A Cd(II) Luminescent Coordination Grid as a Multiresponsive Fluorescence Sensor for Cr(VI) Oxyanions and Cr(III), Fe(III), and Al(III) in Aqueous Medium

Hydro(solvo)thermal reactions of Cd(NO3)2, N-(pyridin-3-ylmethyl)-4-(pyridin-4-yl)-1,8-naphthalimide (NI-mbpy-34), and 5-bromobenzene-1,3-dicarboxylic acid (Br-1,3-H2bdc) afforded a luminescent coordination polymer, {[Cd(Br-1,3-bdc)(NI-mbpy-34)(H2O)]∙2H2O}n (1). Single-crystal X-ray diffraction analysis showed that 1 features a two-dimensional (2-D) gridlike sql layer with the point symbol of (44·62), where the Cd(II) center adopts a {CdO5N2} pentagonal bipyramidal geometry. Thermogravimetric (TG) analysis confirmed the thermal stability of 1 up to about 340 °C, whereas XRPD patterns proved the maintenance of crystallinity and framework integrity of 1 in CH2Cl2, H2O, CH3OH, and toluene. Photoluminescence studies indicated that 1 displayed intense blue fluorescence emissions in both solid-state and H2O suspension-phase. Owing to the good fluorescent properties, 1 could serve as an excellent turn-off fluorescence sensor for selective and sensitive Cr(VI) detection in water, with LOD = 15.15 μM for CrO42− and 14.91 μM for Cr2O72−, through energy competition absorption mechanism. In addition, 1 could also sensitively detect Cr3+, Fe3+, and Al3+ ions in aqueous medium via fluorescence-enhancement responses, with LOD = 2.81 μM for Cr3+, 3.82 μM for Fe3+, and 3.37 μM for Al3+, mainly through an absorbance-caused enhancement (ACE) mechanism.


Introduction
With the advanced development of modern society, rapid industrial and agricultural productions and rich human activities have increasingly brought about severe chemical pollution. Among various chemical pollutants, heavy metal ion based inorganic contaminates are of higher concern compared with other contaminants such as organic pollutants due to their nondegradability and bioaccumulation [1]. Moreover, heavy metal ions are well-known poisonous contaminants in water due to their high toxicity, which could cause serious environmental and ecological harm and cause a detrimental effect on human health [2]. For example, chromium exists in aquatic environments usually in the forms of Cr(VI) oxyanions, i.e., dichromate (Cr 2 O 7 2− ) and chromate (CrO 4 2− ) ions, and/or free cation, i.e., trivalent Cr(III) ion. The Cr(VI) oxyanions are highly carcinogenic and mutagenic, causing hereditary genetic defects and various types of cancers [3][4][5]. While the Cr(III) ion is essentially harmless due to its low toxicity, it may, however, cause mutations and malignant cells when excessive accumulation occurs [5][6][7]. The permissible limit for Cr(VI) in drinking water is set as 50 µg/L by the World Health Organization (WHO) [8]. Iron and aluminum are two ubiquitous metals widely used in daily applications around human living environments [9]. Trivalent Fe(III) and Al(III) cations are the forms of iron and aluminum that can enter human body. As one of the most important elements for living organisms, Fe(III) ion influences a variety of vital bioprocesses such as electron transfer, oxygen storage, oxygen metabolism, among others [10][11][12]. The deficiency or excess of Fe(III) ion is harmful to human health, resulting in some diseases [13][14][15]. Additionally, Al(III) ion in body fluids is toxic to humans and will induce harmful effects that cause diseases such as Alzheimer's disease and Parkinson's disease when its content in the body is over an acceptable standard [9,12,16]. The tolerable daily ingestion of Al 3+ for the human body is about 3-10 mg/day [7] and the permissible level in drinking water is set as 7.41 µM by the WHO [17].

