A new 2D Cu-MOF constructed from carboxylate ligands containing C-H?π interactions as a recyclable responsive luminescent sensor for VOCs

Article abstract of DOI:10.1039/d0dt04363j

A new type of 2D metal-organic framework (MOF), namely Cu-MOF, was constructed from Cu(NO₃)₂·3H₂O and the novel ligand 1,4-bis(4-naphthoic acid)benzene. The C-H?π interactions in the Cu-MOF effectively reduced aggregation-caused quenching (ACQ) due to restrictions in intramolecular motion (RIM), increased fluorescence, and controlled crystal packing properties, resulting in low dimensional parallelogram channels and a responsive “turn-on” fluorescence. The activated Cu-MOF exhibited excellent chemical sensing properties to various volatile organic compounds (VOCs) and showed excellent recyclability.

Full text of DOI:10.1039/d0dt04363j

Dalton
Transactions
View Article Online
View Journal | View Issue
COMMUNICATION
A new 2D Cu-MOF constructed from carboxylate
ligands containing C–H⋯π interactions as a
recyclable responsive luminescent sensor for VOCs†
Cite this: Dalton Trans., 2021, 50,
4124
Received 24th December 2020,
Accepted 3rd March 2021
a
b
c
a
Chengxin Liu, Jin Cui,* Yufang Wang and Mingjie Zhang
*
DOI: 10.1039/d0dt04363j
rsc.li/dalton
1
4
A new type of 2D metal–organic framework (MOF), namely Cu- their high sensitivity, extreme selectivity, and ease of use.
MOF, was constructed from Cu(NO ·3H O and the novel ligand Despite the obvious advantages of using luminescent MOFs to
,4-bis(4-naphthoic acid)benzene. The C–H⋯π interactions in the detect VOCs, there are glaring disadvantages that hinder their
Cu-MOF effectively reduced aggregation-caused quenching (ACQ) widespread application. Only a handful of luminescent MOFs
3
)
2
2
1
1
5
due to restrictions in intramolecular motion (RIM), increased fluor- have been used to detect VOC vapors, such as extremely rare
1
6,17
escence, and controlled crystal packing properties, resulting in low ligand-based luminescent MOFs,
due to the restrictions in
dimensional parallelogram channels and a responsive “turn-on” intramolecular motion (RIM) that occur after their formation
fluorescence. The activated Cu-MOF exhibited excellent chemical and the intrinsically poor fluorescence of the component
1
8
sensing properties to various volatile organic compounds (VOCs) ligands. Yet, these drawbacks do not mitigate the potential
and showed excellent recyclability.
usefulness of ligand-based luminescent MOFs for detecting
VOCs. This trait is attributed to the weaker molecular C–H⋯π
interactions that occur between soft acids and soft bases.
These interactions play significant roles in molecular confor-
Metal–organic frameworks (MOFs) are hybrid materials con-
taining extended, highly porous networks that are formed via
self-assembly processes between organic ligands and metal
19
20
21
22
mation, chiral recognition, and crystal packing. Indeed,
the C–H⋯π interaction-assisted alignment of organic conju-
gated systems significantly impacts the regulation of photo-
1
2
centers. Since MOF 5 was first reported in 1999, it has been
widely used in various research and commercial applications
3
4
5
for gas separation and storage, catalysis, drug delivery, and
23
physical properties. Unlike the conventional hydrogen bond
6
chemical sensing. Compared with traditional micromolecular
and Coulomb forces, C–H⋯π interactions are nonpolar and
effective in water due to their inherent stability, which is attrib-
uted to dispersion interactions and the electrostatic contri-
butions that account for about 20% of the total interactions
fluorescent materials that possess extended π-conjugated struc-
tures, luminescent MOFs offer unique advantages in mole-
cular recognition and sensing because the rigid framework
effectively prevents the formation of π-stacked aggregates with
forbidden fluorescence and reduces aggregation-caused
24
exhibited by these compounds. In this study, Cu-MOF was
successfully prepared using a carboxylate ligand possessing
strong C–H⋯π interactions that made the Cu-MOF product to
exhibit intense fluorescence. Additionally, the newly syn-
thesized Cu-MOF contained low-dimensional channels and
dangling phenyl rings that enhanced the contact interaction
between the framework and the guest molecules. The activated
Cu-MOF showed responsive “turn-on” fluorescence in the pres-
ence of VOC vapors, high selectivity, and excellent
recyclability.
