Silyl Assistance in the Intramolecular Addition of Pyridyl Radicals onto Pyridines and Quinolines

Article abstract of DOI:10.1002/ejoc.201600170

The first example of a stable acylazinium salt obtained by an intramolecular acylpyridine-centered radical attack on azine nitrogen atoms is reported. Depending on the relative position of the bromo substituent on the attacking aminopyridine ring, it is possible to obtain selectively either the N-attack product or a different C-attack derivative. A rationale for this pathway selectivity was obtained from quantum mechanical calculations.

Full text of DOI:10.1002/ejoc.201600170

DOI: 10.1002/ejoc.201600170
Full Paper
Heteroaryl Radical Chemistry
Silyl Assistance in the Intramolecular Addition of Pyridyl
Radicals onto Pyridines and Quinolines
Fabiana Filace,[a] Pedro A. Sánchez-Murcia,[b] David Sucunza,[a] Adrián Pérez-Redondo,[a]
Julio Álvarez-Buílla,[a] Federico Gago,*[b] and Carolina Burgos*[a]
Abstract: The first example of a stable acylazinium salt ob- pyridine ring, it is possible to obtain selectively either the N-
tained by an intramolecular acylpyridine-centered radical attack attack product or a different C-attack derivative. A rationale for
on azine nitrogen atoms is reported. Depending on the relative this pathway selectivity was obtained from quantum mechani-
position of the bromo substituent on the attacking amino- cal calculations.
Introduction
imines have been shown for silyl radicals that are frequently
used as tin-free reducing agents as an alternative to organotin
compounds.[8]
Radical addition onto unsaturated systems has emerged in re-
cent years as a powerful synthetic tool to provide access to
carbon–carbon and carbon–heteroatom bonds.[1] The versatility
of this approach lies, among other factors, in the increasing
number of processes that allow exquisite regioselectivities by
modulating the nature of both the incoming radical and the
acceptor (radicalophile). Most of the reported examples are C-
centered radicals such as alkyl, aryl, vinyl, or acyl radicals.[2] Al-
though radical reactivity is controlled by orbital overlap, it is
well known that this process is highly influenced by polar ef-
fects.[3] Thus, whereas alkyl, aryl, and vinyl radicals are consid-
ered mostly “nucleophilic”,[2] the polarization introduced by at-
tachment of an oxygen proximal to the carbon where the radi-
cal is centered makes acyl radicals “ambiphilic” in reactions of
addition to unsaturated bonds.[4] Whereas acyl radicals behave
as nucleophilic radicals towards C=C bonds,[4b] Ryu and Ko-
matsu[5] demonstrated that acyl radicals show an “N-philic”
character, with complete regioselectivity for nitrogen in addi-
tions to imine bonds, whereas protonation of this nitrogen
changes the regioselectivity to the carbon.[6] Theoretical studies
attempting to unravel the origin of such selectivity have sug-
gested that acyl radicals add to numerous types of π-systems
(e.g., imine/iminium, pyridine/pyridinium or electron-rich ole-
fins)[7] through simultaneous mixing of SOMO–LUMO and
LUMO–HOMO from radical and radicalophile, respectively. Simi-
lar multicomponent orbital interactions with carbonyls and
On the other hand, heterobiaryls, especially bisazines, are
key molecular scaffolds that are represented in several impor-
tant areas of organic chemistry, including natural products and
catalysts.[9] Examples of this class of compound endowed with
pharmacological activity include cytotoxic alkaloids and glycos-
ides,[10] antitumor antibiotics,[11] and synthetic drugs.[12] None-
theless, the synthesis of these compounds remains a difficult
challenge because the “classical cross-coupling” reactions start-
ing from monoheteroaryl precursors suffer from limitations.[13]
Several elegant methodologies such as the direct catalytic aryl-
ation of N-oxides and related derivatives[13b–13d] and the de-
carboxylative cross-coupling reaction[13a] circumvent some of
these problems. A complementary alternative approach that
leads to biaryls is the radical pathway that involves the addition
of an aryl radical to another aromatic nucleus.[14] In this context,
we have been interested for some time in the intramolecular
addition of pyridyl radicals onto another aromatic nucleus un-
der metal-free, thermal conditions in the presence of an excess
of tris(trimethylsilyl)silane (TTMSS) and azobisisobutyronitrile
(AIBN). In particular, we have explored the reactivity of this type
of radical in which the pyridine ring is connected to another
(hetero)aromatic ring by linkers containing pyridinium N-ylides,
sulfonamides, carbamates, and N-substituted amides.[15] Herein,
we describe how the intramolecular addition of pyridyl radicals
onto pyridines and quinolines selectively affords N1- or C3-addi-
tion products depending on the nature of the reactants. Our
proposal of a radical spirocycle intermediate and assistance by
TTMSS in the evolution towards the final products is supported
by results from quantum mechanical calculations.
