-24d. Enantiomeric excess (%ee) of 47 was determined indirectly
by HPLC analysis of derived KOOH oxidation products.
epibatidine was explored. Restricting attention to the most
effective carbenoid precursor found to date, StReCH of bor-
onate (R)-D-47 was evaluated with both enantiomers of
chlorosulfoxide anti-D-24d using the preferred PhLi initia-
tion protocol (Scheme 12, inner reactions). The desired ster-
eodiad products DD-48 were isolated following oxidative
work-up and the absolute stereochemistry of the R,R isomer
was proven by XRD analysis (Figure 2).[29] As likewise
Scheme 12. Synthesis of contiguous stereodiad containing carbinols DD-
48 by either a single StReCH cycle from boronate (R)-D-47 or a one-pot
double StReCH reaction from B-(2-chloro-pyrid-5-yl) pinacol boronate
(17).
ee, an effect most famously articulated by Horeau,[45] and it
may be readily understood in the present context as follows.
Consider an ideal StReCH manifold and the complete ho-
mologation of a chiral boronate substrate of e.r.=x:1 by
a configurationally stable chiral carbenoid of e.r.=y:1. The
most probable event involves the major enantiomer of each
participant reacting together and the least likely event the
opposite scenario: either of these possibilities produces the
targeted relative stereochemistry and this material will exhib-
it e.r.=xy:1. Chemical events of intermediate probability in-
volve the major enantiomer of one component reacting with
the minor enantiomer of the other; these reactions do not
lead to the targeted diastereoisomer and the epimeric mate-
rial produced will exhibit e.r.=x:y. It follows that the ratio
between targeted and untargeted diastereoisomers is [(xy+
Figure 2. An ORTEP diagram for (R,R)-DD-48 showing 30% probability
ellipsoids.[29]
found during the synthesis of 36, a net protodeboronation
pathway dominated, accounting for 51–60% of the mass bal-
ance of D-47, whereas extended adducts DD-48 were gener-
ated in less than 25% yield.[44] Notwithstanding the low
yield of these reactions, the distribution of stereoisomers
produced in each agreed perfectly with expectation and
showed features characteristic of iterative StReCH, namely,
the targeted diastereoisomers exhibited ee (ꢂ98%) signifi-
cantly higher than the substrate (R)-D-47 (89%ee) or the
carbenoid reagents D-18d (ca. 90%ee) employed, and the
minor (untargeted) diastereoisomers were generated in low
ee. This kind of stereochemical outcome is an artifact of the
statistical combination of two chiral molecules of imperfect
1)/ACTHNUGRTENUNG(x+y)]:1 and it is appreciated that the “boost” in ee for
the desired compound is at the expense of yield since some
material is directed toward an unwanted diastereomer. In
the present example, substrate boronate (R)-D-47 and car-
benoid reagents D-18d each exhibited e.r.~95:5 (i.e., x=y=
19); the predicted d.r. of the StReCH reaction is therefore
(192 +1)/
ACTHNUTRGNE(NUG 2ꢂ19):1=90.5:9.5, the desired target is anticipated
to exhibit e.r.=192:1=99.7:0.3, and the untargeted diaste-
reomeric side-product should manifest e.r.=19:19 (racemic).
Data obtained for the targeted syntheses of (R,R)-DD-48
Chem. Eur. J. 2013, 19, 16342 – 16356
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16351

P. R. Blakemore et al.
and (R,S)-DD-48 (Scheme 12) are in excellent agreement
with this prediction.
Given the difficulty of purifying boronate (R)-D-47 with-
out loss, we explored an alternative approach to (R,R)- and
(R,S)-DD-48 by one-pot double StReCH reactions from 17
such that isolation of the delicate intermediate benzylic bor-
onate could be avoided (Scheme 12, outer reactions). The
execution of a sequence of multiple a-chlorosulfoxide-based
StReCH reactions in a single reaction vessel is easy to con-
duct and consists of temperature cycling and reagent/initia-
tor charging events that could in principle be automated.[46]
In this case, the one-pot approach to adducts DD-48 result-
ed in an overall yield of 40–49% for the three transforma-
tions involved from 17 (average yield per step, 74–79%).
The desired compounds were obtained with uniformly excel-
lent enantiopurity but this time (R,S)-DD-48 was generated
in a significantly lower d.r. than its epimer (R,R)-DD-48.
This finding is a likely consequence of unreacted traces of
(SS)-anti-D-24d from the first StReCH cycle compromising
the second cycle, which should involve only (RS)-anti-D-24d
when targeting (R,S)-DD-48. The synthesis of polyfunctional
alcohols DD-48 in good yield directly from chloropyridyl
boronate 17 by a convenient one-pot process was the final
culmination of many findings and provided a sound platform
from which to advance on epibatidine.
Advance toward (ꢀ)-1,2-[2H]2-epibatidine: Opting to intro-
duce the C1 nitrogen atom of epibatidine through stereoin-
vertive amination of an hydroxyl group, StReCH adduct
(R,S)-DD-48 was the appropriate stereodiad for further
elaboration. A speculative plan to form the 7-azabicyclo-
Scheme 14. Synthesis of azidodialdehyde DD-49 and pyrroline DD-53
and conversion of the former to an isotopomer of Coreyꢀs (ꢀ)-epibati-
dine precursor 32 (DD-32).
all and produced azide DD-52 in a form that was easier to
purify by column chromatography. Transacetalization from
ethylene glycol acetals to simple methanol acetals was criti-
cal to allow for acetal hydrolysis to progress under condi-
tions mild enough for the sensitive dialdehyde intermediate
to survive. Thus, treatment of DD-52 with aqueous trifluoro-
acetic acid (TFA) in CHCl3 at RT (2 h) resulted in its
smooth conversion to DD-49; as expected, the azidodialde-
hyde exhibited low stability and it was used immediately
without chromatographic purification in all subsequent ex-
periments. The conversion of DD-49 to DD-51 along the
lines indicated above (Scheme 13) was attempted using SmI2
as a reductant or the same reagent in combination with
PPh3, which was anticipated to yield an initial pyrroline
through the Staudinger reaction. Neither protocol gave a 5-
hydroxyepibatidine product but resulted instead in the gen-
eration of a complex mixture of polar components. It was
discovered that an isolable pyrroline intermediate (DD-53)
retaining one acetal moiety could be readily generated from
azide DD-52 by iminophosphorane formation with PPh3 fol-
lowed by treatment with TFA. Reductive cyclization of this
compound with SmI2 (THF, ꢀ788C to RT) was also briefly
entertained; however, the anticipated azapinacol product (a
5-methoxyepibatidine) was not generated and this line of in-
quiry was not further investigated.