7
quenching (ACQ). The porosity of the framework provides
8
enough room for interactions with guest molecules.
Luminescent MOFs that exhibit tunable fluorescence are exten-
9
sively used for temperature monitoring applications and for
1
0
11
the production of medicines,
explosives,
chemical
1
2
13
10
weapons, solvents, and small organic molecules due to
a
Department of Chemistry, School of Sciences, Tianjin University, Tianjin 30035,
The structure of this previously unreported carboxylate
ligand, namely 1,4-bis(4-naphthoic acid)benzene, was deter-
P. R. China
b
National Foodstuff Inspection Center, Tianjin Product Quality Inspection
1
13
Technology Research Institute, Tianjin 300384, P. R. China
mined via H and C NMR spectroscopy and X-ray crystallo-
graphy; here, the solvent molecules were removed for spectro-
scopic clarity. Details of the synthesis of this carboxylate
ligand can be found in the ESI.† Theoretically, the two dan-
gling phenyl rings that lacked carboxylate groups were able to
c
Scientific Research Department, Shijiazhuang University of Applied Technology,
Shijiazhuang 050081, P. R. China
†
Electronic supplementary information (ESI) available. CCDC 2021999 and
010441. For ESI and crystallographic data in CIF or other electronic format see
2
DOI: 10.1039/d0dt04363j
4124 | Dalton Trans., 2021, 50, 4124–4128
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Communication
vibrate, and the other benzene rings could rotate in extremely responsive “turn-on” fluorescence character because the
dilute solutions of the ligand. However, the spectroscopic ana- unfixed benzene rings are easily affected by the analytes in the
8
lysis revealed that a molecule of the ligand was closely con- pores. The individual crystals of Cu-MOF exhibited a 2D split-
nected with six surrounding ligands, thereby forming a two- level lamellar structure with 1D parallelogram-shaped chan-
dimensional plane that restricted the intramolecular vibration nels; Fig. S6† shows that some of these channels were occu-
and rotation of the ligand’s crystals. The extensive C–H⋯π pied by the dangling benzene rings. The observed C–H⋯π
interactions also suppressed nonradiative decay through intra- interactions suspended the dangling benzene rings in the
molecular motion restrictions (Fig. S3†).
adjacent layers bound to the same channels, thereby resulting
Since the C–H⋯π interactions were most effective in water/ in Cu-MOF channels that were 10.30 Å in diameter (Fig. 1b).
2
4
solutions, the ligand was configured as a dilute solution of
The obtained Cu-MOF was soaked sequentially in MeOH
.0 × 10− mg L using THF/H O mixed solutions of various and CH Cl for 3 days each at room temperature before being
6
−1
1
2
2
2
volume ratios as the solvents during spectroscopic analysis. vacuum-dried at 80 °C to give the activated Cu-MOF as green
The fluorescence of the ligand first intensified after the initial transparent lamellar crystals. Unfortunately, the crystals could
dilution step, but this intensity eventually decreased, presum- not be subjected to X-ray crystallography; thus, the structure of
ably due to ACQ. A subsequent increase in the emission was the activated Cu-MOF was determined via powder X-ray diffrac-
attributed to RIM caused by C–H⋯π interactions and the emis- tion (PXRD), thermogravimetric analysis, and surface area and
sion maximum gradually red-shifted when the water ratio pore size distribution measurements. As shown in Fig. S5,†
increased (Fig. S4†). The ligand’s quantum yield was 14.52%, Cu-MOF contained two kinds of solvent molecules, namely co-
and the fluorescence lifetime was 68.5 ns, which indicated ordinated and free solvent molecules. Each of the aforemen-
that the ligand is a good fluorescent material or a monomer in tioned molecules accounted for 11.7% of the total molecular
luminescent MOFs even in the solid or aggregated state.