Materials and Characterization
All of the chemicals and solvents were acquired from market sources (MATRIX, UL-TRA, ACROS, PanReac Applichem, SHOWA, Fluka, VETEC, ALFA, MACRON, J.T. Baker, SIGMA ALDRICH), and used without further processing. Ligand NI-mbpy-34 was synthesized according to the previously reported literature [40]. Thermogravimetric (TG) analyses were performed using a Thermo Cahn VersaTherm HS TG analyzer (Thermo, Newington, NH, USA) under flow nitrogen with a heating rate of 5 • C/min. X-ray powder diffraction (XRPD) patterns were measured using a Shimadzu XRD-7000 diffractometer (Shimadzu, Kyoto, Japan) with a graphite monochromatized Cu Kα radiation (λ = 1.5406 Å) at 30 kV and 30 mA. Infrared (IR) spectroscopic measurements were performed on a Perkin-Elmer Frontier Fourier transform infrared spectrometer (Perkin-Elmer, Taipei, Taiwan) using attenuated total reflection (ATR) technique. Fluorescence spectroscopic measurements were performed at room temperature using a Hitachi F7000 fluorescence spectrophotometer (Hitachi, Tokyo, Japan) equipped with a 150 W xenon lamp as an excitation source. UV-Vis absorption spectra were recorded at room temperature using a JASCO V-750 UV/VIS spectrophotometer (JASCO, Tokyo, Japan). Elemental microanalyses (C, H, N) were performed on an Elementar Vario EL III analytical instrument (Elementar, Langenselbold, Germany). Ultrasonic agitation of suspensions was conducted using a Qsonica Q125 instrument. X-ray photoelectron spectroscopy (XPS) analyses were performed on an ULVAC-PHI PHI 5000 VersaProbe/Scanning ESCA Microprobe instrument (ULVACPHI Inc., Kanagawa, Japan).  3.09; N, 5.42%. Found: C, 48.24; H, 2.94; N, 5.41%. The phase purity of the bulky sample was confirmed by the closely matched XRPD patterns between the simulated pattern, calculated from single-crystal diffraction data and the experimental pattern of as-synthesized 1 without grinding ( Figure S1). Of particular note, the XRPD patterns of the same microcrystalline sample after grinding showed alternations in intensity, and in some peak positions, compared to the simulated XRPD patterns. This is tentatively attributed to the influences of either the variation in preferred orientation of the powdered sample [30] or the partial crystal structure distortion caused by grinding [41] or both.

Single-Crystal X-ray Structure Determinations
The diffraction data were collected using a Bruker D8 Venture diffractometer configured with a PHOTO100 CMOS detector at 150(2) K, equipped with a graphite monochromated Mo Kα radiation (λ = 0.71073 Å). The structures were solved by direct methods with the SHELXTL program [42] and refined by full-matrix least-squares methods on F 2 using the SHELXL-2014/7 [43], incorporated in WINGX [44] crystallographic collective package. Non-hydrogen atoms were refined with anisotropic displacement parameters, except where noted. Carbon-bound hydrogen atoms were calculated in ideal positions and refined as riding mode. Oxygen-bound hydrogen atoms were structurally evident in the difference Fourier map. All of the hydrogen atoms were refined with isotropic displacement parameters, U iso , constrained to be 1.2 or 1.5 times U eq of the carrier atom. Experimental details for X-ray data collection and the refinements are summarized in Table 1. Hydrogen-bonding parameters are shown in Table S1. CCDC 1991627 (1) contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif (accessed on 19 March 2020). Obsd reflns (I > 2σ (I)) 5201 Params 406 R 1 a (I > 2σ (I)) 0.0289 . Single-crystal X-ray structure analysis reveals that the crystal structure of 1 belongs to the triclinic space group P1. The asymmetric unit contains one Cd(II) center, one Br-1,3-bdc 2− dianion, one NImbpy-34 ligand, and one coordination and two lattice water molecules ( Figure 1a

Chemical Stability and Thermal Properties
The chemical stability of 1 in different solvents including dichloromethane (CH 2 Cl 2 ), N,N -dimethylacetamide (DMAc), N,N -dimethylformamide (DMF), H 2 O, methanol (CH 3 OH), and toluene was checked. After separately immersing 1 in CH 2 Cl 2 , H 2 O, CH 3 OH, and toluene for 24 h, the checked XRPD patterns of so-obtained powdered samples were very similar to the patterns of as-synthesized 1 after grinding, with slight differences in peak intensity ( Figure 2). This might suggest the preferred orientation effect or the minor extent of partial distortion of the long range order in 1. However, the checked XRPD patterns still imply the maintenance of framework integrity and crystallinity, con-firming the high stability of 1. In contrast, 1 displayed low stability after immersing in DMAc and DMF due to the poorly matched XRPD profiles.