[a] Departamento de Química Orgánica y Química Inorgánica, Universidad de
Alcalá,
carrt. Madrid-Barcelona Km 33,700, Alcalá de Henares, Madrid, Spain
E-mail: carolina.burgos@uah.es
[b] Departamento de Ciencias Biomédicas, Unidad Asociada al IQM-CSIC,
Universidad de Alcalá,
Results and Discussion
carrt. Madrid-Barcelona Km 33,700, 28871 Alcalá de Henares, Spain
E-mail: federico.gago@uah.es
http://www3.uah.es/farmamol/
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under http://dx.doi.org/10.1002/ejoc.201600170.
In the course of our studies, we envisaged that N-(bromo-
pyridyl)pyridine-2-carboxamide (1; HET = pyridine, R2 = H)
would be an interesting scaffold for the preparation of biaryl
Eur. J. Org. Chem. 2016, 1891–1896
1891
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
derivatives 3 (Scheme 1). Our initial strategy involved the radi- unstable N-acylazinium derivatives 5, we also aimed to obtain
cal intramolecular reaction of 1 under thermal conditions using the corresponding quinoline derivatives starting from 1d–i
TTMSS/AIBN (2 equiv.) in m-xylene. In light of reports from Cur- (Scheme 2, Table 1) in an attempt to improve the stability of
ran and Ganguly[16] and our own earlier results,[15c] we reasoned the resulting acylquinolinium bromides. In all cases, a single
that this process would afford pyridonaphthyridinone 2, which, reaction product was obtained and only traces of alternative
after subsequent hydrolysis and decarboxylation, would yield compounds were detected, if at all. Starting from 1d and 1e
the 3-(pyridyl)pyridine derivative 3.[17] Thus, we attempted the (Table 1, entries 4 and 5), we could isolate 7-methyl-6-oxo-11-
reaction of dipyridines 1a–c under our optimized conditions[15c] azaquinolino[1,2-c]quinazolin-5-ium bromide (5d; 69 % yield)
(Table 1, see the Supporting Information). The above process, and its 2-Cl-substituted analogue 5e (67 % yield), respectively,
however, resulted in the formation of product 4a starting from as yellow solid products, which did not decompose. To our
1a (X = N) due to a predictable SNAr, whereas 5b and 5c were knowledge, only one other example of a similarly stable N-
obtained from 1b (Y = N) and 1c (W = N) (59 and 47 % yield, acylazinium derivative has been published to date and this was
respectively), without traces of the expected 2a–c derivatives obtained by an alternate route.[18] It is noteworthy that the SNAr
(Table 1, entries 1–3). The acylazinium salts 5b and 5c precipi- process described for the pyridine derivative 1a was not ob-
tated as yellow products within the flask during the course of served for quinoline 1d under the same reaction conditions
the reaction.
but, instead, required heating at reflux for seven days to afford
the corresponding 4d.
Scheme 2. Synthesis of 4–6 from 1 in the presence of TTMSS/AIBN (2 equiv.)
under thermal conditions.
Scheme 1. Initial strategy and formation of 5b and 5c.
Table 1. Results for the reaction of 1 using TTMSS/AIBN.
The identity of 5d was unequivocally confirmed by X-ray
crystallographic analysis (Figure 1).[19] The presence of two het-
erocyclic moieties connected by a C2–C2′ bond in 5d implies
breakage of the original C2–amide bond and covalent attach-
ment of the amide carbonyl to the quinoline N1. This radical
ring-closure onto a heterocyclic ring has few precedents, mainly
reported by Zard et al.[20]
Entry
1
HET[a]
X
Y
W
R1 R2
Product
Yield
[%]
1
2
3
4
5
6
7
8
9
1a
1b
1c
1d
1e
1f
1g
1h
1i
P
P
P
Q
Q
Q
Q
Q
Q
N
CH
N
CH
CH
CH
N
N
N
CH
CH
CH
N
CH
CH
CH
CH
CH
N
–
–
–
H
Cl
H
H
H
H
–
–
–
H
H
H
Cl
OMe
H
4a
5b
5c
5d
5e
6f
6g
6h
6i
99
59
47
69
67
53
46
47
56
CH
CH
N
N
CH
CH
CH
CH
[a] HET: heterocycle, Q: quinoline, P: pyridine.