6 2 2
weight (C N O H12). The thermogravimetric analysis (Fig. 2a)
Cu-MOF, which was isolated as green transparent individ- revealed that Cu-MOF had lost its free solvent molecules at
ual crystals, was obtained via solvothermal reactions between 150 °C (∼11.7 wt%), followed by its coordinated solvent mole-
the carboxylate ligand and Cu(NO ) ·3H O in DMF. Single- cules at 250 °C (∼11.7 wt%). The onset of thermal stability at
3 2 2
crystal X-ray diffraction analysis revealed that the Cu(II) 320 °C prevented further losses in mass. However, the acti-
complex crystallized in the triclinic system with the P ˉ1 space vated Cu-MOF did not undergo mass losses; instead, it vio-
group and exhibited a 2D dinuclear copper-paddlewheel sec- lently decomposed at 320 °C, which indicated that the acti-
ondary building unit (SBU) in its structure. The ligand elimi- vated Cu-MOF had lost all solvent molecules. Thus, the depro-
nated a proton in the copper-paddlewheel SBU, and each car- tonated ligands were linked by Cu–Cu interactions in a
boxylic moiety coordinated with two Cu(II) ions. Since the two solvent-free 2D split-level lamellar structure, as confirmed via
Cu(II) ions formed a Cu–Cu bond with a distance of 2.6326(15) PXRD (Fig. S7†).
Å, the ions simultaneously coordinated with four oxygen
atoms from four deprotonated ligands with two oxygen atoms
from two DMF molecules. The asymmetric unit included two
coordinated DMF molecules, two free DMF molecules, and two
SBU-extended ligands (Fig. S5†). There are six types of benzene
rings in Cu-MOF, forming three types of C–H⋯π interactions.
First, the two sides of the benzene ring at which C13 is located
simultaneously accept the C–H⋯π interactions from H28 or
H31, resulting in their fixed positions. Each carbon atom on
the benzene ring has only one possible position, the occu-
pancy factor of which is 1. Second, one side of the benzene
ring at which C10 is located accepts the C–H⋯π interaction
from H22 or H39, which is simultaneously connected to the
benzene ring at which C13 is located. Hence, some of the
atoms are fixed in position, with occupancy factors of 1. There
are two possible positions for another atom, the occupancy
factors of which are 0.5. Finally, one side of the benzene ring
at which C33 or C18 is located accepts the C–H⋯π interaction
from H6 or H6A. Therefore, all the atoms have two possible
Fig. 1 (a) The C–H⋯π interactions of Cu-MOF, represented as dotted
positions, the occupancy factors of which are 0.5. The benzene lines, measured between H and the adjacent phenyl ring centroids. (b)
The crystal structure of Cu-MOF viewed along the [010] direction. Here,
red and blue represent two neighboring layers, whereas the yellow cap-
sules represent 1D channels with a diameter of 10.30 Å. The solvent
molecules have been removed for clarity. (c) The crystal structure of Cu-
rings of H28, H31, H22, H39, H6, and H6A are all providers
and not acceptors of the C–H⋯π interactions, because of
which all atoms have two possible positions, the occupancy
factors of which are 0.5. The detailed C–H⋯π interaction table
MOF viewed along the [100] direction. The solvent molecules have been
is shown in the ESI (Table S1†). The activated Cu-MOF has a removed for clarity.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 4124–4128 | 4125

View Article Online
Communication
Dalton Transactions
Fig. 3 (a) Fluorescence emission spectrum of the activated Cu-
MOF . guest. The side entrance slits were 3 nm (solid line) and 0.5 nm
Fig. 2 (a) Thermogravimetric curves of Cu-MOF and activated Cu-MOF.
(dotted line). Inset: Changes in the intensity of fluorescence as a func-
(
b) Nitrogen adsorption–desorption isotherm of the activated Cu-MOF.
tion of the emission maximum. (b) “On/off” cycles of the activated Cu-
MOF via alternating treatment with methanol vapor and under vacuum.
Inset: NLDFT pore size distribution of the activated Cu-MOF.
Based on the obtained adsorption and desorption iso-
therms of N , the surface area measurements of Cu-MOF gave MOF was heavily influenced by the attached guest molecules.
2
a typical type I sorption isotherm, indicating the presence of This relationship between the guest molecule and the Cu-
homogeneous micropores inside (Fig. 2b). The existence of MOF could be exploited to produce fluorescent molecular
significant hysteresis loops between the adsorption and de- sensors.
