Harnessing the ortho-directing ability of the azetidine ring for the regioselective and exhaustive functionalization of arenes

Article abstract of DOI:10.1002/chem.201403141

This work demonstrates how the directing ability of the azetidine ring could be useful for regioselective ortho-C-H functionalization of aryl compounds. Robust polar organometallic (lithiated) intermediates are involved in this synthetic strategy. The reagent n-hexyllithium emerged as a safer, yet still effective, basic reagent for the hydrogen/lithium permutation relative to the widely used reagent nBuLi. Two different reaction protocols were discovered for regioselective lithiation at the ortho positions adjacent to the azetidine ring, which served as a toolbox when other competing directing groups were installed on the aromatic ring. The coordinating ability of the azetidine nitrogen atom, as well as the involvement of dynamic phenomena related to the preferential conformations of 2-arylazetidine derivatives, were recognized to be responsible for the observed reactivity and regioselectivity. A site-selective functionalization of the aromatic ring was achieved for aryl azetidines with either coordinatively competent groups (e.g. methoxy) or inductively electron-withdrawing substituents (e.g. chlorine and fluorine). By fine-tuning the reaction conditions, regioselective introduction of several substituents on the aromatic ring could be realized. Several substitution patterns were accomplished, which included 1,2,3-trisubstitution, 1,2,3,4-tetrasubstitution, and 1,2,3,4,5-pentasubstitution, up to the exhaustive substitution of the aromatic ring.

Full text of DOI:10.1002/chem.201403141

DOI: 10.1002/chem.201403141
Full Paper
&
Lithiation
Harnessing the ortho-Directing Ability of the Azetidine Ring for
the Regioselective and Exhaustive Functionalization of Arenes
Marina Zenzola,[a] Leonardo Degennaro,*[a] Piera Trinchera,[a] Laura Carroccia,[a]
Arianna Giovine,[a] Giuseppe Romanazzi,[b] Piero Mastrorilli,[b] Rosanna Rizzi,[c] Luisa Pisano,[d]
and Renzo Luisi*[a]
In memory of Manfred Schlosser
Abstract: This work demonstrates how the directing ability
of the azetidine ring could be useful for regioselective ortho-
CꢀH functionalization of aryl compounds. Robust polar or-
ganometallic (lithiated) intermediates are involved in this
synthetic strategy. The reagent n-hexyllithium emerged as
a safer, yet still effective, basic reagent for the hydrogen/lith-
ium permutation relative to the widely used reagent nBuLi.
Two different reaction protocols were discovered for regiose-
lective lithiation at the ortho positions adjacent to the azeti-
dine ring, which served as a toolbox when other competing
directing groups were installed on the aromatic ring. The co-
ordinating ability of the azetidine nitrogen atom, as well as
the involvement of dynamic phenomena related to the pref-
erential conformations of 2-arylazetidine derivatives, were
recognized to be responsible for the observed reactivity and
regioselectivity. A site-selective functionalization of the aro-
matic ring was achieved for aryl azetidines with either coor-
dinatively competent groups (e.g. methoxy) or inductively
electron-withdrawing substituents (e.g. chlorine and fluo-
rine). By fine-tuning the reaction conditions, regioselective
introduction of several substituents on the aromatic ring
could be realized. Several substitution patterns were accom-
plished, which included 1,2,3-trisubstitution, 1,2,3,4-tetrasub-
stitution, and 1,2,3,4,5-pentasubstitution, up to the exhaus-
tive substitution of the aromatic ring.
Introduction
ple, the nature of the directing group, with reference to either
its ability to coordinate the metal (lithium), or its inductive
electron-withdrawing capability, the nature of the base and
the reaction conditions. In other words, the regioselectivity of
such metalations cannot always be predicted because the
actual combination of thermodynamic and kinetic factors that
govern this aspect cannot be accurately assessed. However,
scope and limitations of this strategy have been widely investi-
gated[2] and mechanistic studies tried to shed light on this pro-
cess. The most accepted theories to help rationalize reactivity
and regioselectivity rely on the complex-induced proximity
effect (CIPE), kinetically enhanced metalation, the triple-ion
mechanism, and the override-base mechanism.[3] The bounda-
ries between such mechanisms in some cases are subtle and
caution should be taken in the evaluation of the reactivity of
aryl compounds in directed metalations, especially if kinetic
and mechanistic evidence are missing. Fortunately, synthetic
chemists are curious, and research efforts focused on directed
metalation still offer the opportunity to discover new directing
groups and synthetic pathways to fill the knowledge gap.[4,5] In
this context, despite the broad range of heteroatom and
carbon-bonded substituents available to perform the hydro-
gen–metal permutation in aryl compounds, small saturated N-
heterocycles have received inadequate attention as potential
directing metalation groups (DMGs). The importance of the
aziridine ring in the regioselective lithiation of 2-aryl derivatives
Directed ortho-metalation (DoM)[1] is an important methodolo-
gy for the functionalization of aromatic and heteroaromatic de-
rivatives versus classical aromatic electrophilic substitution. In
fact, this methodology allows for regioselective functionaliza-
tion of aromatic (or heteroaromatic) rings by the intermediacy
of carbanionic species, usually generated by deprotonation in
the proximity of a directing group. Nevertheless, such hydro-
gen–lithium permutation depends on several factors, for exam-
[a] M. Zenzola, Dr. L. Degennaro, Dr. P. Trinchera, L. Carroccia, Dr. A. Giovine,
Prof. Dr. R. Luisi
Department of Pharmacy—Drug Sciences
University of Bari “A. Moro” Via E. Orabona 4, Bari 70125 (Italy)
E-mail: leonardo.degennaro@uniba.it
renzo.luisi@uniba.it
[b] Dr. G. Romanazzi, Prof. Dr. P. Mastrorilli
DICATECh, Polytechnic of Bari
Via E. Orabona 4, Bari 70125 (Italy)
[c] R. Rizzi
Istituto di Cristallografia (IC-CNR)
Via Amendola 122/o, 70125 Bari (Italy)
[d] Dr. L. Pisano
Dipartimento di Chimica e Farmacia
University of Sassari, Via Vienna 2, 07100 Sassari (Italy)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201403141.
Chem. Eur. J. 2014, 20, 12190 – 12200
12190
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. 2-Arylazetidines used in this study.
Scheme 1. Nitrogen dynamics and nature of the N-substituent as factors
that affect the regioselectivity of the lithiation.
has recently been reported by Luisi, Florio, and co-workers.[6] It
has been proposed that nitrogen dynamics and coordination
effects can play important roles to address the regioselectivity
of the lithiation.[7] In fact, N-alkyl-2-arylaziridines could be
ortho-lithiated at low temperatures, whereas a-benzylic lithia-
tion was observed at higher temperatures, which proved that
the aziridino group could act as a DMG (Scheme 1).[8] More re-
cently, with the aim to develop new stereo- and regioselective
synthetic protocols by using metalated heterocycles, we fo-
cused our attention to the almost unexplored four-membered
heterocycles thietanes and azetidines.[9] Such heterocycles are
appealing, either because of their occurrence in several natural
products and biologically active molecules, or due to the lack
of methodologies for their direct functionalization.[10] In the
case of 2-arylazetidines, in a preliminary communication we re-
ported that the nature of the N-substituent plays an important
role in addressing the regioselectivity of the lithiation reaction.