Chemical Stability and Thermal Properties
The chemical stability of 1 in different solvents including dichloromethane (CH2Cl2), N,N′-dimethylacetamide (DMAc), N,N′-dimethylformamide (DMF), H2O, methanol (CH3OH), and toluene was checked. After separately immersing 1 in CH2Cl2, H2O, CH3OH, and toluene for 24 h, the checked XRPD patterns of so-obtained powdered samples were very similar to the patterns of as-synthesized 1 after grinding, with slight differences in peak intensity ( Figure 2). This might suggest the preferred orientation effect or the minor extent of partial distortion of the long range order in 1. However, the checked XRPD patterns still imply the maintenance of framework integrity and crystallinity, confirming the high stability of 1. In contrast, 1 displayed low stability after immersing in DMAc and DMF due to the poorly matched XRPD profiles.

Chemical Stability and Thermal Properties
The chemical stability of 1 in different solvents including dichloromethane (CH2Cl2), N,N′-dimethylacetamide (DMAc), N,N′-dimethylformamide (DMF), H2O, methanol (CH3OH), and toluene was checked. After separately immersing 1 in CH2Cl2, H2O, CH3OH, and toluene for 24 h, the checked XRPD patterns of so-obtained powdered samples were very similar to the patterns of as-synthesized 1 after grinding, with slight differences in peak intensity ( Figure 2). This might suggest the preferred orientation effect or the minor extent of partial distortion of the long range order in 1. However, the checked XRPD patterns still imply the maintenance of framework integrity and crystallinity, confirming the high stability of 1. In contrast, 1 displayed low stability after immersing in DMAc and DMF due to the poorly matched XRPD profiles. Thermogravimetric (TG) analysis was performed under a nitrogen atmosphere to examine the thermal stability of 1 ( Figure S2). The TG trace of 1 exhibited a weight loss of 5.4% from room temperature to 69 • C, corresponding to the escape of lattice water molecules (calcd. 4.6%). A gradual weight loss of 1.6% corresponding to the removal of coordinated water molecules (calcd. 2.3%) followed when the temperature was raised to approaching 186 • C. Then, the TG trace showed the existence of a stable plateau before the framework began to process a two-step collapse from ca. 340 • C to ca. 633 • C. Dur-ing the decomposition, bromide might react with divalent cadmium to generate CdBr 2 (b.p. = 844 • C), which would escape at higher temperature. The final residue of 8.4% was reasonably assigned to the CdO component (calcd. 8.3%). Thermogravimetric (TG) analysis was performed under a nitrogen atmosphere to examine the thermal stability of 1 ( Figure S2). The TG trace of 1 exhibited a weight loss of 5.4% from room temperature to 69 °C, corresponding to the escape of lattice water molecules (calcd. 4.6%). A gradual weight loss of 1.6% corresponding to the removal of coordinated water molecules (calcd. 2.3%) followed when the temperature was raised to approaching 186 °C. Then, the TG trace showed the existence of a stable plateau before the framework began to process a two-step collapse from ca. 340 °C to ca. 633 °C. During the decomposition, bromide might react with divalent cadmium to generate CdBr2 (b.p. = 844 °C), which would escape at higher temperature. The final residue of 8.4% was reasonably assigned to the CdO component (calcd. 8.3%).

Gas Adsorption Properties of Activated 1
In the crystal structure of 1, there are free void spaces of about 9.8% of the unit cell volume, hence, the porous properties of activated 1 were investigated by gas adsorption studies. Prior to gas adsorption experiments, as-synthesized 1 (about 100 mg) was thermally treated at 100 °C under a reduced pressure for 24 h to remove solvent molecules and thus to give activated 1. For activated 1, N2 adsorption isotherms exhibited no appreciable uptakes of 5.92 cm 3 g -1 STP at P/P0 = 1 and 77 K, whereas CO2 adsorption isotherms showed negligible uptakes of 11.19 cm 3 g -1 STP at P/P0 = 1 and 195 K ( Figure S3). The low N2 and CO2 uptakes of thermally activated 1 might be attributed to the small sufficient solvent-accessible voids and framework distortion induced pore-reduction, which resulted in surface adsorption. The latter assumption was supported by the checked XRPD patterns, which showed obvious differences with the experimental profiles of as-synthesized sample of 1 ( Figure S4).