sorption branches suggests wide channels and/or dynamic
Chemosensing studies were performed by exposing the acti-
structural alteration, such as double bonds, π⋯π interactions, vated Cu-MOF to various solvent vapors (as discussed in the
6
,8
or C–H⋯π interactions. According to previous reports,
ESI†). As illustrated in Fig. 3a, the subsequent photo-
MOFs with such structures may be used in detection and sep- luminescence tests showed that the spectral intensities
2
−1
aration. Its Langmuir surface area is 312 m
g , which is changed with peak shifts (see the inset). The emission
equivalent to the value of the reported 2D MOFs with 1D maximum of the respective appropriately-sized guest mole-
2
5,26
channels and indicates
that the framework retains some cules was preferentially adsorbed into the pores. This means
porosity after the removal of the solvent. The NLDFT pore size that the observed emission maximum values of the activated
distribution shows that the micropore diameter mainly ranges Cu-MOF . guest molecules were significantly influenced by
from 1.2 to 1.4 nm, which is in good agreement with the the size of the guest molecule and followed the order: cyclo-
crystal structure.
hexane > benzene > toluene > DMF > acetone > methanol.
It was theorized that the fluorescence of Cu-MOF was
The solid-state fluorescence of the ligands was determined
before the Cu-MOF was used to conduct fluorescence sensing caused by the attached ligands and that the responsive “turn-
experiments. Here, we noted that the fluorescence lifetime of on” fluorescence of the activated Cu-MOF was derived from
Cu-MOF was 99.5 ns, and the associated quantum yield was the intrinsically strong fluorescent ligands that exhibited
2
.21%, whereas those of the ligand were 68.5 ns and 14.52%, incomplete RIM. The guest molecules filled the pores of the
respectively. It was theorized that the low quantum yield was activated Cu-MOF, thereby restricting the vibration and
due to the dangling phenyl rings in Cu-MOF that were rotation of the dangling benzene ring in the framework; the
unaffected by the C–H⋯π interactions, which, in turn, accompanying fluorescence of the compound is shown in the
encouraged nonradiative decay processes. Fig. 3a shows that inset of Fig. 3a. It is worth noting that when benzene was used
the emission maximum of Cu-MOF was observed at 431 nm, as the guest molecule, the intensity of fluorescence was extre-
which exhibited a clear blue-shift relative to the emission mely low. We theorized that dynamic vapor adsorption (VTI)
maximum of the activated Cu-MOF (i.e., at 444 nm). processes between the activated Cu-MOF and benzene encour-
Additionally, the observed intensity of fluorescence was aged adsorption, even when benzene’s vapor pressure was low
lower, indicating that the fluorescence character of the Cu- (i.e., only 10% of the saturated vapor pressure of the system).
4126 | Dalton Trans., 2021, 50, 4124–4128
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Communication
When benzene’s vapor pressure was 25% of the total saturated
vapor pressure of the system, the absorbance value was 4.9%
Conclusions
of its own weight; generally, the absorbance values of the By judiciously designing the ligand and conducting crystal
system increased continuously under higher vapor pressures. engineering, a new type of 2D split-level fluorescent MOF exhi-
Since Cu-MOF contained free DMF molecules (11.7%) and the biting parallelogram 1D channels was synthesized using Cu
3 2 2
fluorescence of the activated Cu-MOF . benzene was lower (NO ) ·3H O and the previously unreported carboxylate ligand
than that of Cu-MOF and its activated analog, there were few 1,4-bis(4-naphthoic acid)benzene, which contained C–H⋯π
incidents of adsorption. This was because the perpendicular interactions. The strong fluorescence character of the solid-
conformation of benzene tended to weaken the C–H⋯π inter- state Cu-MOF was due to extensive C–H⋯π interactions that
actions, thereby encouraging the coplanar conformation in the restricted the intramolecular vibration and rotation, thereby
system which easily transitioned to ACQ via π-stacked aggrega- controlling the crystal packing. Its activated analog in the solid
tion. As noted in the inset of Fig. 3b, the absorbance values state exhibited responsive “turn-on” fluorescence toward VOC
increased with time at first, but as the reaction progressed, vapors and excellent recyclability. The size of guest molecules
very little change was noted even when subjected to the same significantly influenced the emission maximum of the acti-
vapor pressure. However, once the vapor pressure inevitably vated Cu-MOF . guest molecules, and the associated intensity
increased, again the absorbance values continued to increase. of fluorescence showed the relationship between the ligands
In other words, even though the benzene vapor continued to and the guest molecules. This finding presents a successful
flow and there were still vacancies in the metallic center, the strategy for constructing MOFs with strongly fluorescent
associated MOF did not continue to absorb more benzene ligands containing C–H⋯π interactions. This new method
until the vapor pressure increased despite being subjected to minimized the ACQ formation, which was different from tra-
the same vapor pressure. These findings confirmed the theory ditional aggregation-induced emission (AIE), thereby providing
of the previous speculations, which stated that the fluo- a new direction for designing functional fluorescent molecules
rescence of the Cu-MOF was caused by the ligand. In the case and preparing fluorophore-containing MOFs in the future.
of toluene, the methyl group limits its π–π stacking with the
benzene ring in the MOF. However, in the case of cyclohex-
anes, such π–π stacking is almost nonexistent because of their Author contributions
2
7,28
noncoplanar structure and higher H content.