In particular, with an electron-withdrawing group [EWG; e.g.
tert-butoxycarbonyl (Boc)] as the N-substituent, the lithiation
occurred at the a-benzylic position, whereas with an electron-
donating group (EDG; e.g. alkyl) ortho-aromatic lithiation was
observed, which disclosed the ability of the azetidine ring to
act as a DMG (Scheme 1). However, in striking contrast to aziri-
dines, a-benzylic lithiation was never observed with 2-aryl-N-al-
kylazetidines.
First, we investigated the reactivity of 1-methyl-2-phenylaze-
tidine (1a), chosen as a reference substrate, towards bases
other than n-hexyllithium (nHexLi). We found that nHexLi was
effective in the regioselective ortho-lithiation of 1a under very
mild conditions.[9a] Nevertheless, as reported in Table 1, other
organolithium reagents and lithium amides were taken into
consideration for two main reasons: 1) to explore the possibili-
ty to switch the regioselectivity (i.e. ortho versus a-benzylic);
Table 1. Optimization study for the ortho lithiation of 1a.
Entry Base (equiv) Ligand (equiv) Solvent T [8C] t [h] Yield [%][a]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
LDA (1.5)
LDA (1.5)
sBuLi (1.5)
sBuLi (2.5)
sBuLi (2.5)
sBuLi (1.5)
nBuLi (2.5)
nBuLi (2.5)
nBuLi (2)
–
–
–
–
–
–
–
–
–
–
–
–
THF
THF
THF
THF
toluene
Et2O
THF
THF
Et2O
toluene
Et2O
Et2O
Et2O
toluene
Et2O
Et2O
toluene
Et2O
Et2O
Et2O
ꢀ78
0
ꢀ78
0
1
1
1
1
–
–
–
–
–
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
50
0.5
16
0.5 52
2
16
18
1
4
16
2
1
1
1
2
1
4
1
16
45
60
80
30
40
Herein, we report the results of an investigation aimed at
understanding the factors that affect the reactivity and regio-
selectivity of the lithiation of 2-aryl-N-alkylazetidines, as well as
the use of this azetidine-based DoM tactic for the regioselec-
tive and exhaustive functionalization of arenes.
nBuLi (2)
nHexLi (1.5)
nHexLi (1.5)
nHexLi (2)
nBuLi (2.5)
nBuLi (2)
71
–
93[b]
60
TMEDA (2.5)
TMEDA (2)
TMEDA (1.5)
40
65
65
55
60
83
95[b]
48[c]
sBuLi (1.5)
nHexLi (1.5) TMEDA (1.5)
nHexLi (1.5) TMEDA (0.1)
nHexLi (1.3) TMEDA (0.2)
nHexLi (1.3) TMEDA (0.2)
nHexLi (1.3) TMEDA (1.3)
nHexLi (1.3) TMEDA (1.3)
Results and Discussion
With the aim to harness the role of the azetidine ring as
a DMG, the reactivity of 2-aryl-substituted azetidines 1a–h was
investigated. With the exception of 1a, regioselectivity and re-
activity problems can be envisaged for each system. Azetidines
1a–h were prepared by using two simple strategies from read-
ily available azetidinones or by intermolecular cyclization of
1,3-dichloropropane derivatives (Scheme 2).[11,12]
Et2O
Et2O
[a] Yields calculated by 1H NMR spectroscopy of the crude reaction mix-
ture. [b] Yield of isolated product. [c] Deuterium was used as the electro-
phile. Only ortho-deuterated azetidine was observed. The lower yield was
ascribed to partial decomposition of nHexLi.
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12191
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
2) for benchmarking other bases against nHexLi, which is po-
tentially a more sustainable and safer alternative to other alkyl-
lithium reagents.[13] Azetidine 1a was reacted with common
lithiated bases (lithium diisopropylamide (LDA), sBuLi, and
nBuLi) then alkylated with MeI. The resulting crude reaction
mixture was analyzed by NMR spectroscopy to ascertain yield,
regiochemistry, and conversion. The use of sBuLi and LDA
either at low or high temperature for short reaction times (0.5–
1 h) was ineffective and starting material 1a was recovered
(Table 1, entries 1–5). The use of sBuLi for a longer reaction
time (16 h) in Et2O, furnished ortho-functionalized azetidine 2a
in 45% yield (Table 1, entry 6). Partial ortho-lithiation was ob-
served by using nBuLi as the base and performing the reaction
in a coordinating solvent, such as THF or Et2O (Table 1, en-
tries 7–9). Exclusively ortho-methylated azetidine 2a was ob-
tained in 80% yield by using nBuLi (2 equiv) in Et2O at 208C
(Table 1, entry 9) with a long reaction time (16 h). The use of
a non-coordinating solvent, such as toluene, at 208C (Table 1,
entry 10), resulted in lower conversion and formation of 2a in
30% yield, even after a long reaction time. The use of nHexLi
as the base gave a slightly better yield of 2a than the reaction
with nBuLi (Table 1, entry 13 versus 9). The effect of N,N,N’,N’-
tetramethylethylenediamine (TMEDA) as a ligand on the rate
of the lithiation reaction was evaluated with sBuLi, nBuLi, and
nHexLi (Table 1, entries 14–22). As expected, the use of TMEDA
accelerates the lithiation reaction to give higher conversions,
even when used in substoichiometric amounts (Table 1, en-
tries 18–20).[14] However, the use of both sBuLi/TMEDA (Table 1,
entry 16) and nBuLi/TMEDA (Table 1, entry 15) resulted in
slightly more sluggish reactions with respect to nHexLi/TMEDA
(Table 1, entry 21). In addition, Et2O was found to be the sol-
vent of choice because it gave faster and cleaner reactions
(Table 1, entry 14 versus 15 and entry 17 versus 21).