This procedure allowed the straightforward isolation and pu-
rification of 5b and 5c, but these compounds proved to be only
moderately stable, especially 5c, which slowly decomposed to
N-methyl-3-(2-pyridyl)pyridin-4-amine (7).
Encouraged by the versatility and potential synthetic value
of this reaction for the one-pot preparation of these typically
Figure 1. Unit cell of the crystal structure of the bromide salt of 5d. Note the
layout of the positively charged heterocyclic molecules (sticks) in the crystal
lattice and the interspersed bromide anions (red spheres).
Eur. J. Org. Chem. 2016, 1891–1896
www.eurjoc.org
1892
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
When the substrate was 1f, with an arrangement of 2-amino- Br at the 3-position of the pyridine vs. with a Br at the 2-posi-
3-bromopyridine (Table 1, entry 6), the process led to the isola- tion, due to the repulsion between the lone electron pairs of
tion of 6f. According to Zard's work,[20] and in an attempt to the two nitrogen atoms in intermediates 9.[20d] These differen-
obtain the N-addition derivatives 5 from this kind of starting ces could be key to explaining the reactivity of the starting
bromoaminopyridines, chloro- and methoxy-substituted materials. Whereas compounds 1f–i, containing both a quinol-
amides 1g and 1h were prepared and subjected to the same ine as the acceptor unit and a Br at the 3-position of the pyrid-
reaction conditions (Table 1, entries 7 and 8). Our results ine, react to afford type 6 compounds, 1b–d, which have either
showed that the products of C-addition, 6g and 6h, respec- a pyridine as the acceptor unit or a Br at the 2-position of the
tively, were only obtained, albeit in lower yields. In a similar pyridine, react to form type 5 derivatives. The only example
way, reaction of 1i, with a structure of 4-amino-3-bromopyr- that contains both a pyridine acceptor unit and a Br at the 2-
idine (Table 1, entry 8) afforded 6i, without the corresponding position of the pyridine (1a) did not react through a radical
derivative 5i being detected in the reaction mixture (Table 1, pathway.
entry 9).
Theoretical support for this pathway is provided by quantum
From a mechanistic viewpoint, we reasoned that formation mechanical calculations. Thus, estimated activation energy bar-
of both products 5 and 6 would start with the initial formation riers linking 8 to 9–11 (ΔE‡, Table 2) strongly indicate that for-
of the C-centered pyridyl radical 8, which would then react with mation of 9d, 9f, and 9b (ipso-addition, path A) is favored in all
the heteroaryl fragment through an intramolecular [1,5] ipso- cases over the alternative path B (intermediates 10 and 11,
addition to yield the spirocyclic radical 9 as an intermediate Table 2). These results are consistent with the kinetic preference
(path A, Scheme 3).
for spirocyclization in other related radical additions onto phen-
yls and (hetero)arenes.[23] However, for the construction of 5
and 6, the calculated energy barriers (ΔE‡), although signifi-
cantly high in both cases, would indicate that formation of 5 is
favored over 6 in all cases. For the bipyridine derivative, the
energy barriers that connect 9b with 5b· and 6b· are quite simi-
lar (39.4 and 37.8 kcal/mol, respectively). In contrast, 5b· shows
a much lower energy (ΔE ca. 13 kcal/mol) than 6b· (see the
Supporting Information, Figure S1–S3). This is in accordance
with the isolation of 5d and 5b but does not account for the
observed evolution of 9f to 6f. The computed transition states
are reminiscent of acyl radicals, which have demonstrated their
N-philicity in cyclizations onto C=N double bonds due to partic-
ular SOMOacyl→π*N=C and LPN=C→π*acyl multicomponent or-
bital interactions.[4a,5]
Table 2. DFT computed activation energies (ΔE‡) and enthalpy variations
(ΔH298) (kcal mol–1 [a]
for the proposed reaction mechanism.
)
Entry
Compound
ΔE‡
ΔH298
1
2
3
4
5
6
8d
–
0.0
9d (spirocycle)
10d (C-ortho)
11d (N-ortho)
5d· (N-addition)
6d· (C-addition)
5.7
8.0
8.1
32.5
48.6
–21.1
–24.7
–39.4
–44.7
–28.3
7
8f
–
0.0
8
9
10
11
12
9f (spirocycle)
10f (C-ortho)
11f (N-ortho)
5f· (N-addition)
6f· (C-addition)
3.0
5.1
5.0
33.2
47.7
–28.6
–28.9
–41.3
–49.3
–38.3
Scheme 3. Proposed chemical pathways for the evolution of 1 to 5 and 6.
