G. Kumar et al. / Journal of Catalysis 371 (2019) 298–304
299
In this scenario, the sustainability of organo-photo redox dyes
has been validated by several groups, [26,27] where EOSIN-Y turns
to be very promising, with the inherent difficulty related to its
tough separation from homogeneous conditions. Nevertheless,
the disadvantage of separation can be eradicated by hooking EOSIN
Y on solid support. In fact, Xiaobin Fan covalently hooked EOSIN-Y
on reduced graphene oxide for photo catalytic applications, [28]
while Cooper et al. successfully integrated Rose Bengal into conju-
gated micro porous polymers, [29] Nasalevich et al. implemented
methyl red in a MOF (NH2-MIL-125(Ti) [30] although such reports
are only handful.
modified framework exhibits well matched diffraction peaks, indi-
cating that the integrity and crystallinity of the UiO-66-NH2 frame-
work is well maintained during the coupling reaction. Given the
covalent structure has no obvious effect on the mother framework,
the PXRD result additionally attribute to the relatively low content
of EY with its well dispersed small particle size in EY@UiO-66-NH2.
Fig. S1 (supplementary data) compares the TGA curves of the pris-
tine MOF and EY@ UiO-66-NH2. The first weight losses upto 120 °C
is ascribed to the release of moisture, and solvent both in pristine
and modified MOF. The second weight loss in UiO-66-NH2 is attrib-
uted to the slow decomposition of material from 400 to 700 °C.
EY@UiO-66-NH2 shows weight loss from 150 to 400 °C, which is
ascribed to the degradation of EY, and second weight loss from
400 to 700 °C correlates to the degradation of pristine MOF. There-
fore, modified EY@UiO-66-NH2 shows comparatively higher
weight loss than pristine MOF at 700 °C. Fourier transform infrared
(FT-IR) spectroscopy provides strong evidence for the covalent
immobilization of EY (Fig. S2). For UiO-66-NH2, the absorption
peaks between 600 and 800 cmꢀ1 in the FT-IR correspond to ZrAO
as longitudinal and transverse modes, while intense doublets at
1433 and 1389 cmꢀ1 can be assigned to the stretching modes of
the carboxylic groups in the BDC-NH2 ligand. Two obvious absorp-
tion peaks at 3394 and 3242 cmꢀ1 were attributed to the asym-
metric and symmetric vibrations of NAH bonding in the amino
group. In the spectrum of EY, OAH stretching at 3136 cmꢀ1, C@O
stretching at 1591 cmꢀ1, and OAH bending at 1339 cmꢀ1 prove
the existence of free ACO2H groups. As for the modified EY@UiO-
66-NH2, most of the IR vibration bands are fully consistent with
those of the pristine framework, suggesting the composite have
the similar structure as of UiO-66-NH2. Importantly, the absorption
peaks for free amino groups in the BDC-NH2 ligand disappear after
the covalent functionalization with EY. Besides, the typical signal
of COANH at 1635 cmꢀ1 and the characteristic absorption of the
NAH stretching band at 3344 cmꢀ1 appears in the spectra of the
EY@UiO-66-NH2. Moreover, emergence of two new bands at
1615 cmꢀ1 and 1046–1090 cmꢀ1, corresponding to the vibration
of C@O and CAO from EY, respectively, confirms that ACO2H of
EY successfully reacted with ANH2 moiety in the framework.
It should be noted that due to the introduction of EY by dehy-
drate coupling reaction, the NAH stretching vibration and CAN
stretching vibration become less prominent. Additionally, after
the dehydration, the ACO2H bands from EY disappear (Fig. S2),
indicating the successful conversion of carboxyl groups into
O@CAN. The nitrogen adsorption isotherm of the as-prepared sam-
ple indicates a steep uptake at low relative pressure, and belong to
the typical type I isotherm, revealing microporous structure of the
framework (Fig. 1b). The BET surface area of pristine UiO-66-NH2 is
smaller than UiO-66 due to incorporation of pendent ANH2 groups
in pore structure [48]. After introduction of EY, the BET surface area
of EY@UiO-66-NH2 further decreases to 395.12 m2/g due to encap-
sulation of EY inside the pore. It should be noted that the BET sur-
face area is still adequate to supply more surface active sites for
enhanced photo catalytic performance.
Based on the aforesaid finding, and keeping our target to cova-
lently modify EOSIN-Y for efficient photo-catalytic applications, we
turned our attention to porous metal-organic frameworks (MOFs)
[31]. MOFs have emerged as promising heterogeneous catalysts
because of their (i) well-defined and tunable pores, (ii) versatile
structures, and (iii) easy functionalization [32–34]. Furthermore,
MOFs often endure covalent linkage with organic moieties even
after de novo synthesis, commonly called as post synthetic modifi-
cation (PSM), benefitting a way to further functionalize these
materials [35]. In this context, free amine moiety containing MOFs
are widely used because of its easy accessibility to take part in var-
ious reactions [36]. However, chemical stability of MOFs is of pri-
mary concern, because of the weak coordination bond, that limits
their applications towards such covalent functionalization [37].
To this end, employing Zr(IV) based metal nodes can dramatically
increase the stability of the MOFs [38]. The UiO-66-NH2 framework
is highly chemically stable, even defiant to acids [39], bases [40]
moisture [41] and shows extraordinary mechanical stability [42].
Cohen [43] and Tilset [44] pioneered the amino functionalized
UiO-66-NH2 using 2-amino-1,4-benzenedicarboxylate, where the
–NH2 group remains free and acts as potential functional organic
site (FOS) [45]. Afterwards, several other research groups have uti-
lized the post-synthetic activity of this free amine moiety, by using
anhydrides, [46] phthalocyanines [47] or amide functionalised
pores [48]. Keeping in mind the aforesaid rationale, we aimed to
heterogenise the organo-photo redox EOSIN-Y via PSM between
its free acid group and –NH2 functionality of UiO-66-NH2 by using
DCC as a dehydrating coupling reagent. The post synthetically
modified EY@UiO-66-NH2 was characterized by a battery of exper-
imental evidences including PXRD, FT-IR, XPS, NMR, CHN analysis
and used further for oxidative cyanation of tertiary amines. The
versatility of the photo-catalyst towards various other nucle-
ophiles was also cross-checked that revealed high yield of the
desired product is maintained in all the cases. Recyclability test
shows no loss of activity even upto ten cycles. Thus, heterogenizing
the EOSIN-Y via PSM in UiO-66-NH2 provides two fold benefit, (i)
avoiding the separation difficulty of homogeneous catalysis, and
(ii) controlling the adverse environmental effects of toxic organic
dye.
2. Results and discussion
We also recorded solid state 15N NMR of pristine UiO-66-NH2,
and its post synthetically modified analogue (Fig. 1c). While the
peaks of free NH2 in the pristine framework were located at –
322.79 ppm, the ANH of amide group in EY@UiO-66-NH2 down
field shifted (region ꢀ266.01 ppm) with no residue signal left at
–322.79 ppm that confirms the successful post synthetic covalent
modification. To demonstrate such small shift of the 15N NMR, cor-
responding to modification of amine to amide in the framework,
we conducted quantum chemical calculations using GAUSSIAN
09 [50]. Structures were fully optimized at B3LYP/genecp 6-31+g
(d) lanl2dz level of theory. Gauge-Independent Atomic Orbital
(GIAO) method [51] was used for calculating 15N chemical shift
at same level of theory. Model structures were used for calculation
2.1. Synthesis and characterization of the photocatalyst
The parent framework UiO-66-NH2 was prepared by using the
reported synthetic procedure [49]. For the as-prepared UiO-66-
NH2, the experimental powder X-ray diffraction (PXRD) pattern
exactly matches with the calculated one [48] confirming phase
purity (Fig. 1a).
As already described above, EY was introduced into the frame-
work via single step dehydrate coupling of the carboxyl group of
EOSIN-Y free acid (EY-FA) with primary amine group of UiO-66-
NH2 by using N,N-dicyclohexylcarbodiimide (DCC) as a dehydrat-
ing reagent in DMF solution. The PXRD pattern of the post-