Modification of extended open frameworks with fluorescent tags for sensing explosives: Competition between size selectivity and electron deficiency

Article abstract of DOI:10.1002/chem.201302455

Three new electron-rich metal-organic frameworks (MOF-1-MOF-3) have been synthesized by employing ligands bearing aromatic tags. The key role of the chosen aromatic tags is to enhance the π-electron density of the luminescent MOFs. Single-crystal X-ray st

Full text of DOI:10.1002/chem.201302455

DOI: 10.1002/chem.201302455
Full Paper
&
MOFs
Modification of Extended Open Frameworks with Fluorescent
Tags for Sensing Explosives: Competition between Size Selectivity
and Electron Deficiency
Bappaditya Gole, Arun Kumar Bar, and Partha Sarathi Mukherjee*[a]
Abstract: Three new electron-rich metal–organic frameworks
(MOF-1–MOF-3) have been synthesized by employing
ligands bearing aromatic tags. The key role of the chosen ar-
omatic tags is to enhance the p-electron density of the lumi-
nescent MOFs. Single-crystal X-ray structures have revealed
that these MOFs form three-dimensional porous networks
with the aromatic tags projecting inwardly into the pores.
These highly luminescent electron-rich MOFs have been suc-
cessfully utilized for the detection of explosive nitroaromatic
compounds (NACs) on the basis of fluorescence quenching.
Although all of the prepared MOFs can serve as sensors for
NACs, MOF-1 and MOF-2 exhibit superior sensitivity towards
4-nitrotoluene (4-NT) and 2,4-dinitrotoluene (DNT) compared
to 2,4,6-trinitrotoluene (TNT) and 1,3,5-trinitrobenzene (TNB).
MOF-3, on the other hand, shows an order of sensitivity in
accordance with the electron deficiencies of the substrates.
To understand such anomalous behavior, we have thorough-
ly analyzed both the steady-state and time-resolved fluores-
cence quenching associated with these interactions. Deter-
mination of static Stern–Volmer constants (KS) as well as col-
lisional constants (KC) has revealed that MOF-1 and MOF-2
have higher KS values with 4-NT than with TNT, whereas for
MOF-3 the reverse order is observed. This apparently anom-
alous phenomenon was well corroborated by theoretical cal-
culations. Moreover, recyclability and sensitivity studies have
revealed that these MOFs can be reused several times and
that their sensitivities towards TNT solution are at the parts
per billion (ppb) level.
around 2 ppb.[7] Therefore, the sensing of NACs is a growing
concern for both security reasons and pollution control.[8]
Although various techniques are known for the detection of
explosives,[9] many of them are expensive, less sensitive, and
not suitable for in-field application.[10] Metal detectors are com-
monly used to detect explosives, but by definition are limited
to the detection of metal-based weapons. Canines are also
used for this purpose, being among the most reliable and ef-
fective detectors at our disposal.[3] Fluorescence-based chemo-
sensors offer several advantages over other approaches be-
cause of their high sensitivity, and can be used in hand-held
devices for in-field detection. A wide variety of luminescence-
based sensors for NACs are known in the literature,[11] such as
metal complexes,[12,13] supramolecular polymers,[14] carbon
nanotubes,[15] conjugated polymers,[16] and dendrimers,[17] as
well as metal–organic frameworks as a new generation of sen-
sors for the sensing of explosives.[18] In the majority of cases,
the sensing event mainly relies on an oxidative quenching
mechanism. Generally, sensors act as electron donors and, due
to the presence of their electron-withdrawing nitro group(s),
NACs act as electron acceptors. Upon photo-excitation,
a sensor molecule transfers an electron to the analyte, leading
to oxidation of the excited state, which in turn quenches the
fluorescence. Conjugated polymers have been well studied
and have been shown to be very efficient sensors because of
their higher ability to donate electrons, which is further en-
hanced by their delocalized p* excited states.[16] Moreover,
a conjugated polymeric backbone facilitates exciton migration
Introduction
An increasing number of terror attacks using explosive chemi-
cals in several countries have become recent threats to our
daily life. Although various diverse explosives are commercially
available, a large number of them include nitro- or nitroaro-
matic compounds.[1] The most commonly used explosive ingre-
dients are 2,4,6-trinitrotoluene (TNT), 1,3,5-trinitrobenzene
(TNB), 1,3-dinitrobenzene (DNB), 2,4-dinitrotoluene (DNT), and
so on.[2] Moreover, NACs are the main constituents of many un-
exploded landmines used worldwide during World War II, and
even today these NACs remain as important energetic materi-
als for the preparation of landmines.[3] Notably, NACs are also
widely used in the agrochemical industry.[4] There are many en-
vironmental and safety concerns relating to the use of NACs.[5]
It has been found that exposure to TNT can lead to anaemia
and abnormal liver function.[6] Due to its adverse health effects,
the US Environmental Protection Agency (EPA) has stipulated
a maximum permissible level of TNT in drinking water of
[a] B. Gole, Dr. A. K. Bar, Prof. P. S. Mukherjee
Department of Inorganic and Physical Chemistry
Indian Institute of Science
Bangalore-560012 (India)
Fax: (+91)80-2360-1552
E-mail: psm@ipc.iisc.ernet.in
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201302455.
Chem. Eur. J. 2014, 20, 2276 – 2291
2276
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
upon excitation to enhance their sensitivity. However, real-time
application of these materials is still limited due to various
issues. Firstly, the traditional synthesis of these covalent organ-
ic polymers involves a multi-step process and in many cases it
ends in a low yield of the target product. Secondly, control
over the molecular organization or determination of structure
is not an easy task, mainly because of their non-crystalline
nature and/or poor solubility.
Results and Discussion
We have designed and synthesized three new ligands, H2L1,
H2L2, and H2L3, with two distinct functionalities, namely iso-
phthalic acid (iph) and a fluorophore. The iph unit was used to
generate a coordination polymer backbone with metal ion
connectors, and the fluorophores served as fluorescent tags to
enhance the electron density in the MOFs for sensing electron-
deficient species. Phenyl, naphthyl, and pyrenyl groups were
used as tags to tune the p-electron densities of the resulting
MOFs in increasing order. Additionally, 1,4-di(4-pyridyl)benzene
(dpb) was employed as an interlinking unit to enhance the mi-
croporosity of the three-dimensional (3D) networks. Solvother-
mal reactions of the respective ligands in dimethyl formamide
(DMF) with Zn(NO3)2·6H2O at 1208C for 24 h furnished 3D net-
works for each of the MOFs in excellent yields.
In recent times, metal–organic frameworks have attracted
enormous attention because of their high surface areas, high
thermal stability, and the tunability of their porous struc-
tures.[19] To date, they have been widely used for gas stor-
age,[20] gas separation,[21] drug delivery,[22] imaging,[23] and het-
erogeneous catalysis.[24,25] Luminescent metal–organic frame-
works can also be used for chemosensing.[26] The well-defined
pore structure and the ordered nature of these frameworks fa-
cilitate rapid and selective detection of NACs. The selectivity
and sensitivity mainly depend on the electron density and the
ability of the MOFs to donate electrons. Recently, a few lumi-
nescent MOFs have been reported for the sensing of explo-
sives based on the concept of introducing electron-rich p-con-
jugated fluorescent ligands,[18] which mainly gives rise to lumi-
nescence in the backbone of the MOFs.
Single-crystal X-ray diffraction analyses (Table 1) showed the
3D networks of MOF-1 and MOF-3 to be isostructural (Fig-
ures 1 and 2), although they crystallized in different crystal sys-
tems (triclinic P1 for MOF-1 and orthorhombic Pba2 for
MOF-3). Moreover, they possessed similar 3D porous networks
with moderately solvent-accessible 3D pores of 45% (for
MOF-1) and 9.8% (for MOF-3) of the total volume, as calculat-
ed using PLATON. Though several attempts to obtain suitable
single crystals of MOF-2 were unsuccessful, the PXRD pattern
indicated that the as-synthesized MOF-2 was isostructural with
MOF-1 and MOF-3 (Figure 3).
In the frameworks, the iph moieties of the ligands are in-
volved in the formation of two-dimensional layered networks
connecting distorted octahedral Zn2+ ions. The equatorial sites
of each Zn2+ ion are coordinated to four O atoms of three dif-
ferent iph units with average bond distances of 2.165 ꢁ (for
Herein, we report a new strategy of using polyaromatic moi-
eties as fluorescent tags for systematic tuning of the sensitivity
and selectivity of MOFs. Although the concept of using fluores-
cent tags for tuning selectivity and sensitivity is quite common
in bio-sensing applications, our present approach of using aro-
matic tags to tune the efficiency of MOFs for sensing explo-
sives is new. The ligands 5-(benzyloxy)isophthalic acid (H2L1),
5-(naphthalen-1-ylmethoxy)isophthalic acid (H2L2), and 5-
(pyren-1-ylmethoxy)isophthalic acid (H2L3), with increasing
order of p-electron density in the aromatic tags (phenyl, naph-
thyl, and pyrenyl; Scheme 1) have been synthesized, character-
ized, and successfully employed in the synthesis of three new
MOFs (MOF-1–MOF-3). The introduction of these tags was an-
ticipated to enhance the selectivity and sensitivity for the de-
tection of NACs. In addition, we have also focused on the
mechanisms of the sensing of NACs, which should aid the
future design of better sensors. The present study has also re-
vealed a competition between size selectivity and electron de-
ficiency of the NACs. The observed experimental phenomenon
has been well corroborated by theoretical band structure cal-
culations on the MOFs.
Table 1. Crystallographic parameters of MOF-1 and MOF-3.
MOF-1
MOF-3
Empirical formula
Fw
T [K]
C31H22N2O5Zn
567.89
90(2)
C41H26N2O5Zn
692.03
90(2)
orthorhombic
Pba2
15.749(10)
16.425(11)
15.725(11)
90.00
90.00
90.00
4068(5)
4
1.227
Crystal system
Space group
a [ꢁ]
b [ꢁ]
c [ꢁ]
a [deg]
b [deg]
g [deg]
V [ꢁ3]
triclinic
P1
15.657(2)
15.672(2)
16.339(2)
89.9(2)
89.9(8)
81.9(1)
3968.6(8)
1
Z
1calcd [gcmꢀ3
]
0.950
0.648
m(MoKa) [mmꢀ1
l [ꢁ]
]
0.652
0.71073
1167.0
64323
29368
1.129
0.71073
1554.9
14304
7113
0.971
0.1065
0.2930
F (000)
Collected reflns.