Subsequent formation of either 5 or 6 would require an in-
tramolecular N- or C-ortho attack, respectively, through a Beck-
with–Dowd ring-expansion mechanism,[21] and a following
oxidation step in both cases.[22] These results suggest the exis-
tence of a trend in the reactivity of these compounds. It can be
seen from the results in Table 1 that the tested starting materi-
als differ in two respects: (i) the type of heterocycle in the ac-
ceptor unit, either a quinoline or a pyridine, and (ii) the position
of the Br on the pyridine ring. In both cases, one of the two
possibilities would provide more stable spirocyclic intermedi-
ates 9, i.e., quinolines vs. pyridines, because of the additional
13
14
15
16
17
18
8b
–
0.0
9b (spirocycle)
10b (C-ortho)
11b (N-ortho)
5b· (N-addition)
6b· (C-addition)
3.7
6.2
5.8
39.4
37.8
–26.9
–27.1
–40.8
–50.0
–37.1
[a] BHandHLYP/6-311+G**, including the zero-point correction and solvent
effect (m-xylene, IEFPCM).
Given that there is a large excess of TTMSS in the reaction
stabilizing effect of the extra aromatic ring, and those having a medium and that TTMSSBr is generated in situ, we took into
Eur. J. Org. Chem. 2016, 1891–1896
www.eurjoc.org
1893 © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
account the pioneering work of Minisci[24] and reports concern- optimized geometries of these N-silylated TS12→5/6, the forming
ing the silylation of heteroaromatic bases in the presence of bonds are shorter than those that are broken (Figure 3), in con-
silyl halides[25] to postulate that formation of both products 5 trast to the Beckwith–Dowd geometries observed for TS9→5/6
and 6 could start with the initial formation of the N-silylated (Scheme 3 and Figure S1–S3).
spirocyclic radicals 12d·+ and 12f·+. When we included these
species in the calculations, not only did the energy barriers be-
come lower but we could also explain the formation of 6f and
the experimentally observed regioselectivity (Table 3 and Fig-
ure 2).
Table 3. DFT-computed activation energies (ΔE‡) and enthalpy variations
(ΔH298) (kcal mol–1) for the TTMSS-assisted reaction mechanism (BHandHLYP/
6-311+G**, including the zero-point correction and solvent effect – m-xylene,
IEFPCM).
Entry
Compound[a]
ΔE‡
ΔH298
1
2
12d·+
–
0.00
7.09
5dNTTMSS·+
36.90
3
4
12f·+
–
0.00
–8.39
6fNTTMSS·+
14.63
[a] 5dNTTTMSS·+: N-silylated 5·+; 6fNTTTMSS·+: N-silylated 6·+.
Scheme 4. Silyl assistance in the intramolecular formation of 5d and 6f.
Figure 2. Optimized geometries of the N1-silylated cation radical intermedi-
ates 12d·+ and 12f·+ and 2D depiction to explain the regioselectivity in their
evolution to 5d and 6f, respectively.
Figure 3. Optimized structures of the transition states that connect 12d·+ with
N-silylated 5d·+ (5dNTTTMSS·+) (left) and 12f·+ with N-silylated 6f·+
(6fNTTTMSS·+) (right).
Regarding the topicity of the reaction, we observed that the
carbonyl oxygen helps to orient the TTMSS group so as to af-
ford 12f·+, whereas it is the pyridine nitrogen N1′ that drives
the orientation of TTMSS on the opposite face to yield 12d·+
(Figure 2).
Therefore, in stark contrast to the expectations based on the
proposed unassisted mechanism (Table 2 and Figure S1–S3),
the optimized 3D structures of 12d·+ and 12f·+ do provide a
clear explanation for the experimentally observed regioselectiv-
ity. In 12f·+ the steric effects of the bulky TTMSS moiety (syn to
the carbonyl) avoids addition to the quinoline N1 so that access
to 5f is prevented, whereas in 12d·+ (silyl group in trans to the
carbonyl) the inability to rotate around the C2–C2′ bond blocks
access to 6d.
Several considerations support our mechanistic proposal.