ACHTUNGTRENNUNG[2.2.1]heptane core of the target by reductive cyclization of
an azidodialdehyde derived from this compound (DD-49)
was pursued first. Thus, it was envisioned that single-elec-
tron transfer to a cyclic iminoaldehyde generated in situ
from 49 by reduction of the azide, could conceivably lead to
a 5-hydroxyepibatidine (51) by aza-pinacol coupling of the
intermediate diradical 50 (Scheme 13).[47,48] In attempting to
Scheme 13. A hypothetical “azapinacol” reductive cyclization of an azi-
dodialdehyde (49) to a 5-hydroxyepibatidine isomer (51); Ar=2-chloro-
pyrid-5-yl.
reduce this plan to practice, an immediate precursor to alde-
hyde DD-49, azido bisacetal DD-52, was prepared from
(R,S)-DD-48 by either a one-pot Mitsunobu azidation[49]
/
acidic methanolysis reaction, or alternatively, by a three-
pot sequence involving mesylation, SN2 displacement with
sodium azide, and then again transacetalization
(Scheme 14). The second approach was higher yielding over-
Having failed to realize a concise azapinacol coupling ap-
proach to the target, recourse was made to the original syn-
16352
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16342 – 16356

Functionalized a-Chloroalkyllithiums
FULL PAPER
thetic plan, which called for intersection upon Coreyꢀs inter-
mediate 32 by ring-closing olefin metathesis. For this pur-
pose, azidodialdehyde DD-49 was converted to dienyl
amide DD-54 by exposure to a threefold excess of methyle-
netriphenylphosphorane (generated in situ from nBuLi and
Ph3PCH3I) followed by the addition of trifluoroacetic anhy-
dride. The course of this complex process remains uncertain;
however, it was evident that the azide functional group
within DD-49 engaged with the phosphorane reagent and
was reduced, possibly yielding an iminophosphorane.[50]
Whatever the true nature of the intermediate so generated,
it reacted with the anhydride to deliver the desired dienyl
amide DD-54. Finally, treatment of the diene with Grubbsꢀ
first generation olefin-metathesis catalyst[51] resulted in its
clean conversion to cyclohexene DD-32. Aside from expect-
ed differences due to its isotopic composition, the spectral
signature of DD-32 was in excellent agreement with its
known all protio isotopomer 32, the compound previously
converted to (ꢀ)-epibatidine (15) by Corey and co-work-
ers.[19b] Further advancement of DD-32 to (ꢀ)-1,2-[H2]2-epi-
batidine was not conducted and thus whether or not this
effort constitutes a formal synthesis of the alkaloid is
a matter for debate.
ated upon the addition of RLi to chlorosulfoxides are of
course also a type of carbenoid, and it is conceivable that
these fleeting species are responsible for loading the chlor-
oalkyl moiety onto the boronate. Studies to elucidate the
role of sulfurane intermediates in organolithium-initiated
StReCH reactions involving chlorosulfoxides and boronates
are in progress and will be reported in due course.
Experimental Section
See the Supporting Information for general experimental and analytical
techniques and for synthetic procedures and characterization data for all
other compounds not listed below and not previously described in
ref. [22].
Representative example of sulfoxide synthesis (Scheme 6)
2-[(SS)-2-(4-Methyl-phenyl)sulfinylethyl]-1,3-dioxolane [(SS)-23d]: This
was prepared by the method of Jackson et al.[28] A 100 mL RB flask
equipped with a magnetic stirrer bar was charged with (S)-tert-leucinol
derived ligand (S)-25 (158 mg, 0.334 mmol) and vanadyl acetoacetonate
(59 mg, 0.223 mmol) followed by CHCl3 (20 mL). The resulting mixture
was stirred at RT for 2 h during which time the initially emerald-green
solution became a caramel-brown color. A solution of thioether 21d
(5.00 g, 22.3 mmol) in CHCl3 (15 mL) was then added all at once and the
mixture was stirred for 30 min at RT before being cooled to 28C (in a re-
frigerator). H2O2 (30 wt.% aq., 2.70 mL, d=1.11, 3.00 g, 26.5 mmol) was
added and stirring at 28C was continued until TLC analysis indicated
that all of the starting material had been consumed (19.5 h). The result-
ing dark-orange colored solution was quenched with 1 wt.% aq. Na2S2O3
(20 mL) and the layers were shaken and separated. The aqueous phase
was extracted with CH2Cl2 (2ꢂ20 mL) and the combined organic phases
were washed with brine (20 mL), dried (Na2SO4), and concentrated in
vacuo. The dark-brown residue (ca. 6.4 g) was purified by column chro-
matography (SiO2, eluting with 35–75% EtOAc in hexanes) to afford
sulfoxide (SS)-23d (4.30 g, 17.9 mmol, 80%) as a colorless oil, which
Conclusion
The work described herein has established that the stereoge-
nicity of substituted a-chloroalkyllithiums can be used to
good advantage in the assembly of nontrivial stereochemi-
cally complex molecules through the iterative StReCH para-
digm. Although issues of stereocontrol are for the most part
easily handled, successful management of the many nonpro-
ductive pathways available to these archetypal carbenoid
species is a significant concern. In the present case, two key
devices were employed to achieve efficient chain extension
of a particularly challenging boronate substrate: 1) the in-
trinsic chemical stability of the a-chloroalkyllithium was im-
proved by incorporation of a Lewis basic site within the car-
benoid side-chain, and 2) a deuterium atom was introduced
into the chlorosulfoxide carbenoid precursor to slow the
rate of carbenoid quenching. Each of these tactics can be
viewed as a contrivance, and the second device has awkward
philosophical implications in that it means that deuterium
rich targets are produced rather than their naturally abun-
dant isotopic forms. Notwithstanding these issues, sulfoxide–
lithium-exchange-based StReCH chemistry enjoys an opera-
tional simplicity that is unrivalled by alternate direct metal-
lation-based approaches. The recent discovery by OꢀBrien
and co-workers that Hoppe-type metallated carbamates can
be efficiently generated from stereodefined a-carbamoyloxy
sulfoxides[23a] means that the advantages of sulfoxide–metal-
exchange-based StReCH can now be interfaced with the use
of a more robust carbenoid type.[9]
slowly crystallized on standing. M.p.: 53–548C (TBME-hexanes); [a]D23
ꢀ97.3 (c=1.00, CHCl3); IR (neat): n˜ =2880, 1492, 1401, 1135, 1083, 1047,
811 cmꢀ1 1H NMR (400 MHz, CDCl3): d=7.51 (d, J=8.3 Hz, 2H), 7.32
=
;
(d, J=8.0 Hz, 2H), 4.97 (t, J=4.2 Hz, 1H), 3.96–3.90 (m, 2H), 3.89–3.83
(m, 2H), 2.93 (ddd, J=13.2, 9.9, 5.8 Hz, 1H), 2.87 (ddd, J=13.2, 9.9,
5.7 Hz, 1H), 2.41 (s, 3H), 2.09 (dddd, J=14.1, 9.8, 5.9, 4.2 Hz, 1H),
1.89 ppm (dddd, J=14.2, 9.9, 5.7, 4.2 Hz, 1H); 13C NMR (100 MHz,
CDCl3): d=141.6 (0), 140.6 (0), 130.1 (2C, 1), 124.4 (2C, 1), 102.9 (1),
65.3 (2C, 2), 51.2 (2), 26.4 (2), 21.6 ppm (3); LRMS (EI+): m/z (%): 240
(12) [M]·+, 223 (100), 137 (24), 100 (32), 73 (60); HRMS (EI+) m/z calcd
for C12H16O3S: 240.0820; found: 240.0816. XRD analysis of a recrystal-
lized sample of (RS)-23d produced in the same fashion using ligand (R)-
25 confirmed the stereochemical outcome of this reaction.[29]
HPLC analysis of an authentic sample of (ꢃ)-23d, prepared by oxidation
of 21d with 30 wt.% aq. H2O2 in MeOH, performed with a Daicel Chir-
alcel OD column (4.6 mm IDꢂ250 mm), was eluted with 10% iPrOH in
hexanes at 0.5 mLminꢀ1 and monitored by UV at 210 nm, showed re-
solved peaks: tR [(+)-(RS)-23d]=25.6 min, tR [(ꢀ)-(SS)-23d]=29.0 min.
