RSC Advances
Communication
using 1,2-dimethoxyethane (DME) as a solvent does not lead to
any conversion of the starting materials (Table 1, entry 2). To
optimize the solvent-free reaction conditions, three different
families of [Pd(NHC)] pre-catalysts (NHC = N-heterocyclic carbene)
were tested: [Pd(NHC)(cin)Cl],9b,9c,10 [Pd(NHC)(acac)Cl]11 and
[Pd(NHC)(PEPPSI)],12 where acac = acetylacetonato and PEPPSI =
pyridine-enhanced pre-catalyst preparation, stabilization and
initiation (Fig. 1). These are reported to be amongst the best for
the Buchwald–Hartwig amination cross-coupling reaction. Three
different bases were also screened: KOtBu and KOtAm (KOtAm =
potassium tert-amylate), which are the most popular and efficient
bases for arylamination, and LiHMDS (HMDS = hexamethyldisi-
lazane) that has already shown high efficiency in this C–N bond
forming reaction.11c,13 As observed in Table 1, and as it was
postulated, the reaction is strongly influenced by the nature of the
ligand. It is clear that bulky NHCs (IPr* and IPr*Tol, IPr*Tol = 1,3-
bis(2,6-bis(di-p-tolylmethyl)-4-methylphenyl)imidazol-2-ylidene)
promote the reaction most efficiently. On the other hand, IPr and
SIPr (IPr = 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene; SIPr =
1,3-bis(2,6-diisopropylphenyl)-4,5-dihydroimidazol-2-ylidene) lead
to poor conversions in each case, whatever the nature of the
palladium complex (1–12%, Table 1, entries 4, 5, 7, 8 and 11). The
leaving group on the palladium pre-catalyst also plays a crucial
role. [Pd(NHC)(cin)Cl] and [Pd(NHC)(acac)Cl] derivatives promote
the reaction efficiently. On the other hand, the [Pd(NHC)(PEPPSI)]
family does not lead to any significant conversion of the starting
materials, even when the bulky IPr* ligand (Table 1, entries 7–9) is
the NHC. Finally, the best reaction conditions make use of
[Pd(IPr*)(cin)Cl] (1 mol%), using KOtAm as the base. The desired
product is then obtained in an excellent isolated yield (98%,
Table 1, entry 1). It is noteworthy that 1 mol% of the pre-catalyst is
necessary to reach complete conversion. Indeed, if the amount of
palladium is reduced by half, the conversion is proportionately
divided by nearly two (59%, Table 1, entry 3). It is important to
state here that the reaction is initially carried out at room
temperature.14 Nevertheless, during the course of the reaction, a
strong self-generated exotherm is detected in the reaction
medium. Notably, the formation of fumes can be observed in
the vial. A probe placed inside the reaction vessel indicated an
increase of temperature from 25 uC to approximately 80 uC in just
a few seconds. The subsequent decrease of temperature was then
as fast as the increase was, and the system stabilised at 25 uC after
a few seconds. After this thermal event, reaction completion was
reached. This phenomenon explains why the reaction occurs
without solvents. In the presence of DME, the heat generated by
the reaction is probably absorbed and dissipated by the solvent,
thereby disfavouring the C–N bond formation. It should also be
stated clearly here that there is a clear danger associated with such an
exotherm if the reaction were to be carried out on a very large scale.
The scope and limitations of the solvent-free amination process
were next investigated. Results are reported in Table 2. Both
arylbromides and chlorides are suitable substrates for the
reaction. To the best of our knowledge, this is the first report of
a palladium catalysed solvent-free amination of aryl chlorides
without the addition of an external source of heat. Arylbromides
displayed higher activities leading to the formation of the desired
products in slightly shorter times (Table 2, entries 2, 4 and 7 vs.
entries 1, 3 and 6). Primary amines were coupled very efficiently to
neutral aryl halides, including sterically hindered congeners, in
short reaction times (93–99%, Table 2, entries 1–9). In fact, the use
of bulkier amines appears to correlate to faster reactions.15 This
statement can notably be highlighted by the case of the simplest
aniline. No conversion was observed after 10 min and a reaction
time of 24 h was required to reach complete conversion and an
excellent isolated yield (95%, Table 2, entry 10). Electron-rich
chlorides were next involved in the coupling reaction. Excellent
conversions were obtained at longer reaction times. After 24 h, the
expected coupled products were consequently obtained in
excellent isolated yields (94–98%, Table 2, entries 11 and 12).
Using a non-aromatic primary amine does not appear to be a
limitation of the system. However this specific example required
an increase of the reaction time (93%, Table 2, entry 13). The
reactivity of the secondary amines was next explored.
N-Methylaniline and morpholine were shown to be superb
coupling partners, yielding the desired products in excellent yields
and reaction times ranging from 10 minutes to 24 h (83–97%,
Table 2, entries 14–17). Noteworthy here is that in all these
examples, the aryl chloride was a liquid, solubilising the palladium
complex and KOtAm prior to the addition of the amine.
Nevertheless, when the reaction required extended reaction times,
the exotherm was not observed, the reaction mixture became solid
after a few minutes and the coupling occurred mainly as a solid
state reaction. To confirm that the reaction was possible using
solid halides, the reactivity of 1-bromopentamethylbenzene was
finally examined. After 24 h, a conversion of 73% was measured by
Fig. 1 Pre-catalyst screening for solvent-free amination.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 3840–3843 | 3841
Chartoire, Anthony