First, when substrates 1d and 1f were used and the amount of
TTMSS/AIBN was decreased down to sub-stoichiometric ratios,
under the same experimental conditions, the yield dramatically
dropped to the extent that almost all the unreacted starting
materials could be isolated. Second, the stereoelectronic differ-
ences within spirocycles 9d–f due to the inductive and reso-
nance effects exerted by the nitrogen of the attacking pyridine
did not significantly affect either the energy or geometry of the
frontier molecular orbitals (Table 4). Third, the calculated energy
barrier in going from 12·+ to N-silylated 6·+ (6NTTMSS·+) is
lower for the silyl-assisted mechanism than for the unassisted
mechanism (from 9) and is similar in the case of 9 → 5·
Taken together, our findings strongly suggest an assisted silyl (Tables 2 and 3). Finally, our invocation of the steric effects of
free-radical addition/cyclization cascade process (Scheme 4) in TTMSS highlights the importance of the size of the incoming
which the cation radical species 12d·+ and 12f·+ would drive radical to trap the quinoline nitrogen N1 in a particular confor-
the evolution towards 5d and 6f through, in this case, two mation. Indeed, when we ran similar calculations for 9d and 9f
carbamoyl radical transition states (Table 3 and Figure 3). In the using H+ as the quencher (the other possible electrophile in
Eur. J. Org. Chem. 2016, 1891–1896
www.eurjoc.org
1894
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
H5′), 7.59–7.67 (m, 2 H, H3 and H4), 7.80 (d, J = 7.9 Hz, 1 H, H4′),
8.12 (d, J = 4.6 Hz, 1 H, H6 or H6′), 8.18 (dd, J = 4.6, 1.6 Hz, 1 H, H6
or H6′) ppm; 13C NMR (75 MHz, CDCl3; Me4Si): δ = 37.1, 123.0, 124.3,
124.6, 136.5, 138.4, 141.3, 142.5, 147.5, 148.3, 152.6, 167.7 ppm.
HRMS (ESI-TOF): m/z calcd. for C12H1179BrN3O [M + H]+ 292.0080;
found 292.0084.
the reaction mixture) discrimination was not possible. A similar
mechanism in the case of pyridine compounds 1b and 1c can-
not be ruled out.
Table 4. Energy (eigenvalues) of the frontier molecular orbitals SUMO, SOMO,
HOMO, and HOMO-1 of the spirocyclic intermediates 9d,f (quinoline deriva-
tives) and 9b (pyridine derivative).
Preparation of N-Acylazinium Salts 5b–e and Naphthyridin-6-
ones 6f–i. General Procedure: A solution of TTMSS (248 mg,
1 mmol) and AIBN (164 mg, 1 mmol) in m-xylene (10 mL) was added
dropwise, by using a syringe pump during 13 h, to a solution of
the corresponding N-methylamide 1 (0.5 mmol) in m-xylene (2 mL)
at 80 °C (bath temperature). When the addition was finished, the
reaction mixture was stirred for an additional 24 h at the same
temperature. For acylazinium salts, the precipitate was filtered and
washed with hexane. For naphthyridin-6-one derivatives, the result-
ing solution was concentrated and the crude mixture was separated
by flash chromatography on silica gel (hexane/ethyl acetate).
Molecular orbital
(MO)
Energy (eigenvalues)[a]
9d
9f
9b
SUMO
SOMO
HOMO
HOMO
HOMO–1
HOMO–1
69ꢀ
69α
68ꢀ
68α
67ꢀ
67α
–0.0697
–0.2456
–0.2744
–0.2758
–0.2835
–0.2970
–0.0740
–0.2499
–0.2769
–0.2781
–0.2878
–0.3012
–0.0649
–0.2569
–0.2788
–0.2793
–0.3268
–0.3357
[a] Calculated by using the BHandHLYP/6-311+G** method including the
zero-point correction and taking into account the solvent effect (m-xylene) by
means of the polarizable continuum model (PCM) and the integral equation
formalism variant (IEFPCM).
7-Methyl-6-oxo-11-azaquinolino[1,2-c]quinazolin-5-ium Brom-
ide (5d): The general procedure starting from 1d (171 mg) gave 5d
(118 mg, 69 %) as a yellow solid; m.p. > 250 °C (dec.); IR (KBr): νmax
=
˜
3433, 1730, 1611, 1297, 783 cm–1 1H NMR (500 MHz, CD3OD;
.