Photoluminescence Properties
When excited at λex = 370 nm, the solid-state emission spectrum of NI-mbpy-34 showed an intense emission band centered at 462 nm, which was overlapped with two further bands as shoulders at around 433 and 480 nm ( Figure S5). After irradiation at λex = 360 nm, Br-1,3-H2bdc emitted an extremely weak solid-state emission centered at around 468 nm. Comparably, 1 emitted intense blue fluorescence with an emission band centered at 436 nm in solid-state and 422 nm in H2O suspension-phase upon excitation at λex = 365 nm. From the band position and shape, the emissions were tentatively assigned to the intraligand charge transfer of the NI-mbpy-34 ligand perturbed by metal coordination; this is further supported by the high resemblance in excitation spectra between NI-mbpy-34 and 1.

Gas Adsorption Properties of Activated 1
In the crystal structure of 1, there are free void spaces of about 9.8% of the unit cell volume, hence, the porous properties of activated 1 were investigated by gas adsorption studies. Prior to gas adsorption experiments, as-synthesized 1 (about 100 mg) was thermally treated at 100 • C under a reduced pressure for 24 h to remove solvent molecules and thus to give activated 1. For activated 1, N 2 adsorption isotherms exhibited no appreciable uptakes of 5.92 cm 3 g −1 STP at P/P 0 = 1 and 77 K, whereas CO 2 adsorption isotherms showed negligible uptakes of 11.19 cm 3 g −1 STP at P/P 0 = 1 and 195 K ( Figure S3). The low N 2 and CO 2 uptakes of thermally activated 1 might be attributed to the small sufficient solvent-accessible voids and framework distortion induced pore-reduction, which resulted in surface adsorption. The latter assumption was supported by the checked XRPD patterns, which showed obvious differences with the experimental profiles of as-synthesized sample of 1 ( Figure S4).

Photoluminescence Properties
When excited at λ ex = 370 nm, the solid-state emission spectrum of NI-mbpy-34 showed an intense emission band centered at 462 nm, which was overlapped with two further bands as shoulders at around 433 and 480 nm ( Figure S5). After irradiation at λ ex = 360 nm, Br-1,3-H 2 bdc emitted an extremely weak solid-state emission centered at around 468 nm. Comparably, 1 emitted intense blue fluorescence with an emission band centered at 436 nm in solid-state and 422 nm in H 2 O suspension-phase upon excitation at λ ex = 365 nm. From the band position and shape, the emissions were tentatively assigned to the intraligand charge transfer of the NI-mbpy-34 ligand perturbed by metal coordination; this is further supported by the high resemblance in excitation spectra between NI-mbpy-34 and 1.

Detection of Anions
In view of the high water stability and excellent fluorescence properties of 1 in H 2 O suspension-phase, its potential ability to detect anions was explored in water. Anion detection studies were carried out by separately adding aqueous solutions of NaF and

Detection of Anions
In view of the high water stability and excellent fluorescence properties of 1 in H2O suspension-phase, its potential ability to detect anions was explored in water. Anion detection studies were carried out by separately adding aqueous solutions of NaF and KmX (X m− = Cl − , Br − , I − , ClO4 − , CO3 2− , Cr2O7 2− , CrO4 2− , NO3 − , and PO4 3− , m = 1, 2, 3) into the well-prepared H2O suspensions of 1, with a concentration of 1 mM. Upon excitation at λex = 365 nm, the fluorescence detection results showed that most anions had an inconspicuous fluorescence intensity change effect (<10% change) on 1, except for the CO3 2− , Cr2O7 2− , CrO4 2− , and PO4 3− ions (Figure 3). It is noted that CO3 2− and PO4 3− caused significant fluorescence enhancement by 63% and 44%, respectively. However, the two Cr(VI) oxyanions, Cr2O7 2− and CrO4 2− , showed obviously high quenching effects with an efficiency up to 97% and 99%, respectively, (quenching efficiency (%) = (I0 − I)/I0 × 100%, where I0 and I are the maximum fluorescence intensities of 1 before and after addition of a quencher). Therefore, it is suggested to use the fluorescence quenching of 1 to detect trace amounts of Cr2O7 2− and CrO4 2− ions in water media. The anti-interference ability of 1 toward Cr2O7 2− and CrO4 2− was examined with the existence of different competitive anions in equal concentration. The competition experiments clearly indicated that Cr2O7 2− and CrO4 2− both retained high quenching ability to almost completely turn off the fluorescence of 1 when the other interference anions are present (Figure 4), implying high selectivity of 1 toward Cr2O7 2− and CrO4 2− over others perturbed anions in water media.