Unique reflns.
GOF (F2)
[a]
R1
0.1251
0.3198
[b]
wR2
2
[a] R1 =SjFoꢀFc j/SjFo j ; [b] wR2 =S[w(Fo ꢀFc2)2]/w(Fo2)2]1/2
.
Scheme 1. Acids used in this study.
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2277
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ure 1b and Figure 2b). These 2D
layers are interconnected by 1,4-
di(4-pyridyl)benzene units coor-
dinated at the axial positions of
the Zn2+ ions, giving rise to 3D
MOFs with large porous chan-
nels. Interestingly, the phenyl or
pyrenyl groups attached to the
polymeric backbone of the
frameworks occupied the pores
and acted as fluorescent tags.
The phase purities and ther-
mal stabilities of the MOFs were
checked by PXRD and thermo-
gravimetric analysis (TGA). Com-
parison of the simulated PXRD
patterns of MOF-1 and MOF-3
with that of the as-synthesized
MOF-2 indicated that MOF-2
Figure 1. Crystal structure of MOF-1. (a) Two-dimensional layered network involving iph units of H2L1 and Zn2+
ions. (b) The functionalized phenyl moieties residing above the layer. (c) Space-filling representation of MOF-1.
The fluorescent tags are located inside the channels. The phenyl moieties are shown in a light-gray color.
adopted
a
similar molecular
structure to those of MOF-1 and
MOF-3 (Figure 3). The PXRD pat-
terns of the activated samples
(details of the procedure for acti-
vation of the MOFs are given in
the Experimental Section) indi-
cated stability of their networks
even after removal of the sol-
vent molecules from the pores.
TGA revealed
a
consistent
weight loss of about 20% for all
three MOFs up to 2008C, which
is attributed to the removal of
water and DMF. All of the MOFs
showed similar thermal stability
up to around 2808C. The activat-
Figure 2. Crystal structure of MOF-3. (a) Two-dimensional layered network involving iph units of H2L3 and Zn2+
ions. (b) Projection of functionalized pyrene moieties residing above the layer. (c) Space-filling representation of
MOF-3. The pyrene tags are situated inside the channels. The pyrene moieties are shown in a light-gray color.
ed samples displayed similar thermal stabilities to
those of the as-synthesized samples (Figure 4).
Photophysical properties
The ligands H2L1, H2L2, and H2L3 showed their charac-
teristic photoluminescence emission maxima at 335,
340, and 375 nm upon excitation at 306, 283, and
345 nm, respectively. The photoluminescence behav-
iors of H2L2 and H2L3 closely resembled those of
naphthalene and pyrene, respectively. This confirmed
that these fluorescent tags were the origin of the
photoluminescence of these ligands. MOF-1–MOF-3,
constructed from d10 (Zn2+) ions, showed excellent
photoluminescence (PL) properties, making them
promising as photoactive materials. The PL spectra of
dispersions of MOF-1–MOF-3 in ethanol at room
temperature (Supporting Information) showed strong
emission bands at lem =415, 420, and 480 nm upon
Figure 3. PXRD patterns of as-synthesized and activated MOF-1–MOF-3, and simulated
PXRD patterns of MOF-1 and MOF-3.
MOF-1) and 2.145 ꢁ (for MOF-3). In both cases, the fluoro-
phores reside below (or above) the 2D layered networks (Fig-
excitation at 310, 300, and 360 nm, respectively. The strong
red-shifts of the emission maxima of the MOFs compared to
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2278
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 4. TGA plots of MOF-1–MOF-3 along with their activated forms.
their parent ligands is most probably due to substantial elec-
tronic coupling between neighboring ligands through the
Zn2+ ions.
Sensing of nitroaromatics
The chemical structures of the MOFs incorporated electron-rich
fluorophores with increasing electron density: phenyl, naph-
thyl, and pyrenyl for MOF-1–MOF-3, respectively. To probe the
applicability of these MOFs for the sensing of electron-deficient
nitroaromatics (NACs), the PL properties of their dispersions in
ethanol were investigated. The PL intensities of the MOFs were
found to increase in the order MOF-1<MOF-2<MOF-3 under
similar experimental conditions (slit width 2 nm). This trend
can be mainly ascribed to the incorporation of electron-rich
fluorophores that essentially enhance the electron density in
the MOFs. Addition of small amounts of different NACs to dis-
persions of MOF-1–MOF-3 resulted in quenching of the photo-
luminescence intensities.
The NACs used in this study were 1,3,5-trinitrobenzene
(TNB), 2,4,6-trinitrotoluene (TNT), 2,4-dinitrotoluene (2,4-DNT),
3,4-dinitrotoluene (3,4-DNT), 3,5-dinitrobenzoic acid (3,5-
DNBA), 3-nitrobenzoic acid (3-NBA), 4-nitrobenzoic acid (4-
NBA), 4-nitrotoluene (4-NT), and nitrobenzene (NB). A few
simple electron-rich aromatic compounds (p-xylene, p-cresol,
and 4-methoxybenzoic acid) were also examined. A few elec-
tron-deficient non-nitro-aromatic compounds, namely chloro-
benzene, 1,2-dichlorobenzene, and benzoic acid, were also
used to assess the selectivity of the sensing of nitroaromatics.
To gain a better understanding of the ability of these MOFs
to sense NACs, PL quenching titrations were performed by
gradual addition of 10 mm stock solutions of different analytes
to the MOFs dispersed in ethanol. Changes in the PL intensities
of all of the MOFs upon titration with TNT are shown in
Figure 5, and the results with other NACs are provided in the
Supporting Information.
Figure 5. Reduction in the PL emission intensities of the MOFs upon gradual
addition of TNT. Inset: MOFs before and after titration with TNT under UV
light.
Interestingly, selectivity experiments on each of the MOFs re-
vealed that all of the NACs showed significant quenching re-
sponses over the other non-nitro analytes (Figure 6). The in-
ability of the non-nitro analytes to quench the fluorescence
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2279
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ciencies, MOF-3 was especially selective for TNT, TNB, and
DNTs, which are the main constituents of commonly used
chemical explosives. An opposite effect was observed when
the electron-rich analytes (p-xylene, 1,2-dichlorobenzene, and
p-cresol) were added to the MOFs. Unlike NACs, these elec-
tron-rich analytes appeared to enhance the fluorescence inten-
sity of the MOFs, as observed in a few previous cases.[18d,h] The
fluorescence quenching and enhancement can be attributed
to the donor–acceptor electron transfer between the MOFs
and the analytes. Due to the extended network structure of
the MOFs, they possess narrow energy bands because of
highly localized electronic states, especially in the present
cases involving d10 metal ions. As demonstrated by Li and co-
workers,[18d] for an extended structure, calculation of electronic
band structure is much more appropriate compared to molec-
ular orbital (MO) calculations. Indeed, the valence band (VB)
and conduction band (CB) energies can be described in a simi-
lar manner to the MOs of discrete molecules. One can envisage
that electron-deficient analytes (e.g., nitroaromatics) will
quench the luminescence of the MOFs if the lowest unoccu-
pied MOs (LUMOs) of the analytes, which are p*-type orbitals,
reside between the VB and CB of the luminescent MOFs. Upon
excitation, effective charge transfer can take place from the CB
of the MOFs to the LUMO of the NACs, thereby quenching the
fluorescence intensity. Hence, for more electron-deficient ana-
lytes with more stable LUMOs, electron transfer from the CB of
a particular MOF to the LUMOs of the analytes becomes ther-
modynamically more favorable. However, with electron-rich an-
alytes, for which the LUMO is located above the CB of the
MOFs, excited-state electrons from the LUMO of the analytes
will be transferred to the CB of the MOFs, leading to enhance-
ment of the PL intensity.
The feasibility of the fluorescence quenching mechanism
can be better understood by Rehm–Weller analysis according
to Equation (1):
DG0 ¼ EM0ðOoFxÞ ꢀ E0ðredÞ
analyte ꢀ DE00ðMOFÞ
ð1Þ
0ðoxÞ
0ðredÞ
where E
is the oxidation potential of the MOF, E
is the
MOF
analyte
reduction potential of the analyte, and DE00 is the difference in
energy between the lowest vibrational levels of the excited
0ðoxÞ
0ðredÞ
state and the ground state of the MOF. E
and E
were
analyte
MOF
calculated with respect to the standard calomel electrode
(SCE); DE00 was calculated from the absorption and emission
spectra of the MOFs. The reduction potentials of all three
MOFs and various analytes are given in Table S1 in the Sup-
porting Information. It is noted that the free energy change
(DG0) associated with the electron-transfer process was deter-
mined to be negative, which is the driving force of the fluores-
cence quenching process for all of the MOFs (Table 2).