Conclusions
Me4Si): δ = 4.08 (s, 3 H, CH3), 8.10–8.14 (m, 2 H, H6 and H5′), 8.29
(ddd, J = 9.0, 7.0, 1.5 Hz, 1 H, H7), 8.37 (dd, J = 8.8, 1.2 Hz, 1 H,
H4′), 8.50 (dd, J = 8.0, 1.5 Hz, 1 H, H5), 9.00 (dd, J = 4.3, 1.2 Hz, 1
H, H6′), 9.37 (d, J = 8.8 Hz, 1 H, H4), 9.58 (d, J = 8.8 Hz, 1 H, H3),
9.79 (dd, J = 9.0, 0.53 Hz, 1 H, H8) ppm; 13C NMR (125 MHz. CD3OD;
Me4Si): δ = 34.0, 118.9, 124.2, 125.2, 131.0, 131.1, 131.7, 132.0, 132.8,
136.3, 137.8, 139.1, 148.4, 148.9, 150.8, 153.7 ppm. HPLC (0 to 100 %
of water in 10 min): tR = 4.56 min (95 %). HRMS (ESI-TOF): m/z calcd.
for C16H12N3O [M]+ 262.0975; found 262.0966.
The one-pot reactions described here grant easy access to N-
acylquinolinium salts 5 of exceptional stability through an intra-
molecular radical addition onto the quinoline nitrogen of N-(2-
bromo-3-pyridyl)-N-methylquinoline-2-carboxamides in the
presence of excess TTMSS. The same procedure led to less sta-
ble N-acylpyridinium salts when the corresponding pyridine de-
rivatives were employed. In contrast, when the same methodol-
ogy was applied to N-(3-bromo-2-pyridyl)-N-methylquinoline-2-
carboxamide, the alternative C3-addition products 6 were ob-
tained. This regioselectivity from a common radical spirocycle
is made possible by the steric bulk of the TTMSS imposing two
distinct and very well-defined geometries on the reactants.
5-Methylquinolino[3,2-c][1,8]naphthyridin-6-one (6f): The gen-
eral procedure starting from 1f (171 mg) gave the crude product,
which was purified by flash chromatography (silica gel; hexane/
EtOAc, 9:1) to give pure 6f (69 mg, 53 %) as a white solid; m.p. 189–
191 °C; IR (KBr): νmax = 3419, 1665, 1339, 791 cm–1 1H NMR
.
˜
(300 MHz, CDCl3; Me4Si): δ = 3.93 (s, 3 H, CH3), 7.33 (dd, J = 7.7,
4.8 Hz, 1 H, H5′), 7.62 (ddd, J = 8.2, 6.8, 1.1 Hz, 1 H, H6), 7.89 (ddd,
J = 8.4, 6.8, 1.4 Hz, 1 H, H7), 8.07 (dt, J = 8.2, 0.8 Hz, 1 H, H5), 8.24
(dt, J = 8.4, 1.1 Hz, 1 H, H8), 8.65 (dd, J = 4.8, 2.0 Hz, 1 H, H6′), 9.27
(dd, J = 7.7, 2.0 Hz, 1 H, H4′), 9.34 (s, 1 H, H4) ppm; 13C NMR
(125 MHz, CDCl3; Me4Si): δ = 31.4, 119.4, 121.5, 122.2, 129.7, 130.1,
131.9, 132.0, 135.0, 136.7, 141.4, 150.5, 152.9, 153.0, 153.1,
165.3 ppm. HPLC (10 to 100 % of water in 10 min): tR = 5.25 min
(95 %). HRMS (ESI-TOF): m/z calcd. for C16H12N3O [M + H]+ 262.0975;
found 262.0979.
Experimental Section
Typical Experimental Procedures
Preparation of N-Methylamides 1. General Procedure: The corre-
sponding amide 14 (see the Supporting Information; 1.0 mmol) was
dissolved in anhydrous DMF (3 mL), and the solution was cooled to
0 °C in an ice/water bath. Sodium hydride (29 mg, 1.2 mmol) was
added in portions such that the internal temperature was main-
tained <5 °C (vigorous stirring was required to keep the suspension
fluid). Upon complete addition, the suspension was stirred vigor-
ously for an additional 20 min while maintaining the temperature
below 5 °C and then methyl iodide (156 mg, 1.1 mmol, 0.07 mL)
was added dropwise. After stirring for 30 min, the reaction mixture
was brought to room temperature over a 5 h period. The mixture
was then treated with water (2.5 mL) and extracted with EtOAc. The
combined organic layers were washed with brine before being
dried (MgSO4) and concentrated. The resulting residue was then
purified by chromatography (silica gel; hexane/ethyl acetate), to fur-
nish the pure product 1.