Detection of Anions
In view of the high water stability and excellent fluorescence properties of 1 in H2O suspension-phase, its potential ability to detect anions was explored in water. Anion detection studies were carried out by separately adding aqueous solutions of NaF and KmX (X m− = Cl − , Br − , I − , ClO4 − , CO3 2− , Cr2O7 2− , CrO4 2− , NO3 − , and PO4 3− , m = 1, 2, 3) into the well-prepared H2O suspensions of 1, with a concentration of 1 mM. Upon excitation at λex = 365 nm, the fluorescence detection results showed that most anions had an inconspicuous fluorescence intensity change effect (<10% change) on 1, except for the CO3 2− , Cr2O7 2− , CrO4 2− , and PO4 3− ions (Figure 3). It is noted that CO3 2− and PO4 3− caused significant fluorescence enhancement by 63% and 44%, respectively. However, the two Cr(VI) oxyanions, Cr2O7 2− and CrO4 2− , showed obviously high quenching effects with an efficiency up to 97% and 99%, respectively, (quenching efficiency (%) = (I0 − I)/I0 × 100%, where I0 and I are the maximum fluorescence intensities of 1 before and after addition of a quencher). Therefore, it is suggested to use the fluorescence quenching of 1 to detect trace amounts of Cr2O7 2− and CrO4 2− ions in water media. The anti-interference ability of 1 toward Cr2O7 2− and CrO4 2− was examined with the existence of different competitive anions in equal concentration. The competition experiments clearly indicated that Cr2O7 2− and CrO4 2− both retained high quenching ability to almost completely turn off the fluorescence of 1 when the other interference anions are present (Figure 4), implying high selectivity of 1 toward Cr2O7 2− and CrO4 2− over others perturbed anions in water media.   The sensitivity of Cr(VI) oxyanions detection in water media can be evaluated by quantitative analysis and limit of detection (LOD) values. The concentration-dependent fluorescence intensity of 1 was determined by gradually adding different concentrations of Cr(VI) oxyanion into well-dispersed H2O suspensions of 1. As observations, the fluorescence intensity of 1 gradually decreased with increasing concentration of Cr2O7 2− and CrO4 2− (Figure 5a,b), indicating that the fluorescence quenching of 1 caused by the introduction of Cr2O7 2− and CrO4 2− ions can be quantified. It is also noted that the fluorescence intensity (λem = 414 nm) versus the concentration of Cr2O7 2− and CrO4 2− can be well fitted to the first-order exponential decay (Figure S6), suggesting a diffusioncontrolled fluorescence quenching. The fluorescence quenching efficiencies were further analyzed by a Stern−Volmer analysis based on the equation: I0/I = 1 + KSV [Q], where I0 and I are the fluorescence intensities before and after the addition of analytes, respectively, [Q] is the molar concentration of the analyte (mM), and KSV is the Stern−Volmer quenching constant (M −1 ). The Stern-Volmer plots exhibited good linear correlations at low concentrations (inset in Figure 5c,d), which gave the KSV values of 5.56 × 10 3 M -1 for Cr2O7 2− and 1.32 × 10 4 M -1 for CrO4 2− . Furthermore, the Stern-Volmer plots displayed upward deviation from linearity at high concentrations (Figure 5c,d), as a result of the combination of a dynamic energy transfer mechanism and a static self-absorption mechanism [46,47].
The LOD values for Cr2O7 2− and CrO4 2− were determined using the equation: LOD = 3σ/k, where σ is the standard deviation of five blank measurements of fluorescence for the H2O suspensions of 1 and k is the absolute value of the slope of the calibration curve at low concentrations. The LOD values for Cr2O7 2− and CrO4 2− were determined to be 14.91 (3.22) and 15.15 (1.76) μM (ppm), respectively, ( Figure S7). Owing to the excellent antiinterference ability and high KSV and lower LOD values, 1 can be an effective fluorescence sensor displaying high detection selectivity and sensitivity toward Cr2O7 2− and CrO4 2− in water media.

