Chemistry of Materials
Article
istics were determined by NH3- and CO2-temperature-
programmed desorption (TPD), respectively. As shown in
Figure S14, YS-TpPa@UiO-66-(COOH)2 exhibited the CO2
desorption peak within the range of 150−300 °C. It can
therefore be inferred that basic sites existed in the TpPa. The
adsorbed CO2 could be assigned to the type of CO2 that has
interacted with the NH species in the TpPa. Two main NH3
desorption peaks were observed in the NH3-TPD of YS-
TpPa@UiO-66-(COOH)2, which are caused by the combina-
tion of physical adsorption and chemical adsorption on the
acid sites. The TpPa shell with periodic pores and the cavity
between the shell and core did not hinder the exposure of acid
sites in UiO-66-(COOH)2. YS-TpPa@UiO-66-(COOH)2 can
effectively catalyze the one-pot tandem D−K condensation of
benzaldehyde dimethyl acetal and malononitrile (Figure 5a).
The tandem D−K condensation involves two reaction steps.
The first step is the deacetalization of benzaldehyde dimethyl
acetal with water to generate benzaldehyde over the acid sites.
In the second step, Knoevenagel condensation of benzalde-
hyde with malononitrile proceeds over the basic sites, yielding
the unsaturated product, which is an important intermediate in
the synthesis of natural products, pharmaceuticals, and
functional polymers. The conversion of the reactant and the
selectivity of the target product were monitored by gas
chromatography and mass spectrometry. As expected, the
result showed that the tandem reaction can be efficiently
catalyzed by YS-TpPa@UiO-66-(COOH)2 with 99% yield of
the benzylidene malononitrile product (Figure 5b). The
catalytic performance is among the best compared to other
heterogeneous catalytic systems that have been reported so far
(Table S2). To better understand the catalytic activity of yolk−
shell TpPa@UiO-66-(COOH)2, several control experiments
were investigated. As shown in Figure 5c, no product was
detected without the addition of YS-TpPa@UiO-66-
(COOH)2. When UiO-66-(COOH)2 was added into the
reaction system, the deacetalization took place smoothly while
the Knoevenagel reaction almost did not occur. When YS-
TpPa@UiO-66, core−shell TpPa@NH2-UiO-66, and TpPa
were employed in the catalytic reaction, respectively, very low
conversion of benzaldehyde dimethyl acetal was observed,
owing to the absence of acid sites in these materials, implying
the importance of the cooperativity between acid and base sites
for the D−K one-pot tandem reaction. The core−shell-
structured Am-TpPa@UiO-66-(COOH)2 showed a relatively
low ability to fulfill the one-pot reaction with 99% conversion
of benzaldehyde dimethyl acetal, 75% yield of benzaldehyde,
and 24% yield of benzylidene malononitrile under identical
reaction conditions, indicating the fact that the crystalline COF
shell with ordered pores and the hollow cavity to facilitate the
mass transfer and inhibit the annihilation of acid−base sites is
crucial for the D−K condensation reaction. The use of a
random mixture of UiO-66-(COOH)2 and TpPa leads to a
99% conversion of benzaldehyde dimethyl acetal and only 52%
yield of benzylidene malononitrile mainly because of the
annihilation of the opposite groups on the surface, thus
resulting in destruction of cooperative catalysis.60 These results
confirmed that the separation of the acidic and basic sites
within the yolk−shell-structured COF@MOF is quite
important to achieve the high catalytic performance. The
catalyst can be used at least 5 times without obvious loss of
catalytic activity (Figure 5d). The PXRD peaks and FTIR of
the recovered YS-TpPa@UiO-66-(COOH)2 match well with
those of the as-synthesized sample (Figure S15), indicating the
intact integrity of the structure.
The possible catalytic mechanism of YS-TpPa@UiO-66-
(COOH)2 was proposed (Figure 5e). First, the starting
material of benzaldehyde dimethyl acetal was passed through
the ordered pores in the COF shell. Then, the deacetalization
of benzaldehyde dimethyl acetal catalyzed by carboxylic groups
in the UiO-66-(COOH)2 core afforded a benzaldehyde
intermediate. Finally, the Knoevenagel condensation of
benzaldehyde and malononitrile took place in the TpPa
shell, yielding the benzylidene malononitrile product. The
Knoevenagel condensation is supposed to be the rate-
determining step.61 On account of the diversity and
tailorability of COFs, yolk−shell COF@MOF containing
stronger base sites with higher efficiency could be expected
to be developed.
CONCLUSIONS
■
Series of yolk−shell-structured COF@MOF nanocomposites
were successfully constructed through a template-free
solvothermal method. Time-dependent experiments and
characterization findings showed that the formation of a
hollow cavity between the core and the shell resulted from the
transformation of an amorphous external shell with a
disordered structure to a crystalline network with an ordered
structure along with the in situ shrinkage of the shell. The
representative yolk−shell-structured YS-TpPa@UiO-66-
(COOH)2 nanocages possessing spatially distributed acidic
and basic sites showed synergistically enhanced catalytic
performances in the one-pot D−K tandem reaction compared
to that of its homologues. Importantly, benefiting from the
diversity and adjustable nature of MOFs and COFs, the
systematic design of functionality of COFs@MOFs can be
facilely targeted. We envision that the future research on
combinations of MOFs and COFs in yolk−shell composites
will allow for various novel functions.
EXPERIMENTAL SECTION
■
Synthesis of YS-UiO-66-(COOH)2@TpPa. The YS-TpPa@UiO-
66-(COOH)2 was prepared using the two-step method. In the first
step, core−shell microspheres Am-TpPa@UiO-66-(COOH)2 consist-
ing of UiO-66-(COOH)2 in the core and an amorphous polyimine
network in the shell were prepared. Typically, the dispersion of UiO-
66-(COOH)2 (16 mg) in tetrahydrofuran (11 mL) was mixed with Pa
(9 mg) under ultrasonication for 30 min, resulting in a milk-white
homogeneous solution. The obtained solution was kept at 50 °C for
30 min under vigorous stirring allowing for the interaction between Pa
and UiO-66-(COOH)2. Subsequently, a solution of 1,3,5-triformyl-
phloroglucinol (Tp, 12 mg) in tetrahydrofuran (4 mL) was added to
the abovementioned dispersion at a feeding rate of 0.4 mL/min, and
the reaction was proceeded for 12 h. The solvent was removed by
rotary evaporation to obtain the core−shell Am-TpPa@UiO-66-
(COOH)2 as red powder. In the second step, the Am-TpPa@UiO-66-
(COOH)2 microspheres were dispersed in a mixed solvent of 1,2-
dichlorobenzene and n-butanol (1.35 mL/0.15 mL) in a Pyrex tube
(body length of 20 cm). Pyrrolidine (0.15 mL) as a catalyst was added
into the Pyrex tube. The tube was then flash-frozen at 77 K (liquid N2
bath) and degassed by three freeze−pump−thaw cycles, before
evacuating to a pressure of 0.5 mbar. The tube was flame-sealed,
reducing the total length by ca. 15 cm. After gradual warming up to
room temperature, the tube was heated at 120 °C for 3 days. The
solid was collected by filtration, washed with acetone, and dried at 60
°C under vacuum for 12 h to yield YS-TpPa@UiO-66-(COOH)2 as
crimson powder.
5696
Chem. Mater. 2021, 33, 5690−5699