Acknowledgments
Financial support from the Spanish Ministry of Science and In-
novation (CTQ2014-52488-R), the Spanish Ministry of Economy
and Competitiveness (SAF2012-39760-C02-02), the Comunidad
de Madrid (S2010-BMD-2457), the Universidad de Alcalá
(projects CCG2015/EXP-011, CCG2015/EXP-026 and CCG2014/
EXP-031) and the Instituto de Salud Carlos III (REDinREN, RD12/
0021/0014) is gratefully acknowledged. F. F. is the recipient of a
research scholarship from the Spanish Ministry of Education
(grant number FPU-02045-2012).
N-(2-Bromo-3-pyridyl)-N-methylpyridine-2-carboxamide (1a):
The general procedure, starting from 14a (278 mg) gave the crude
product, which was purified by flash chromatography (silica gel;
hexane/EtOAc, 7:3) to give pure 1a (201 mg, 69 %) as a white solid;
m.p. > 250 °C (decomp); IR (KBr): νmax = 3432, 1693, 1458, 1338,
Keywords: Radical reactions · Density functional calculations ·
Nitrogen heterocycles · Silanes
˜
1
807 cm–1. H NMR (300 MHz, CDCl3; Me4Si): δ = 3.40 (s, 3 H, CH3),
7.11 (dd, J = 7.6, 4.6 Hz, 1 H, H5), 7.19 (dd, J = 7.9, 4.6 Hz, 1 H,
Eur. J. Org. Chem. 2016, 1891–1896
www.eurjoc.org
1895
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] D. Liu, S. Tang, H. Yi, C. Liu, X. Qi, Y. Lan, A. Lei, Chem. Eur. J. 2014, 20,
15605–15610.
[2] G. K. Friestad, Tetrahedron 2001, 57, 5461–5496.
Mendicuti, C. Burgos, J. Álvarez-Buílla, J. Org. Chem. 2008, 73, 8800–8807;
c) A. Núñez, A. Sánchez, C. Burgos, J. Álvarez-Buílla, Tetrahedron 2007,
63, 6774–6783; d) A. Sánchez, A. Núñez, J. Álvarez-Buílla, C. Burgos, Tetra-
hedron 2004, 60, 11843–11850.
[3] J. Fossey, D. Lefort, J. Sorba, Free Radicals in Organic Chemistry, John
Wiley & Sons, Chichester, UK, 1995, p. 49.
[16] a) F. González-López de Turiso, D. P. Curran, Org. Lett. 2005, 7, 151–154;
b) A. K. Ganguly, C. H. Wang, T. M. Chan, Y. H. Ing, A. V. Buevich, Tetrahe-
dron Lett. 2004, 45, 883–886.
[4] a) C. H. Schiesser, U. Wille, H. Matsubara, I. Ryu, Acc. Chem. Res. 2007, 40,
303–313; b) C. Chatgilialoglu, D. Crich, M. Komatsu, I. Ryu, Chem. Rev.
1999, 99, 1991–2069.
[5] a) T. Fukuyama, N. Nakashima, T. Okada, I. Ryu, J. Am. Chem. Soc. 2013,
135, 1006–1008; b) I. Ryu, K. Matsu, S. Minakata, M. Komatsu, J. Am.
Chem. Soc. 1998, 120, 5838–5839.
[6] J. Tauber, D. Imbri, T. Opatz, Molecules 2014, 19, 16190–16222.
[7] a) R. I. J. Amos, J. A. Smith, B. F. Yates, C. H. Schiesser, Tetrahedron 2009,
65, 7653–7657; b) S. H. Kyne, C. H. Schiesser, H. Matsubara, J. Org. Chem.
2008, 73, 427–434.
[8] a) U. Wille, J. C.-S. Tan, E.-K. Mucke, J. Org. Chem. 2008, 73, 5821–5830;
b) C. Chatgilialoglu, Chem. Rev. 1995, 95, 1229–1251.
[17] C. González-Bello, L. Castedo, in: Modern Heterocyclic Chemistry, vol. 3
(Eds.: J. Álvarez-Buílla, J. J. Vaquero, J. Barluenga), Wiley-VCH Verlag
GmbH & Co. KGaA, Weinheim, Germany, 2011, p. 1506–1507.
[18] M. D. Markey, T. R. Kelly, Tetrahedron 2008, 64, 8381–8388.
[19] Crystallographic data are summarized in Figure S4 and Table S1. CCDC
1401002 (for 5d) contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre.