Detection of Metal Ions
The influence of different metal ions on the fluorescent properties of 1 in H 2 O suspension-phase was also investigated through similar procedures by adding wellprepared aqueous solutions of metal ions, including M(NO 3 ) n (M n+ = Ag + , Al 3+ , Mg 2+ , Ca 2+ , Co 2+ , Cr 3+ , Cu 2+ , Fe 3+ , Na + , K + , Mn 2+ , Ni 2+ , and Pb 2+ , n = 1, 2, 3), into the H 2 O suspension of 1, with a concentration of 1 mM. Upon excitation at λ ex = 365 nm, interestingly, trivalent metal ions of Cr 3+ , Fe 3+ , and Al 3+ have striking enhancement responses of 10.9, 5.4, and 5.4 times on the fluorescence intensity of 1 in H 2 O suspension-phase, while other mono-and divalent metal ions have no or only minor effects (intensity change ≤ 10%) on the fluorescence intensity ( Figure 6). It is also noteworthy that the addition of metal ions did not cause significant wavelength shift. To examine the selectivity of 1 toward Cr 3+ , Fe 3+ , and Al 3+ , the interference experiments were carried out. After adding Cr 3+ , Fe 3+ , and Al 3+ ions into the H 2 O suspension of 1 containing other competitive metal ions in equal concentration, the fluorescence intensity of 1 in H 2 O suspension-phase increased immediately but showed a certain degree of reduction in some cases, with a maximum of about 43% reduction compared to that without the co-existence of competitive metal ions (Figure 7). The observations indicated that changes in intensity for sensing Cr 3+ , Fe 3+ , and Al 3+ were somewhat negatively affected, but still within acceptable limits, by the competitive metal ions. As a result, 1 displayed accessible anti-interference ability and detection selectivity for recognizing Cr 3+ , Fe 3+ , and Al 3+ over other competitive metal ions.
Cr2O7 2− and CrO4 2− for 1 in H2O suspension-phase. Inset: the linear Stern−Volmer plot at low Cr2O7 2− and CrO4 2− ion concentrations, where the solid line represents the linear fit to the Stern−Volmer equation.

Detection of Metal Ions
The influence of different metal ions on the fluorescent properties of 1 in H2O suspension-phase was also investigated through similar procedures by adding wellprepared aqueous solutions of metal ions, including M(NO3)n (M n+ = Ag + , Al 3+ , Mg 2+ , Ca 2+ , Co 2+ , Cr 3+ , Cu 2+ , Fe 3+ , Na + , K + , Mn 2+ , Ni 2+ , and Pb 2+ , n = 1, 2, 3), into the H2O suspension of 1, with a concentration of 1 mM. Upon excitation at λex = 365 nm, interestingly, trivalent metal ions of Cr 3+ , Fe 3+ , and Al 3+ have striking enhancement responses of 10.9, 5.4, and 5.4 times on the fluorescence intensity of 1 in H2O suspension-phase, while other mono-and divalent metal ions have no or only minor effects (intensity change ≤ 10%) on the fluorescence intensity ( Figure 6). It is also noteworthy that the addition of metal ions did not cause significant wavelength shift. To examine the selectivity of 1 toward Cr 3+ , Fe 3+ , and Al 3+ , the interference experiments were carried out. After adding Cr 3+ , Fe 3+ , and Al 3+ ions into the H2O suspension of 1 containing other competitive metal ions in equal concentration, the fluorescence intensity of 1 in H2O suspension-phase increased immediately but showed a certain degree of reduction in some cases, with a maximum of about 43% reduction compared to that without the co-existence of competitive metal ions (Figure 7). The observations indicated that changes in intensity for sensing Cr 3+ , Fe 3+ , and Al 3+ were somewhat negatively affected, but still within acceptable limits, by the competitive metal ions. As a result, 1 displayed accessible anti-interference ability and detection selectivity for recognizing Cr 3+ , Fe 3+ , and Al 3+ over other competitive metal ions.

Detection of Metal Ions
The influence of different metal ions on the fluorescent properties of 1 in H2O suspension-phase was also investigated through similar procedures by adding wellprepared aqueous solutions of metal ions, including M(NO3)n (M n+ = Ag + , Al 3+ , Mg 2+ , Ca 2+ , Co 2+ , Cr 3+ , Cu 2+ , Fe 3+ , Na + , K + , Mn 2+ , Ni 2+ , and Pb 2+ , n = 1, 2, 3), into the H2O suspension of 1, with a concentration of 1 mM. Upon excitation at λex = 365 nm, interestingly, trivalent metal ions of Cr 3+ , Fe 3+ , and Al 3+ have striking enhancement responses of 10.9, 5.4, and 5.4 times on the fluorescence intensity of 1 in H2O suspension-phase, while other mono-and divalent metal ions have no or only minor effects (intensity change ≤ 10%) on the fluorescence intensity ( Figure 6). It is also noteworthy that the addition of metal ions did not cause significant wavelength shift. To examine the selectivity of 1 toward Cr 3+ , Fe 3+ , and Al 3+ , the interference experiments were carried out. After adding Cr 3+ , Fe 3+ , and Al 3+ ions into the H2O suspension of 1 containing other competitive metal ions in equal concentration, the fluorescence intensity of 1 in H2O suspension-phase increased immediately but showed a certain degree of reduction in some cases, with a maximum of about 43% reduction compared to that without the co-existence of competitive metal ions (Figure 7). The observations indicated that changes in intensity for sensing Cr 3+ , Fe 3+ , and Al 3+ were somewhat negatively affected, but still within acceptable limits, by the competitive metal ions. As a result, 1 displayed accessible anti-interference ability and detection selectivity for recognizing Cr 3+ , Fe 3+ , and Al 3+ over other competitive metal ions.   To further quantitatively evaluate the fluorescence response of 1 in H 2 O suspensionphase toward Cr 3+ , Fe 3+ , and Al 3+ , fluorescence titration for the above three metal ions was also investigated. It was observed that the fluorescence intensity of 1 in H 2 O suspensionphase increased gradually upon incremental addition of the aqueous solutions of Cr 3+ , Fe 3+ , and Al 3+ ions (Figure 8a-c). The fluorescence intensity (λ em = 414 nm) obeyed the first-order exponential decay relationship with the concentration of Cr 3+ , Fe 3+ , and Al 3+ (Figure 8d-f (Figure S8), suggesting that 1 could potentially quantitatively determine Cr 3+ , Fe 3+ , and Al 3+ ions. Accordingly, the LOD is calculated to be 2.81 (146.2) µM (ppb) for Cr 3+ detection, 3.82 (198.7) µM (ppb) for Fe 3+ detection, and 3.37 (90.9) µM (ppb) for Al 3+ detection, through the equation of 3σ/k.

Fluorescence-Responsive Sensing Mechanisms
Herein, the possible fluorescence-responsive sensing mechanisms are elucidated. For Cr(VI) oxyanions detection, the XRPD patterns of 1 treated with Cr2O7 2− and CrO4 2− in water were closely matched to that of as-synthesized 1 ( Figure S9), implying that the framework integrity of 1 was retained. Thus, the turn-off sensing mechanism could not be due to the collapse of the framework. Ultimately, the mechanism of the quenching effect could be interpreted by UV−vis absorption spectroscopy. The UV−vis absorption band of Cr(VI) oxyanions and the fluorescence emission band of 1 show a small degree of overlap, suggesting that fluorescence quenching was not caused by energy transfer mechanism ( Figure S10). However, the large extent of overlap between the UV−vis absorption spectra of Cr2O7 2− /CrO4 2− and the fluorescence excitation spectrum of 1 demonstrates that the competition of absorption of irradiated light source energy between the Cr2O7 2− and CrO4 2− ions and 1 leads to the high fluorescence quenching efficiencies.

Fluorescence-Responsive Sensing Mechanisms
Herein, the possible fluorescence-responsive sensing mechanisms are elucidated. For Cr(VI) oxyanions detection, the XRPD patterns of 1 treated with Cr 2 O 7 2− and CrO 4 2− in water were closely matched to that of as-synthesized 1 ( Figure S9), implying that the framework integrity of 1 was retained. Thus, the turn-off sensing mechanism could not be due to the collapse of the framework. Ultimately, the mechanism of the quenching effect could be interpreted by UV−vis absorption spectroscopy. The UV−vis absorption band of Cr(VI) oxyanions and the fluorescence emission band of 1 show a small degree of overlap, suggesting that fluorescence quenching was not caused by energy transfer mechanism ( Figure S10). However, the large extent of overlap between the UV−vis absorption spectra of Cr 2 O 7 2− /CrO 4 2− and the fluorescence excitation spectrum of 1 demonstrates that the competition of absorption of irradiated light source energy between the Cr 2 O 7 2− and CrO 4 2− ions and 1 leads to the high fluorescence quenching efficiencies.
The XRPD patterns of 1 after being treated with Cr 3+ , Fe 3+ , and Al 3+ in water were very similar to that of as-synthesized 1 ( Figure S11), suggesting that 1 kept the framework integrity and crystallinity. This excluded the possibility of framework collapse caused fluorescence enhancement. Literature has shown that framework-analyte interactions usually demonstrated one of the most possible mechanisms for fluorescence turn-on detection toward metal ions [12,13,31,40,48,49]. To verify this assumption, IR and X-ray photoelectron spectroscopy (XPS) spectra were measured. Unfortunately, the IR spectra of 1 did not show significant changes after immersion ( Figure S12). Meanwhile, the O 1s and N 1s peaks in the XPS spectra were almost unchanged ( Figure S13). These findings indicate that there might be no framework-analyte interactions or that such interactions are too weak to cause spectra change. Thus, the turn-on detection mechanism is not due to the framework-analyte interactions. Interestingly, the UV-vis absorption spectra of 1 after addition of Cr 3+ , Fe 3+ , and Al 3+ exhibited significant increase in the absorbance within the sphere of 300-400 nm ( Figure S14), corresponding to the excitation energy used. The results illustrated that the turn-on detection of 1 toward Cr 3+ , Fe 3+ , and Al 3+ can be explained by the absorbance caused enhancement (ACE) mechanism [12,50], meaning that the M 3+ -treated 1 would release more energy than pristine 1 to demonstrate turn-on effect in the emission process.

Conclusions
In summary, a luminescent Cd(II) coordination polymer 1 adopting a 2-D gridlike sql layer has been successfully synthesized. Coordination polymer 1 shows strong fluorescence emissions in solvent suspension-phase, making them potential candidates to be employed in sensing Cr 2 O 7 2− and CrO 4 2− ions via fluorescence quenching effect and in detecting Cr 3+ , Fe 3+ , and Al 3+ ions via fluorescence enhancement response, with high sensitivity and selectivity. From the fluorescence-sensing mechanism studies, absorption energy competition and absorbance caused enhancement can, respectively, interpret the quenching effect toward Cr(VI) oxyanions and the enhancement effect toward Cr 3+ , Fe 3+ , and Al 3+ metal ions.

Supplementary Materials:
The following are available online, Figure S1. Experimental and simulated XRPD patterns for as-synthesized 1. Table S1. Hydrogen-bonding parameters in 1. Figure S2. TG curve of 1. Figure S3. Gas adsorption isotherms for thermally activated 1 conducted at 77 K for N 2 and 195 K for CO 2 . Figure S4. XRPD patterns of 1: simulated, as-synthesized, and after N 2 and CO 2 adsorption. Figure S5. Fluorescence excitation and emission spectra of NI-mbpy-34, Br-1,3-H 2 bdc, and 1 in solid-state, as well as 1 in H 2 O suspension-phase at room temperature.  Figure S11. XRPD patterns of 1 before and after immersing in Cr 3+ , Fe 3+ , and Al 3+ aqueous solutions for 24 h. Figure S12. IR spectra of 1 before and after immersing in Cr 3+ , Fe 3+ , and Al 3+ aqueous solutions for 24 h. Figure S13. XPS high resolution spectra of O 1s and N 1s of 1 before and after immersing in Cr 3+ , Fe 3+ , and Al 3+ aqueous solutions for 24 h. Figure S14. UV-vis spectra of 1 before and after addition of Cr 3+ , Fe 3+ , and Al 3+ .