Methanol,
C. X. Liu designed the material and performed the experi-
ments; C. X. Liu and Y. F. Wang wrote the paper; J. Cui and
M. J. Zhang contributed to the supervision and editing; all
authors contributed to the general discussion.
DMF, and acetone, which contain O or N in their structures,
can easily form weak interactions with the ligands or the
copper-paddlewheel SBU of the MOF, restricting the intra-
molecular movement of the benzene ring that is not fixed in
2
9,30
the MOF, thereby enhancing its fluorescence intensity.
Since the emission maximum of the activated Cu-
MOF . methanol was significantly different from the numeri-
cal value associated with the activated Cu-MOF, methanol was
used as the guest in cyclic experiments. Before this, the time-
dependent intensity quenching experiment was conducted
with the result shown in Fig. S10.† The activated Cu-MOF was
exposed to methanol vapor. When the vapor of the guest mole-
cules gradually entered the channels of the MOF, the fluo-
rescence intensity increased and the emission maximum blue
shifted and reached the extreme value after adsorption satur-
ation (24 h). To ensure sufficient contact, the activated Cu-
MOF was placed in methanol vapor for 2 days when perform-
ing cyclic experiments. The emission spectrum was obtained
after heating the activated Cu-MOF . methanol to 80 °C under
vacuum for 3 h. Emission spectra were obtained after sub-
sequent exposure to methanol vapor for 48 h at 40 °C. No
obvious changes were noted in the resulting emission
maximum and the PXRD data, even after the process was
repeated six times (Fig. S7†). After the cyclic experiments, the
Conflicts of interest
The authors declare no competing financial interest.
Acknowledgements
This work was supported by the Natural Science Foundation of
Tianjin (17JCQNJC05500), P. R. of China. CCDC 2021999 and
2010441 contain the supplementary crystallographic data
associated with this paper.
Notes and references
1 S. Bi, H. Banda, M. Chen, L. Niu, M.-Y. Chen, T.-Z. Wu,
J.-S. Wang, R.-X. Wang, J.-M. Feng, T.-Y. Chen, M. Dincă,
A.-A. Kornyshev and G. Feng, Nat. Mater., 2020, 19, 552.
2 H. Li, M. Eddaoudi, M. O’Keeffe and O.-M. Yaghi, Nature,
1999, 402, 276.
2
−1
Langmuir surface area of the MOF was 248 m g ; further-
more, the adsorption and desorption measurements of N still
3 W.-F. Wu, J.-Y. Su, M.-M. Jia, Z.-J. Li, G.-Q. Liu and W.-B. Li,
Sci. Adv., 2020, 6, 7270.
4 J.-S. Seo, D. Whang, H. Lee, S.-I. Jun, J. Oh, Y.-J. Jeon and
K. Kim, Nature, 2000, 404, 982.
2
showed a typical type I sorption isotherm (Fig. S15†). This
observation confirmed the new compound’s excellent repeat-
ability and recyclability (Fig. 3b).
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 4124–4128 | 4127

View Article Online
Communication
Dalton Transactions
5
6
P. Horcajada, C. Serre, M. Vallet-Regi, M. Seban, F. Taulelle 16 Y. Yang, L. Chen, F.-L. Jiang, X.-Y. Wan, M.-X. Yu, Z. Cao,
and G. Ferey, Angew. Chem., Int. Ed., 2006, 45, 5974. T. Jing and M.-C. Hong, J. Mater. Chem. C, 2017, 5, 4511.
C.-A. Bauer, T.-V. Timofeeva, T.-B. Settersten, B.-D. Patterson, 17 J.-H. Wang, M. Li and D. Li, Chem. Sci., 2013, 4, 1793.