Scheme 3. Regioselective lithiation of 2-arylazetidines.
entry 21) followed by reaction with MeI. As expected, exclusive
ortho-lithiation was observed for 1b and 1c, and both reac-
tions had good regioselectivity (2b/3 and 2c/4 ratios=70:30)
with proton Ha preferentially extracted. These results are in line
with what has been reported for the open-chain analogue b-
dimethylaminomethyl naphthalene (Scheme 3).[15] Attempts to
improve the regioselectivity of the lithiation of 1b–c were
made by using different reaction conditions and more-hin-
dered lithiating agents (Scheme 3). The use of nHexLi without
the TMEDA additive (for conditions see Table 1, entry 13) re-
sulted in the same regioselectivity (2b/3=70:30) and a lower
yield. The use of lithium amides such as LDA and lithiumtetra-
methylpiperidide (LTMP) were ineffective, whereas the use of
sBuLi in the presence of TMEDA for 1 h gave slightly better re-
gioselectivity (2b/3=80:20 and 2c/4=82:18) but lower yields
(Scheme 3). In the case of 1b, lithiation with tBuLi in the pres-
ence of TMEDA at 208C for 1 h provided a better regioselectivi-
ty (2b/3=90:10) but a low yield. Attempts to prolong the lith-
iation time with both sBuLi and tBuLi, either in the presence or
absence of TMEDA, gave even lower yields, likely because of
degradation of the base at 208C. Quite surprisingly, in the case
of naphthylazetidine 1d, for which peri-lithiation might be con-
sidered, product 2d was isolated in 96% yield as the result of
a very selective ortho-lithiation (Scheme 3). This result is, in our
opinion, remarkable because it reveals an opposite regiochem-
istry to that observed with the open-chain analogue a-dimeth-
ylaminomethylnaphthalene. For the latter, peri-lithiation pre-
dominates by treatment with nBuLi in ether/hexane.[16] Howev-
er, because our protocol is dissimilar due to the presence of
the ligand TMEDA, for the sake of comparison and to assess
the role of the azetidine ring, we ran the lithiation of 1 d in
the absence of TMEDA (conditions B in Scheme 3). Trapping of
the putative lithiated intermediates of 1d with MeI again oc-
This comparative study also demonstrated that nHexLi rep-
resented a good alternative to nBuLi (or sBuLi) because it gave
similar or better reaction performance. The best conditions for
the straightforward ortho-lithiation of 1a are reported in
Table 1, entry 21. In this case, the use of TMEDA (1.3 equiv) is
mandatory to speed up the deprotonation reaction. Further,
with the aim to switch the regioselectivity, lithiation of 1a was
conducted by raising the temperature to 508C (Table 1,
entry 22). However, even with lower yield, only ortho-deuterat-
ed azetidine [D]-1a was detected. In our opinion, it is interest-
ing that two different sets of reaction conditions (Table 1, en-
tries 13 and 21) could be used for the effective ortho-lithiation
of azetidine 1a. The usefulness of two different protocols avail-
able to conduct the ortho-lithiation will be highlighted in the
functionalization of aryl azetidines (see below) with competi-
tive DMGs as aryl substituents. By using the optimized condi-
tions (Table 1, entry 21), we turned out our attention to study
the regioselectivity of the lithiation of 2-arylazetidines 1b–h.
Even though the regiocontrolled functionalization of an aro-
matic ring could be challenging, success would give the op-
portunity to obtain new target molecules by manipulation of
the introduced functionalities. Azetidines 1b–d, with two po-
tential sites for deprotonation (Ha and Hb; Scheme 3) were sub-
jected to lithiation under the optimized conditions (Table 1,
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12192
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
curred regioselectively to lead
exclusively to 2d. Thus, different
behavior can be envisaged for
the cyclic versus open-chain ter-
tiary amine.
It is worth mentioning that
even with 1-methoxynaphtha-
lene competition between ortho-
and peri-lithiation has been ob-
served (Scheme 3).[17] Aware that
halogens could promote ortho-
lithiation of the aromatic ring
mainly by means of their induc-
tive effect,[1a,3f] we also evaluated
the regioselectivity in the lithia-
tion of chloro-, bromo-, and
fluoro-substituted
azetidines
Scheme 5. Regioselective lithiation of 4-fluorophenylazetidine (1g).
1e–g (Schemes 4 and 5) by
using the optimized conditions
(Table 1, entry 21). The inductive
effect of the chlorine substituent combined with the ortho-di-
recting ability of the azetidine ring, regioselectively provided
product 2e in high yield (Scheme 4).[18] In the case of 1 f, a bro-
mine–lithium exchange occurred when nHexLi was used as the
base; a trial with lithium amide was unsuccessful.[19] The use of
the zincate base tBu2Li-Zn(TMP) (TMP=2,2,6,6-tetramethylpi-
peridide), developed by Kondo and co-workers,[20] was also in-
effective and resulted in only partial dehalogenation
(Scheme 4). Attempts to deprotonate the more challenging
bis-azetidine 1i (prepared by a Suzuki–Miyaura coupling reac-
tion from 1 f) under various lithiation conditions also failed,
likely due to its very low solubility, and only starting material
was recovered (Scheme 4).
regioselective metalation was achieved by reaction of 1g with
Schlosser’s superbase LIC-KOR (nBuLi/tBuOK) at ꢀ788C in THF
(Scheme 5).[22] Under these reaction conditions, the ortho posi-
tions with respect to the azetidine ring remained untouched
and the metalation reaction occurred smoothly at the position
adjacent to the fluorine atom to provide the methylated prod-
uct 10 in 98% yield.
Nevertheless, lithiation of 1g revealed an interesting scenar-
io to give unexpected products 5 (45%) and 6 (11%). The intri-
guing presence of 5 prompted us to suggest the involvement
of aryne 7 (Scheme 5). The ability of 7 to capture nucleophilic
species, such as nHexLi, produced the cine- (meta-6) and ipso-
substitution (para-6) products in a 70:30 ratio, respectively. It is
likely that 7 could also intercept TMEDA to give the anilinium
salt (intermediate 8), which affords 5 after b-elimination.[23,24]
Next, we considered the reactivity of methoxy-substituted
arylazetidine 1h (Scheme 6). In this case, competition between
the azetidine ring and the methoxy group was observed. In
particular, the lithiation/trapping sequence of 1h by using
either nHexLi and nBuLi in the presence of TMEDA resulted in
a mixture of products 2 f and 11. The reaction run in the pres-
Different results were obtained in the lithiation of 1g. In this
case, the fluorine substituent behaves as a better directing
group with respect to the azetidine ring. However, it is well-
documented that fluoroarenes undergo ortho-metalation.[21]
A
Scheme 4. Regioselective lithiation of 4-chlorophenyl- (1e) and 4-bromo-
phenyl- (1 f) azetidines. dppf=1,1’-bis(diphenylphoshino)ferrocene.
Scheme 6. Regioselective lithiation of methoxyphenylazetidine (1h).
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12193
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ence of TMEDA disclosed a competition between the azetidine
ring and the methoxy group in directing the lithiation reaction.
