Radical-triggered three-component coupling reaction of alkenylboronates, α-halocarbonyl compounds, and organolithium reagents: The inverse ylid mechanism

Article abstract of DOI:10.1002/chem.201802384

An operationally simple protocol to affect a radical addition to alkenylboronates that spontaneously undergo a [1,2]-metalate shift is described. Overall, the reaction is a three-component coupling of an organolithium, alkenylboronic ester, and halide which takes place with broad scope and good to excellent yields. Experimental mechanistic investigations support the formation of a boron inverse ylid intermediate.

Full text of DOI:10.1002/chem.201802384

A Journal of
Accepted Article
Title: Radical Triggered Three-Component Coupling Reaction
of Alkenylboronates, α-Halocarbonyl Compounds and
Organolithium Reagents: The Inverse Ylid Mechanism
Authors: Philippe Renaud, Nicholas David Charles Tappin, and Manuel
Gnägi-Lux
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201802384
Link to VoR: http://dx.doi.org/10.1002/chem.201802384
Supported by

10.1002/chem.201802384
Chemistry - A European Journal
Radical Triggered Three-Component Coupling Reaction of Alkenylboronates, α-
Halocarbonyl Compounds and Organolithium Reagents: The Inverse Ylid Mechanism
Nicholas David Charles Tappin, Manuel Gnägi-Lux, Philippe Renaud*
University of Bern, Department of Chemistry and Biochemistry, Freiestrasse 3, 3012 Bern,
Switzerland
philippe.renaud@dcb.unibe.ch
Abstract. An operationally simple protocol to affect a radical addition to alkenylboronates
that spontaneously undergo a [1,2]-metalate shift is described. Overall, the reaction is a
three-component coupling of an organolithium, alkenylboronic ester, and halide which
takes place with broad scope and good to excellent yields. Experimental mechanistic
investigations support the formation of a boron inverse ylid intermediate.
R1
R1
R2
radical
initiator
Bpin
R3
R3
R1
Bpin
Bpin
R2
R2–Li
R3–X
inverse
boron ylid
Introduction
Radical reactions have proven over the last decades to be a powerful tool in organic
synthesis.[1-4] The intermolecular addition of carbon-centered radicals to alkenes is one of
the mildest approaches for carbon–carbon bond formation.[5] Three component reactions
involving either two consecutive radical reactions, or one radical reaction followed by an
ionic reaction are currently attracting a lot of attention.[6-7] Our long-standing interest for
radical reactions involving organoboron species prompted us recently to investigate the use
of a-chloroalkylboronates in radical reactions and this led to the discovery of an
intramolecular cyclopropanation reaction.[8-9] In these studies, the a -chloroalkylboronates
were prepared according to Matteson's procedure, i.e. addition of the
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
dichloromethyllithium to an alkylboronate. In order to find a different way to prepare a -
haloalkylboronates, we decided to investigate the atom transfer mediated radical addition
to alkenylboron derivatives. This approach was first investigated by Matteson who reported
Kharasch type addition of radicals to vinylboronates.[10-11] Interestingly, Matteson
discovered that the product of CCl3Br radical addition to dibutyl vinylboronate afforded an
α-arylated boronate upon treatment with aryl Grignard reagents such as 1-
mesitylmagnesium bromide (Scheme 1, a).[12] The discovery of this now eponymous [1,2]-
metalate rearrangement[13-14] opened a fruitful field of research as demonstrated by the
work of Aggarwal,[15-18] Fu,[19] and Morken.[20-21] Recently, Zard reported a radical mediated
xanthate transfer addition to MIDA-vinyl boronates and demonstrated the beneficial effect
of complexing the Lewis acidic boron atom by a tertiary amino group (Scheme 1, b).[22] The
complex formation speeds-up the addition of electrophilic radicals to the vinyl group.
