N. Kinarivala, P.C. Trippier / Tetrahedron Letters 55 (2014) 5386–5389
5387
true chemoselectivity was achieved and the large amounts of Pd/C
employed represent a significant safety hazard. The synthesis of
deuterated pyridines via hydrogenation of dichloropyridine-N-oxi-
vided a 100% conversion to 6 (Table 1: entry 3), while 1 h stirring
again resulted in 25% isolation of chlorinated starting material
(Table 1: entry 4).
des employing Pd/C, D
but with no investigation of chemoselectivity. Chelucci reported
2
O, and K
2
CO
3
at 190 °C has been reported
To ensure that hydrochloric acid produced as a byproduct in
this reaction was not acting to deactivate the Pd/C catalyst we used
15
chemoselectivity in hydrodehalogenation of pyridine and quino-
stoichiometric NaHCO
additive to counter the acid. While other reports cite NaHCO
acting as a poison at low catalyst loading we observed no such
effect. Results with this additive were identical to those without
after 1 and 2 h(s) reaction times (Table 1: compare entries 5 and 6
to 3 and 4), although a pleasing improvement in sharpness of the
NMR spectra was observed attributed to the prevention of the acid
salt formation. Finally, the use of palladium on barium sulfate (Pd/
3
, relative to the produced HCl, as a basic
lone derivatives using NaBH
dium catalysts including Pd(OAc)
range of temperatures from 25 °C to 60 °C and reaction times up
4
, TMEDA, PPh
3
and a variety of palla-
3
as
2
, PdCl
2
(dppf), and PdCl at a
2
2
1
1
6
to three days. Under these conditions 4-chloro-2-cyanopyridine
was hydrodechlorinated without concomitant reduction of the
cyano group. The methodology was also extended to a range of hal-
ogenated heterocycles.17
Given the synthetic advantages of retaining reduction-sensitive
functionality and protecting groups in pyridine intermediates
obtained from hydrodechlorination of 2-chloropyridine moieties
we wished to establish the relative chemoselectivity of reduc-
tion-sensitive functionalities commonly encountered in synthetic
routes to bioactive molecules. Successful hydrodechlorination of
unfunctionalized 2-chloropyridines using catalytic quantities of
4
BaSO ) was investigated to ascertain if poisoning the catalyst
would result in observable chemoselectivity (Table 1: entry 7).
However, analysis of the product by NMR revealed a large number
of degradation products. Based on this collection of experiments
we set the conditions for hydrodechlorination as 1 mol % Pd/C with
3
NaHCO additive for 2 h. It is also evident that no chemoselectivity
exists between hydrodechlorination and reduction of the nitro
group within 5 with the reaction rates for both processes appar-
ently equal.
A series of 2-chloropyridine derivatives was assembled through
commercial sources and standard synthetic procedures as sub-
strates to examine the relative chemoselectivity of hydrodechlori-
nation under these conditions (Table 2). In all the scope
experiments the aromatic ring of the substrate was not reduced,
attributed to the mild conditions employed. However, it is notable
that a recent report describing the use of these exact conditions
1
0% w/w Pd/C, hydrogen gas, and elevated pressures (2–3 atms)
1
8
has been reported. Reasoning that elevated pressure would
increase the rate of both hydrodechlorination and reduction, 2-
chloro-5-nitro-6-aminopyridine was stirred at 1 atmosphere for
2
4 h with Pd/C and hydrogen gas. These conditions succeeded in
fully reducing the aromatic ring, the nitro group, and the chlo-
rine–carbon bond. The effect of solvent in catalytic hydrogenations
1
9
is known to have a significant influence on rate of reaction there-
fore to eliminate this variable methanol was employed as the sol-
vent in all reactions. Similar conditions, utilized with
a
with a ClCH
the aromatic ring to the corresponding piperidine.
2 2
CHCl additive did result in full hydrogenation of
2
2
triethylamine additive, have been reported to be a mild and gen-
eral procedure to achieve hydrodechlorination with no concomi-
tant loss of aromaticity in a variety of phenyl and naphthyl
To ensure hydrodechlorination was achievable we began by
exposing 2-chloropyridine (7) (Table 2, entry 1) to the reaction
conditions, as expected after 2 h complete conversion to pyridine
was obtained. In order to investigate if selectivity of hydrodechlo-
rination is preferred at the 2-position we next exposed 2,3-dichlo-
ropyridine (9) to the reaction conditions (Table 2, entry 2).
Complete conversion to pyridine (8) in quantitative yield demon-
strated that no positional selectivity was evident. Indeed, carefully
following the reaction by NMR provided no evidence of the 3-chlo-
ropyridine product that would be expected if the 2-chloro position
is hydrodechlorinated at a faster rate.
2
0
chlorides.
Initial investigation into optimal conditions for hydrodechlori-
nation employed 2,6-dichloro-3-nitropyridine (5) as the substrate
(Table 1), reasoning that conditions that would remove two equiv-
alents of chlorine would be efficient for all subsequent substrates.
Reaction of 5 in the presence of 10 mol % Pd/C for 4 h provided
complete conversion to amine 6 (Table 1: entry 1). Decreasing
the molar percentage of the catalyst by 20-fold provided a 75%
conversion to amine 6 with unreacted starting material remaining
(Table 1: entry 2). Setting the catalyst loading at 1 mol % we next
We next wanted to examine possible selectivity between halo-
investigated the effect of reduced reaction time. 2 h stirring pro-
gens. It has been reported that the rate of hydrodehalogenation
increases with increasing electronegativity (I < Br < Cl).2
3,24
Meth-
odology similar to the investigated hydrodechlorination condi-
tions, but using a 6.4-fold excess of NaHCO has been reported to
allow the selective reduction of phenyl bromines over phenyl chlo-
rines. Intrigued by this apparent conflict within the literature we
investigated the chemoselectivity of hydrodechlorination condi-
tions on 5-bromo-2-chloro-3-nitropyridine (10) (Table 2, entry 3)
which resulted in an 80% yield of 3-nitropyridine (11). The other
product isolated was a 20% yield of 3-aminopyridine (6). Interest-
Table 1
3
Optimization of reaction conditions for hydrodechlorination
2
3
Entry Pd source Pd
mol %
Additive
Time
(min)
Conversion (%)
3
ingly, when two equivalents of NaHCO are employed to counter
both halogens a 1:1 mixture of the nitro and amino pyridine prod-
ucts are obtained. In both cases no isolation of any 3-bromopyri-
dine or 2-chloropyridine indicates no selectivity between the two
halogens. It is possible that the adjacent pyridine nitrogen plays
a role in withdrawing electron density from the 2-chloro group
reducing electronegativity to a value comparable to that of bro-
mine, thus accounting for the lack of selectivity between the two
halogens.
1
2
3
4
5
10% Pd/C
10% Pd/C
10% Pd/C
10% Pd/C
10% Pd/C
10
0.5
1
1
1
n/a1
n/a
n/a
n/a
240
240
120
60
100
75:25 (P/SM)2
100
75:25 (P/SM)
75:25 (P/SM)
NaHCO
3
60
(
2 equiv)
6
7
10% Pd/C
5% Pd/
1
1
NaHCO
3
120
240
100
(
2 equiv)
n/a
Degradation
products
It is intriguing to note that the major product of this reaction is
the non-reduced 3-nitropyridine. The quantity of catalyst used in
the reaction conditions is sufficient to reduce all three groups
BaSO
4
1
2
n/a = not applicable.
P = product; SM = starting material.
(Table 1). It would appear that the presence of bromine within
Kinarivala, Nihar