However, a slight preference in favor of the azetidine ring was
observed (Scheme 6). This result is in striking contrast to what
has been reported for the corresponding open-chain analogue
p-methoxy-N,N-dimethylbenzyl amine, for which a marked
preference for direction by the methoxy group (87:13 ratio of
regioisomers) occurred under similar reaction conditions.[25,3b]
Nevertheless, when the lithiation reaction was performed with-
out TMEDA (see Table 1, entry 12 for conditions) we were
pleased to find that azetidine 2 f was the sole reaction product
after trapping with MeI. In this case, the regioselectivity
matched that observed for lithiation of the open-chain ana-
logue p-methoxy-N,N-dimethylbenzyl amine (Scheme 6).
The scope of the methodology was assessed with the azeti-
dines able to undergo the ortho-lithiation/electrophilic-trap-
ping sequence. This protocol was successfully applied to the
regioselective ortho functionalization of lithiated azetidines
1a–e and 1h with several electrophiles, such as MeI,
Ph2MeSiCl, Br(CH2)2Br, C2Cl6, ketones, imines, aldehydes, Wein-
reb amides, Ph2PCl, and isocyanates (Scheme 7). Functionalized
azetidines 2g–z were obtained in good-to-excellent yields.
Nevertheless, poor stereoselectivity was observed for the reac-
tion with prochiral electrophiles, such as benzaldehyde and N-
Boc-benzylidenimine, which led to 2n–o, whereas the use of
cyclohexenone gave 2p as a single stereoisomer. A single re-
gioisomers, likely for steric reasons, was observed in the reac-
tions of lithiated 1b and 1c with benzophenone to afford 2t
and 2v, respectively.
Trapping ortho-lithiated azetidines with boropinacolate
(iPrOBpin) afforded the corresponding arylboronates 12a–
e (Scheme 8).[26] In all cases, the reaction proceeded with good
yields and a mixture of regioisomers (12b and 12e) was ob-
served in the lithiation/borylation sequence for azetidines 1b–
c. Such derivatives were isolated in good yields by flash chro-
matography and 11B NMR spectroscopic analysis confirmed an
intramolecular B···N dative bond, recently demonstrated by
Shipman and co-workers for azetidines tethered to boronate
esters (see the Supporting Information).[27]
Scheme 7. Scope of the azetidine-directed ortho-lithiation. [a] Inseparable
mixture of diastereoisomers. [b] The structure of the major isomer and rela-
tive configuration was assigned by single-crystal X-ray analysis. [c] A single
diastereomer was detected in the H NMR spectrum of the crude reaction
mixture.
1
Encouraged by these results, we envisioned further synthetic
applications for ortho-lithiated 2-arylazetidines. Since the cou-
pling with allyl halides is problematic because of a lithium–hal-
ogen exchange reaction, we undertook a transmetalation of
lithiated compounds of 1a and 1e with CuCN·2LiCl. The puta-
tive aryl cuprates cleanly coupled with allylic bromides to give
the desired allyl derivatives 13a–c in good yields (Scheme 9).
The reactivity described so far proves the role of the azeti-
dine ring as DMG, and reveals the reaction conditions can ad-
dress the regioselectivity when additional substituent are in-
stalled on the aromatic ring. The observed ortho-directing abili-
ty of the azetidine ring could be explained by assuming the
coordinating role of the nitrogen atom, as seen in the case of
2-arylaziridines or other Lewis basic heteroatoms that bear an
aromatic ring.[28,3–5] However, as in the case of N-alkyl-2-aryl-
aziridines (Scheme 1), the importance of dynamic phenomena
on the reactivity should be taken into consideration. We envis-
aged that N-alkyl-2-arylazetidines could also show dynamic
phenomena associated with both nitrogen inversion (NI) and
additional ring puckering (RP).[29] In particular, nitrogen inver-
sion is expected to be faster (lower barrier) with respect to the
corresponding three-membered ring, but slower (higher barri-
er) with respect to ring puckering.[30] Such a dynamic process
(Scheme 10) involves four different conformational isomers (A–
D) for N-alkyl-2-arylazetidines. Nevertheless, conformers C and
D, with the phenyl ring in a pseudo-axial arrangement, are ex-
pected to be unstable because of A1,3 ring strain (as demon-
strated for other azetidines).[31] Thus, analogously to the three-
membered ring aziridines, the most stable N-invertomers A
and B should be considered. We proposed this picture in view
of the fact that VT-NMR spectroscopy and NOESY experiments
in the range T=298–200 K indicate the presence of one main
invertomer with a trans arrangement between the N-methyl
group and the phenyl ring (see the Supporting Information).
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12194
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
former was more stable by about 4.5 kcalmolꢀ1 (see the Sup-
porting Information). For these reasons, we came to the con-
clusion that conformer A (Scheme 10) would be involved in
the lithiation reaction. If this assumption is made, then it is
likely that the nitrogen lone pair could play a role in facilitating
the lithiation, either making pre-lithiation complexes or stabi-
lizing the lithiated intermediate by chelation.[32]
To assess a possible coordinating role of the azetidine nitro-
gen atom during the deprotonation process, intra- and inter-
molecular kinetic isotope effects (KIE) were evaluated. As dem-
onstrated by the groups of Beak,[33] Collum,[34] Clayden,[35] KIE
experiments are informative as to the number of steps in-
Scheme 8. Borylation of ortho-lithiated azetidines.
Scheme 9. Allylation of ortho-lithiated azetidines.
Scheme 10. Azetidine dynamics.
Single-crystal X-ray analysis of ortho-functionalized azetidines
2k, 2o, and 2w also confirmed a trans arrangement between
the N-methyl group and the aromatic group and the puckered
conformation for the azetidine ring (Figure 1). DFT calculations
at the B3LYP/6-311++G(d,p) level of theory converged into
two minima that corresponded to conformers A and B; the
Figure 1. X-ray analysis of 2k (top), 2o (middle), and 2w (bottom); some
hydrogen atoms are omitted for clarity.
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12195
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
volved in a directed lithiation process. In particular, a difference
between the intra- and intermolecular KIE indicated that the
process likely proceeds by a two-step mechanism and that
pre-lithiation complexes could be involved. Another option
could be the kinetic acceleration imparted by the directing
group in the transition state. In both cases, complexation phe-
nomena come into play. The intramolecular KIE for the ortho-
lithiation of azetidine 1a-[D1], was evaluated either in the pres-
ence or absence of TMEDA (Scheme 11). In the experiment
executed without the ligand TMEDA, kH/kD =10. A similar value
was observed for the experiment run in the presence of
TMEDA (kH/kD =10). The intermolecular KIE was evaluated by
using almost equimolar mixtures of 1a-[D2] and 1a, again in
the presence or absence of TMEDA (Scheme 11). In these
cases, the analysis was a little more complicated but from
1H NMR spectroscopy and MS analysis it clearly emerged that
product 2a-[D1] formed predominantly and that the unreacted
starting material was enriched in 1a-[D2]. The apparent KIE
k0H/k0D =10 was found for the reaction performed without
TMEDA, whereas k0H/kD0 =5 was calculated for the reaction per-
formed in the presence of TMEDA (see the Supporting Infor-
mation). On the basis of the KIE experiments, we assume that
the ortho-lithiation of 2-arylazetidines likely proceeds by a two-
step mechanism with a fast, reversible complexation followed
by a rate-limiting deprotonation step.[36] In addition, the fact
that benzylic lithiation, which leads to a-1a-Li (Scheme 10),
was never observed, even when the lithiation reaction was
executed at high temperature (Table 1, entry 21), strengthens
the hypothesis that conformer A (Scheme 10) could be respon-
sible for the observed reactivity.[37]
atom of the azetidine ring. In all three structures, the aryl
groups occupy a pseudo-equatorial position and are trans ori-
ented with respect to the nitrogen substituent. In fact, regard-
less of the nature of the primarily introduced electrophile, in
the solid state the residual ortho proton of the aryl ring ap-
pears oriented towards the lone pair of the azetidine nitrogen.