[20] a) D. Liu, C. Liu, A. Lei, Chem. Asian J. 2015, 10, 2040–2054; b) M. El Qa-
cemi, L. Petit, B. Quiclet-Sire, S. Z. Zard, Org. Biomol. Chem. 2012, 10,
5707–5719; c) Y. Laot, L. Petit, S. Z. Zard, Chem. Commun. 2010, 46, 5784–
5786; d) M. El Qacemi, L. Ricard, S. Z. Zard, Chem. Commun. 2006, 4422–
4424.
[21] a) J. Hierold, D. W. Lupton, Org. Lett. 2012, 14, 3412–3415; b) J. R. Rodrí-
guez, L. Castedo, J. L. Mascareñas, Org. Lett. 2001, 3, 1181–1183.
[22] A. L. J. Beckwith, V. W. Bowry, W. R. Bowman, E. Mann, J. Parr, J. M. D.
Storey, Angew. Chem. Int. Ed. 2004, 43, 95–98; Angew. Chem. 2004, 116,
97–100.
[23] a) R. T. McBurney, J. C. Walton, Beilstein J. Org. Chem. 2013, 9, 1083–
1092; b) R. T. McBurney, A. Eisenschmidt, A. M. Z. Slawin, J. C. Walton,
Chem. Sci. 2013, 4, 2028–2035.
[9] P. Gros, Y. Fort, Curr. Org. Chem. 2003, 7, 629–648.
[10] a) P. Fu, Y. Zhu, X. Mei, Y. Wang, H. Jia, C. Zhang, W. Zhu, Org. Lett. 2014,
16, 4264–4267; b) X. Qu, B. Pang, Z. Zhang, M. Chen, Z. Wu, Q. Zhao, Q.
Zhang, Y. Wang, Y. Liu, W. Liu, J. Am. Chem. Soc. 2012, 134, 9038–9041;
c) P. Fu, S.-X. Wang, K. Hong, X. Li, P. P. Liu, Y. Wang, W. M. Zhu, J. Nat.
Prod. 2011, 74, 1751–1756.
[11] J. Donohoe, C. R. Jones, A. F. Kornahrens, L. C. A. Barbosa, L. J. Walport,
M. R. Tatton, M. O'Hagan, A. H. Rathi, D. B. Baker, J. Org. Chem. 2013, 78,
12338–12350.
[12] A. L. Illert, A. K. Seitz, C. Rummelt, S. Kreutmair, R. A. Engh, S. Goodstal,
C. Peschel, J. Duyster, N. von Bubnoff, PLoS One 2014, 9, e112318/1–
e112318/27.
[13] a) J.-B. Rouchet, C. Schneider, C. Spitz, J. Lefevre, G. Dupas, C. Fruit, C.
Hoarau, Chem. Eur. J. 2014, 20, 3610–3615; b) H. Bonin, M. Sauthier, F.-X.
Felpin, Adv. Synth. Catal. 2014, 356, 645–671; c) S. Duric, F. D. Sypaseuth,
S. Hoof, E. Svensson, C. C. Tzschucke, Chem. Eur. J. 2013, 19, 17456–
17463; d) F. Gosselin, S. J. Savage, N. Blaquiere, S. T. Staben, Org. Lett.
2012, 14, 862–865.
[24] a) J. Tauber, D. Imbri, T. Opatz, Molecules 2014, 19, 16190–16222; b) F.
Minisci, R. Bernardi, F. Bertini, R. Galli, M. Perchinummo, Tetrahedron
1971, 27, 3575–3579.
[25] a) A. Bartoszewicz, M. Kalek, J. Stawinski, Tetrahedron 2008, 64, 8843–
8850; b) M. Zirngast, J. Baumgartner, C. Marschner, Eur. J. Inorg. Chem.
2008, 1078–1087; c) J. Minamikawa, A. Brossi, Tetrahedron Lett. 1978, 19,
3085–3086.
[14] A. Studer, D. P. Curran, Angew. Chem. Int. Ed. 2011, 50, 5018–5022–5127;
Angew. Chem. 2011, 123, 5122, and references cited therein.
[15] a) M. Camacho-Artacho, V. Abet, L. M. Frutos, F. Gago, J. Álvarez-Buílla,
C. Burgos, J. Org. Chem. 2011, 76, 1452–1455; b) V. Abet, A. Núñez, F.
Received: February 16, 2016
Published Online: March 11, 2016
Eur. J. Org. Chem. 2016, 1891–1896
www.eurjoc.org
1896
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim