H. Shen et al.
Applied Catalysis A, General 623 (2021) 118244
S4, in which the typical peak value (2 θ = 23◦) of the two-dimensional
material was clearly observed. However, no obvious palladium signal
was observed because of its low loading and high dispersion. Raman
spectra in Fig. 3g showed further characterization. Compared with the
precursor, these two-dimensional materials showed widened peaks or
bands. The Raman shifts of NGDY and GDY in D and G bands are
different, but there is little difference in the vibration band of dialkynyl
group, indicating that nitrogen doping in NGDY induced more electron
changes and defects. Metal nanoparticles anchored hybrid materials,
such as Pd-NGDY, showed typical D-band, G-band and dialkynyl vi-
bration modes at 1349 cmꢀ 1, 1540 cmꢀ 1, and 2172 cmꢀ 1. Pd NPs
loading enhanced D/G ratio and slightly shifted D and G bands to lower
wave number as compared with GDY and NGDY.
reaction between bromobenzene and phenyl boric acid. Ethanol / water
(7:4), K2CO3, and 80 ℃ were optimized as the best solvents, base and
temperature for this reaction, without the need of phase transfer agents.
Only one biphenyl product was observed in all reactions. The lines
chart in Fig. 5 comprehensively showed the catalytic activity of different
catalysts, and the conversion rate changes with time (taking the
coupling of iodobenzene and phenylboronic acid as the model reaction).
The catalytic performance of Pd-NGDY is better than that of Pd-C and
Pd-GDY. With the catalysis of Pd-NGDY, the conversion reached 75 % in
30 min, then over 90 % in 1 h, with the TOF number of 1287 hꢀ 1, and the
substrate/catalyst ratio was 497. In contrast, the coupling reaction ef-
ficiency of commercial Pd-C and Pd-GDY catalyst was relatively low.
The conversion rate was only about 2/3 after 1 h, and the TOF value is
about 700 hꢀ 1 for Pd-C. N-doping and chemical functional groups of
supporting materials could promote the adsorption and activation for
reactants [44], therefore Pd-NGDY showed superior catalytic perfor-
mance than commercial Pd-C catalyst. At the same time, there were
more borate coupling by-products at the end of the reaction catalyzed by
Pd-GDY, which indicated that the N-doping of graphdiyne was benefi-
cial to heterocoupling to a certain extent and inhibited homocoupling.
Then, more than ten experiments of catalyst recovery were carried out.
Fig. 5 also showed the progress of the 12th cycle of the coupling reac-
tion, and Fig. S6 provided the high-resolution TEM image of Pd-NGDY
after the 12th cycle. The catalytic activity of Pd NPs on NGDY was
almost unchanged after 12th recycle, and there was no obvious aggre-
gation of Pd NPs, indicating that Pd-NGDY was quite stable and
recyclable.
K-edge XANES profiles and Fourier transformed EXAFS spectra
provide coordination information between metal NPs (taking Pd as an
example) and substrate (Fig. 3h, I). Pd foil was set as the standard
(EXAFS fitting parameters are showed in Table S6). Compared with Pd-
GDY, the XANES profile of Pd-NGDY (Fig. 3h) was less similar to that of
Pd foil, and more similar to that of Pd-N [51], which meant higher
Pd2+/Pd◦ ratio and less Pd atoms aggregation in Pd-NGDY (in line with
XPS results), that is, N doping is beneficial to NPs dispersion. The Fourier
transformed EXAFS spectra (Fig. 3i) provided Pd-Pd and Pd-N bonding
details for Pd-NGDY and Pd-GDY. In Pd-NGDY, Pd-N and Pd-Pd bond
distances were 0.203 and 0.275 nm, respectively, and the coordination
numbers (CN) for Pd-N and Pd-Pd were 1.6 and 3.4, respectively.
However, Pd-GDY showed a slightly shorter Pd-Pd bond distance (0.275
nm), and the coordination number was fitted as 6.6. At the same time,
Pd-NGDY showed a different Pd-N peak profile at ~ 0.15 nm as
compared with Pd-GDY and Pd foil, which also indicated the different
coordination environment in Pd-NGDY, thus influence of the charge
transfer between metal NPs and NGDY could be expected.
Brunauer-Emmet-Teller (BET) analysis was performed to determine
the specific surface area of NGDY and Pd-NGDY. As shown in Fig. 4, N2
adsorption and desorption isotherms indicated type IV curves for both
NGDY and Pd-NGDY, and the BET surface area were 535 m2 gꢀ 1 and 152
m2 gꢀ 1, respectively, which were at high level for carbon materials [37],
implying high porosity of NGDY. Anchoring of Pd NPs on NGDY caused
large decrease of N2 adsorption of the material. The BET surface area of
GDY was determined to be 666 m2 gꢀ 1 (Fig. S5), a bit larger than NGDY.
3.2. Catalytic performance in the Suzuki-Miyaura reaction
Among all the prepared nanohybrid materials, Pd-NGDY was
selected as the catalyst to investigate its activity in Suzuki-Miyaura re-
action, and Pd-C was used as the control catalyst. Pd-GDY, an analogue
with Pd NPs anchored on all carbon material GDY, was also involved in
order to understand the role of substrate materials in catalysis. In order
to obtain the best catalyst conditions, various solvents (Table S3), bases
(Table S4) and temperatures (Table S5) were considered in the coupling
Fig. 5. Progress of Suzuki-Miyaura reactions under the catalyzing of different
catalysts. Conditions: 0.5 mmol iodobenzene, 0.75 mmol phenylboronic acid, 1
mmol K2CO3, Ethanol/Water (7:4), 80 ℃; 10 mg Pd-NGDY, reused Pd-NGDY,
and Pd-C catalysts. Inset: TOF number of reactions using different catalysts.
Fig. 4. N2 adsorption and desorption isotherms of a) NGDY and b) Pd-NGDY.
6