For a better understanding of the fluorescence quenching
and enhancement mechanisms, we carried out density of
states (DOS) calculations on the MOFs. The band gaps of
MOF-1 and MOF-3 were calculated as 1.76 and 2.07 eV, respec-
tively (Figure 7). A schematic diagram of the fluorescence
quenching mechanism of the MOFs by a representative ana-
Figure 6. Reductions in fluorescence intensities (plotted as quenching effi-
ciencies) observed upon addition of 400 mL of 10 mm stock solutions of sev-
eral quenchers to dispersions of the MOFs in ethanol (left). Enhancements
(plotted as an enhancement efficiency %) in fluorescence intensity upon ti-
tration with electron-rich analytes (right); 1,3,5-trinitrobenzene (TNB), 2,4,6-
trinitrotoluene (TNT), 2,4-dinitrotoluene (2,4-DNT), 3,4-dinitrotoluene (3,4-
DNT), 3,5-dinitrobenzoic acid (3,5-DNBA), 3-nitrobenzoic acid (3-NBA), 4-ni-
trobenzoic acid (4-NBA), 4-nitrotoluene (4-NT), nitrobenzene (NB), 4-meth-
oxybenzoic acid (4-MeO-BA), benzoic acid (BA), chlorobenzene (CB), and 1,2-
dichlorobenzene (DCB or 1,2-DCB).
emission indicated that these MOFs are biased towards NACs
and should therefore offer higher selectivity for the detection
of nitro-explosives. Although MOF-1 and MOF-2 showed high
selectivity towards all of the NACs with similar quenching effi-
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2280
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Although all of the MOFs showed high selectivity towards
NACs, it was more important to determine the rate of quench-
ing, which is essentially related to the Stern–Volmer constant
(KSV). Most previously reported studies related to MOFs for NAC
detection have mainly been concerned with selectivity without
any focus on quenching rate.[18] It is important to mention
here that an effective sensing material must show a finite and
measurable response to the analytes at a particular concentra-
tion. The rates of PL quenching of the MOFs by a few selected
NACs are shown in Figure 8. To our surprise, the rates of
quenching of MOF-1 and MOF-2 were faster with 2,4-DNT, 3,4-
DNT, and 4-NT than with TNT and TNB, which was not fully in
accordance with the trend of electron deficiency. With MOF-3,
however, the rates of quenching followed the expected order,
those with TNT and TNB being faster. This apparent anomaly
could be clearly understood by detailed analysis of the experi-
mental quenching mechanism, which was again supported by
theoretical investigations. For this, we chose two extreme ana-
lytes, TNT and 4-NT, and the mechanisms were elaborated in
detail. The quenching rate is determined by the Stern–Volmer
Table 2. Free energy changes (DG0) associated with the fluorescence
quenching processes.
MOFs
DG0TNT
[kcalmolꢀ1
DG02,4-DNT
DG0NB
[kcalmolꢀ1
DG04-NT
]
[kcalmolꢀ1
]
]
[kcalmolꢀ1
]
MOF-1
MOF-2
MOF-3
ꢀ26.46
ꢀ24.26
ꢀ14.60
ꢀ19.55
ꢀ17.34
ꢀ7.68
ꢀ16.09
ꢀ13.88
ꢀ4.22
ꢀ14.93
ꢀ12.73
ꢀ3.07
lyte, TNT, is shown in Figure 7c. Interestingly, the theoretical
calculations indicated that the LUMO of TNT (ꢀ4.54 eV) is situ-
ated between the VB and CB of the MOFs. Hence, excited-state
electron transfer would be expected from the CB of the MOFs
to the LUMO of TNT, resulting in fluorescence quenching.
Moreover, theoretical calculations on the same MOFs and a rep-
resentative electron-rich analyte (p-xylene) showed that the ex-
cited-state electrons from the LUMO of p-xylene (located
above the CBs of the MOFs; Figure 7) would be expected to
populate the CB of the MOFs, resulting in fluorescence en-
hancement. Notably, the theoretical calculations are in accord
with the experimental observations.
Figure 7. (a) The calculated densities of states of MOF-1 and MOF-3. (b) Schematic representation of fluorescence quenching and enhancement mechanisms.
The LUMO of TNT is located between the VB and the CB of the MOFs, leading to quenching. The LUMO of p-xylene is situated above the CB of the MOF,
which is responsible for the fluorescence enhancement.
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2281
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
I0
I
ð2Þ
¼ 1 þ KSV ½Qꢁ
If the I0/I versus [Q] plot is linear, KSV can be estimated accu-
rately. However, it does not give any insight into the quench-
ing mechanism operating between the sensing material and
analytes. Moreover, in most cases a plot of I0/I versus [Q] devi-
ates from linearity, which is typically due to the presence of
two distinct quenching mechanisms. In static quenching,
a non-emissive ground-state complex is formed between the
fluorophore and quencher, whereas in dynamic quenching,
electron transfer takes place between the analyte and quench-
er in the excited state through collisions. The two mechanisms
can be differentiated by time-resolved measurements of PL
decays of the sensing material at different analyte concentra-
tions. If the PL lifetime of a sensor remains unchanged upon
increasing addition of a quencher, the quenching mechanism
is considered to be static. However, collisional quenching
offers an additional relaxation pathway for the excited mole-
cule and therefore reduces the PL lifetime.
To explore the quenching mechanism, the excited-state life-
times of the MOFs were measured before and after multiple
additions of 4-NT and TNT solutions. The decreases in the PL
lifetimes of the MOFs upon addition of 4-NT and TNT suggest-
ed the presence of dynamic quenching. The lifetime decays
were analyzed by the Stern–Volmer equation given by [Eq. (3)]:
t0
¼ 1 þ KC½Qꢁ
ð3Þ
t
where t0 is the PL lifetime of the MOF before addition of the
analyte, t is the PL lifetime of the MOF at a given analyte con-
centration [Q], and KC is the Stern–Volmer collisional constant.
Time-resolved Stern–Volmer plots for collisional quenching of
the MOFs with 4-NT and TNT are shown in Figure 9. The values
of the collisional constant KC were extracted by fitting the
plots using Equation (3). MOF-1 showed the highest KC values
with both 4-NT and TNT, at 193mꢀ1 and 311mꢀ1, respectively.
For MOF-2 and MOF-3, the values were much lower at
10.2mꢀ1, 19.6mꢀ1 and 16.4mꢀ1, 13.7mꢀ1, respectively.
For steady-state PL quenching involving both static and col-
lisional quenching, the change of fluorescence quenching is
given by Equation (4):
I0
I
ð4Þ
¼ ð1 þ KC½QꢁÞð1 þ KS½QꢁÞ
which contains both collisional (KC) and static (KS) terms. The
apparent deviation from linearity of the steady-state PL
quenching can be rationalized by expanding the above equa-
tion [Eq. (5)].
Figure 8. Rates of fluorescence quenching of (a) MOF-1, (b) MOF-2, and
(c) MOF-3 with selected nitroaromatics.
I0
I
2
ð5Þ
¼ 1 þ KC½Qꢁ þ KS½Qꢁ þ KCKS½Qꢁ
binding constant (KSV), and in most cases it has been obtained
from steady-state fluorescence quenching experiments.
If only one kind of quenching mechanism is operative, the
quenching can be represented by the Stern–Volmer equation
[Eq. (2)], where I0 is the initial PL intensity before addition of
the analyte, I is the PL intensity for any given concentration of
quencher [Q], and KSV is the Stern–Volmer constant.
When the analyte concentration is very low, the contribution
of the [Q]2 term is less prominent and Equation (5) would yield
a linear plot. However, at higher concentrations, the plot devi-
ates from linearity and the collisional constant (KC) has a signifi-
cant effect. The steady-state PL quenching data were fitted
with Equation (5) using the collisional constants (KC) obtained
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2282
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 3. Lifetimes and Stern–Volmer constants. The Stern–Volmer con-
stants were obtained by combining static and collisional quenching con-
stants of the corresponding analytes.
4-NT
SV
[Mꢀ1
TNT
SV
[Mꢀ1
]
MOFs
Lifetime
[s]
K
K
]
MOF-1
MOF-2
MOF-3
9.3483ꢂ10ꢀ9
1.2728ꢂ10ꢀ8
2.1663ꢂ10ꢀ8
5702
6865.2
579.4
2859
1554.6
1624.7
spectively; Figure 10). Surprisingly, the opposite trend was ob-
served for MOF-3 with KS =563 and 1611mꢀ1 for 4-NT and TNT,
respectively. The abovementioned higher KS for 4-NT is mainly
responsible for the observed anomaly in the quenching rates
of MOF-1 and MOF-2. The extent of excited-state energy trans-
fer depends on the distance between fluorophore and analyte,
and becomes prominent when they are in close proximity.
Thus, after excitation of the MOFs, the excited-state energy
can migrate within the particle until it is completely dissipated
by a radiative or nonradiative process. The energy migration
within a particle is typically monitored by time-dependent
fluorescence anisotropy decay analysis. This method can
detect both molecular rotation as well as energy migration res-
onance energy transfer (EM-RET). For extended 3D MOFs, how-
ever, the fluorescence anisotropy decay by EM-RET is expected
to be much faster than that by molecular rotation. The initial
change in the slope of the anisotropy decay curves (Fig-
ure S12) of the MOFs in ethanol in the presence of TNT or 4-NT
indicated energy migration within the particles.
The experimentally observed static quenching phenomenon
can be satisfactorily rationalized by band structure calculations
of the free MOFs. If a nitroaromatic compound (with LUMO be-
tween the VB and the CB of the MOF) forms a ground-state
charge-transfer complex with the MOF, the energy of the CB of
the newly formed complex will be lowered, whereas the VBs
would remain unchanged. The greater the electron deficiency
of the analytes, the greater the lowering of the CB of the
charge-transfer complex. This phenomenon is clearly reflected
in the variation of the band gaps of the MOFs upon complexa-
tion with different electron-deficient analytes. The band gap of
MOF-1 (1.76 eV) gradually decreased upon the encapsulation
of electron-deficient analytes. Theoretical studies indicated
that the band gap for MOF-1 decreased from 1.76 to 1.25 eV
upon complexation with one 4-NT molecule per unit cell (Sup-
porting Information), and further decreased to 1.21 eV with
two 4-NT molecules per unit cell. A sharp decrease in the band
gap (1.76 to 0.84 eV) was calculated when two 3,4-DNT mole-
cules per unit cell were encapsulated into the pores of MOF-1.
According to our calculations, the energy of the top of the VB
of MOF-1 remains almost unchanged (from ꢀ5.85 to ꢀ5.88 to
ꢀ5.89 eV), whereas the energy of the CB lowers from ꢀ4.09 to
ꢀ4.67 to ꢀ5.05 eV in the presence of two 4-NT and two 3,4-
DNT molecules, respectively (Figure 11).
Figure 9. Time-resolved Stern–Volmer plots for MOF-1–MOF-3 with 4-NT and
TNT. The solid lines represent fits to the time-resolved data using Equa-
tion (2).
from the time-resolved experiments. The values of KC and KS
for 4-NT and TNT with all of the MOFs are given in the corre-
sponding Stern–Volmer plots. It is important to mention that
the combination of KC and KS results in the Stern–Volmer con-
stant (KSV). The overall KSV values for all of the MOFs with 4-NT
and TNT are provided in Table 3.
The calculated static quenching constants (KS) for
MOF-1 and MOF-2 with 4-NT (5509 and 6855mꢀ1, respectively)
are much higher than those with TNT (2548 and 1535mꢀ1, re-
The experimentally determined higher KS values for 4-NT
compared to those for TNT with MOF-1 and MOF-2 may be at-
tributed to the solvent-accessible porosity and pore width of
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2283
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 11. (a) Calculated superimposed plots of density of states of
MOF-1 along with the presence of two 4-NT and two 3,4-DNT molecules per
unit cell of the MOF-1. The position of the VB of the MOF-1 remains almost
unchanged in the presence of the analytes. Lowering of the CB is observed
in the presence of these analytes. (b) Schematic representation of the
change of band gap in the presence of the abovementioned analytes.
spectively. Thus, 4-NT can easily enter into the channels so as
to better interact with the fluorescent tags (phenyl and naph-
thyl) in MOF-1 and MOF-2. However, the larger size of TNT
prevents it from entering the channels, thus resulting in lower
KS values. Thus, the observed fluorescence quenching of MOFs
by the NACs can be attributed to two distinct features: (i) for
MOF-1 and MOF-2, the quenching process is mainly due to
the encapsulation of smaller NAC molecules (4-NT, 2,4-DNT)
over larger molecules (TNT), and (ii) for MOF-3, the process
occurs primarily because of surface adsorption of the NACs on
the MOF particles.
Figure 10. Steady-state Stern–Volmer plots for MOF-1–MOF-3 with 4-NT and
TNT. The solid lines represent fits to the steady-state data using Equation (3).
The corresponding static quenching constants are given.
the MOFs indicated by PLATON calculations as well as by BET
surface area measurement. The PLATON calculations indicated
that MOF-1 (45%) had a higher solvent-accessible volume
compared to MOF-3 (9.8%). This was also reflected in the BET
surface area measurements for all of the MOFs. Although all of
the MOFs showed typical type II adsorption isotherms,
MOF-1 (217 m2 gꢀ1) and MOF-2 (185 m2 gꢀ1) had higher surface
areas than MOF 3 (37.6 m2 gꢀ1) (Figure S15). The calculated
pore diameters were 11 ꢁ, 9 ꢁ, and 2 ꢁ for MOF-1–MOF-3, re-
The apparently anomalous behavior and host (MOF)–guest
(analyte) interaction could be adequately rationalized on the
basis of theoretical optimization of the structures in the pres-
ence of various analytes. Due to the large number of atoms
present in the unit cell of activated MOF-1, only one unit cell
with periodic boundary conditions was considered for all of
the calculations. Two 4-NT and 3,4-DNT molecules were ran-
domly inserted into the cavity of the unit cell segment of the
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2284
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The faster quenching rates of
MOF-1 and MOF-2 with 4-NT
may be better understood on
the basis of solvent-dependent
PL quenching experiments. In
a typical process, MOF-1 was
dispersed in cyclohexane in
a similar manner as in ethanol
and titrated with a solution of 4-
NT in chloroform.
A drastic
change in the quenching rate
with 4-NT was observed when
MOF-1 was dispersed in cyclo-
hexane (Figure 14) instead of
ethanol.
A similar experiment
with a solution of TNT in chloro-
form showed almost no change
in the quenching rate upon
using cyclohexane instead of
ethanol. In cyclohexane, the rate
of quenching with 4-NT was
Figure 12. (a) Extended network view of MOF-1 showing the presence of one-dimensional channel-like pores and
(b) corresponding space-filling view, which clearly indicates that MOF-1 can encapsulate small molecules through
size selection. (c) Extended molecular view of MOF-1 after accommodation of two 4-NT molecules inside the
pores. Two types of 4-NT are present based on their positions, marked as A and B. Type A 4-NT molecules are
very close to the phenyl tag, whereas type B 4-NT does not have much interaction with the phenyl ring.
(d) Space-filling representation of 4-NT encapsulated in MOF-1.
markedly
decreased
even
activated MOF-1 and a quantum mechanical geometry optimi-
zation was carried out to investigate the most favorable host–
guest interactions. As we anticipated, 4-NT (as well as 3,4-DNT)
interacted mainly with the fluorescent tag moiety (-Ph). The
positions of the two 4-NT molecules were different (Figure 12),
with one (marked as A) close to the -Ph tag, and the other
(marked as B) not significantly interacting with the -Ph tag. A
similar observation for the analyte 3,4-DNT (Supporting Infor-
mation) indicated that this molecule could also be accommo-
dated in the cavity of MOF-1 during the fluorescence titration.
However, similar calculations using TNT as the analyte indicat-
ed destabilization of the system. Hence, one may assume that
the cavity of MOF-1 cannot accommodate TNT or larger ana-
lytes. Thus, the observed fluorescence quenching of
MOF-1 with TNT can mainly be ascribed to surface interaction.
Diffusion into the pores of MOF-1 and MOF-2 is controlled by
the size of the analytes. Attempted geometry optimizations of
MOF-3 with 4-NT and 3,4-DNT were unsuccessful because the
cavity of this MOF is almost occupied by the pyrenyl fluores-
cent tag, leaving insufficient space to accommodate a guest.
An extended molecular network view is given in Figure 13,
along with the corresponding space-filling model, to support
this conjecture. The space-filling model clearly indicates a lack
Figure 13. (a) Extended molecular representation of MOF-3 and (b) corre-
sponding space-filling model, which indicates that the pores are blocked by
the pyrene moieties.
of molecular porosity in MOF-3 due to the presence of the pyr-
enyl moieties. Thus, the observed fluorescence quenching phe-
nomenon for MOF-3 can only be due to an electronic effect.
Therefore, as expected, the most electron-deficient analyte,
TNT, exhibited the highest quenching rate. Hence, for
MOF-1 and MOF-2 the fluorescence quenching takes place
through size selection, whereas for MOF-3 the quenching phe-
nomenon follows the trend of the electron deficiencies of the
NACs.
though the quenching efficiency reached almost 80% upon
addition of 400 mL of the analyte solution. This can be ascribed
to the fact that the less polar nature of 4-NT makes it well sol-
vated in cyclohexane such that it has less interaction with the
MOF. However, in more polar ethanol, it is less solvated and
may enter the almost nonpolar pores of the MOFs, interacting
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2285
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 15. (a) Recyclability of MOF-3 dispersed in ethanol for TNT detection.
The MOF was filtered and washed several times with ethanol after each ex-
periment. The black bars represent the initial fluorescence intensity and the
grey bars represent the intensity upon addition of 400 mL of a 10 mm solu-
tion of TNT. (b) Determination of the detection limit of MOF-3 towards TNT
solution. The detection limit was calculated from the intercept of the two
linear fits of the data.
Figure 14. Solvent-dependent fluorescence quenching of MOF-1 and MOF-3
with 4-NT and TNT. A drastic change in fluorescence quenching rate was ob-
served for MOF-1 in the presence of 4-NT, but in the presence of TNT the
rate remained almost unchanged (top). No prominent solvent effect was ob-
served for MOF-3 (bottom).
implied their high photostability, making them potentially ap-
plicable for long-term in-field explosive detection or environ-
mental monitoring applications.
As discussed above, all of the prepared MOFs were highly
selective towards NACs over other aromatic compounds. How-
ever, in the present case, the rates of quenching not only
depend on the electronic properties of the MOFs and analytes,
but also on the porosity of the MOFs in relation to the size of
the analytes. Since the examined fluorescence quenching
mainly takes place through the static mechanism, MOF-1 and
MOF-2 are better sensors for small NACs (less electron defi-
cient) due to the their higher porosities. For MOF-3, however,
the rates of quenching follow the order of the electron defi-
ciencies of the analytes because in this case the quenching
takes place through surface adsorption on the MOF particles.
Sensitivity studies showed that MOF-3 was able to sense TNT
at a level as low as 0.9 ppb, much lower than with MOF-1 and
MOF-2. The order of sensitivity from MOF-1 (3.63 ppb) to
MOF-3 (0.9 ppb) indicates that a more electron-rich MOF pre-
fers more electron-deficient NACs over less electron-deficient
analytes. Thus, by tuning the porosity as well as increasing the
more extensively with the fluorescent tags. A similar experi-
ment on MOF-3 revealed that the quenching rates were not
significantly altered for either 4-NT or for TNT upon replacing
the polar solvent (ethanol) with a less polar solvent (cyclohex-
ane). This observation further implied that the quenching of
the PL of MOF-3 by nitroaromatics takes place through surface
adsorption rather than by encapsulation.
The sensitivity of the MOFs towards TNT solution was found
to be of the order of parts per billion (ppb). However, the sen-
sitivity was higher for MOF-3 (0.9 ppb) compared to the other
two MOFs (3.63 ppb and 2.27 ppb for MOF-1 and MOF-2, re-
spectively; Figure 15). The detection limit for TNT with MOF-3
falls below the permissible level of TNT in drinking water es-
tablished by the US EPA.[7] Moreover, all of the MOFs could be
reused over a significant number of cycles after repeated
washing with ethanol. The almost complete recovery of the ini-
tial fluorescence intensity of all of the MOFs over several cycles
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2286
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
a Horiba Scientific DAS6 instrument using spectroscopic grade sol-
vents.
electron density of the MOFs, we may be able to discriminate
NACs and increase the sensitivity towards TNT.
Synthesis
Conclusion
Dimethyl 5-(benzyloxy)isophthalate (1): A two-necked flask was
charged with dimethyl 5-hydroxyisophthalate (992.9 mg, 4 mmol),
potassium carbonate (1658.0 mg, 12 mmol), and potassium iodide
(663.5 mg, 4 mmol) under a nitrogen atmosphere, and then dry
acetonitrile (60 mL) was added. Finally, benzyl bromide (821.0 mg,
4.8 mmol) was added dropwise by means of a syringe. The reac-
tion mixture was stirred at reflux under nitrogen for 24 h. The sol-
vent was then removed under vacuum and the residue was dried
completely. The crude product was partitioned between dichloro-
methane and water. The organic phase was washed three times
with brine, dried over anhydrous Na2SO4, filtered, and concentrated
under reduced pressure. The crude product was purified by
column chromatography on silica gel (60–120 mesh) eluting with
chloroform/hexane (1:1) to afford compound 1 as a white crystal-
In conclusion, three new luminescent Zn-MOFs have been syn-
thesized by employing new isophthalic acid derivatives bear-
ing aromatic tags with different degrees of p-conjugation. The
ligands consist of two distinct functionalities: the dicarboxylate
moiety for assembly of the coordination polymer and the at-
tached aromatic tags to enhance the electron density into
MOFs. The aromatic tags were used strategically to enhance
the electron density of the MOFs for the detection of NACs.
Dispersions of the MOFs in ethanol exhibited strong fluores-
cence emissions, but their initial PL intensities were efficiently
quenched upon addition of a small amount of a nitroaromatic
explosive. Our study has revealed that all of the prepared
MOFs were highly selective towards explosive nitroaromatics
such as DNT, TNT, and TNB over the other electron-deficient ar-
omatics. However, the rates of quenching of MOF-1 and
MOF-2 were higher with DNT and 4-NT compared to those
with TNT and TNB, whereas MOF-3 showed a quenching order
in accordance with that expected based on the electron defi-
ciencies of the analytes. This apparent anomaly in quenching
rates has been fully investigated by the determination of static
quenching constants (KS) as well as collisional quenching con-
stants (KC) from steady-state PL titration and time-resolved PL
decays, respectively. The experimental observations have been
well corroborated by theoretical studies. Moreover, the recycla-
bility and very high sensitivity of the MOFs towards TNT make
them potential sensors for the detection of NACs. Thus, this
methodology of designing electron-rich fluorescent MOFs for
the selective and efficient sensing of electron-deficient NAC ex-
plosives and complete investigation of the quenching mecha-
nism may enable the discovery of a new generation of much
improved sensors for in-field explosives sensing.
1
line solid (85% yield). H NMR (CDCl3): d=8.291 (d, 1H), 7.839 (d,
2H), 7.461–7.343 (m, 5H), 5.145 (s, 1H), 3.936 ppm (s, 6H); 13C NMR
(CDCl3): d=166.542, 159.222, 136.497, 132.254, 129.122, 128.703,
128.017, 123.661, 120.614, 70.912, 52.873 ppm.
5-(Benzyloxy)isophthalic acid (H2L1) (2): Dimethyl 5-(benzyloxy)-
isophthalate (1201.0 mg, 4 mmol) was taken up in MeOH/H2O (1:1,
v/v; 50 mL) in a round-bottomed flask. NaOH (480 mg, 12 mmol)
was then added and the mixture was heated at reflux for 12 h. The
clear solution was then concentrated under vacuum, cooled to
room temperature, and filtered. The aqueous phase was acidified
with dilute hydrochloric acid to afford a white solid, which was col-
lected by filtration, washed several times with distilled water, and
finally dried under vacuum. Yield: 95%. FTIR: n˜ =2828 (w), 2537
(w), 1689 (w), 1594 (m), 1461 (m), 1431 (s), 1313 (s), 1275 (s), 1118
(w), 1027 (s), 942 (s), 907 (m), 839 (w), 755 (s), 731 (w), 694 (s), 667
1
(w), 610 (m), 549 (w), 505 (m), 437 cmꢀ1 (w); H NMR ([D6]DMSO):
d=13.297 (s, 2H), 8.091 (s, 1H), 7.738 (d, 2H), 7.490 (d, 2H), 7.391
(m, 2H), 7.343 (d, 1H), 5.244 ppm (s, 2H); 13C NMR ([D6]DMSO): d=
167.272, 159.377, 137.421, 133.530, 129.404, 128.872, 128.539,
123.347, 120.371, 70.588 ppm.
Dimethyl 5-(naphthalen-1-ylmethoxy)isophthalate (3): A two-
necked flask was charged with dimethyl 5-hydroxyisophthalate
(496.5 mg, 2 mmol), potassium carbonate (829.3 mg, 6 mmol), and
potassium iodide (331.7 mg, 2 mmol) under a nitrogen atmos-
phere, and then dry acetonitrile (30 mL) was added. 1-(Bromome-
thyl)naphthalene (530.6 mg, 2.4 mmol) was then added dropwise
and the resulting mixture was heated under reflux for 24 h with
stirring. Thereafter, the solvent was removed under vacuum and
the crude product was extracted with ethyl acetate. The organic
phase was washed three times with brine, dried over anhydrous
Na2SO4, and concentrated to dryness. The crude product was puri-
fied by column chromatography on silica gel (60–120 mesh) elut-
ing with ethyl acetate/hexane (1:9). The pure product was ob-
tained as a white solid (80% yield). 1H NMR (CDCl3): d=8.332 (s,
1H), 8.044 (d, 1H), 7.925 (d, 2H), 7.895 (m, 2H), 7.631 (d, 1H), 7.553
(m, 2H), 7.482 (m, 1H), 5.575 (s, 2H), 3.942 ppm (s, 6H); 13C NMR
(CDCl3): d=166.546, 159.304, 134.251, 132.349, 131.951, 131.883,
129.813, 129.200, 127.311, 127.062, 126.466, 125.716, 124.034,
123.794, 120.662, 69.648, 52.887 ppm.
Experimental Section
Materials and methods
All chemicals and solvents were obtained from commercial sources
and were used without further purification, unless otherwise men-
tioned. 1-(Bromomethyl)naphthalene, 1-(bromomethyl)pyrene, and
1,4-di(pyridin-4-yl)benzene were prepared according to the rele-
vant literature procedures.[27] NMR spectra were recorded on
a Bruker 400 MHz spectrometer. Chemical shifts (d) in the H NMR
spectra are reported in ppm relative to tetramethylsilane (TMS) as
an internal standard (d=0.0 ppm). IR spectra were recorded on
a Bruker ALPHA FTIR spectrometer over the range 4000–400 cmꢀ1
Elemental analyses were carried out on a Perkin-Elmer 240C CHNS
analyzer. Powder X-ray diffraction (XRD) data were recorded on
a Philips X’pert Pro using CuKa radiation (l=1.5406 ꢁ). Thermogra-
vimetric (TG) analyses of the MOFs were carried out on a Mettler-
Toledo thermal gravimetric analyzer under a nitrogen flow. Elec-
tronic absorption spectral measurements were performed on
a Perkin-Elmer Lambda 750 UV/Vis spectrophotometer and fluores-
cence emission studies were carried out on a Horiba Jobin Yvon
Fluoromax-4 spectrometer. Lifetime analyses were carried out on
1
.
5-(Naphthalen-1-ylmethoxy)isophthalic acid (H2L2) (4): Dimethyl
5-(naphthalen-1-ylmethoxy)isophthalate (3; 700.7 mg, 2 mmol) was
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2287
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
taken up in MeOH/H2O (1:1, v/v; 50 mL) in a round-bottomed flask.
NaOH (200 mg, 5 mmol) was then added and the mixture was
heated at reflux for 12 h. The clear solution was then concentrated
under vacuum, cooled to room temperature, and filtered. The
aqueous phase was acidified with dilute hydrochloric acid to afford
a white solid, which was collected by filtration, washed several
times with distilled water, and finally dried under vacuum. Yield:
95%. FTIR: n˜ =2822 (w), 2543 (w), 1686 (s), 1592 (s), 1462 (m), 1421
(s), 1319 (m), 1279 (s), 1128 (w), 1071 (s), 1011 (w), 908 (s), 878 (s),
787 (s), 758 (s), 693 (s), 659 (m), 615 (w), 565 (w), 525 (s), 483 cmꢀ1
room temperature at a rate of 158Chꢀ1. Colorless crystals of the
product (10 mg, 88%) were collected by filtration and washed with
fresh DMF (2 ꢂ 3 mL). FTIR: n˜ =3060 (w), 2336 (m), 2064 (w), 2023
(w), 1961 (w), 1700 (s), 1606 (s), 1553 (s), 1444 (m), 1374 (s), 1310
(m), 1254 (m), 1119 (w), 1030 (s), 850 (m), 785 (s), 715 (s), 615 (w),
588 (w), 486 cmꢀ1 (m); elemental analysis calcd (%) for
C31H22N2O5Zn [L1Zn(dpb)] (activated sample): C 65.56, H 3.90, N
4.93; found: C 65.23, H 3.45, N 5.12.
Synthesis of MOF-2: Zn(NO3)2·6H2O (6 mg, 0.02 mmol), H2L2
(6.5 mg, 0.02 mmol), and 1,4-di(pyridin-4-yl)benzene (4.6 mg,
0.02 mmol) were placed in an 8 mL scintillation vial. DMF (3 mL)
was added and the mixture was stirred for 10 min at room temper-
ature. The reaction vial was then tightly capped, placed in a pro-
grammable oven, and heated at 1208C for 24 h. It was then slowly
cooled to room temperature at a rate of 158Chꢀ1. Colorless crystals
of the product (11 mg, 89%) were collected by filtration and
washed with fresh DMF (2 ꢂ 3 mL). FTIR: n˜ =2355 (s), 2179 (w),
2142 (w), 2046 (w), 1551 (s), 1448 (m), 1370 (s), 1224 (s), 1125 (w),
1061 (w), 1024 (m), 782 (s), 718 (s), 545 (w), 463 cmꢀ1 (m); elemen-
tal analysis calcd (%) for C35H24N2O5Zn [L2Zn(dpb)] (activated
sample): C 68.03, H 3.91, N 4.53; found: C 68.53, H 3.74, N 4.42.
1
(m); H NMR ([D6]DMSO): d=13.325 (s, 2H), 8.141 (d, 1H), 8.124 (d,
1H), 7.999 (m, 2H), 7.830 (d, 2H), 7.710 (d, 1H), 7.620 (m, 3H),
5.698 ppm (s, 2H); 13C NMR ([D6]DMSO): d=167.306, 159.474,
134.210, 133.583, 132.894, 132.013, 129.765, 129.399, 127.723,
127.435, 126.946, 126.298, 124.871, 123.458, 120.485, 69.307 ppm.