V.-H. Liu, B.-A. Simmons and M.-D. Allendorf, J. Am. Chem. 18 X.-G. Liu, H. Wang, B. Chen, Y. Zou, Z.-G. Gu, Z.-J. Zhao
Soc., 2007, 129, 7137.
and L. Shen, Chem. Commun., 2015, 51, 1677.
7
8
W.-J. Ji, G.-F. Liu, B.-Q. Wang, W.-B. Lu and Q.-G. Zhai, 19 M. Nishio, CrystEngComm, 2004, 6, 130.
CrystEngComm, 2020, 22, 4710.
20 M. Kitamura, K. Nakano, T. Miki, M. Okada and R. Noyori,
J. Am. Chem. Soc., 2001, 123, 8939.
M. Zhang, G.-X. Feng, Z.-G. Song, Y.-P. Zhou, H.-Y. Chao,
D.-Q. Yuan, T.-T.-Y. Tan, Z.-G. Guo, Z.-G. Hu, B.-Z. Tang, 21 M. Aazome, Y. Ueno, H. Ooiso and K. Ogura, J. Org. Chem.,
B. Liu and D. Zhao, J. Am. Chem. Soc., 2014, 136, 7241. 2000, 65, 68.
Y.-J. Cui, H. Xu, Y.-F. Yue, Z.-Y. Guo, J.-C. Yu, Z. X.-X. Chen, 22 T. Kobayashi, R.-C. Yanagita and K. Irie, Tetrahedron, 2020,
9
J.-K. Gao, Y. Yang, G.-D. Qian and B.-L. Chen, J. Am. Chem.
Soc., 2012, 134, 3979.
0 X.-Y. Xu and B. Yan, Sens. Actuators, B, 2016, 230, 463.
30, 127657.
23 Y. Sooda, N. Tohnai, Y. Zhou, Y. Shimoi and R. Azumi,
Cryst. Growth Des., 2018, 18, 6477.
1
1
1 F.-G. Moscoso, J. Almeida, A. Sousaraei, T.-L. Costa, 24 H. Sezawa, T. Yoshida, Y. Umezawa, S. Tsuboyama and
A.-M.-G. Silva, J.-C. Gonzalez, L.-C. Silva and J. M. Pedrosa,
J. Mater. Chem. C, 2020, 8, 3626.
2 C. Montoro, F. Linares, E.-Q. Procopio, I. Senkovska,
M. Nishio, Eur. J. Inorg. Chem., 2002, 12, 3148.
25 W. Kim, K. Yeon Lee, E. Ryu, J. Gu, Y. Kim, S. J. Lee and
S. Huh, Eur. J. Inorg. Chem., 2013, 24, 4228.
1
1
1
1
S. Kaskel, S. Galli, N. Masciocchi, E. Barea and 26 T. H. Chen, I. Popov and O. S. Miljanic, Chem. – Eur. J.,
J.-A.-R. Navarro, J. Am. Chem. Soc., 2011, 133, 11888. 2017, 23, 286.
3 Z.-Q. Liu, Y. Zhao, P. Wang, Y.-S. Kang, M. Azam, S.-I. Al- 27 S. L. Jackson, A. Rananaware, C. Rix, S. V. Bhosale and
Resayes, X.-H. Liu, Q.-Y. Lu and W.-Y. Sun, Dalton Trans.,
017, 46, 9022.
4 X.-J. Liu, Y.-H. Zhang, Z. Chang, A.-L. Li, D. Tian,
K. Latham, Cryst. Growth Des., 2016, 16, 3067.
28 J. Q. Dong, A. K. Tummanapelli, X. Li, S. M. Ying, H. Hirao
and D. Zhao, Chem. Mater., 2016, 28, 7889.
2
Z.-Q. Yao, Y.-Y. Jia and X.-H. Bu, Inorg. Chem., 2016, 55, 29 Y. Yu, J. P. Ma, C. W. Zhao, J. Yang, X. M. Zhang, Q. K. Liu
326. and Y. B. Dong, Inorg. Chem., 2015, 54, 11590.
5 W. Yan, C. Zhang, S. Chen, L. Han and H. Zheng, ACS Appl. 30 G. Mehlana, G. Ramon and S. A. Bourne, CrystEngComm,
7
Mater. Interfaces, 2017, 9, 1629.
2014, 16, 8160.
4128 | Dalton Trans., 2021, 50, 4124–4128
This journal is © The Royal Society of Chemistry 2021