Moreover, it accelerates the xanthate transfer step since the radical adduct is no longer
stabilized by a vacant p-orbital. These results strongly suggest that boronate complexes,
that are negatively charged and are missing the empty p-orbital at boron should also act as
efficient radical traps for halogen atom transfer processes. Based on this assumption, we
decided to investigate the reactivity of alkenylboron ate complexes as radical trap in order
to develop a three-component reaction involving a radical addition to a vinylboron ate
complex followed by a halogen transfer and a final [1,2]-metalate rearrangement (Scheme
1, c). At an advanced stage of our study, Studer, [23-24] Aggarwal[25-26] and Morken[27]
reported very closely related transformations. We report here a single-pot approach
involving neither solvent switch nor the use of any additive (photoredox catalyst or co-
solvent). Mechanistic evidence for the formation of an inverse ylid intermediate is provided.
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
a) 1963: Matteson (3 components, 2 steps)
Br
Ar
ArMgBr
CCl3Br
On-Bu
B
Cl3C
On-Bu
Cl3C
On-Bu
[1,2]-shift
AIBN
B
B
On-Bu
On-Bu
On-Bu
b) 2016: Zard (2 components, 1 step)
O
S
S
O
S
OEt
NC
S
OEt
B
N
Me
DLP
NC
B(MIDA)
O
O
c) 2017: Studer, Aggarwal, and this work (3 components, 1 step)
R1
R1
R2
2) R3–X
initiation
1) R2–Li
Li
BPin
R2
R3
BPin
R1
BPin
Scheme 1. Radical mediated two and three component coupling processes involving
vinylboronate.
Results and Discussion
Optimization and scope of the reaction. The coupling reaction between ethyl iodoacetate,
vinylboronic acid pinacol ester 1a (vinylBpin) and phenyllithium was attempted first by
addition of PhLi (1.05 equiv) to a solution of the vinylBpin in THF at 0 °C. Ethyl
iododoacetate (2 equiv) was added to the mixture at ambient temperature followed by Et3B
(2 equiv) and the reaction flask was open to air to initiate the radical process. Under these
simple reaction conditions, the boronate 2aa resulting from the desired three-component
coupling reaction was obtained with an encouraging 71% yield (Scheme 2).
1) PhLi (1.05 equiv), 0 °C to rt
2) ICH2CO2Et (2 equiv)
Et3B (2 equiv), air, rt
BPin
THF
Ph
EtO2C
Bpin
2aa 71%
1a
Scheme 2. Discovery of reaction.
The reaction was then extended to other organolithium derivatives (eq. 2, Table 1). The use
of n-butyllithium afforded the boronate 4aa in 75% yield under the same conditions (Table
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
1, entry 2). However, with secondary and tertiary alkyllithiums the boronates 6aa and 7aa
were obtained in low yields (Table 1, entries 5,6). The used of di-tert-butylhyponitrite
(DTBHN) at 60 °C increased the yield and the reproducibility (Table 1, entry 7). The lower
yields observed with secondary and tertiary alkyllithiums relative to the primary ones
suggest that a competing reaction is operating with the bulkier substrates. We hypothesized
that the ethoxycarbonylmethyl radical was abstracting a hydrogen atom from
tetrahydrofuran leading to the formation of HI by decomposition of the unstable 2-
iodotetrahydrofuran. This process is leading to non-productive consumption of the starting
iodoester and more importantly to decomposition of the vinylboron-ate complex by
protonation. The fact that this competing reaction is more important for t-BuLi than for n-
BuLi suggests that the t-BuLi·1a boronate complex is less reactive towards radicals than the
n-BuLi·1a complex. This was experimentally confirmed by a competition experiment
involving ethyl iodoacetate, vinylboronate 1a (4.4 equivalents) and 2.2 equivalents of both
n-BuLi and t-BuLi. The reaction afforded a 9:1 mixture of 5aa and 7aa in 40% yield (see
supporting information).[28] A straightforward solution is to use a solvent with lower
hydrogen atom donor properties.[29] Switching to the weakest possible hydrogen atom
donor etheral solvent, i.e. tert-butyl methyl ether (TBME) gave enhanced results both with
n-BuLi (Table 1, entry 3, 92% yield) and t-BuLi (Table 1, entry 8, 68% yield).[30] Testing the
reaction with simply AIBN as the initiator and benzene as the solvent gave good results
except in the case of t-BuLi complexes (Table 1, entries 4, 9).[31]
Table 1. Optimization of the initiation and the solvent for the one-pot three component
coupling reaction between vinylBPin, ICH2CO2Et, and RLi species.
initiating system
1) RLi (1.05 equiv), 0 °C to rt
2) ICH2CO2Et (2.0 equiv)
initiating conditions
R
A: BEt3/ air
B: BEt3/ DTBHN (30 mol%)
C: AIBN (10-30 mol%)
BPin
EtO2C
Bpin
solvent (0.2 M)
1a
Entry
RLi
Product
Initiating
Solvent
Temp.
Reaction Yield
system
time
16 h
16 h
1
2
PhLi
2aa
4aa
A
A
THF
THF
25 °C
25 °C
71%
75%
n-BuLi
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
3
4
5
6
7
8
9
n-BuLi
n-BuLi
s-BuLi
t-BuLi
t-BuLi
t-BuLi
t-BuLi
4aa
4aa
6aa
7aa
7aa
7aa
7aa
B
C
A
A
B
B
C
TBME
benzene
THF
60 °C
85 °C
25 °C
25 °C
60 °C
60 °C
85 °C
1.5 h
92%
78%
17%
21%
44%
68%
23%
2 h
16 h
16 h
1.5 h
1.5 h
2 h
THF
THF
TBME
benzene
The reaction conditions optimized above (conditions B in TBME) have been tested with
vinylboronic and 2-propenylboronic acid pinacol esters 1a and 1b, different halides (usually
iodides) and organolithium reagents. The results are summarized in Scheme 3. The reaction
of iodoacetonitrile with PhLi·1a afforded the desired nitrile 2ac in only 33% yield.
Performing the reaction at room temperature (air initiation) dramatically improved the yield
to 76%. Remarkably, the reaction of diethyl 2-bromo-2-methylmalonate with PhLi·1a
worked fine and afforded the diester 2ad in 68% yield.[32] The vinylboronate derived from
(+)-pinanediol afforded 4ca as 1:1 mixture of diastereomers. Reactions involving secondary
and tertiary organolithium were then investigated. Pleasingly, our method works fine with
these stubborn couplings reactions. Reactions involving s-BuLi and i-PrLi and ethyl
iodoacetate provided the coupling products in 52–85% yield. Similarly, the reaction of
PhSO2CH2I and EtO2CCH2I with t-BuLi·1a and t-BuLi·1b afforded 7aa, 7ba and 7ab in 62–69%
yield.[33] Careful product analysis of the coupling leading to 7aa showed the presence of the
more polar borinic ester 8 (7%) which was isolated and characterized. Such a borinic ester
has been identified in the homologation of a bulky tertiary boronic ester with
chloromethyllithium.[34] The reaction with iodoacetonitrile gave a low yield with the
vinylboronate 1a (7ac, 24%) but a satisfactory yield was obtained with the more reactive
isopropenylBpin 1b (54% for 7bc).[35] As can be generally seen in all the examples, and
especially when comparing 7ac to 7bc, the methyl group increases the radical trap ability of
the boronate complex. This effect was experimentally confirmed by a competition
experiment involving a 1:1 mixture of 1a/1b (excess), n-BuLi and EtO2CCH2I that gave
4aa/4ba as 1:1.5 mixture (see supporting information).[36]
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
1) R2Li (1.05 equiv), 0 °C to rt