Dimethyl 5-(pyren-1-ylmethoxy)isophthalate (5): An oven-dried,
two-necked flask was charged with dimethyl 5-hydroxyisophthalate
(462.4 mg, 2.2 mmol), potassium carbonate (829.3 mg, 6 mmol),
and potassium iodide (364.9 mg, 2.2 mmol) under a nitrogen at-
mosphere, and then dry acetonitrile (30 mL) was added. 1-(Bromo-
methyl)pyrene (590.3 mg, 2 mmol) was added in one portion and
the resulting mixture was stirred under nitrogen at reflux for 24 h.
The solvent was then removed under vacuum and the residue was
dried completely. The crude product was partitioned between
ethyl acetate and water. The organic phase was washed three
times with brine, dried over anhydrous Na2SO4, filtered, and con-
centrated under reduced pressure in a rotary evaporator. The
crude product was purified by column chromatography on silica
gel (60–120 mesh) eluting with ethyl acetate/hexane (1:9) to afford
compound 5 as a white solid (80% yield). 1H NMR (CDCl3): d=
8.333 (s, 1H), 8.286 (d, 1H), 8.229 (m, 3H), 8.209 (d, 2H), 8.167 (m,
2H), 8.048 (m, 1H), 7.976 (d, 2H), 5.812 (s, 2H), 3.944 ppm (s, 6H);
13C NMR (CDCl3): d=166.587, 159.371, 132.369, 132.235, 131.671,
131.174, 129.832, 129.232, 128.720, 128.298, 127.819, 127.452,
126.572, 126.014, 125.416, 125.117, 123.850, 123.325, 120.750,
69.795, 52.919 ppm.
Synthesis of MOF-3: Zn(NO3)2·6H2O (6 mg, 0.02 mmol), H2L3
(7.9 mg, 0.02 mmol), and 1,4-di(pyridin-4-yl)benzene (4.6 mg,
0.02 mmol) were placed in an 8 mL scintillation vial. DMF (3 mL)
was added and the mixture was stirred for 10 min at room temper-
ature. The reaction vial was then tightly capped, placed in a pro-
grammable oven, and heated at 1208C for 24 h. It was then slowly
cooled to room temperature at a rate of 158Chꢀ1. Colorless crystals
of the product (8 mg, 58%) were collected by filtration and
washed with fresh DMF (2 ꢂ 3 mL). FTIR: n˜ =2362 (s), 2330 (s),
2202 (s), 1995 (m), 1873 (w), 1605 (m), 1555 (s), 1371 (s), 1228 (m),
1123 (w), 1026 (m), 847 (m), 795 (s), 717 (s), 675 (m), 609 (w),
487 cmꢀ1 (w); elemental analysis calcd (%) for C41H26N2O5Zn[L3Zn-
(dpb)] (activated sample): C 71.16, H 3.79, N 4.05; found: C 71.52,
H 4.02, N 4.41.
5-(Pyren-1-ylmethoxy)isophthalic acid (H2L3) (6): Dimethyl 5-
(pyren-1-ylmethoxy)isophthalate (5; 500 mg, 1.2 mmol) was taken
up in MeOH/H2O (1:1, v/v; 50 mL) in a round-bottomed flask.
NaOH (240 mg, 6 mmol) was then added and the mixture was
heated to reflux for 12 h. The clear solution was then concentrated
under vacuum, cooled to room temperature, and filtered. The
aqueous phase was acidified with dilute hydrochloric acid to afford
a pale-yellow solid, which was collected by filtration, washed sever-
al times with distilled water, and finally dried under vacuum. Yield:
95%. FTIR: n˜ =2845 (w), 2634 (w), 1696 (s), 1592 (s), 1459 (m), 1412
(m), 1269 (s), 1181 (m), 1122 (m), 1067 (m), 1040 (m), 908 (m), 877
(m), 837 (s), 753 (s), 692 (s), 661 (m), 617 (w), 582 (w), 517 (w), 483
(w), 454 cmꢀ1 (w); 1H NMR (CD3OD): d=13.290 (s, 2H), 8.211 (m,
3H), 8.162–7.983 (m, 7H), 7.753 (s, 2H), 5.755 ppm (s, 2H); 13C NMR
([D6]DMSO): d=167.464, 159.267, 133.367, 131.819, 131.455,
130.952, 130.162, 129.540, 128.829, 128.529, 128.118, 127.348,
126.464, 126.379, 125.519, 124.763, 124.479, 123.900, 123.453,
120.573, 69.318 ppm.
X-ray crystallographic data collection and refinements
Suitable X-ray diffraction quality single crystals of MOF-1 and
MOF-3 were obtained by solvothermal reaction of the correspond-
ing ligands and Zn(NO3)2·6H2O. The single crystals slowly lost their
crystalline nature when removed from their mother liquors.
Though we examined the crystals at low temperature (90 K), the
quality of the collected diffraction data was less than satisfactory.
MOF-1 and MOF-3 (CCDC-947143 and -947144) crystallized in the
triclinic crystal system with the P1 space group and the ortho-
rhombic crystal system with the Pba2 space group, respectively.
The diffraction data of both the complexes were collected on
a Bruker Kappa CCD diffractometer with MoKa radiation (l=
0.71073 ꢁ). Data reduction was carried out using the SMART/SAINT
program.[28] The SADABS program was used for empirical absorp-
tion correction.[29] The structures were solved by direct methods
(SHELXS-97) and standard Fourier techniques, and refined against
F2 by full-matrix least-squares procedures (SHELXL-97) implement-
ed in WinGX.[30,31] All hydrogen atoms were assigned idealized posi-
tions and given thermal parameters equivalent to either 1.5
(methyl hydrogen atoms) or 1.2 (all other hydrogen atoms) times
the thermal parameter of the carbon atom to which they were at-
tached. The relatively high values of R(int.) and R(sigma) are due to
the weak diffraction of the crystals. The aromatic rings in the 1,4-
di(pyridin-4-yl)benzene, phenyl, and pyrene moieties (the tags) re-
Synthesis of MOF-1: Zn(NO3)2·6H2O (6 mg, 0.02 mmol), H2L1
(5.5 mg, 0.02 mmol), and 1,4-di(pyridin-4-yl)benzene (4.6 mg,
0.02 mmol) were placed in an 8 mL scintillation vial. DMF (3 mL)
was added and the mixture was stirred for 10 min at room temper-
ature. The vial was then tightly capped, placed in a programmable
oven, and heated at 1208C for 24 h. It was then slowly cooled to
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2288
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
mained thermally disordered. Several atoms were refined with par-
tial anisotropy. A few atoms in the MOF-1 were refined isotropical-
ly due to the high disorder. Several fragments in the main structure
deviate from their ideal geometry owing to high thermal vibration,
thus leading to a high Uiso(max)/Uiso(min) ratio for non-solvent H
atoms (10.0) upon affixation of H atoms to complete the models of
the electronic structures. Weak diffraction strength and high ther-
mal vibration led to relatively low CꢀC bond precision (as low as
0.2249 ꢁ for MOF-3 and 0.0259 ꢁ for MOF-1). The solvent mole-
cules are thermally too disordered to be located and hence,
squeezed out completely from MOF-1 using PLATON software.
While one DMF molecule is located in the structure of MOF-1 and
rest of the solvents are squeezed out. The squeezed out solvents
may comprise of DMF and water. Solvent-accessible void volumes
predicted using PLATON software amounted to 1802 ꢁ3 and 399 ꢁ3
for MOF-1 and MOF-3, respectively. The weak diffraction and high
thermal distortion can be attributed to fact that the crystals de-
cayed by losing crystallization solvents when taken out of mother
liquor, even at low temperatures. Moreover, the large-framework
cavities allowed dislocation of the crystallization solvent molecules
and flipping of the conformationally flexible fragments (especially
the fluorescent tags) of the structures. The refined crystallographic
parameters are given in Table 1.
measured once more. The material was recovered by centrifuga-
tion after each quenching experiment, washed several times with
ethanol, dried, and reused for further cycles. For sensitivity meas-
urements, a 10 mm TNT solution was gradually added to the MOFs
dispersed in ethanol. Initially, the fluorescence intensity of the
MOFs did not change, but with further addition of the substrate
quenching was observed. The detection limit was calculated from
the intercept of the two linear fits of the data.
Electrochemistry
The reduction potentials of the three MOFs and selected analytes
were measured using a three-electrode cell at room temperature.
An indium tin oxide (ITO) electrode was used as the working elec-
trode, platinum as the counter-electrode, and a standard calomel
electrode (SCE) as the reference. Electrochemical measurements of
the analytes were carried out using 0.01 mm solutions of each in
1:1 acetonitrile/1.0m aqueous tetrabutylammonium nitrate solu-
tion. For the MOFs, the powdered materials were coated on ITO
electrodes. The reduction potentials of the compounds were ob-
tained from the cyclic voltammograms and corrected with respect
to the SCE.
Computational details
Activation of the MOFs
First principles DFT calculations were carried out by using the
DMOL3 module implemented in Accelrys.[32–34] The unit cells of
MOF-1 and MOF-3, with and without analytes, were optimized
using a local density approximation with Perdew–Wang correlation
(LDA/PWC) functional. A double numerical basis set with d polari-
zation (DND), which is comparable to the 6–31G* basis set, was
used for all calculations. A Fermi smearing of 0.005 Hartree was
employed to improve the computational performance. The conver-
gence criteria for energy, gradient, and displacement were 2.0ꢂ
10ꢀ5 Hartree, 4.0ꢂ10ꢀ3 Hartreeꢁꢀ1, and 5.0ꢂ10ꢀ3 ꢁ, respectively.
The self-consistent field convergence criterion for all calculations
was 1.0ꢂ10ꢀ5. The projected density of states (PDOS) was calculat-
ed for the optimized structures.
As-synthesized MOF samples (about 200 mg) were soaked in meth-
anol and the supernatant was discarded at intervals of 8 h (three
times) and replaced by fresh methanol. After methanol exchange,
the sample was further treated in the same way with acetone and
dichloromethane to remove methanol and acetone, respectively.
Finally, the dichloromethane was decanted off and the sample was
dried under a dynamic vacuum at 1208C for 6 h.