We wondered whether such a stereochemical preference
could be retained in solution and, most importantly, if it could
be responsible for a further selective ortho-lithiation. To verify
this hypothesis we searched for this preferential conformation
1
in solution by analysis of the H NMR spectra of monofunction-
alized azetidine 2a. In this case, VT-NMR spectroscopy (T=
370–200 K) did not show any significant line broadening or
signal splitting. However, 1D-NOESY experiments (mixing
time=750 ms) with 2a carried out at 200 K revealed proximity
interactions in line with conformer A (Figure 2).[38]
To strengthen the hypothesis for a preferential conforma-
tion, we ran a conformational DFT analysis and the nature of
the more stable conformers were elucidated by comparison to
experimental and calculated NMR spectroscopic data (see the
Supporting Information). The conformations A and B, found to
be the most stable by conformational analysis, were subjected
to a full geometry optimization at the DFT-SMD/B3LYP/6-311+
+G(d,p) level of theory.[39] Vibrational analysis applied to the
stationary points proved that all of them were true minima
and provided values of free energy. Conformer A was the most
stable conformer; conformer B was higher in energy by about
2.5 kcalmolꢀ1. Subsequently, we wanted to learn whether the
conformational picture obtained from DFT geometry computa-
tions agreed with the situation observed in NMR experiments.
1
Assuming the directing ability of the azetidine ring, we ex-
plored the possibility of further functionalization of the aro-
matic group. A careful look at the X-ray structures of mono-
functionalized azetidines 2k, 2o and 2w (Figure 1), reveal a pe-
culiar arrangement of the aryl group linked to the C2 carbon
For this purpose, the H NMR spectra were calculated for both
conformers and compared with the experimental ones. Shield-
ing constants were calculated by using the GIAO-DFT method
at the SMD/MPW1PW91//B3LYP/6-311++G(d,p) level of theory;
for spin–spin coupling constants J(H,H) the B3LYP functional
with the B3LYP/6-311++G(d,p) basis set was used, for consis-
tency with our previous studies.[40]
1
The H NMR spectra calculated for the two conformers re-
vealed important differences. In particular, the simulated
1H NMR spectrum of conformer A was very similar to the ex-
perimental spectrum (see the Supporting Information). Most
likely a consequence of a preferential conformation, a marked
difference in chemical shifts can be seen for the residual ortho
protons (Ho’ in Figure 2) of conformers A and B.
Other evidence to support the role of the azetidine nitrogen
atom and the hypothesis of a preferential conformation came
from the lithiation/trapping experiment with ortho-methyl-sub-
stituted azetidines 2a and 2a-[D1] (Scheme 12), in which a com-
petitive lateral benzylic lithiation could occur. In fact, in two
separate experiments, 2a and 2a-[D1] were subjected to lithia-
tion/electrophilic trapping under the optimized conditions
(Table 1, entry 21) and only 14, the product of double function-
alization, was observed (Scheme 12). Nevertheless, the lithia-
tion of 2a-[D1] occurred much more slowly with respect to
that of 2a. This was likely a KIE and no traces of lateral trap-
1
ping product were detected by H NMR spectroscopy of the
Scheme 11. KIE studies.
crude reaction mixture. This result is in striking contrast to that
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12196
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Top) 1H NMR, middle) 1D-NOESY (Ha irradiated), and bottom) 1D-NOESY (Ho’ irradiated) spectra of 2a in [D8]toluene at 200 K.
observed for ortho-tolylaziridines, for which lateral lithiation oc-
curred even at low temperature. Such lateral regioselectivity
has been observed for other ortho-methyl-substituted arenes
with a methoxy or N,N-dimethylaminomethyl directing group
(Scheme 12).[41] We again ascribe the lack of lateral lithiation to
the preferential conformation of the aromatic ring that places
the ortho-methyl-substituent distant from the azetidine nitro-
gen coordination site.[42–44] In our opinion, the presence of
a preferential conformation would guarantee a proximity rela-
tionship to favor an ortho-lithiation.
The preferential ortho-lithiation observed for 2-arylazetidines
was further exploited for polysubstitution of the aromatic ring.
In particular, we envisaged the possibility to introduce new
substituents at a predictable position by fine-tuning the reac-
tion conditions. In the case of azetidines 2g and 2e, with an
inductively electron-withdrawing chlorine substituent on the
aryl ring, the regioselectivity could be addressed by exploita-
tion of the coordinating azetidine ring (Scheme 13). By using
the optimized conditions (Table 1, entry 21), tri- and tetra-sub-
stituted azetidines 15a–c were easily obtained in good-to-ex-
cellent yields (Scheme 13). It is worth noting that in the func-
tionalization of 2e, again the lateral benzylic lithiation did not
occur.[45]
Scheme 12. Attempts to induce lateral lithiation in ortho-tolylazetidines.
Chem. Eur. J. 2014, 20, 12190 – 12200 www.chemeurj.org
12197
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 13. DoM strategies for the functionalization of azetidine-bearing
arenes.
More challenging was the case of azetidine 2y, which bears
three coordinatively competent directing groups on the aro-
matic ring, (Scheme 13). The use of nHex-Li in THF at 208C al-
lowed for a very selective deprotonation at the most activated
position (simultaneous presence of methoxy and amide
groups). The ortho-position adjacent to the azetidine ring was
unreactive under these conditions. Trapping of the ortho-lithi-
ated intermediate 2y-Li with electrophiles furnished azetidines
17a,b with a tetra-substituted aromatic ring with a 1,2,3,4-sub-
stitution pattern. Interestingly, the use of benzophenone as
the electrophile gave rise to an intramolecular cyclization lead-
ing to functionalized phthalide derivative 17b. Aware of the
importance of fluorinated aryl compounds,[46] we used the site-
selective metalation as a tool for an exhaustive functionaliza-
tion of the aromatic ring (Scheme 14). Starting from azetidine
10 (obtained as reported in Scheme 5), reiteration of the met-
alation/trapping sequence with the superbase LIC-KOR ach-
ieved a straightforward regioselective functionalization at the
ortho positions adjacent to the fluorine atom. Trapping of the
metalated azetidine generated from 10 with electrophiles fur-
nished azetidines 18a,b in very good yields. As proof of con-
cept, the exhaustive functionalization of the aromatic ring was
attempted with azetidine 18a, which has two equivalent resid-
ual aromatic protons. Application of the ortho-lithiation/elec-
trophilic trapping sequence to 18a under the reaction condi-
tions that favor deprotonation at the ortho-position adjacent
to the azetidine ring (Table 1, entry 21), penta-substituted aze-
tidine 19 was obtained in very good yield. Surprisingly, 1D-
NOESY experiments for 19 revealed proximity interactions ac-
cording to a preferential conformation (Scheme 14) as seen in
the case of 2a (see the Supporting Information). Again such
a preferential conformation could favor the hydrogen–lithium
permutation to allow, in the case of 19, an exhaustive func-
tionalization. Thus, the aromatic ring of 19 was exhaustively
Scheme 14. Example of exhaustive functionalization of fluoroarylazetidines.
functionalized simply by application of the ortho-lithiation/
electrophilic-trapping protocol. In this way, azetidines 20a,b
with an aromatic ring fully substituted with different groups
were prepared (Scheme 14).
The last part of this work aimed to harness the role of the
azetidine ring as a DMG. We evaluated the regioselective elab-
oration of the aromatic ring of enantioenriched azetidines (S)-
1a and (S)-1e. Chiral nonracemic azetidines (S)-1a and (S)-1e
were obtained from the corresponding b-chloroketones 21a,b
by using the synthetic strategy depicted in Scheme 15. Enan-
tioselective Corey–Bakshi–Shibata (CBS) reduction[47] of 21a,b
furnished chloro alcohols 22a,b, which were transformed into
the corresponding 1,3-dichloropropane derivatives 23a,b by
a retentive nucleophilic substitution with SOCl2.[48] Cyclization
with a solution of MeNH2 (30%) in alcohol, furnished chiral aze-
tidines (S)-1a and (S)-1e, which gave functionalized azetidines
(S)-2k, (S)-2q, (S)-2s, and (S)-2z after the lithiation/electrophilic
trapping sequence without loss of optical purity (demonstrat-
ed by chiral HPLC analysis).
Conclusion
This study attempted to introduce a synthetic strategy for the
regioselective ortho-CꢀH functionalization of aryl compounds
mediated by the azetidine ring. Robust lithiated intermediates
that reacted with a broad range of electrophiles under very
mild conditions are involved in this synthetic strategy. The de-
veloped protocol uses the safer reagent nHexLi as an effective
lithiating agent at room temperature. Two different reaction
protocols were optimized for regioselective lithiation at the
ortho-positions adjacent to the azetidine ring. These conditions
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12198
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[2] For recent examples of the application of DoM in synthesis see: a) Z.
Zhao, V. Snieckus, Org. Lett. 2005, 7, 2523–2526; b) M. A. Pena, J. P. Ses-
telo, L. A. Sarandeses, J. Org. Chem. 2007, 72, 1271–1275; c) S. Wunder-
lich, P. Knochel, Org. Lett. 2008, 10, 4705–4707; d) S. N. Kessler, H. A.
Wegner, Org. Lett. 2012, 14, 3268–3271; e) C. Schneider, E. David, A.
Toutov, V. Snieckus, Angew. Chem. 2012, 124, 2776–2780; Angew. Chem.
Int. Ed. 2012, 51, 2722–2726; f) C. Schneider, E. Broda, V. Snieckus, Org.
Lett. 2011, 13, 3588–3591; g) T. E. Hurst, T. K. Macklin, M. Becker, E. Hart-
mann, W. Kugel, J.-C. Parisienne-La Salle, A. S. Batsanov, T. B. Marder, V.
Snieckus, Chem. Eur. J. 2010, 16, 8155–8161.
[3] a) M. C. Whisler, S. MacNeil, V. Snieckus, P. Beak, Angew. Chem. 2004,
116, 2256–2276; Angew. Chem. Int. Ed. 2004, 43, 2206–2225; b) J. Clay-
den in Chemistry of Organolithium Compounds, Vol. 1 (Ed.: Z. Rappo-
port), John Wiley & Sons: UK, 2004, pp. 495–646; c) T.-H. Nguyen,
N. T. T. Chau, A.-S. Castanet, K. P. P. Nguyen, J. Mortier, J. Org. Chem.
2007, 72, 3419–3429; d) S. T. Chadwick, A. Ramirez, L. Gupta, D. B.
Collum, J. Am. Chem. Soc. 2007, 129, 2259–2268; e) H. J. Reich, W. H. Si-
korski, B. O. Gudmundsson, R. R. Dykstra, J. Am. Chem. Soc. 1998, 120,
4035.
[4] For recent examples see: a) C. E. Houlden, G. C. Lloyd-Jones, K. I. Brook-
er-Milburn, Org. Lett. 2010, 12, 3090–3092; b) S. L. MacNeil, O. B. Fami-
loni, V. Snieckus, J. Org. Chem. 2001, 66, 3662–3670; c) T. K. Macklin, V.
Snieckus, Org. Lett. 2005, 7, 2519–2522.
Scheme 15. ortho Functionalization of enantioenriched 2-arylazetidines.
served as toolbox when competing directing groups were in-
stalled on the aromatic ring. The role of the azetidine nitrogen
atom was assessed, as well as the effect of dynamic phenom-
ena related to nitrogen inversion and preferential conforma-
tions of the 2-aryl substituent. Both factors account for the
lack of benzylic lithiation that occurred with the lower 2-aryl-
aziridines analogues. The possibility to plan a site-selective
functionalization of the aromatic ring has been demonstrated
with aryl azetidines that bear a coordinatively competent
group (e.g. the methoxy group in 2y), or inductively electron-
withdrawing substituents (e.g. chlorine and fluorine in 2g, 2e,
and 10). By using appropriate reaction conditions in terms of
solvent, temperature and metalating agent, a regioselective in-
troduction of substituents has been realized even in the pres-
ence of competing directing groups. Several substitution pat-
terns for the aromatic ring could be accomplished, including
1,2,3-trisubstitution, 1,2,3,4-tetrasubstitution, 1,2,3,4,5-penta-
substitution, and the exhaustive substitution with six groups
(derivatives 20a,b).
[5] The challenging meta substitution has also been recently accomplished,
see: a) D. Leow, G. Li, T.-S. Mei, J.-Q. Yu, Nature 2012, 486, 518–522;
b) H. A. Duong, R. E. Gilligan, M. L. Cooke, R. J. Phipps, M. J. Gaunt,
Angew. Chem. 2011, 123, 483–486; Angew. Chem. Int. Ed. 2011, 50,
463–466; c) R. J. Phipps, M. J. Gaunt, Science 2009, 323, 1593–1597.
[6] V. Capriati, S. Florio, R. Luisi, B. Musio, Org. Lett. 2005, 7, 3749–3752.
[7] a) S. Florio, R. Luisi, Chem. Rev. 2010, 110, 5128–5157; b) R. Luisi, V. Cap-
riati, S. Florio, B. Musio, Org. Lett. 2007, 9, 1263–1266; c) V. Capriati, S.
Florio, R. Luisi, A. Mazzanti, B. Musio, J. Org. Chem. 2008, 73, 3197–
3204; d) R. Luisi, V. Capriati, P. Di Cunto, S. Florio, R. Mansueto, Org. Lett.
2007, 9, 3295–3298.
[8] a) F. Affortunato, S. Florio, R. Luisi, B. Musio, J. Org. Chem. 2008, 73,
9214–9220; b) M. C. de Ceglie, B. Musio, F. Affortunato, A. Moliterni, A.
Altomare, S. Florio, R. Luisi, Chem. Eur. J. 2011, 17, 286.
[9] a) L. Degennaro, M. Zenzola, P. Trinchera, L. Carroccia, A. Giovine, G. Ro-
manazzi, A. Falcicchio, R. Luisi, Chem. Commun. 2014, 50, 1698–1700;
b) L. Carroccia, L. Degennaro, G. Romanazzi, C. Cuocci, L. Pisano, R.
Luisi, Org. Biomol. Chem. 2014, 12, 2180–2184.
[10] a) Y. Han, M. Han, D. Shin, C. Song, H.-G. Hahn, J. Med. Chem. 2012, 55,
8188–8192; b) H. Brꢁuner-Osborne, L. Bunch, N. Chopin, F. Couty, G.
Evano, A. A. Jensen, M. Kusk, B. Nielsen, N. Rabasso, Org. Biomol. Chem.
2005, 3, 3926; c) D. Honcharenko, C. Zhou, J. Chattopadhyaya, J. Org.
Chem. 2008, 73, 2829.
[11] A. Brandi, S. Cicchi, F. M. Cordero, Chem. Rev. 2008, 108, 3988–4035.
[12] See the Supporting Information for experimental details.
[13] We believe that nHexLi is a safer substitute for nBuLi, with several ad-
vantages: 1) it gives more stable and safer solutions; it decomposes
slowly with respect to nBuLi; 2) it is a non-pyrophoric strong base;
3) the deprotonation reaction gives n-hexane which is less volatile and
has a higher flash point than butane generated with nBuLi; 4) it is safer
for large-scale applications. NMR spectroscopic investigations of the ag-
gregation state of nHexLi (expected to be similar to nBuLi) are under-
way.
[14] For the use of catalytic amount of TMEDA see: D. W. Slocum, T. K. Rein-
scheld, C. B. White, M. D. Timmons, P. A. Shelton, M. G. Slocum, R. D.
Sandlin, E. G. Holland, D. Kusmic, J. A. Jennings, K. C. Tekin, Q. Nguyen,
S. J. Bush, J. M. Keller, P. E. Whitley, Organometallics 2013, 32, 1674–
1686.
Acknowledgements
We thank National Project “FIRB - Futuro in Ricerca” (code:
CINECA RBFR083M5N); Regional project “Apulian food finger-
prints” Code: 68, Interuniversity Consortium CINMPIS. We are
grateful to Dr. Giovanna Parisi for preparation of the starting
materials used in the present study.
Keywords: azetidines · dynamic reactivity · heterocycles ·
lithiation · regioselectivity
[15] For a recent example see: J. T. B. H. Jastrzebski, A. M. Arink, H. Kleijn,
T. W. Braam, M. Lutz, A. L. Spek, G. van Koten, J. Am. Chem. Soc. 2013,
135, 13371–13378.
[16] a) R. L. Gay, C. R. Hauser, J. Am. Chem. Soc. 1967, 89, 2297–2303; b) J.
Clayden, C. S. Frampton, C. McCarthy, N. Westlund, Tetrahedron 1999,
55, 14161–14184.
[17] D. A. Shirley, C. F. Cheng, J. Organomet. Chem. 1969, 20, 251.
[18] For an example of chlorine-directed lithiation see: M. Iwao, J. Org.
Chem. 1990, 55, 3622–3627.
[1] a) C. Hartung, V. Snieckus, in Modern Arene Chemistry, (Ed.: D. Astruc),
WILEY, New York, 2002, p. 330; b) T. K. Macklin, V. Snieckus, in Handbook
of CꢀH Transformations, Vol. 1, (Ed.: G. Dyker) WILEY-VCH, Weinheim,
2005, pp. 106–118; c) V. Snieckus, Chem. Rev. 1990, 90, 879–933; d) P.
Beak, A. Basu, D. J. Gallagher, Y. S. Park, S. Thayumanavan, Acc. Chem.
Res. 1996, 29, 552–560; e) P. Beak, A. I. Meyers, Acc. Chem. Res. 1986,
19, 356–363; f) M. Schlosser, Organometallics in Synthesis: A Manual,
Wiley, Chichester, 1994.
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12199
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[19] Lithium amides can be employed in ortho-lithiations, see: a) A. C.
Hoepker, L. Gupta, Y. Ma, M. F. Faggin, D. B. Collum, J. Am. Chem. Soc.
2011, 133, 7135–7151; b) A. J. Bridges, A. Lee, E. C. Maduakor, C. E.
Schwartz, Tetrahedron Lett. 1992, 33, 7499–7502.
[20] a) Y. Kondo, M. Shilai, M. Uchiyama, T. Sakamoto, J. Am. Chem. Soc.
1999, 121, 3539–3540; b) D. Tilly, F. Chevallier, F. Mongin, P. C. Gros,
Chem. Rev. 2014, 114, 1207–1257.
[21] a) M. Schlosser, Angew. Chem. 2005, 117, 380–398; Angew. Chem. Int.
Ed. 2005, 44, 376–393; b) G. W. Gribble, Lithium Compounds. In Science
of Synthesis Section 8.1.14, Thieme: Stuttgart, 2006, pp. 357–426.
[22] A similar competition has been reported for fluoroaryls that bear a me-
thoxy substituent, see: a) M. Schlosser, G. Katsoulos, S. Takagishi, Synlett
1990, 747; b) G. Katsoulos, S. Takagishi, M. Schlosser, Synlett 1991, 731.
[23] A mixture of isomers were detected by GC-MS analysis and 1H NMR
spectroscopy of the crude reaction mixture. However, only the meta
isomer (cine substitution) was isolated. This reaction does not occur at
low temperatures and all attempts to capture the aryne 7 with a diene
(furan) failed.
[35] J. Clayden, J. H. Pink, N. Westlund, F. X. Wilson, Tetrahedron Lett. 1998,
39, 8377–8380.
[36] In the intermolecular KIE experiment, an effect of the ligand TMEDA
was observed. The role of TMEDA is not well understood at the
moment and a change in the mechanism for the H/Li permutation
cannot be ruled out. For similar studies see: M. Stratakis, J. Org. Chem.
1997, 62, 3024–3025. However, structural studies by NMR spectroscopy,
DOSY, and X-ray crystallography are underway, with the aim to shed
light on this process and on the nature of the lithiated intermediates.
[37] F. Couty, G. Evano, Synlett 2009, 3053–3064.
[38] a) K. Scott, J. Stonehouse, J. Keeler, A. J. Shaka, J. Am. Chem. Soc. 1995,
117, 4199–4200; b) D. Neuhaus, M. Williamson, The Nuclear Overhauser
Effect in Structural and Conformational Analysis; VCH, Weinheim 1989,
p. 264.
[39] R. Ribeiro, A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Chem. Theory
Comput. 2009, 5, 2284–2300.
[40] U. Azzena, G. Dettori, L. Pisano, B. Musio, R. Luisi, J. Org. Chem. 2011,
76, 2291–2295.
[41] a) F. N. Jones, M. F. Zinn, C. R. Hauser, J. Org. Chem. 1963, 28, 663;
b) R. L. Vaulx, F. N. Jones, C. R. Hauser, J. Org. Chem. 1964, 29, 1387.
[42] Restricted rotations and preferential conformations have been reported
for several systems, see: a) C. Wolf, Dynamic Stereochemistry of Chiral
Compounds. Principles and Applications 2008, RSC, Cambridge (UK),
Chapter 3; b) Example of preferential conformations in substituted aziri-
dines have been recently reported by us: M. C. de Ceglie, L. Degennaro,
A. Falcicchio, R. Luisi, Tetrahedron 2011, 67, 9382–9388.
[43] We also observed a similar deshielding for the residual ortho protons in
most of the prepared functionalized azetidines. In some cases, such
a preferential conformation has been confirmed by NOESY experiments;
see the spectra of derivatives 2a, 2g, and 19 (see the Supporting Infor-
mation).
[24] For recent examples of aryne formation and addition of nucleophiles to
benzyne see: a) E. R. Biehl, S. P. Khanapure, Acc. Chem. Res. 1989, 22,
275; b) H. Hart, In The Chemistry of Functional Groups, Suppl. C2: The
Chemistry of Triple-bonded Functional Groups; Potai, S., Ed.; Wiley: Chi-
chester 1994; pp. 1017–1134; c) J. Pawlas, M. Begtrup, Org. Lett. 2002,
4, 2687–2690; d) S. Tripathy, R. LeBlanc, T. Durst, Org. Lett. 1999, 1,
1973.
[25] D. W. Slocum, C. A. Jennings, J. Org. Chem. 1976, 41, 3653.
[26] In the lithiation/borylation of 1b–c a 70:30 mixture of regioisomers was
obtained.
[27] a) C. T. Hoang, I. Prokes, G. J. Clarkson, M. J. Rowland, J. H. R. Tucker, M.
Shipman, T. R. Walsh, Chem. Commun. 2013, 49, 2509–2511; b) R. Luisi,
A. Giovine, S. Florio, Chem. Eur. J. 2010, 16, 2683–2687.
[44] O. Kitagawa, M. Fujita, M. Kohriyama, H. Hasegawa, T. Taguchi, Tetrahe-
dron Lett. 2000, 41, 8539–8544.
[45] Even in this case we assume that a preferential conformation is respon-
sible for the lack of benzylic lithiation.
[28] D. Tilly, J. Magolan, J. Mortier, Chem. Eur. J. 2012, 18, 3804–3820.
[29] a) J. B. Lambert, W. L. Oliver, B. S. Packard, J. Am. Chem. Soc. 1971, 93,
933–937; b) F. A. L. Anet, J. Am. Chem. Soc. 1985, 107, 4335–4337; c) R.
Dutler, A. Rauk, T. S. Sorensen, J. Am. Chem. Soc. 1987, 109, 3224–3228.
[30] The energy required for the N-inversion is expected to be in the range
9–12 kcalmolꢀ1, much higher than that required for the ring puckering
(2–3 kcalmolꢀ1), see: A. R. Katritzky, C. A. Ramsden, J. Joule, V. V. Zhdan-
kin, in Handbook of Heterocyclic Chemistry (Third Edition). Elsevier,
Oxford, 2010; pp. 210–237.
[31] J. Catalꢂn, O. Mo, M. Yꢂn¯ez, J. Mol. Struct. 1978, 43, 251–257.
[32] For structural studies on chelated ortho-lithiated arenes see: a) K. L.
Jantzi, I. A. Guzei, H. J. Reich, Organometallics 2006, 25, 5390–5395;
b) H. J. Reich, W. S. Goldenberg, B. O. Gudmundsson, A. W. Sanders, K. J.
Kulicke, K. Simon, I. A. Guzei, J. Am. Chem. Soc. 2001, 123, 8067–8079;
c) H. Reich, Chem. Rev. 2013, 113, 7130–7178.
[46] a) C. Hollingworth, V. Gouverneur, Chem. Commun. 2012, 48, 2929–
2942; b) C. T. Furuya, J. E. M. N. Klein, T. Ritter, Synthesis 2010, 1804–
1821; c) J. Wang, M. Sꢂnchez-Rosellꢃ, J. L. AceÇa, C. del Pozo, A. E. Soro-
chinsky, S. Fustero, V. A. Soloshonok, H. Liu, Chem. Rev. 2014, 114,
2432–2506; d) I. Ojima, J. Org. Chem. 2013, 78, 6358–6383.
[47] E. J. Corey, C. J. Helal, Angew. Chem. 1998, 110, 2092–2118; Angew.
Chem. Int. Ed. 1998, 37, 1986–2012.
[48] a) W. A. Cowdrey, E. D. Hughes, C. K. Ingold, S. Masterman, A. D. Scott, J.
Chem. Soc. 1937, 1252–1271; b) E. S. Lewis, C. E. Boozer, J. Am. Chem.
Soc. 1952, 74, 308; c) D. J. Cram, J. Am. Chem. Soc. 1953, 75, 332–338;
d) C. C. Lee, J. W. Clayton, D. G. Lee, A. J. Finlayson, Tetrahedron 1962,
18, 1395–1402.
[33] D. R. Anderson, N. C. Faibish, P. Beak, J. Am. Chem. Soc. 1999, 121,
7553–7558.
[34] R. A. Rennels, A. J. Maliakal, D. B. Collum, J. Am. Chem. Soc. 1998, 120,
421–422.
Received: April 17, 2014
Published online on August 5, 2014
Chem. Eur. J. 2014, 20, 12190 – 12200
www.chemeurj.org
12200
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim