Inorganic Chemistry
Article
Similarly, DQ with the highest reduction potential and hence
lowest LUMO energy should be the easiest to reduce to the
corresponding radical cation. The observed trend in the kq
(1010 M−1 s−1) values, as mentioned before, is DQ (6.41) >
PQ (1.41) > OQ (0.46) > TBPQ (0.051). Thus, a ready
correlation is observed for kq with reduction potentials of the
dications with the exception of TBPQ. In other words, for the
three dications DQ, PQ, and OQ, the fluorescence quenching
rates are diffusion limited and are guided by their reduction
potentials, i.e., the higher the electron deficiency of the analyte,
the greater is the value of kq, cf. Figure 9. For example, the
highest fluorescence quenching rate (6.41 × 1010 M−1 s−1) is
observed for DQ, which has the highest Ered° (vs SCE) of
−0.35 V, while the quenching rates are comparatively lower for
PQ (1.41 × 1010 M−1 s−1) and OQ (0.46 × 1010 M−1 s−1) with
Ered° values of −0.45 V and −0.60 V, respectively. In stark
contrast, one observes that TBPQ with almost a similar
reduction potential (−0.37 V) as that of DQ shows the lowest
fluorescence quenching rate of 5.1 × 108 M−1 s−1. The reason
as to why TBPQ with a respectably high reduction potential
exhibits such an unusually low fluorescence quenching rate
should be reconciled from structural attributes. This is
evidently due to the fact that TBPQ with two 3,5-di-tert-
butylbenzyl groups is enormously large structurally such that it
is not suitable to transport into the relatively smaller voids of
the MOF to exchange the DMA cations. This clearly points to
the fact that the fluorescent Zn-DBC MOF is size-selective in
terms of cationic guest exchange.
To further corroborate the relative abilities of the dications
to quench the fluorescence of Zn-DBC, change in the
fluorescence quenching efficiency (η%) at rt was monitored
with increasing concentration of each of the dications, cf. SI. As
shown in Figure 9, η increases progressively with increasing
concentrations of the quenchers in that the highest η value is
observed for DQ. For instance, at a concentration of ca. 12
mM, one observes a quenching efficiency of ca. 95.2% for DQ,
while the same turns out being ca. 55.2% for PQ, 40.6% for
OQ, and only 1.7% for TBPQ, cf. Table 1. Thus, the trend of η
values is the same as that observed for kq, cf. Figure 9. In other
words, for dications of almost similar size, i.e., DQ, PQ, and
OQ, the fluorescence quenching rate/efficiency is dictated by
the electron deficiency, i.e., reduction potential of the analyte.
However, for TBPQ, a quencher with almost similar reduction
potential as that of DQ but bulky in size, the lowest
fluorescence quenching rate/efficiency is observed, under-
standably due to difficulty in its transportation to the
fluorophoric dibenzochrysene sites in the crystals due to
relatively smaller voids.
concentration as low as 1 μM in DMSO, due possibly to
hydrogen-bonding interactions involving the carboxylic acid
functionalities in the presence of water; complicated/abrupt
fluorescence intensity changes were observed while carrying
out the fluorescence titrations with aqueous solutions of the
dications. Steady-state fluorescence quenching experiments
were thus carried out with the solution (1 μM) of 3 in DMSO
(λex = 350 nm; λem = 420 nm) and aqueous solutions of the
the basis of the Stern−Volmer quenching constants thus
derived and the singlet lifetime of 3 (1τ = 7.0 ns) determined
from a time-resolved study (Figure S11), bimolecular
quenching rate constants (kq) were calculated for the dications.
The trend of kq (109 M−1 s−1) for fluorescence quenching of 3
thus turned out to be DQ (10.6) > TBPQ (8.0) > PQ (5.4) >
OQ (3.9), which is in accordance with their redox potentials.
This suggests that for the model system, the fluorescence
quenching is dictated solely by the e-deficiency of the
dicationic analytes and that the model system can not
distinguish the latter by their size; in other words, a sterically
bulky analyte (TBPQ) is sensed with a much faster rate than
analytes of smaller size (PQ and OQ).
Overall, Zn-DBC MOF offers the following advantages over
the pure organic system 3 in terms of fluorescence sensing of
dicationic analytes: (i) The unique stability of the MOF in
water permits sensing of the analytes in aqueous medium
without the need of an organic solvent. (ii) Bimolecular
quenching rates are much faster in case of the MOF (ca. 1010
M
−1 s−1) as compared to the model system (kq on the order of
109 M−1 s−1), which points to better sensitivity of detection.
(iii) Fluorescence sensing of the analytes with the MOF is
controlled by both size and reduction potential, while for the
model system, the same is dictated only by the reduction
potential, and (iv) the MOF can distinguish between the two
herbicides DQ and PQ better than the model system in terms
of the ratio of fluorescence quenching rate constants, i.e.,
kq(DQ)/kq(PQ); the latter are ca. 4.5 and 2.0 for the MOF
and model system, respectively. Two plausible reasons24−27 for
the observed contrasting differences between MOF and the
organic model system are (i) rigidification of the fluorophore
when present as part of the MOF and (ii) confinement of the
cationic guests within the pores of certain size/shape.
To assess the sensitivity limits of Zn-DBC for detection of
the dicationic herbicides, fluorescence quenching experiments
were carried out at very low concentrations of DQ, a
representative case, in water, cf. Figure S14. Accordingly, a
plot of quenching efficiency (η%) versus the quencher
concentration reveals that the minimum concentration for
detection of DQ is ca. 15 μM, cf. Figure S14. In other words,
the Zn-DBC MOF permits fluorescence sensing of DQ with a
detection limit of ca. 2.8 ppm. As mentioned at the outset,
although LMOFs with tunable luminescence properties have
been employed for diverse sensing applications, they have
heretofore been not explored quite inexplicably for fluores-
cence detection of toxic herbicides. The results unveiled herein
offer intriguing mechanistic insights of temperature-dependent
emission behavior of an LMOF based on the dibenzo[g,p]-
chrysene fluorophore and application of the latter for “turn-off”
fluorescence sensing of toxic herbicides in water with a
detection limit of ca. 2.8 ppm. The results thus constitute an
invaluable addition to the gamut of ever-increasing applications
being uncovered for MOFs in general.
It is noteworthy that the Zn-DBC MOF offers a unique
advantage over the precursor organic linker H4DBC in terms of
fluorescence sensing of the bipyridinium dications. For
instance, the latter could be sensed by the MOF suspension
in water without employing any organic solvent, whereas the
same was not possible in the case of the organic linker due to
its complete insolubility and obvious aggregation in aqueous
media. For further comparison of the advantageous attributes
of the MOF versus organic system, fluorescence quenching
experiments were carried out with a model system, namely,
tetramethoxy ether of dibenzo[g,p]chrysene, i.e., 3 in Scheme
1, which is devoid of carboxylic acid functionalities at the
periphery. Notably, the reason for employing the tetraether 3
instead of the tetra-acid H4DBC itself for quenching studies is
that the latter was found to undergo aggregation even at a
I
Inorg. Chem. XXXX, XXX, XXX−XXX