Fluorescence quenching titrations in dispersed medium
The fluorescence properties of MOF-1–MOF-3 were investigated in
ethanol emulsions at 293 K. A stock suspension was prepared by
dispersing MOF powder (3 mg) in ethanol (3 mL). For fluorescence
measurements, 1 mL of the stock suspension was diluted to 2 mL
with fresh ethanol, sonicated for 30 min, and subsequently placed
in a quartz cell of pathlength 1 cm. All titrations were carried out
by gradually adding methanolic solutions of the aromatic analytes
(10mm) in an incremental fashion. Each titration was repeated at
least three times to obtain reliable titres. For all measurements, the
excitation wavelengths (lex) were 310 nm for MOF-1, 300 nm for
MOF-2, and 360 nm for MOF-3. The corresponding emissions were
monitored over the wavelength ranges lem =320–600 nm for
MOF-1, 310–600 nm for MOF-2, and 380–700 nm for MOF-3. Both
the excitation and emission slit widths were 2 nm for all measure-
ments. No change in shape of the emission spectra was observed
upon gradual addition of the quenchers; only quenching of the ini-
tial fluorescence emission intensity occurred upon titration with
electron-deficient nitroaromatic quenchers, whereas fluorescence
enhancement was obtained with electron-rich analytes. To check
the selectivity of the MOFs, aliquots (400 mL) of 10mm solutions of
each quencher were added to the MOFs dispersed in ethanol. The
fluorescence efficiency was calculated according to [(I0ꢀI)/I0]ꢂ
100%, where I0 is the initial fluorescence intensity. The PL decays
were fitted with double exponential convolved with instrumental
reference. The recyclability of the MOFs for TNT-sensing was
checked. As above, the fluorescence of the MOFs was recorded
from dispersions in ethanol. An aliquot (400 mL) of a 1ꢂ10ꢀ3 m TNT
solution was added to the dispersion and the fluorescence was
Acknowledgements
P.S.M. thanks the IISc-STC for financial support. B.G. sincerely
thanks Dibyendu Mallick for his assistance with the theoretical
calculations. B.G. also thanks the Indian Institute of Science for
a Bristol Mayer Squibb fellowship and research grant. A.K.B.
thanks Rajat Saha for discussion on the crystal structures.
Keywords: coordination polymers
fluorescence quenching metal–organic frameworks
nitroaromatics · sensors · zinc(II) complexes
·
explosives sensing
·
·
·
[1] W. C. Trogler, in NATO ASI Workshop, Electronic Noses & Sensors for the
Detection of Explosives (Eds.: J. W. Gardner, J. Yinon), Kluwer Academic
Publishers, Dordrecht, The Netherlands, 2004.
[2] a) J. Yinon, in Forensic and Environmental Detection of Explosives, John
Wiley & Sons, Chichester, 1999; b) S. J. Toal, W. C. Trogler, J. Mater.
Chem. 2006, 16, 2871–2883.
[3] a) A. W. Czarnik, Nature 1998, 394, 417–418; b) L. C. Shriver-Lake, B. L.
Donner, F. S. Ligler, Environ. Sci. Technol. 1997, 31, 837–841; c) J. Lu, Z.
Zhang, Anal. Chim. Acta 1996, 318, 175–179; d) M. Dock, M. Fisher, C.
Cumming, in Fifth International Symposium of Mine Warfare Association,
Monterey, CA, 2002, pp. 1–5; e) V. George, T. F. Jenkins, D. C. Leggett,
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2289
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
J. H. Cragin, J. Phelan, J. Oxley, J. Pennington, Proc. SPIE 1999, 3710,
258–269.
f) M. Guo, O. Varnavski, A. Narayanan, O. Mongin, J. P. Majoral, M. Blan-
chard-Desce, T. Goodson III, J. Phys. Chem. A 2009, 113, 4763–4771.
[18] a) A. J. Lan, K. H. Li, H. H. Wu, D. H. Olson, T. J. Emge, W. Ki, M. C. Hong,
J. Li, Angew. Chem. 2009, 121, 2370–2374; Angew. Chem. Int. Ed. 2009,
48, 2334–2338; b) C. Zhang, Y. Che, Z. Zhang, X. Yang, L. Zang, Chem.
Commun. 2011, 47, 2336–2338; c) Z. Zhang, S. Xiang, X. Rao, Q. Zheng,
F. R. Fronczek, G. Qian, B. Chen, Chem. Commun. 2010, 46, 7205–7207;
d) S. Pramanik, C. Zheng, X. Zhang, T. J. Emge, J. Li, J. Am. Chem. Soc.
2011, 133, 4153–4155; e) B. Gole, A. K. Bar, P. S. Mukherjee, Chem.
Commun. 2011, 47, 12137–12139; f) J. H. Wang, M. Li, D. Li, Chem. Sci.
2013, 4, 1793–1801; g) S. S. Nagarkar, B. Joarder, A. K. Chaudhari, S. Mu-
kherjee, S. K. Ghosh, Angew. Chem. 2013, 125, 2953–2957; Angew.
Chem. Int. Ed. 2013, 52, 2881–2885; h) G.-l. Liu, Y.-j. Qin, L. Jing, G.-y.
Wei, H. Li, Chem. Commun. 2013, 49, 1699–1701.
[19] a) O. M. Yaghi, M. O’Keeffe, N. W. Ockwig, H. K. Chae, M. Eddaoudi, J.
Kim, Nature 2003, 423, 705–714; b) S. Kitagawa, R. Kitaura, S. Noro,
Angew. Chem. 2004, 116, 2388–2430; Angew. Chem. Int. Ed. 2004, 43,
2334–2375; c) M. W. Bennett, M. P. Shores, M. G. Beauvais, J. R. Long, J.
Am. Chem. Soc. 2000, 122, 6664–6668; d) M. O’Keeffe, O. M. Yaghi,
Chem. Rev. 2012, 112, 675–702; e) N. Stock, S. Biswas, Chem. Rev. 2012,
112, 933–969.
[20] a) M. P. Suh, H. J. Park, T. K. Prasad, D. W. Lim, Chem. Rev. 2012, 112,
782–835; b) K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D.
Bloch, Z. R. Herm, T. H. Bae, J. R. Long, Chem. Rev. 2012, 112, 724–781;
c) M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe, O. M.
Yaghi, Science 2002, 295, 469–472; d) S. S. Y. Chui, S. M. F. Lo, J. P. H.
Charmant, A. G. Orpen, I. D. Williams, Science 1999, 283, 1148–1150;
e) A. M. Shultz, O. K. Farha, J. T. Hupp, S. T. Nguyen, J. Am. Chem. Soc.
2009, 131, 4204–4205; f) L. J. Murray, M. Dinca˘, J. Yano, S. Chavan, S.
Bordiga, C. M. Brown, J. R. Long, J. Am. Chem. Soc. 2010, 132, 7856–
7857; g) A. O. Yazaydın, R. Q. Snurr, T. H. Park, K. Koh, J. Liu, M. D. LeVan,
A. I. Benin, P. Jakubczak, M. Lanuza, D. B. Galloway, J. L. Low, R. R. Willis,
J. Am. Chem. Soc. 2009, 131, 18198–18199.
[4] a) J. H. Fletcher, J. Hamilton, I. Hechenbleikner, E. Hoegberg, B. J. Sertl, J.
Cassaday, J. Am. Chem. Soc. 1950, 72, 2461–2464; b) G. W. Ware, The
Pesticide Book, Thomson Publications, Fresno, CA, 2000; c) D. C. Leggett,
T. F. Jenkins, R. P. Murrmann, US Army Engineer Research and Develop-
ment Center, CRREL, 1977, Special Report 77; d) R. Dalpozzo, G. Bartoli,
Curr. Org. Chem. 2005, 9, 163–178.
[5] G. R. Lotufo, J. D. Farrar, L. S. Inouye, T. S. Bridges, D. B. Ringelberg, Envi-
ron. Toxicol. Chem. 2001, 20, 1762–1771.
[6] a) Toxicological Profile for 2,4,6-Trinitrotoluene, Agency for Toxic Substan-
ces and Disease Registry, US Department of Health and Human Serv-
ices, Atlanta, GA, 1995.
[7] 2004 Edition of the Drinking Water Standards and Health Advisories, US
Environmental Protection Agency, Washington DC, 2004.
[8] Approaches for the Remediation of Federal Facility Sites Contaminated
with Explosive or Radioactive Wastes, US Environmental Protection
Agency, Washington DC, 1993.
[9] a) D. S. Moore, Rev. Sci. Instrum. 2004, 75, 2499–2512; b) G. E. Spangler,
J. Carrico, D. Campbell, J. Test. Eval. 1985, 13, 234–240; c) L. M. Dorozh-
kin, V. A. Nefedov, A. G. Sabelnikov, V. G. Sevastjanov, Sens. Actuators B
2004, 99, 568–570; d) E. S. Forzani, D. L. Lu, M. J. Leright, A. D. Aguilar,
F. Tsow, R. A. Iglesias, Q. Zhang, J. Lu, J. H. Li, N. J. Tao, J. Am. Chem. Soc.
2009, 131, 1390–1391; e) R. A. McGill, T. E. Mlsna, R. Chung, V. K.
Nguyen, J. Stepnowski, Sens. Actuators B 2000, 65, 5–9.
[10] a) K. Hꢃkansson, R. V. Coorey, R. A. Zubarev, V. L. Talrose, P. Hakansson, J.
Mass Spectrom. 2000, 35, 337–346; b) J. M. Sylvia, J. A. Janni, J. D. Klein,
K. M. Spencer, Anal. Chem. 2000, 72, 5834–5840; c) V. P. Anferov, G. V.
Mozjoukhine, R. Fisher, Rev. Sci. Instrum. 2000, 71, 1656–1659; d) R. D.
Luggar, M. J. Farquharson, J. A. Horrocks, R. J. Lacey, X-Ray Spectrometry.
1998, 27, 87–94.
[11] a) M. S. Meaney, V. L. McGuffin, Anal. Chim. Acta 2008, 610, 57–67;
b) M. S. Meaney, V. L. McGuffin, Anal. Bioanal. Chem. 2008, 391, 2557–
2576.
[21] a) J. R. Li, R. J. Kuppler, H. C. Zhou, Chem. Soc. Rev. 2009, 38, 1477–
1504; b) J. R. Li, J. Sculley, H. C. Zhou, Chem. Rev. 2012, 112, 869–932;
c) H. S. Choi, M. P. Suh, Angew. Chem. 2009, 121, 6997–7001; Angew.
Chem. Int. Ed. 2009, 48, 6865–6869; d) J. An, S. J. Geib, N. L. Rosi, J. Am.
Chem. Soc. 2010, 132, 38–39; e) Y. E. Cheon, J. Park, M. P. Suh, Chem.
Commun. 2009, 5436–5438; f) C. Y. Lee, Y.-S. Bae, N. C. Jeong, O. K.
Farha, A. A. Sarjeant, C. L. Stern, P. Nickias, R. Q. Snurr, J. T. Hupp, S. T.
Nguyen, J. Am. Chem. Soc. 2011, 133, 5228–5231; g) R. Matsuda, R. Ki-
taura, S. Kitagawa, Y. Kubota, R. V. Belosludov, T. C. Kobayashi, H. Saka-
moto, T. Chiba, M. Takata, Y. Kawazoe, Y. Mita, Nature 2005, 436, 238–
241; h) J. P. Zhang, X. M. Chen, J. Am. Chem. Soc. 2008, 130, 6010–6017.
[22] a) J. An, S. J. Geib, N. L. Rosi, J. Am. Chem. Soc. 2009, 131, 8376–8377;
b) P. Horcajada, C. Serre, M. Vallet-Regi, M. Sebban, F. Taulelle, G. Ferey,
Angew. Chem. 2006, 118, 6120–6124; Angew. Chem. Int. Ed. 2006, 45,
5974–5978; c) P. Horcajada, T. Chalati, C. Serre, B. Gillet, C. Sebrie, T.
Baati, J. F. Eubank, D. Heurtaux, P. Clayette, C. Kreuz, J. S. Chang, Y. K.
Hwang, V. Marsaud, P. N. Bories, L. Cynober, S. Gil, G. Fꢄrey, P. Couvreur,
R. Gref, Nat. Mater. 2010, 9, 172–178.
[23] a) K. M. L. Taylor, W. J. Rieter, W. Lin, J. Am. Chem. Soc. 2008, 130,
14358–14359; b) J. Della Rocca, D. M. Liu, W. B. Lin, Acc. Chem. Res.
2011, 44, 957–968.
[24] a) L. Q. Ma, C. Abney, W. B. Lin, Chem. Soc. Rev. 2009, 38, 1248–1256;
b) A. Corma, H. Garcia, F. X. L. Xamena, Chem. Rev. 2010, 110, 4606–
4655; c) M. Yoon, R. Srirambalaji, K. Kim, Chem. Rev. 2012, 112, 1196–
1231; d) D. Farrusseng, S. Aguado, C. Pinel, Angew. Chem. 2009, 121,
7638–7649; Angew. Chem. Int. Ed. 2009, 48, 7502–7513; e) Y. Liu, W. M.
Xuan, Y. Cui, Adv. Mater. 2010, 22, 4112–4135.
[25] a) B. L. Chen, S. C. Xiang, G. D. Qian, Acc. Chem. Res. 2010, 43, 1115 –
1124; b) L. E. Kreno, K. Leong, O. K. Farha, M. Allendorf, R. P. V. Duyne,
J. T. Hupp, Chem. Rev. 2012, 112, 1105–1125; c) Z. Z. Lu, R. Zhang, Y.-Z.
Li, Z. J. Guo, H. G. Zheng, J. Am. Chem. Soc. 2011, 133, 4172–4174; d) Y.
Takashima, V. M. Martinez, S. Furukawa, M. Kondo, S. Shimomura, H.
Uehara, M. Nakahama, K. Sugimoto, S. Kitagawa, Nat. Commun. 2011, 2,
168; e) B. Gole, A. K. Bar, A. Mallick, R. Banerjee, P. S. Mukherjee, Chem.
Commun. 2013, 49, 7439–7441; <lit f>R. K. Das, A. Aijaz, M. K.
Sharma, P. Lama, P. K. Bharadwaj, Chem. Eur. J. 2012, 18, 6866–6872.
[26] a) Y. Cui, Y. Yue, G. Qian, B. Chen, Chem. Rev. 2012, 112, 1126–1162; b) J.
An, C. M. Shade, D. A. Chengelis-Czegan, S. P. Petoud, N. L. Rosi, J. Am.
[12] a) T. L. Andrew, T. M. Swager, J. Am. Chem. Soc. 2007, 129, 7254–7255;
b) M. E. Germain, M. J. Knapp, Chem. Soc. Rev. 2009, 38, 2543–2555;
c) M. E. Germain, T. R. Vargo, P. G. Khalifah, M. J. Knapp, Inorg. Chem.
2007, 46, 4422–4429; d) M. E. Germain, M. J. Knapp, J. Am. Chem. Soc.
2008, 130, 5422–5423; e) M. E. Germain, T. R. Vargo, B. McClure, J. J.
Rack, P. G. Van Patten, M. Odoi, M. J. Knapp, Inorg. Chem. 2008, 47,
6203–6211; f) A. D. Hughes, I. C. Glenn, A. D. Patrick, A. Ellington, E. V.
Anslyn, Chem. Eur. J. 2008, 14, 1822–1827.
[13] a) S. Ghosh, P. S. Mukherjee, Organometallics 2008, 27, 316–319; b) S.
Shanmugaraju, H. Jadhav, Y. P. Patil, P. S. Mukherjee, Inorg. Chem. 2012,
51, 13072–13074; c) S. Shanmugaraju, S. A. Joshi, P. S. Mukherjee, Inorg.
Chem. 2011, 50, 11736–11745; d) A. K. Bar, S. Shanmugaraju, K. W. Chi,
P. S. Mukherjee, Dalton Trans. 2011, 40, 2257–2267.
[14] a) B. Gole, S. Shanmugaraju, A. K. Bar, P. S. Mukherjee, Chem. Commun.
2011, 47, 10046–10048; b) S. Shanmugaraju, H. Jadhav, R. Karthik, P. S.
Mukherjee, RSC Adv. 2013, 3, 4940–4950.
[15] a) M. E. Kose, B. A. Harruff, Y. Lin, L. M. Veca, F. S. Lu, Y. P. Sun, J. Phys.
Chem. B 2006, 110, 14032–14034; b) E. S. Snow, F. K. Perkins, E. J.
Houser, S. C. Badescu, T. L. Reinecke, Science 2005, 307, 1942–1945.
[16] a) J. S. Yang, T. M. Swager, J. Am. Chem. Soc. 1998, 120, 11864–11873;
b) D. T. McQuade, A. E. Pullen, T. M. Swager, Chem. Rev. 2000, 100,
2537–2574; c) S. W. Thomas, G. D. Joly, T. M. Swager, Chem. Rev. 2007,
107, 1339–1386; d) H. Sohn, R. M. Calhoun, M. J. Sailor, W. C. Trogler,
Angew. Chem. 2001, 113, 2162–2163; Angew. Chem. Int. Ed. 2001, 40,
2104–2105; e) H. Sohn, M. J. Sailor, D. Magde, W. C. Trogler, J. Am.
Chem. Soc. 2003, 125, 3821–3830; f) H. Cavaye, P. E. Shaw, X. Wang, P. L.
Burn, S.-C. Lo, P. Meredith, Macromolecules 2010, 43, 10253–10261;
g) K. K. Kartha, S. S. Babu, S. Srinivasan, A. Ajayaghosh, J. Am. Chem. Soc.
2012, 134, 4834–4841.
[17] a) H. Cavaye, H. Barcena, P. E. Shaw, P. L. Burn, S.-C. Lo, P. Meredith, Proc.
SPIE. 2009, 7418, 741803; b) H. Cavaye, A. R. G. Smith, M. James, A.
Nelson, P. L. Burn, I. R. Gentle, S. C. Lo, P. Meredith, Langmuir 2009, 25,
12800–12805; c) D. A. Olley, E. J. Wren, G. Vamvounis, M. J. Fernee, X.
Wang, P. L. Burn, P. Meredith, P. E. Shaw, Chem. Mater. 2011, 23, 789–
794; d) S. Richardson, H. S. Barcena, G. A. Turnbull, P. L. Burn, I. D. W.
Samuel, Appl. Phys. Lett. 2009, 95, 063305; e) A. Narayanan, O. P. Varnav-
ski, T. M. Swager, T. Goodson III, J. Phys. Chem. C 2008, 112, 881–884;
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2290
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Chem. Soc. 2011, 133, 1220–1223; c) K. Jayaramulu, P. Kanoo, S. J.
George, T. K. Maji, Chem. Commun. 2010, 46, 7906–7908.
[27] a) A. Saha, S. Ramakrishnan, Macromolecules 2009, 42, 4956–4959;
b) T. B. Faust, V. Bellini, A. Candini, S. Carretta, G. Lorusso, D. R. Allan, L.
Carthy, D. Collison, R. J. Docherty, J. Kenyon, J. Machin, E. J. L. McInnes,
C. A. Muryn, H. Nowell, R. G. Pritchard, S. J. Teat, G. A. Timco, F. Tuna,
G. F. S. Whitehead, W. Wernsdorfer, M. Affronte, R. E. P. Winpenny, Chem.
Eur. J. 2011, 17, 14020–14030.
[31] a) L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837; b) L. J. Farrugia,
WinGX, version 1.65.04, Department of Chemistry, University of Glas-
gow, Glasgow, Scotland, 2003.
[32] B. Delley, J. Chem. Phys. 1990, 92, 508–517.
[33] B. Delley, J. Chem. Phys. 2000, 113, 7756–7764.
[34] Material Studio DMOL3, Version 6.0. Accelrys, Inc., San Diego, USA.
[28] SMART/SAINT, Bruker AXS, Inc., Madison, WI, 2004.
[29] G. M. Sheldrick, SADABS, University of Gçttingen, Gçttingen, Germany,
1999.
Received: June 26, 2013
Revised: November 15, 2013
Published online on January 23, 2014
[30] G. M. Sheldrick, SHELX-97, University of Gçttingen, Gçttingen, Germany,
1998.
Chem. Eur. J. 2014, 20, 2276 – 2291
www.chemeurj.org
2291
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim