2
F. Chahdoura et al. / Catalysis Communications xxx (2014) xxx–xxx
2
. Experimental
the solvent evaporated under reduced pressure. The product was puri-
fied by short-column chromatography on silica gel. The corresponding
product was identified by comparison of its H and C NMR spectra
and GC–MS data with those of an authentic sample.
1
13
General part including instruments and synthesis of ligands and
PdNP are provided in the Supporting Information.
2
.1. Pd-catalysed C\C cross-couplings
2.4. Catalytic phase recycling
The methodology corresponding to the Heck–Mizoroki coupling is
here described; for the other couplings, see the Supporting Information.
mmol of aryl halide (187 mg for 4-bromoanisole; 234 mg for 4-
iodoanisole), 1.2 mmol of the corresponding alkene (124.8 mg for
styrene; 84 mg for 3-buten-2-one) and tBuOK (235 mg, 2.1 mmol)
were consecutively added to a 1 mL of solution of preformed nano-
particles in glycerol (0.01 mmol Pd, 1.0 mol%). The resulting mixture
was heated at 100 °C for 2 h and then cooled to room temperature.
The organic products were extracted from the catalytic mixture with
dichloromethane (3 × 10 mL) and the combined organic phases were
A typical experimental procedure to recycle the catalytic phase is
described. The glycerol phase from the previous run was maintained
for 30 min under dynamic vacuum while stirring at 100 °C. The corre-
sponding reagents were then added to the glycerol phase. The catalytic
mixture was treated under the corresponding conditions applied for the
first run. The product was extracted with dichloromethane and
analysed by GC and NMR.
1
3. Results and discussion
2 4
dried over anhydrous Na SO , filtered and the solvent evaporated
under reduced pressure. The product was purified by short-column
chromatography on silica gel. All the products were identified by com-
parison of their H and C NMR spectra and GC–MS data with those
3.1. Synthesis of palladium nanoparticles, PdLx
1
13
In our previous work, we used TPPTS as stabiliser [17], which is lim-
ited in regard to the ligand structure tuning. With the aim to study the
effect of the ligand on both the NP stabilisation and catalytic behaviour,
we decided to prepare a series of N-substituted PTA derivatives, 1–8
of authentic samples.
2
.2. Pd-catalysed hydrogenation
(Scheme 1 and Scheme S1) (ligands 1 [26] and 8 [27] were previously
1
mmol of alkene (146 mg for (E)-4-phenyl-but-3-en-2-one;
reported). These ligands were fully characterised, including X-ray dif-
fraction studies on single crystal for 1–7 (Table S1 and Figure S1); all
of them are soluble in glycerol, in contrast to their precursor, PTA.
PdNP were synthesised in neat glycerol, starting from Pd(II) precur-
1
46 mg for coumarin; 165 mg for 4-nitroacetophenone; 180 mg for
trans-stilbene) was added to a 1 mL of solution of preformed nanopar-
ticles in glycerol (0.01 mmol Pd, 1.0 mol%). The resulting mixture was
stirred at room temperature under argon in a Fisher–Porter bottle. The
system was then pressurized with dihydrogen (3 bar) and stirred at
00 °C for 3 h. The mixture was then cooled to room temperature. The
catalytic mixture was extracted with dichloromethane (3 × 10 mL)
and the combined organic phases were dried over anhydrous Na SO
filtered and the solvent evaporated under reduced pressure. The prod-
uct was purified by short-column chromatography on silica gel. All the
products were identified by comparison of their H and C NMR spectra
and GC–MS data with those of authentic samples.
sors (Pd(OAc)
2 2
(a); [PdCl (cod)] (b) where cod = 1,5-cyclooctadiene),
at 60 °C under dihydrogen atmosphere (3 bar), in the presence of
the corresponding N-alkylated PTA-based ligand (1–8) in a ratio Pd/L
of 1/1 (Scheme 1) [17].
The colloidal solutions obtained for both palladium precursors with
the different ligands were analysed by TEM (Tables S2 and S3), because
of the negligible vapour pressure of glycerol (b1 mmHg at 50 °C) which
permits to work under high vacuum conditions without isolating the
1
2
4
,
1
13
2
nanoparticles at solid state [28]. For [PdCl (cod)] (Table S2), we could
notice that the best dispersion and homogeneity in terms of size
and shape of nanoparticles were observed for ligands with non-
functionalised aryl substituents, Pd1a and Pd3a. PdNP containing
the benzyl group, Pd1a, were the most regular in size (Fig. 1). How-
ever those PTA-based ligands containing alcohol (2), thioether (4)
and ether (5 and 6) groups were not well-dispersed. Ligand 7,
involving a long alkyl chain, favoured the trend to agglomeration
probably due to the hydrophobic interactions with glycerol, while
the ligand bis-PTA 8 gave non homogenous particles. A similar be-
2
.3. Pd-catalysed sequential process: C\C cross-coupling followed by
hydrogenation
4
-Iodoanisole (234 mg, 1 mmol), phenyl boronic acid (145 mg,
1
.2 mmol) and tBuOK (235 mg, 2.1 mmol) were consecutively added
to a 1 mL of solution of preformed nanoparticles in glycerol
0.01 mmol Pd, 1.0 mol%). The resulting mixture was heated at 100 °C
(
during 2 h and then cooled to room temperature. The system was
then pressurized with dihydrogen (3 bar) and stirred at 100 °C for 2 h.
The mixture was then cooled to room temperature. The catalytic mix-
ture was extracted with dichloromethane (3 × 10 mL) and the com-
2
haviour could be noted for Pd(OAc) used as metallic precursor
(Table S3), but the particles with ligand 1 (Pd1b) were less homoge-
neous in size than that observed for Pd1a (Fig. 1). Ligands containing
ether groups, 6 and 7, gave agglomerated (Pd6b) and micelle-type
2 4
bined organic phases were dried over anhydrous Na SO , filtered and
Scheme 1. Synthesis of palladium nanoparticles stabilized by N-substituted PTA derivatives, PdLx.