2) EWG–CX2–I (2.0 equiv)
Et3B (2.0 equiv), DTBHN, 60 °C
R1
R2
X X
EWG
Bpin
1a (R1 = H)
2a (R1 = Me)
Bpin
R1
TBME (0.2 M), 1.5 h
PhLi
Ph
Ph
Ph
Ph
Ph
EtO2C
EtO2C
EtO2C
Bpin
EtO2C
Bpin
Bpin
PhO2S
2ab 64%
Bpin
NC
Bpin
Bpin
2ac 33%, 76%a
2ba 95%
2aa 87%
2ad 68%b
n-RLi
n-Bu
Bpin
n-Bu
n-Bu
Bpin
EtO2C
Bpin
EtO2C
EtO2C
4aa 92%
EtO2C
4ba 93%
Bpin
PhO2S
3ba 94% (27%)b
3aa 87%
4ab 85%
n-Bu
F F
EtO2C
n-Bu
n-Bu
Bpin
n-Bu
Bpin
BnO2C
n-Bu
Bpin
EtO2C
EtO2C
Bpin
NC
Bpin
4ac 80%
BnO2C
4ae 68%
4af 59%
4be 89%
4ad 69%b
n-Bu
n-Bu
n-Bu
n-Bu
F F
Cl3C
O
Cl3C
EtO2C
B
Bpin
Bpin
EtO2C
Bpin
O
4ca 86%
d.r. = 1:1
4bg 79%b
4ag 59%b
4bf 76%
s-RLi
i-Pr
i-Pr
EtO2C
5aa 52%
d.r. = 1:1
Bpin
EtO2C
6aa 54%
Bpin
EtO2C
6ba 85%
Bpin
t-RLi
t-Bu
t-Bu
t-Bu
O
PhO2S
Bpin
EtO2C
Bpin
EtO2C
Bpin
O
EtO2C
B
7aa 68%
7ba 62%c
7ab 69%
t-Bu
8 7%
t-Bu
t-Bu
t-Bu
t-Bu
F F
EtO2C
Cl3C
NC
Bpin
7bc 54%c
Bpin
NC
Bpin
7ac 24%
Bpin
7bg 19%b,c
7bf 30%
Scheme 3. Scope of the three-component coupling reaction. Reactions were performed on a
1.0 mmol scale and all yields reported are isolated after silica gel chromatography. a)
Initiated with air at 25 °C; b) starting from the corresponding bromide; c) boronate complex
prepared at –78 °C to avoid allylic deprotonation.
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
Mechanistic investigations. The three-component coupling process may follow different
(competing) pathways as summarized in Scheme 4. Pathway 1 involves an atom transfer
radical addition reaction leading to an a-haloboronate B that undergoes the classical [1,2]-
metalate rearrangement. In pathway 2, the radical anion A can reduce the starting halide
via a single electron transfer (SET) process that leads to an inverse ylid C which subsequently
undergoes a fast migration reaction. Recombination of C with the halide anion would
provide B and cannot be excluded, but is highly unlikely (intermolecular process vs.
intramolecular 1,2-shift, low iodide concentration). Alternatively, in pathway 3, A could
undergo a [1,2]-alkyl- or aryl-migration[37][38-39] leading to a boryl radical-anion D which
could then propagate the chain via a reductive SET. Indeed, boryl radical anions such as
•–
BH3 and BH2CN•– are known to abstract efficiently iodine atom presumably via a SET
process.[40]
R3–X
R3
X
Li
R3
Bpin
R2
atom transfer
[1,2]-migration
B (α-haloboronate)
pathway 1
LiX
R3–X
R3 + LiX
R2
Li
R3
R3
R3
Li
R3
Bpin
R2
Bpin
R2
Bpin
R2
[1,2]-migration
Bpin
SET
A
C (inverse boron ylid)
pathway 2
pathway 3
R2
R3–X
SET
R3 + LiX
Li
R3
Bpin
rad. [1,2]-migration
D (boryl radical-anion)
Scheme 4. Three possible mechanisms.
In order to gain more insight into the mechanism, a radical clock experiment with the 3,3-
(dimethylprop-2-en-1yl)malonate 9 was used to probe the lifetime of the radical
intermediate A. This reagent has been used recently for efficient radical annulations.[41-43]
The reaction between 1a, n-BuLi, and bromomalonate 9 afforded exclusively the product of
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
[1,2]-metalate rearrangement in 63% yield (Scheme 5, I). No product resulting from a radical
cyclization process could be detected. Since the 5-exo-trig cyclization of the intermediate
hexenyl radical is expected to be very fast, this result supports the SET mechanism (Scheme
4, pathway 2 or 3). Indeed, the rate constant for the radical cyclization can be estimated to
be around 107 s–1 at room temperature[44] and the rate constant for the atom transfer step
involving a primary alkyl radical and diethyl 2-bromo-2-methylmalonate has been reported
to be 0.75–1.3 × 106 M–1 s–1 at 50 °C.[45] Under our reaction conditions ([9] ≤ 0.4 M), the
bromine atom transfer process is not expected to compete with the cyclization process. The
second experiment supporting pathway 2 was the coupling of 1a, n-BuLi and either the
bromide 10a or the selenide 10b (Scheme 5, II). The bromide 10a afforded the product 4aj
in 18% yield whereas neither 4aj nor a product derived from a PhSe group transfer process
was obtained from the reaction of 1a with 10b and the starting selenoacetate was not fully
consumed. The lack of product resulting from SePh transfer process and low conversion
speaks against the atom transfer pathway 1. A reaction via pathway 3 should have been
working with both substrates since the rate transfer of bromine atoms and SePh groups are
very close.[46] However, the reduction potentials of these two substrates are very different.
A bromide being much easier to reduce than a selenide explains why 10b cannot efficiently
react via a SET mechanism. Then, an experiment was designed to probe the feasibility of a
radical 1,2-migration with R2 = phenyl group to maximize the chance to observe a radical
1,2-migration (neophyl type rearrangement).[47-48] For this purpose, the selenide 11 was
converted to its ate complex by addition of PhLi and treated with Bu3SnH and AIBN (Scheme
5, III). Using PhLi resulted in no radical coupling products. However with the use of n-BuLi,
and under these reductive conditions, tetrahydrofurans 12 and 13 resulting from a 5-exo-
trig cyclization followed by reduction of the α-boronate radical were the only isolated
products. Protonolysis of the cyclized ate complex was expected to be unselective leading,
after oxidative work-up, to 12 and 13. No product arising from a [1,2]-migration such as 14
could be observed. Finally, the formation of the borinate byproduct 8 during the coupling
involving 1a, t-BuLi and Et2OCH2I is compatible with the inverse ylid pathway 2 but not with
the radical 1,2-migration proposed in pathway 3. A concerted SET–[1,2]-migration
mechanism cannot be completely ruled out at the moment. However, the isolation of the O-
migration product 8 in the reaction involving the bulky tert-butyllithium (Scheme 5, IV)
tends to support the formation of an inverse ylid intermediate followed by a fast [1,2]-shift.
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
Indeed, the borinate 8 is much less stable than the boronate 7aa (a B–O bond is much
stronger than a B–C bond). In a concerted mechanism, the migrating B–C respectively the B–
O bonds should be already partly cleaved in the transition state. Therefore, C-migration is
expected to be highly favored. On the other hand, the chemoselectivtiy of the [1,2]-
migration of the inverse ylid is expected to be controlled by its conformation. The presence
of a bulky tert-butyl group at boron destabilizes the major conformation leading to C-
migration making the minor conformation leading to O-migration competitive. Similar
effects have been observed in the case of 1-chloro- and 1-bromoalkylboronate complex, the
