D. Kundu et al. / Reactive & Functional Polymers 79 (2014) 8–13
11
furnish the palladium loaded monolith (Pd-AAFM). Successful metal
loading on the monolith was suggested by brownish discoloration.
Infrared (IR) spectra of the original PAN monolith, AAFM and
Pd-AAFM are compared in Fig. 2. In the spectrum of the PAN mono-
lith, the peaks attributed to aliphatic hydrocarbons in the polymer
chain are seen around 2900 cmÀ1 (CAH stretching). The peaks at
2250 cmÀ1 (C„N stretching) and 1732 cmÀ1 (C@O stretching) are
characteristic of nitrile and ester groups, respectively [48]. Note
that the commercial PAN of the present use contained 7 wt% of vi-
nyl acetate as comonomer. The reaction with EDA caused signifi-
cant changes in the spectrum as expected. The intensity of the
nitrile signal at 2250 cmÀ1 decreased, which was accompanied by
the emergence of new peaks around 3300 and 1613 cmÀ1 due to
the formation of amidine/amine groups [43]. The disappearance
of the carbonyl signal at 1732 cmÀ1 was caused by deacetylation
of the acetate [48]. The 1613 cmÀ1 peak shifted to 1593 cmÀ1 in
the spectrum of Pd-AAFM, which is a sign of metal coordination
to the amidine group [49]. The metal coordination was further sup-
ported by the NMR results shown below.
Fig. 3. Solid-state 13C MAS NMR spectra of (a) the original PAN monolith (b) AAFM
Fig. 3 shows solid-state 13C magic-angle-spinning nuclear mag-
netic resonance (MAS NMR) spectra of the three monoliths. The
large peak around 30 ppm includes the signals of all aliphatic car-
bons in the PAN main chain. The peak at 121 ppm is assigned to
the nitrile carbon. The small peaks at 67 and 171 ppm came from
the comonomer. Several new signals appeared in the spectrum of
AAFM, whereby the introduction of amidine/amine groups was con-
firmed here as well; the peaks at 45 and 52 ppm are characteristic of
the ethylene group from EDA. The signals at 162 and 173 ppm are
assigned to the carbons of amidine groups that are nonconjugated
and conjugated, respectively (see Fig. 1) [43]. Noteworthy is the
162 ppm peak significantly shifted downfield and broadened in
Pd-AAFM, while the position of the 173 ppm peak remained un-
changed. These results point to a possibility that the bidentate metal
coordination of the nonconjugated amidines was dominant in the
present case over the other coordination modes [49–51].
Cross-sectional analyses of the PAN monolith, AAFM and
Pd-AAFM were performed by scanning electron microscopy
(SEM) (Fig. 4). The three dimensionally interconnected porous
structures were found for all materials, showing that the mono-
lithic structure was largely maintained throughout the chemical
modifications. In addition, cross-sections of Pd-AAFM were ana-
lyzed by energy dispersive X-ray spectrometry (EDX), proving that
Pd-AAFM was mainly composed of carbon, nitrogen, palladium and
chloride as expected (Fig. S1). Elemental mapping on the cross-sec-
tions indicated homogeneous distribution of palladium over the
polymer matrix (Fig. S2). The palladium amount was estimated
and (c) Pd-AAFM.
by inductively coupled plasma atomic emission spectroscopy
(ICP-AES) to be 69.3 mg (0.651 mmol) per gram of AAFM.
In order to assess mesoscopic porosity of the monolith, nitrogen
adsorption–desorption experiments were performed. The adsorp-
tion–desorption isotherms of the PAN monolith and Pd-AAFM are
shown in Figs. S3 and 5, respectively. These isotherms followed a
typical type IV isotherm with a H2 type hysteresis loop in the P/
P0 range from 0.4 to 0.9. The isotherm of Pd-AAFM showed a grad-
ual increase of the adsorption amount at the relative pressure of P/
P0 = 0.01–0.35 but exhibited a sharp increase in the P/P0 range of
0.4–0.9. After that the adsorption saturated, the specific surface
area and the pore volume of Pd-AAFM were estimated by the Bru-
nauer–Emmett–Teller (BET) method to be 124.3 m2 gÀ1 and
0.108 cm3 gÀ1, respectively. These values are slightly smaller than
139.2 m2 gÀ1 and 0.128 cm3 gÀ1 estimated for the PAN monolith,
respectively. A remarkable difference between the two monoliths
was found in the pore size distribution estimated by the non-local
density functional theory (NLDFT) method [22]. The pore size of
the original PAN monolith was distributed around 5.3 nm, whereas
the distribution became bimodal with peaks around 3.8 and 1.5 nm
for Pd-AAFM. These results suggest that the smaller pores were
opened up by the chemical treatments.
Suzuki–Miyaura cross-coupling reaction was demonstrated by
using Pd-AAFM as catalyst. 4-Iodoanisole and 4-methylboronic
acid were first chosen as substrates. The reaction was performed
under various conditions using a water/ethanol mixed solvent with
an inorganic base at 55 °C (Table 1) [52]. In the presence of Pd-
AFFM the reaction proceeded to afford the product in the yield of
60–96%, while no reaction occurred in its absence. A decreased
amount of Pd-AAFM lowered the yield. These results evidenced
that Pd-AAFM catalyzed the reaction. The reaction in water gave
the lowest yield (60%), which could be explained by poor solubility
of the substrates. The yield was improved by increasing the ethanol
content to be water/ethanol 1:1 (96%). However, further increases
to 1:2 and 0:1 lowered the yields (78% and 75%, respectively),
which may be caused by decreased solubility of the inorganic base.
The use of weaker bases (K2CO3 and CsCO3) resulted in higher
yields than stronger bases (NaOH and KOH) that are prone to cause
more side reactions such as protodeboronation [53,54].
The reaction was also tested by using other aryl halides and
boronic acids as substrates (Table 2). Despite the difference in
halogenide group (bromide or iodide) and para-substituent (elec-
tron donating or withdrawing group), the reaction gave the corre-
sponding product in a high yield (94–97%). These results point to
the versatility of the present protocol using Pd-AAFM as catalyst.
Fig. 2. IR spectra of (a) the original PAN monolith (b) AAFM and (c) Pd-AAFM.