Analysis of the enantioenriched material prepared as described above
using ligand (S)-25, revealed an enantiomeric excess of >98% in favor
of the S enantiomer.
Representative example of syn-chlorosulfoxide synthesis (Scheme 6)
2-[(SS,2S)-2-Chloro-2-(4-methylphenyl)sulfinylethyl]-1,3-dioxolane [(SS)-
syn-H-24d]: This was prepared by the method of Yamakawa and co-
workers.[30] A stirred solution of the scalemic sulfoxide (SS)-23d (4.45 g,
18.5 mmol) in anhydrous CH2Cl2 (50 mL) at RT was treated with solid
K2CO3 (1.50 g, 10.9 mmol) and then N-chlorosuccinimide (NCS, 12.0 g,
89.9 mmol). The resulting suspension was stirred vigorously at RT until
TLC analysis revealed that a majority of the starting material had been
consumed (10 d). The reaction mixture was then filtered through a Celite
pad and the filtrate was treated with 4 wt.% aq. NaI (50 mL) followed
In closing and as a caveat, it is noted that it is not certain
that a-chloroalkyllithiums are the true reactive intermedi-
ates involved in this work. Sulfurane species initially gener-
Chem. Eur. J. 2013, 19, 16342 – 16356
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16353

P. R. Blakemore et al.
by 10 wt.% Na2S2O3 (80 mL) to reduce the excess NCS. The layers were
separated and the aqueous phase was extracted with CH2Cl2 (2ꢂ20 mL).
The combined organic phases were washed with brine (20 mL), dried
(Na2SO4), and then concentrated in vacuo to yield 5.61 g of a brown
solid. The residue was further purified by column chromatography (SiO2,
eluting with 60% EtOAc in hexanes) to afford the desired chlorosulfox-
treated with
a
second portion of PhLi (0.44 mL, 1.69m in nBu2O,
0.744 mmol) during 20 s. The reaction was again initially stirred for 1 h at
ꢀ788C and then warmed to RT and stirred for a further 2 h. The contents
of the flask, which were now very darkly colored, were then cooled to
08C and treated with 2m aq. KOH (0.5 mL) followed by 30 wt.% aq.
H2O2 (0.25 mL), and the biphasic mixture was stirred vigorously for 1.5 h.
The mixture was then partitioned between EtOAc (10 mL) and H2O
(10 mL) and the layers were separated. The aqueous phase was extracted
with EtOAc (10 mL) and combined organic phases were washed with
brine (10 mL), dried (Na2SO4), and concentrated in vacuo. The residue
was purified by column chromatography (SiO2, eluting with 70% EtOAc
in hexanes) to afford the targeted carbinol (R,R)-DD-48 (84 mg,
0.253 mmol, 40%, d.r.=85:15) as a colorless oil, which solidified on
standing. M.p.: 113–1148C (TBME, colorless needles); [a]2D3 =ꢀ52.9 (c=
0.90, CHCl3); IR (KBr): n˜ =3481, 2888, 1734, 1559, 1458, 1406, 1369,
ide (SS)-syn-H-24d (4.36 g, 15.9 mmol, 86%) as
a colorless solid.
1H NMR analysis indicated that this material was a 84:16 mixture of syn
and anti diastereoisomers, respectively. Recrystallization from tBuOMe
(TBME) gave (SS)-syn-H-24d with d.r.ꢂ95:5 as long colorless needles.
M.p.: 126–1288C (TBME); [a]D23 = +108.6 (c=1.15, CHCl3); IR (KBr):
n˜ =2883, 1596, 1492, 1398, 1145, 1083, 1057, 810 cmꢀ1
;
1H NMR
(400 MHz, CDCl3): d=7.56 (d, J=8.2 Hz, 2H), 7.34 (d, J=8.3 Hz, 2H),
5.14 (dd, J=6.6, 3.3 Hz, 1H), 4.78 (dd, J=10.3, 3.4 Hz, 1H), 4.02–3.94
(m, 2H), 3.93–3.85 (m, 2H), 2.48 (ddd, J=14.3, 6.6, 3.4 Hz, 1H), 2.43 (s,
3H), 1.91 ppm (ddd, J=13.8, 10.3, 3.3 Hz, 1H); 13C NMR (100 MHz,
CDCl3): d=142.8 (0), 135.5 (0), 129.8 (2C, 1), 126.0 (2C, 1), 101.5 (1),
71.4 (1), 65.3 (2), 65.1 (2), 35.4 (2), 21.7 ppm (3); LRMS (EI+): m/z (%):
1140, 1028, 944 cmꢀ1 1H NMR (400 MHz, CDCl3): d=8.26 (dd, J=2.5,
;
0.7 Hz, 1H), 7.58 (dd, J=8.3, 2.5 Hz, 1H), 7.27 (dd, J=8.2, 0.7 Hz, 1H),
4.99 (dd, J=4.9, 3.6 Hz, 1H), 4.65 (dd, J=6.6, 3.3 Hz, 1H), 4.01–3.73 (m,
8H), 3.29 (s, 1H), 2.24 (dd, J=14.2, 6.6 Hz, 1H), 2.08 (dd, J=14.2,
3.2 Hz, 1H), 1.74 (dd, J=14.5, 3.5 Hz, 1H), 1.68 ppm (dd, J=14.6,
5.0 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=150.2 (1), 149.9 (0), 138.8
(1), 136.8 (0), 124.1 (1), 103.5 (1), 103.0 (1), 70.3 (CDOH, 1:1:1 multiplet,
1JCD =20.8 Hz), 65.2 (2), 65.1 (2), 64.92 (2), 64.90 (2), 44.0 (CDR, 1:1:1
37
+
35
276 (4) [Mꢀ Cl] , 274 (12) [Mꢀ Cl] +, 257 (44), 134 (60), 73 (100);
HRMS (EI+): m/z calcd for C12H15O335ClS: 274.0430; found: 274.0430.
XRD analysis of (SS)-syn-H-24d confirmed the absolute and relative con-
figuration of this material.[29]
C
C
Representative example of syn- to anti-chlorosulfoxide epimerization
(Table 2)
2-[2H]-2-[(SS,2R)-2-Chloro-2-(4-methylphenyl)sulfinylethyl]-1,3-dioxolane
mulitplet, 1JCD =19.5 Hz), 37.8 (2), 35.0 ppm (2); HRMS (EI+): m/z
35
calcd for C15H18D2 ClNO5: 331.11554 [M] +; found: 331.11849. XRD
C
analysis of a recrystallized sample of (R,R)-DD-48 confirmed the abso-
lute and relative configuration of this contiguous stereodiad (Figure 2).[29]
[(SS)-anti-D-24d]:
A
stirred solution of (SS)-syn-H-24d (2.23 g,
8.12 mmol) in anhydrous THF (25 mL) at ꢀ788C under Ar was treated
with NaHMDS (8.00 mL, 2.0m in THF, 16.0 mmol). This reagent was
added during 1 min down the cold flask side-wall to chill it before it com-
bined with the reaction mixture. The resulting dark orange solution of
metallated sulfoxide was stirred for 5 min at ꢀ788C before being deuter-
ated with CD3OD (2.25 mL). After 5 min, sat. aq. NH4Cl (10 mL) was
added and the quenched reaction mixture was allowed to warm to RT.
EtOAc (30 mL) and H2O (20 mL) were added and the layers were
shaken and separated. The aqueous phase was extracted with EtOAc
(10 mL) and the combined organic phases were washed with brine
(10 mL), dried (Na2SO4), and concentrated in vacuo. The residue was pu-
rified by column chromatography (SiO2, eluting with 60% EtOAc in hex-
anes) to afford the epimerized product (SS)-anti-D-24d (1.91 g,
6.93 mmol, 85%, anti/syn=86:14, 84%D) as a colorless solid. The mate-
rial was triturated with THF (20 mL) resulting in precipitation of a small
quantity (ꢁ50 mg) of highly insoluble (ꢃ)-anti-D-24d (m.p.: 134–1368C
(TBME)]. The THF triturate was concentrated in vacuo and the residue
was recrystallized from tBuOMe (TBME) to afford (SS)-anti-D-24d with
heightened stereoisomeric purity (d.r.=95:5). M.p.: 86–878C (TBME);
[a]2D3 = +133.1 (c=0.96, CHCl3); IR (KBr): n˜ =2961, 2886, 1595, 1495,
One-pot double StReCH from boronic ester 17 (Scheme 12)
(2S,3R)-2,3-[2H]2-1,4-Bis(1,3-dioxolan-2-yl)-3-(2-chloropyrid-5-yl)butan-2-
AHCTUNGTRENNUNG
ol [(R,S)-DD-48]: The stereodiad (R,S)-DD-48 was prepared in an analo-
gous fashion to its epimer (R,R)-DD-48 (above) from boronate 17
(600 mg, 2.50 mmol)[52] by a one-pot double StReCH reaction involving
initial use of chlorosulfoxide (SS)-anti-D-24d (827 mg, 3.00 mmol, d.r.=
95:5) followed by its antipode (RS)-anti-D-24d (827 mg, 3.00 mmol, d.r.=
93:7) in the second stage. The desired product (R,S)-DD-48 (405 mg,
1.22 mmol, 49%, d.r.=79:21) was obtained as a colorless oil. [a]D23
ꢀ35.2 (c=0.60, CHCl3); IR (neat): n˜ =3488, 2886, 1559, 1458, 1140, 1109,
1024, 944 cmꢀ1 1H NMR (400 MHz, CDCl3): d=8.26 (dd, J=2.5, 0.7 Hz,
=
;
1H), 7.73 (dd, J=8.3, 2.5 Hz, 1H), 7.26 (dd, J=8.3, 0.6 Hz, 1H), 4.98
(dd, J=5.2, 3.4 Hz, 1H), 4.69 (dd, J=6.3, 3.6 Hz, 1H), 4.01–3.77 (m,
8H), 3.12 (s, 1H), 2.21 (ddm, J=14.1, 3.4 Hz, 1H), 2.12 (dd, J=14.1,
6.2 Hz, 1H), 1.72 (dd, J=14.4, 3.4 Hz, 1H), 1.39 ppm (dd, J=14.6,
5.4 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=150.6 (1), 150.1 (0), 139.8
(1), 135.6 (0), 123.9 (1), 103.6 (1), 103.0 (1), 69.4 (CDOH, 1:1:1 multiplet,
1JCD ~20 Hz), 65.2 (2), 65.1 (2), 65.0 (2), 64.9 (2), 43.5 (CDR, 1:1:1 multip-
1
let, JCD ~20 Hz), 38.7 (2), 36.8 ppm (2); HRMS (CI+) m/z calcd for
C15H19D235ClNO5: 332.1234 [M+H]+; found: 332.1223.
1401, 1138, 1086, 1044, 810 cmꢀ1 1H NMR (400 MHz, CDCl3): d=7.63
;
A diastereomerically biased racemic sample of DD-48 (d.r.~4:1), favor-
ing the “unlike” [R,S or S,R] relative stereochemical configuration, was
prepared from boronate 17 by two separate sequential StReCH reactions
employing racemic chlorosulfoxide (ꢃ)-anti-D-24d (note: reaction se-
quence gave d.r.~1:1; this ratio was deliberately perturbed by subsequent
column chromatography). HPLC analysis of this sample of DD-48 iso-
mers using a Daicel Chiralcel OD column (4.6 mm IDꢂ250 mm) and
eluting with 10% iPrOH in hexanes, was monitored by UV at 210 nm,
and showed resolved peaks: tR [(S,R)-DD-48]=27.3 min, tR [(S,S)-DD-
48]=37.8 min, tR [(R,S)-DD-48]=50.6 min, and tR [(R,R)-DD-48]=
63.6 min. Samples of DD-48 generated as indicated in Scheme 12 using
scalemic chlorosulfoxides anti-D-24d were analyzed using the same
HPLC conditions to obtain stereoisomeric ratios (d.r. and ee for each dia-
stereoisomer).