more polarized 1-chloro derivative favoring the O-migration compared to the 1-
bromoderivative.[34] Based on these results, a mechanism involving an electron transfer
process leading to the formation of an inverse boron ylid of type C that undergoes a fast
rearrangement is privileged (pathway 2). Both Studer[23] and Aggarwal[25] proposed a similar
mechanism for their reactions but the electron transfer process could only be
experimentally demonstrated when the strongly oxidizing Togni's reagent was used to
generate a trifluoromethyl radical.[23]
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
I
1) n-BuLi (1.05 equiv)
2) 9 (2.0 equiv)
Et3B (2.0 equiv)
DTBHN, 60 °C
n-Bu
Bpin
Bpin
MeO2C
MeO2C
MeO2C
MeO2C
1a
TBME
9
Br
4ai 63%
no cyclization
observed
Li+
Bpin
MeO2C
MeO2C
n-Bu
II
1) n-BuLi (1.05 equiv)
2) 10a or 10b (2.0 equiv)
Et3B (2.0 equiv)
n-Bu
Bpin
NC
X
10a X = Br
10b X = SePh
AIBN, 85 °C
Bpin
1a
NC
TBME
4aj
from 10a: 21%
from 10b: no product
III
1) PhLi (1.05 equiv)
2) Bu3SnH (1.3 equiv)
AIBN, 85 °C
OH
O
O
+
Ph
Ph
O
Bpin
3) H2O2, NaOH
12 34%
trans/cis 10:3
13 5–10%
trans/cis 12:3
PhSe
Ph
11
Ph
H2O2
NaOH
BPin
Li+
Bu3SnH
O
12 + 13
H
Ph
Ph
Ph
BPin
Li+
OH
Ph
O
O
H2O2
NaOH
Ph
BPin
Li+
Ph
O
Ph
14 (not observed)
IV
t-Bu
O
t-BuLi, ICH2CO2Et
+
Bpin
O
Scheme 3
EtO2C
7aa 68%
Bpin
EtO2C
B
8 7% t-Bu
1a
steric repulsion
+
O
+
O
+
H
R
O
R
+
R
H
B
B
O
O
H
R
O
O
H
O
O-migration favored
C-migration favored
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
Scheme 5. Mechanistic investigations.
Conclusions
We have developed an experimentally simple procedure to couple an alkyl iodide (or
bromide) activated by an electron withdrawing group to a alkenylboronic ester and an
organolithium species. The reaction affords functionalized α-branched alkylboronate esters
in good to excellent yield. This procedure does not require the use of a catalyst and it can be
performed in a one pot-process without solvent switch. Strong evidence for the formation
of an inverse boron ylid intermediate is presented and provides an attractive foundation for
future developments.
Experimental Section
General Procedure (DTBHN initiation)
To a solution of the vinyl or isopropenyl boronic acid pinacol ester (1.00 mmol) in TBME (5
mL) at 0 °C was added dropwise a solution of organolithium (1.05 mmol). After stirring for
30 min at 0 °C, the reaction mixture was allowed to warm to rt and to it were added BEt3
solution (2.0 M in benzene, 1.0 mL, 2.0 mmol), the radical precursor (2.0 mmol), and DTBHN
(52 mg, 0.30 mmol) in that order. The reaction mixture was heated to 60 °C, and gently
refluxed for 1.5 h. The reaction was cooled to rt and partitioned between TBME (20 mL) and
water (15 mL), acidified with 1.0 M aq. HCl (2–5 mL). The organic phase was washed once
more with water (20 mL), and the combined aq. phases were extracted once with TBME (5-
10 mL). The combined ethereal phases were washed with sat. aq. NaHCO3 (20 mL) and brine
(20 mL). After drying over Na2SO4, the solution was filtered through a silica plug (i.d. = 20
mm; depth = 20–30 mm) protected with a cotton wad. The crude mixture was concentrated
in vacuo and directly submitted to FC.
Coupling product 2ba
General procedure was followed and the crude was submitted to FC (gradient = 2–5%
AcOEt/heptanes) to afford 2ba (314 mg, 0.945 mmol, 95%) as a clear, viscous oil (Rf = 0.38 in
AcOEt/heptanes 15:85). 1H NMR (300 MHz, CD2Cl2): δ 7.34 – 7.22 (m, 4H), 7.19 – 7.07 (m,
1H), 4.03 (q, J = 7.1 Hz, 2H), 2.27 – 1.86 (m, 4H), 1.31 (s, 3H), 1.19 (m, 15H) [interpreted as
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
1.21 (br s, 6H), 1.22 (br s, 6H) + 1.19 (t, J = 7.1 Hz, 3H)]. 13C NMR (75 MHz, CD2Cl2): δ 174.2
(Cq), 146.5(Cq), 128.5 (CH), 127.3 (CH), 125.7 (CH), 83.9 (Cq), 60.4 (CH2), 34.6 (CH2), 30.6
(CH2), 24.8 (CH3), 24.7 (CH3), 21.1 (CH3), 14.4 (CH3). 11B NMR (96 MHz, CD2Cl2): δ 33.9 (s). IR
(thin film): ν = 1732 (s) cm–1. HRMS (ESI): calculated for C19H29BO4Na [M+Na]+: 355.2051;
found: 355.2053.
Acknowledgements
The Swiss National Science Foundation (Project 200020_172621) and the University of Bern
are gratefully acknowledged for financial support.
Conflict of interest
The authors declare no conflict of interest.
Keywords
boronic esters, radical reaction, single electron transfer, inverse boron ylid, Matteson
rearrangement
[1]
[2]
[3]
[4]
[5]
[6]
Encyclopedia of Radicals In Chemistry, Biology and Materials, Vol. 1-4 (Eds.: C.
Chatgilialoglu, A. Studer), Wiley, 2012.
Radicals in Synthesis I: Methods and Mechanisms., Vol. 263 (Ed.: A. Gansaeuer),
Springer, Berlin, 2006.
Radicals in Synthesis II: Complex Molecules., Vol. 264 (Ed.: A. Gansaeuer), Springer,
Berlin, 2006.
Radicals in Organic Synthesis, Vol. 1-2 (Eds.: P. Renaud, M. P. Sibi), Wiley-VCH,
Weinheim, 2001.
H. Subramanian, Y. Landais, M. P. Sibi, in Comprehensive Organic Synthesis II, 2nd ed.
(Ed.: P. Knochel), Elsevier, Amsterdam, 2014, pp. 699-741.
A. Fusano, I. Ryu, in Science of Synthesis, Multicomponent Reactions, Vol. 2 (Ed.: T. J.
J. Müller), Thieme, 2014, pp. 409-437.
[7]
[8]
[9]
E. Godineau, Y. Landais, Chemistry - A European Journal 2009, 15, 3044-3055.
G. Xu, P. Renaud, Angew. Chem. Int. Ed. Engl. 2016, 55, 3657-3661.
G. Xu, M. Luthy, J. Habegger, P. Renaud, J. Org. Chem. 2016, 81, 1506-1519.
[10] D. S. Matteson, J. Am. Chem. Soc. 1960, 82, 4228-4233.
[11] D. S. Matteson, G. D. Schaumberg, J. Org. Chem. 1966, 31, 726-731.
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
[12] D. S. Matteson, R. W. H. Mah, J. Am. Chem. Soc. 1963, 85, 2599-2603.
[13] D. S. Matteson, in Boron Reagents in Synthesis (Ed.: A. Coca), ACS Symposium Series,
Washington, 2016, pp. 173-208.
[14] D. S. Matteson, J. Org. Chem. 2013, 78, 10009-10023.
[15] C. Sandford, V. K. Aggarwal, Chem. Commun. 2017, 83, 486.
[16] D. Leonori, V. K. Aggarwal, Acc. Chem. Res. 2014, 47, 3174-3183.
[17] C. G. Watson, P. J. Unsworth, D. Leonori, V. K. Aggarwal, in Lithium Compounds in
Organic Synthesis, Vol. 102 (Eds.: R. Luisi, V. Capriati), Wiley-VCH, Weinheim,
Germany, 2014, pp. 397-422.
[18] Y. Wang, A. Noble, C. Sandford, V. K. Aggarwal, Angew. Chem. Int. Ed. Engl. 2017, 56,
1810-1814.
[19] J. Schmidt, J. Choi, A. T. Liu, M. Slusarczyk, G. C. Fu, Science 2016, 354, 1265-1269.
[20] G. J. Lovinger, M. D. Aparece, J. P. Morken, J. Am. Chem. Soc. 2017, 139, 3153-3160.
[21] L. Zhang, G. J. Lovinger, E. K. Edelstein, A. A. Szymaniak, M. P. Chierchia, J. P. Morken,
Science 2016, 351, 70-74.
[22] B. Quiclet-Sire, S. Z. Zard, J. Am. Chem. Soc. 2015, 137, 6762-6765.
[23] M. Kischkewitz, K. Okamoto, C. Mück-Lichtenfeld, A. Studer, Science 2017, 355, 936-
938.
[24] C. Gerleve, M. Kischkewitz, A. Studer, Angew. Chem. Int. Ed. 2018, 57, 2441-2444.
[25] M. Silvi, C. Sandford, V. K. Aggarwal, J. Am. Chem. Soc. 2017, jacs.7b02569.
[26] M. Silvi, R. Schrof, A. Noble, V. K. Aggarwal, Chem. Eur. J. 2018, 24, 4279-4282.
[27] G. J. Lovinger, J. P. Morken, J. Am. Chem. Soc. 2017, 139, 17293-17296.
[28] A similar experiment involving PhLi and n-BuLi gave 2a/2b as a1:1.2 mixture in 67%
yield
[29] Probably for the same reason, Studer and Aggarwal are both using a solvent switch
afer the formation of the ate complex to acetonitrile and 1,3-dimethyl-2-
imidazolidinone, respectively.
[30] Benzene gave similar results but the more benign TBME was kept for the rest of our
study.
[31] Studer and coworkers have also reported that, after a solvent switch from THF to
acetonitrile, Et3B (ref. 23) and light (ref. 24) were both efficiently initiating reactions
involving α-iodoesters.
[32] Attempt to use diethyl bromomalonate was unsuccessful. This radical precursor was
found to protonate the ate complex before the radical reaction could take place.
[33] For the reaction involving 1b, the formation of the ate complex with tert-
butyllithium had to be performed at –78 °C to avoid competing allylic deprotonation.
[34] R. P. Sonawane, V. Jheengut, C. Rabalakos, R. Larouche-Gauthier, H. K. Scott, V. K.
Aggarwal, Angew. Chem. Int. Ed. 2011, 50, 3760-3763.
[35] In this case, the competing formation of an azaboron enolate by reaction of the
cyanomethyl radical with Et3B cannot be excluded
[36] The presence of the methyl substitutent is also expected to enhance the reducing
power of the radical anion intermediate and favor the SET recution of the iodide
regenerating the starting radical.
[37] Radical 1,2-aryl shift (neophyl rearrangement) is well-documented in the literature.
[38] F. H. Seubold, J. Am. Chem. Soc. 1953, 75, 2532-2533.
[39] Z. M. Chen, X. M. Zhang, Y. Q. Tu, Chem. Soc. Rev. 2015, 44, 5220-5245.
[40] T. Kawamoto, I. Ryu, Org. Biomol. Chem. 2014, 12, 9733-9742.
This article is protected by copyright. All rights reserved.

10.1002/chem.201802384
Chemistry - A European Journal
[41] N. Zhou, J. Liu, Z. Yan, Z. Wu, H. Zhang, W. Li, C. Zhu, Chem. Commun. 2017, 53,
2036-2039.
[42] H. Zhu, X. Nie, Q. Huang, G. Zhu, Tetrahedron Lett. 2016, 57, 2331-2335.
[43] C. Che, Q. Huang, H. Zheng, G. Zhu, Chem. Sci. 2016, 7, 4134-4139.
[44] A. L. J. Beckwith, C. H. Schiesser, Tetrahedron 1985, 41, 3925-3941.
[45] D. P. Curran, E. Bosch, J. Kaplan, M. Newcomb, J. Org. Chem. 1989, 54, 1826-1831.
[46] D. P. Curran, A. A. Martin-Esker, S. B. Ko, M. Newcomb, J. Org. Chem. 1993, 58, 4691-
4695.
[47] A. Effio, D. Griller, K. U. Ingold, J. C. Scaiano, S. J. Sheng, J. Am. Chem. Soc. 1980, 102,
6063-6068.
[48] W. H. Urry, M. S. Kharasch, J. Am. Chem. Soc. 1944, 66, 1438-1440.
This article is protected by copyright. All rights reserved.