(d, J=8.2 Hz, 2H), 7.34 (d, J=8.3 Hz, 2H), 5.18 (dd, J=6.4, 3.3 Hz, 1H),
4.03–3.84 (m, 4H), 2.43 (s, 3H), 2.38 (dd, J=14.5, 6.4 Hz, 1H), 2.24 ppm
(ddm, J=14.5, 3.2 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=143.1 (0),
137.3 (0), 129.9 (2C, 1), 126.1 (2C, 1), 101.4 (1), 72.6 (CDCl, 1:1:1 multip-
let, 1JCD =25.6 Hz), 65.3 (2), 65.1 (2), 35.8 (2), 21.7 ppm (3); HRMS
(ES+) m/z calcd for C12H14DO3Na37ClS: 300.0361; found: 300.0353; m/z
calcd for C12H14DO3Na35ClS: 298.0391; found: 298.0377. XRD analysis of
(SS)-anti-D-24d confirmed the absolute and relative configuration of this
material.[29]
One-pot double StReCH from boronic ester 17 (Scheme 12)
ACHTUNGTRENNUNG
A
0.626 mmol)[52] and chlorosulfoxide (SS)-anti-D-24d (207 mg, 0.751 mmol,
d.r.=95:5) in anhydrous THF (3.0 mL, 0.21m in boronate) at ꢀ788C
under Ar, was treated dropwise with PhLi (0.44 mL, 1.69m in nBu2O,
0.744 mmol) during 20 s. The resulting light-yellow mixture was allowed
to stir for 1 h at ꢀ788C, then warmed to RT and stirred for 2 h, during
which time the color darkened. The reaction mixture was then recooled
to ꢀ788C and (SS)-anti-D-24d (207 mg, 0.751 mmol, d.r.=95:5) in dry
THF (1.0 mL) was added. After a period of 10 min to ensure that the
flask content had returned to about ꢀ788C, the reaction mixture was
Acknowledgements
Prof. Albert Eschenmoser (ETH, Zꢃrich) is thanked for sharing insights
concerning the reactivity of a-haloalkyllithiums during a visit to OSU in
April 2007. Financial support for this work by National Science Founda-
tion (NSF) grant CHE-0906409 is gratefully acknowledged. The NSF
16354
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16342 – 16356

Functionalized a-Chloroalkyllithiums
FULL PAPER
(CHE-0722319) and the Murdock Charitable Trust (2005265) are also
thanked for support of the OSU NMR spectroscopy facility.
[18] M. I. Damaj, K. R. Creasy, A. D. Grove, J. A. Rosecrans, B. R.
Martin, Brain Res. 1994, 664, 34–40.
[19] Syntheses of epibatidine: a) C. A. Broka, Tetrahedron Lett. 1993, 34,
3251–3254; b) E. J. Corey, T.-P. Loh, S. AchyuthaRao, D. C. Daley,
S. Sarshar, J. Org. Chem. 1993, 58, 5600–5602; c) D. F. Huang, T. Y.
Shen, Tetrahedron Lett. 1993, 34, 4477–4480; d) S. C. Clayton, A. C.
Regan, Tetrahedron Lett. 1993, 34, 7493–7496; e) S. R. Fletcher, R.
Baker, M. S. Chambers, R. H. Herbert, S. C. Hobbs, S. R. Thomas,
H. M. Verrier, A. P. Watt, R. G. Ball, J. Org. Chem. 1994, 59, 1771–
1778; f) K. Senokuchi, H. Nakai, M. Kawamura, N. Kasube, S.
Nonaka, H. Sawaragi, N. Hamanaka, Synlett 1994, 343–344; g) S. Y.
Ko, J. Lerpiniere, I. D. Linney, R. Wrigglesworth, J. Chem. Soc.
Chem. Commun. 1994, 1775–1776; h) K. Sestanj, E. Melenski, I. Jir-
kovsky, Tetrahedron Lett. 1994, 35, 5417–5420; i) K. Okabe, M. Nat-
sume, Chem. Pharm. Bull. 1994, 42, 1432–1436; j) P. L. Kotian, F. I.
Carroll, Synth. Commun. 1995, 25, 63–71; k) A. Hernandez, M.
Marcos, H. Rapoport, J. Org. Chem. 1995, 60, 2683–2691; l) D. Bai,
R. Xu, G. Chu, X. Zhu, J. Org. Chem. 1996, 61, 4600–4606; m) C.
Szꢄntay, Z. Kardos-Balogh, I. Moldvai, C. Szantay, Jr., E. Temes-
vari-Major, G. Blasko, Tetrahedron 1996, 52, 11053–11062; n) C.
Zhang, M. L. Trudell, J. Org. Chem. 1996, 61, 7189–7191; o) B. M.
Trost, G. R. Cook, Tetrahedron Lett. 1996, 37, 7485–7488; p) E. Al-
bertini, A. Barco, S. Benetti, C. De Risi, G. P. Pollini, V. Zanirato,
Tetrahedron Lett. 1997, 38, 681–684; q) H. Kosugi, M. Abe, R. Hat-
suda, H. Uda, M. Kato, Chem. Commun. 1997, 1857–1858; r) N. S.
Sirisoma, C. R. Johnson, Tetrahedron Lett. 1998, 39, 2059–2062; s) S.
Aoyagi, R. Tanaka, M. Naruse, C. Kibayashi, J. Org. Chem. 1998,
63, 8397–8406; t) G. M. P. Giblin, C. D. Jones, N. S. Simpkins, J.
[1] Selected reviews: a) V. Capriati, S. Florio, Chem. Eur. J. 2010, 16,
4152–4162; b) G. Boche, J. C. W. Lohrenz, Chem. Rev. 2001, 101,
697–756; c) H. Siegel, Top. Curr. Chem. 1982, 106, 55–78.
[2] a) G. L. Closs, L. E. Closs, Angew. Chem. 1962, 74, 431–431; Angew.
Chem. Int. Ed. Engl. 1962, 1, 334–335; b) G. L. Closs, R. A. Moss, J.
Am. Chem. Soc. 1964, 86, 4042–4053.
[3] For a summary of much of Kçbrichꢀs work and a useful primer on
the chemistry of carbenoids, see: G. Kçbrich, Angew. Chem. 1967,
79, 15–27; Angew. Chem. Int. Ed. Engl. 1967, 6, 41–52.
[4] a) G. Kçbrich, W. Goyert, Tetrahedron 1968, 24, 4327–4342;
b) K. G. Taylor, J. Chaney, J. C. Deck, J. Am. Chem. Soc. 1976, 98,
4163–4167; c) M. Topolski, M. Duraisamy, J. Rachon, J. Gawronski,
K. Gawronska, V. Goedken, H. M. Walborsky, J. Org. Chem. 1993,
58, 546–555.
[5] For studies establishing microscopic configurational stability for acy-
clic a-bromoalkyllithums, see: a) R. W. Hoffmann, T. Ruhland, M.
Bewersdorf, J. Chem. Soc. Chem. Commun. 1991, 195–196; b) R. W.
Hofmann, M. Julius, F. Chemla, T. Ruhland, G. Frenzen, Tetrahe-
dron 1994, 50, 6049–6060; c) A. Wieczorek, F. Hammerschmidt, J.
Am. Chem. Soc. 2012, 134, 10021–10034.
[6] a) P. R. Blakemore, S. P. Marsden, H. W. Vater, Org. Lett. 2006, 8,
773–776; b) P. R. Blakemore, M. S. Burge, J. Am. Chem. Soc. 2007,
129, 3068–3069.
[7] D. C. Kapeller, F. Hammerschmidt, J. Am. Chem. Soc. 2008, 130,
2329–2335.
Chem. Soc. Perkin Trans.
1 1998, 3689–3698; u) A. Palmgren,
A. L. E. Larsson, J.-E. Bꢅckvall, P. Helquist, J. Org. Chem. 1999, 64,
836–842; v) J. C. Namyslo, D. E. Kaufmann, Synlett 1999, 114–116;
w) M. T. Barros, C. D. Maycock, M. R. Ventura, Tetrahedron Lett.
1999, 40, 557–560; x) C. Zhang, C. J. Ballay, M. L. Trudell, J. Chem.
Soc. Perkin Trans. 1 1999, 675–676; y) J. Habermann, S. V. Ley, J. S.
Scott, J. Chem. Soc. Perkin Trans. 1 1999, 1253–1256; z) H. Nakashi-
ma, K. Hiroya, T. Taniguchi, K. Ogasawara, Synlett 1999, 1405–
1406; aa) H. F. Olivo, M. S. Hemenway, J. Org. Chem. 1999, 64,
8968—8969; ab) W. F. J. Karstens, M. J. Moolenaar, F. P. J. T. Rutjes,
U. Grabowska, W. N. Speckamp, H. Hiemstra, Tetrahedron Lett.
1999, 40, 8629–8632; ac) A. Avenoza, C. Cativiela, M. A. Fernan-
dez-Recio, J. M. Peregrina, Tetrahedron: Asymmetry 1999, 10, 3999–
4007; ad) A. Hall, P. D. Bailey, D. C. Rees, G. M. Rosair, R. H.
Wightman, J. Chem. Soc. Perkin Trans. 1 2000, 329–342; ae) I.
Cabanal-Duvillard, J.-F. Berrien, J. Royer, Tetrahedron: Asymmetry
2000, 11, 2525–2529; af) B. Roy, H. Watanabe, T. Kitahara, Hetero-
cycles 2001, 55, 861–871; ag) G. Pandey, S. K. Tiwari, R. S. Singh,
R. S. Mali, Tetrahedron Lett. 2001, 42, 3947–3949; ah) D. A. Evans,
K. A. Scheidt, C. W. Downey, Org. Lett. 2001, 3, 3009–3012; ai) T.
Tachihara, H. Watanabe, H. T. Kitahara, Heterocycles 2002, 57, 781–
785; aj) A. J. Moreno-Vargas, P. Vogel, Tetrahedron: Asymmetry
2003, 14, 3173–3176; ak) C.-L. K. Lee, T.-P. Loh, Org. Lett. 2005, 7,
2965–2967; al) M.-Y. Chang, H.-P. Chen, Heterocycles 2005, 65,
1705–1711; am) V. K. Aggarwal, B. Olofsson, Angew. Chem. 2005,
117, 5652–5655; Angew. Chem. Int. Ed. 2005, 44, 5516–5519; an) Y.
Hoashi, T. Yabuta, P. Yuan, H. Miyabe, Y. Takemoto, Tetrahedron
2005, 61, 365–374; ao) H. Kimura, T. Fujiwara, T. Katoh, K. Nish-
ide, T. Kajimoto, M. Node, Chem. Pharm. Bull. 2006, 54, 399–402;
ap) A. Armstrong, Y. Bhonoah, S. E. Shanahan, J. Org. Chem. 2007,
72, 8019–8024; aq) J. Bexrud, M. Lautens, Org. Lett. 2010, 12, 3160–
3163; ar) D. R. Boyd, N. D. Sharma, M. Kaik, P. B. A. McIntyre, P. J.
Stevenson, Org. Biomol. Chem. 2012, 10, 2774–2779; as) K. L.
Jensen, C. F. Weise, G. Dickmeiss, F. Morana, R. L. Davis, K. A. Jor-
gensen, Chem. Eur. J. 2012, 18, 11913–11918.
[8] For general discourse on the configurational stability of organolithi-
ums, see: A. Basu, S. Thayumanavan, Angew. Chem. 2002, 114, 740–
763; Angew. Chem. Int. Ed. 2002, 41, 716–738.
[9] Selected work: a) G. Besong, K. Jarowicki, P. J. Kocienski, E. Sliwin-
ski, F. T. Boyle, Org. Biomol. Chem. 2006, 4, 2193–2207; b) J. L.
Stymiest, G. Dutheuil, A. Mahmood, V. K. Aggarwal, Angew. Chem.
2007, 119, 7635–7638; Angew. Chem. Int. Ed. 2007, 46, 7491–7494;
c) J. K. Stymiest, V. Bagutski, R. M. French, V. K. Aggarwal, Nature
2008, 456, 778–783; d) V. Bagutski, R. M. French, V. K. Aggarwal,
Angew. Chem. 2010, 122, 5268–5271; Angew. Chem. Int. Ed. 2010,
49, 5142–5145. Review: e) S. P. Thomas, R. M. French, V. Jheengut,
V. K. Aggarwal, Chem. Rec. 2009, 9, 24–39.
[10] R. Larouche-Gauthier, C. J. Fletcher, I. Couto, V. K. Aggarwal,
Chem. Commun. 2011, 47, 12592–12594.
[11] E. Vedrenne, O. A. Walker, M. Vitale, F. Schmidt, V. K. Aggarwal,
Org. Lett. 2009, 11, 165–168.
[12] F. Schmidt, F. Keller, E. Vedrenne, V. K. Aggarwal, Angew. Chem.
2009, 121, 1169–1172; Angew. Chem. Int. Ed. 2009, 48, 1149–1152.
[13] I. Coldham, J. J. Patel, S. Raimbault, D. T. E. Whittaker, H. Adams,
G. Y. Fang, V. K. Aggarwal, Org. Lett. 2008, 10, 141–143.
[14] For discourse on the related but conceptually distinct stereoinduc-
tive substrate-controlled homologation method involving prochiral
LiCHX2 species developed by Matteson and co-workers, see ref. [9e]
and: a) D. S. Matteson, in Boronic Acids (Ed.: D. G. Hall), Wiley-
VCH, Weinheim, 2005, pp. 305–342; b) D. S. Matteson, Tetrahedron
1998, 54, 10555–10607; c) D. S. Matteson, Chem. Rev. 1989, 89,
1535–1551; d) D. S. Matteson, J. Organomet. Chem. 1999, 581, 51–
65.
[15] For examples of homologation of boronic esters with other achiral
or racemic a-chloroalkyllithiums, see: with LiCH2Cl: a) K. M.
Sadhu, D. S. Matteson, Organometallics 1985, 4, 1687–1689; b) R.
Soundararajan, G. Li, H. C. Brown, Tetrahedron Lett. 1994, 35,
8957–8960; with Me3SiCHLiCl c) D. S. Matteson, D. Majumdar, Or-
ganometallics 1983, 2, 230–236.
[20] Review: H. F. Olivo, M. S. Hemenway, Org. Prep. Proced. Int. 2002,
34, 1–26.
[21] Reviews: a) F. I. Carroll, Heterocycles 2009, 79, 99–120; b) W. H.
Bunnelle, M. J. Dart, M. R. Schrimpf, Curr. Top. Med. Chem. 2004,
4, 299–334. Selected recent work with leading references: c) P. On-
dachi, A. Castro, C. W. Luetje, M. I. Damaj, S. W. Mascarella, H. A.
[16] a) T. F. Spande, H. M. Garraffo, M. W. Edwards, H. J. C. yeh, L.
Pannell, J. W. Daly, J. Am. Chem. Soc. 1992, 114, 3475–3478;
b) J. W. Daly, T. Tokuyama, T. Fujiwara, R. J. Highet, I. L. Karle, J.
Am. Chem. Soc. 1980, 102, 830–836.
[17] a) J. W. Daly, J. Nat. Prod. 1998, 61, 162–172; b) H. M. Garraffo,
T. F. Spande, M. Williams, Heterocycles 2009, 79, 207–217.
Chem. Eur. J. 2013, 19, 16342 – 16356
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16355

P. R. Blakemore et al.
Navarro, F. I. Carroll, J. Med. Chem. 2012, 55, 6512–6522; d) W. H.
Bunnelle, J. F. Daanen, K. B. Ryther, M. R. Schrimpf, M. J. Dart, A.
Gelain, M. D. Meyer, J. M. Frost, D. J. Anderson, M. Buckley, P.
Curzon, Y.-J. Cao, P. Puttfarchen, X. Searle, J. Ji, C. B. Putman, C.
Surowy, L. Toma, D. Barlocco, J. Med. Chem. 2007, 50, 3627–3644.
[22] C. R. Emerson, L. N. Zakharov, P. R. Blakemore, Org. Lett. 2011,
13, 1318–1321.
[23] Recent examples of sulfoxide–lithium exchange for the synthesis of
non-trivial organolithiums: a) P. J. Rayner, P. OꢀBrien, R. A. J.
Horan, J. Am. Chem. Soc. 2013, 135, 8071–8077; b) K. Jarowicki, P.
Kocienski, Z. Komsta, A. Wojtasiewicz, Synthesis 2012, 44, 946–952;
c) A. L. Barsamian, P. R. Blakemore, Organometallics 2012, 31, 19–
22; d) F. Cai, X. Pu, X. Qi, V. Lynch, A. Radha, J. M. Ready, J. Am.
Chem. Soc. 2011, 133, 18066–18069; e) F. Brebion, G. Vincent, S.
Chelli, O. Kwasnieski, F. Najera, B. Delouvrie, I. Marek, E. Derat,
J.-P. Goddard, M. Malacria, L. Fensterbank, Chem. Asian J. 2011, 6,
1825–1833.
[36] a) R. Robiette, G. Y. Fang, J. N. Harvey, V. K. Aggarwal, Chem.
Commun. 2006, 741–743; b) V. K. Aggarwal, G. Y. Fang, X. Ginesta,
D. M. Howells, M. Zaja, Pure Appl. Chem. 2006, 78, 215–229.
[37] More recent work from Aggarwal and co-workers has likewise es-
tablished that B-pyridyl and other types of heteroaryl boronic esters
are compatible with lithiated carbamate based StReCH chemistry,
see: C. G. Watson, V. K. Aggarwal, Org. Lett. 2013, 15, 1346–1349.
[38] a) T. Satoh, Heterocycles 2012, 85, 1–33; b) T. Satoh, Chem. Soc.
Rev. 2007, 36, 1561–1572.
[39] An exception to this statement is possible if a given sample of a-
chlorosulfoxide is composed of a mixture of syn and anti diastereo-
isomers that exhibit opposite absolute configuration at sulfur. The
mode of synthesis of a-chlorosulfoxides in this work would not lead
to such a situation, however.
[40] S. Roesner, J. Mansilla Casatejada, T. G. Elford, R. P. Sonawane,
V. K. Aggarwal, Org. Lett. 2011, 13, 5740–5743.
[41] M. Schlosser, V. Ladenberger, Tetrahedron Lett. 1964, 5, 1945–1949.
[42] Other examples of Z selective enyne synthesis from homopropargyl-
ic halides through low temperature base induced b-elimination have
been reported, see: J. M. Wright, G. B. Jones, Tetrahedron Lett.
1999, 40, 7605–7609.
[43] Decomposition pathways available to 18d (e.g., Scheme 1) presuma-
bly account for the lower isolated yield of alkylchloride 46 as com-
pared to by-product 34.
[24] For seminal work involving the generation of scalemic a-chloroal-
kylmagnesium reagents via sulfoxide–metal exchange, see: R. W.
Hoffmann, P. G. Nell, R. Leo, K. Harms, Chem. Eur. J. 2000, 6,
3359–3365.
[25] J. Drabowicz, M. Mikolajczyk, Synth. Commun. 1981, 11, 1025–
1030.
[26] G. Tsuchihashi, K. Ogura, Bull. Chem. Soc. Jpn. 1971, 44, 1726.
[27] a) C. Bolm, F. Bienewald, Angew. Chem. 1995, 107, 2883–2885;
Angew. Chem. Int. Ed. Engl. 1995, 34, 2640–2642; b) D. A. Cogan,
G. Liu, K. Kim, B. J. Backes, J. A. Ellman, J. Am. Chem. Soc. 1998,
120, 8011–8019.
[28] C. Drago, L. Caggiano, R. F. W. Jackson, Angew. Chem. 2005, 117,
7387–7389; Angew. Chem. Int. Ed. 2005, 44, 7221–7223.
[29] CCDC-837747 [(RS)-23d], 945138 [(RS)-syn-H-24a], 837748 [(SS)-
syn-H-24d], 837749 [(SS)-anti-D-24d], 945137 [(ꢃ)-anti-H-24d]
and837750 [(R,R)-DD-48] contain the supplementary crystallograph-
ic data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via www.ccdc.ca-
m.ac.uk/data_request/cif. For determination of absolute configura-
tion by single-crystal X-ray crystallographic analysis using the reso-
nant scattering technique, see: a) J. M. Bijvoet, A. F. Peerdeman,
A. J. van Bommel, Nature 1951, 168, 271–272; b) R. Srinivasan,
Adv. Struct. Res. Diffr. Methods 1972, 4, 105–197.
[44] The benzylic nature of boronates 33 and 47 makes these compounds
highly susceptible to deboronation, a problem exacerbated by the
electron-withdrawing chloropyridine unit, which may promote frag-
mentation of ate-complexes. For synthetically useful protodeborona-
tion from benzylic positions, see: S. Nave, R. P. Sonawane, T. G.
Elford, V. K. Aggarwal, J. Am. Chem. Soc. 2010, 132, 17096–17098.
[45] J. P. Vigneron, M. Dhaenens, A. Horeau, Tetrahedron 1973, 29,
1055–1059.
[46] It has proved possible to synthesize contiguous stereotriad contain-
ing materials by single vessel triple sequential StReCH reactions
using chlorosulfoxide anti-D-24d, see: X. Sun, P. R. Blakemore,
Org. Lett. 2013, 15, 4500–4503.
[47] For examples of the pinacol coupling of imines and related species
(e.g., nitrones, N-sulfinyl imines) with carbonyl compounds using
SmI2, see: a) F. Machrouhi, J.-L. Namy, Tetrahedron Lett. 1999, 40,
1315–1318; b) G. Masson, S. Py, Y. Vallee, Angew. Chem. 2002, 114,
1850–1853; Angew. Chem. Int. Ed. 2002, 41, 1772–1775; c) Y.-W.
Zhong, Y.-Z. Dong, K. Fang, K. Izumi, M.-H. Xu, G.-Q. Lin, J. Am.
Chem. Soc. 2005, 127, 11956–11957.
[30] T. Satoh, T. Oohara, Y. Ueda, K. Yamakawa, Tetrahedron Lett.
1988, 29, 313–316.
[48] 5-Hydroxyepibatidine derivatives offer superior discrimination be-
tween nAChR receptor subtypes as compared to the parent alka-
loid, see: Z.-L. Wei, Y. Xiao, C. George, K. J. Kellar, A. P. Kozikow-
ski, Org. Biomol. Chem. 2003, 1, 3878–3881.
[49] B. Lal, B. N. Pramanik, M. S. Manhas, A. K. Bose, Tetrahedron Lett.
1977, 18, 1977–1980.
[50] Alkyl azides do not generally react with alkylidenetriphenylphos-
phoranes and Wittig reactions of azidoaldehydes have been reported
to yield azidoalkenes. In the present example, extra care was taken
to ensure that the phosphonium salt employed (Ph3PCH3I) was not
contaminated by PPh3. For representative examples of Wittig olefi-
nation of azidoaldehydes, see: a) V. U. Pawar, S. T. Chavan, S. G.
Sabharwal, V. S. Shinde, Bioorg. Med. Chem. 2010, 18, 7799–7803;
b) C. Herdeis, P. Kuepper, S. Ple, Org. Biomol. Chem. 2006, 4, 524–
529; c) T. Markidis, G. Kokotos, J. Org. Chem. 2001, 66, 1919–1923;
d) D. R. Williams, D. L. Brown, J. W. Benbow, J. Am. Chem. Soc.
1989, 111, 1923–1925.
[31] Other non-racemizing sulfoxide a-chlorination methods: with SiO2
supported NCS, a) J. Drabowicz, Synthesis 1986, 831–833; with
PhICl2 and AgNO3: b) M. Cinquini, S. Colonna, R. Fornasier, F.
Montanari, J. Chem. Soc. Perkin Trans. 1 1972, 1886–1889; c) P. Cal-
zavara, M. Cinquini, S. Colonna, R. Fornasier, F. Montanari, J. Am.
Chem. Soc. 1973, 95, 7431–7436. For mechanistic insights, see: d) J.
Klein, Chem. Lett. 1979, 359–362.
[32] M. Cinquini, S. Colonna, D. Landini, J. Chem. Soc. Perkin Trans. 2
1972, 296–300.
[33] The low temperature racemization of dichloromethyl p-tolyl sulfox-
ide following its treatment with metal amide bases (LDA, LiHMDS,
NaHMDS, KHMDS) was contemporaneously noted by Satoh et al.
For a description of these results and some interesting mechanistic
conjecture, see: a) T. Satoh, H. Momochi, T. Noguchi, Tetrahedron:
Asymmetry 2010, 21, 382–384; b) T. Kimura, T. Tsuru, H. Momochi,
T. Satoh, Heteroat. Chem. 2013, 24, 131–137.
[34] The gauche effect in a-halogenated (and related) sulfoxides is prece-
dented and often attributed to hyperconjugative stabilization with
the s*SO orbital playing a major role, see: a) M. P. Freitas, J. Org.
Chem. 2012, 77, 7607–7611; b) R. Ulshçfer, J. Podlech, J. Am.
Chem. Soc. 2009, 131, 16618–16619; c) A. L. Schwan, M. R. Roche,
J. F. Gallagher, G. Ferguson, Can. J. Chem. 1994, 72, 312–324; d) E.
Juaristi, M. Ordonez, Tetrahedron 1994, 50, 4937–4948.
[51] P. E. Schwab, R. H. Grubbs, J. W. Ziller, J. Am. Chem. Soc. 1996,
118, 100–110.
[52] Boronate 17 can be obtained from a variety of chemical suppliers
(e.g., Aldrich, cat.: 659843), or prepared from inexpensive 5-bromo-
2-chloro-pyridine by the method of Rault et al., see: A. Bouillon, J.-
C. Lancelot, V. Collot, P. R. Bovy, S. Rault, Tetrahedron 2002, 58,
2885–2890.
[35] S. N. Mlynarski, A. S. Karns, J. P. Morken, J. Am. Chem. Soc. 2012,
134, 16449–16451.
Received: June 28, 2013
Published online: October 14, 2013
16356
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16342 – 16356