- Email: [email protected]
Highly selective CHEF-type chemosensor for lutetium (III) recognition in semi-aqueous media
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Highly selective CHEF-type chemosensor for lutetium (III) recognition in semi-aqueous media R. Selva Kumar a, S.K. Ashok Kumar a,⁎, Kari Vijayakrishna a, Akell Sivaramakrishna a, C.V.S. Brahmnanda Rao b, N. Sivaraman b, Suban K. Sahoo c a b c
Department of Chemistry, School of Advanced Sciences, Vellore Institute of Technology, Vellore, 632014, Tamil Nadu, India Indira Gandhi Centre for Atomic Research, Homi Bhabha National Institute (HNBI), Kalpakkam 603102, Tamil Nadu, India Department of Applied Chemistry, S. V. National Institute Technology, Surat 395007, Gujarat, India
a r t i c l e
i n f o
Article history: Received 18 November 2018 Received in revised form 29 January 2019 Accepted 2 February 2019 Available online 5 February 2019 Keywords: Fluorescent chemosensors Lu3+ CHEF AIE DFT
a b s t r a c t A simple phosphoryl quinolone (L) based sensor has been synthesized for the selective recognition of Lu3+ by spectrofluorimetric method. In methanol-water (1:1, v/v), the ligand L exhibits a weak emission peak at 400 nm upon excitation at 280 nm. Upon interaction with various f-metal and other selected metals from s, p, and d-block elements, the fluorescence of L is selectively enhanced in the presence of Lu3+ due to the chelation enhanced fluorescence (CHEF) effects. The quantum yield (φ) of L (φ = 0.063) is enhanced to φ = 0.118 upon chelation with Lu3+ ion. From the titration experiment, the limit of detection (LOD) of sensor L to recognize Lu3+ is estimated down to 24.2 nM, which is much lower than the WHO guidelines (76 μM) in drinking water. The formation of host-guest complexation between L and Lu3+ in 2:1 binding stoichiometry is studied by Job's method and the binding constant is estimated by band fit analysis (logKf = 5.1). Further, the coordination behaviour between L and Lu3+ is well supported by FT-IR, 1H NMR, 13C NMR, 31P NMR, ESI mass spectral data and the theoretical results. © 2019 Published by Elsevier B.V.
1. Introduction Quinoline based fluorescent derivatives are widely applied for the sensing of metal ions, where the quinolone unit is used as a central fluorescent signalling unit [1–4]. Quinoline derivatives are quite attractive due to their relative synthetic versatility and the facile introduction of substituent to quinoline heterocyclic scaffolds which allows subsequent tuning of photophysical properties of probes [5–8]. The development of probes undergoing fluorescence enhancement in the presence of metal ions has always been attractive as it is easier and more efficient mode of detection [9]. In this view, the design and synthesis of fluorescence turn-on sensor with high selectivity, still remain an important topic of research. Due to similarity in chemical and physical properties of the lanthanides, their analysis in a mixture of several members of this group is extremely time-consuming and complicated [10]. Lutetium is an interesting metal in materials applications and it can be found in household items such as colour television, fluorescent bulbs, glasses, magnetic resonance image contrast agents and catalyst in petroleum refinery units. An increase in lutetium content in the environment is due to the improper disposal of such exhausted products. Eventually, this leads ⁎ Corresponding author. E-mail address: [email protected] (S.K. Ashok Kumar).
https://doi.org/10.1016/j.saa.2019.02.003 1386-1425/© 2019 Published by Elsevier B.V.
to the contamination of soil and water with significant concentration of Lu3+, and further in human and animal philological systems. It can cause the cell membrane damage, creating a negative influence on the reproduction and nervous system function [11]. Therefore, it extremely desirable to find highly selective and sensitive analytical methods for the trace level quantification of Lu3+. The conventional method for the determination of Lu3+ includes neutron activation analysis [12], spectrophotometric determinations [13], resonance ionization spectroscopy determination [14], extractions chromatography-atomic emission spectrometric [15], atomic spectroscopy [16], inductive coupled plasma-atomic spectroscopy [17], and ion-selective electrodes [18]. Most of these methods suffer from a high limit of detection, narrow working concentration range, serious interferences from various metal ions and the cost involved. Therefore, there is a continuous growth in the development of highly selective and sensitive florigenic sensors for quantitative detection of lanthanides. But a very few reports are available for Lu3+ determination using fluorescence method such as 8-hydroxyquinoline functionalized mesoporous silica [19,20], dipicolinamide based chemosensor [21] and salicyloylhydrazone base Schiff base [22]. These reported Lu3+ selective probes suffer from serious interference from other lanthanides (Table 1S). In order to overcome the drawbacks discussed above in the detection of Lu3+, herein, we have designed a new receptor (3-((diphenylphosphoryl)(hydroxy) methyl)-6-methylquinolin-2(1H)-one) (L) for the selective detection
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
and quantification of Lu3+ without any interferences from other lanthanides. 2. Experimental section 2.1. Materials and instrumentations The chloride salts of lanthanides and the nitrate salts of other metals were procured from Sigma Aldrich and SD Fine chemicals, respectively. All other solvents used in various experiments were procured from Merck chemicals (AR grade) and were used without further purification. Fluorescence measurements were carried out in a Hitachi F7000 Fluorescence spectrophotometer equipped with a xenon lamp. Instrumental parameters were controlled with FL Winlan software. The absorption spectra were recorded in a JASCO V730 UV–visible spectrometer using 1 cm quartz cell. Fluorescence lifetime analysis was done with JOBIN-VYON with pulsed-diode excitation sources of 295 nm LED. Infrared spectra were recorded in Shimadzu Affinity 1 FTIR spectrometer in the range of 4000–400 cm−1. NMR spectra were recorded using a 400 MHz Advance Bruker NMR spectrometer. 1H, 13C, and DEPT chemical shift were recorded in ppm downfield from TMS. 31P NMR chemical shift was recorded in ppm using 30% H3PO4 as an internal standard. All theoretical calculations were done by using Gaussian G09W [23,24]. The ground state geometry was optimized by employing DFT at the B3LYP level with the basis set, 6–31(d,p) for L and the SDD effective core potential for Lu. All the structures corresponding to true minima of the potential energy surface were confirmed by the vibrational frequency calculations. The excited state parameters were calculated using the TDDFT method [25]. 2.2. Synthesis of L and LuL2 complex Initially, 6-methyl-2-oxo-1,2-dihydroquinoline-3-carbaldehyde and diphenyl phosphine oxide were synthesized as per literature without any modification [26,27]. The toluene solution of 6-methyl-2-oxo-1,2dihydroquinoline-3-carbaldehyde (0.374 g, 2 mmol) was added to an equimolar diphenyl phosphine oxide (0.4 g, 2 mmol) in 10 mL of toluene in the presence of piperidine. The reaction mixture stirred for overnight. The pale white solid precipitate was filtered, washed with toluene and finally dried under a vacuum over anhydrous CaCl2. A pale white solid of L was obtained in 92% yield. The L was characterized through various spectroscopy techniques such as FTIR, 1H, 13C, 31P NMR and ESI-mass (for spectral details, see Figs. 1S–5S in SI). FT-IR (ATR, cm−1): 3224 (N\\H), 3020 (Ar-CH), 2843 (alk-CH2), 1645 (C_O), 1571 (C_C), 1450 (CH2-bending), 1180 (P_O), 694 (N\\H bending), 515 (P\\C). 1H NMR (400 MHz, DMSO‑d6): 2.31 (s, 3H), 5.98–6.00 (t,1H), 6.40–6.45 (dd,1H), 7.16 (d,1H), 7.28 (s,1H), 7.44 (t,2H), 7.49–7.55 (m,3H), 7.56 (d, 1H), 7.70 (d,1H), 7.73–7.77 (t,2H), 7.80–7.85 (t, 2H), 11.71 (s,1H). 13C NMR (100 MHz, DMSO‑d6): 20.80, 64.61, 65.48, 115.28, 119.21, 127.70, 128.68, 128.78, 129.37, 130.52, 131.30, 131.42, 131.45, 131.50, 132.14, 132.26, 132.34, 132.50, 133.43, 136.46, 138.57, 138.62, 161.07. DEPT135 (100 MHz, DMSO‑d6): 20.80, 64.61, 65.47, 115.21, 127.70, 128.68, 128.79, 131.42, 131.50, 132.09, 132.15, 132.26, 132.34, 138.57, 138.63. 31 P NMR (162 MHz, DMSO‑d6): 28.46. HR-MS: 389.1210 and calculated 389.1200. Complex of LuL2 was synthesized by adding a 10 mL of methanolic solution of LuCl3·6H2O (0.25 g, 0.64 mmol) slowly to the 10 mL methanolic solution of L (0.5 g, 1.28 mmol). Subsequently, the mixture was further stirred at room temperature for an overnight. The solvent was removed under reduced pressure and washed several times with diethyl ether. The colour of the complex is yellow with a yield of 95%. The complex was characterized through various spectroscopic and analytical techniques including; 1H, 13C, 31P NMR FTIR and ESI-mass (for spectral details, see Figs. 6S–10S). FTIR (ATR, cm−1): 3140 (OH, H2O), 3055 (Ar-CH), 2964 (alk-CH2), 1637, 1614 (C_N), 1562 (C_C),
33
1435 (CH2 bending), 1143 (P_O). 1H NMR (400 MHz, DMSO-d6): 2.29 (s, 3H), 6.11 (d, 1H), 7.34 (d, 2H), 7.40 (s, 3H), 7.46 (m, 2H), 7.65 (s, 3H), 7.75–7.78 (m, 4H). 13C NMR (100 MHz, DMSO-d6): 161.12, 138.63, 136.42, 132.37, 132.29, 131.56, 131.48, 130.45, 128.81, 128.70, 127.68, 119.27, 115.37, 65.55, 6470, 20.82. 31P NMR (162 MHz, DMSO-d6): 32.88. ESI-TOF mass: 950.9247, calculated mass: 951.7302. 3. Results and discussion The synthesis of desired ligand L was accomplished in high yield by reacting 6-methyl-2-oxo-1,2-dihydroquinoline-3-carbaldehyde with diphenyl phosphine oxide (Scheme 1). The compound was systematically characterized by 1H NMR, 13C NMR, 31P NMR, FT-IR and HR-mass analysis. 3.1. Absorption and emission studies The ligand L contains several oxygen donor atoms such as phosphoryl, hydroxyl, and carbonyl groups [28], which can participate in coordinate bonding with metal ions while the methyl quinolinone unit can provides analytically useful optical response. The sensing ability of L (50 μM) was investigated in the presence of various metal ions such as Na+, Cs+, Mg2+, Ca2+, Mn2+, Fe3+, Pb2+, Cd2+, Hg2+, Zn2+, Cr3+, Co2+, Cu2+, Al3+, La3+, Ce3+, Ce4+, Pr3+, Nd3+, Sm3+, Eu3+, Gd3+, Tb3+, Dy3+, Ho3+, Er3+, Tm3+, Yb3+, Lu3+, Th4+, and UO22+ (50 μM) by recording the UV–visible spectra in CH3OH-H2O (1:1, v/v). The neat ligand L exhibits two electronic bands appeared at 280 nm and 340 nm with a molar absorptivity of 9740 L·mol−1 cm−1 and 7380 L·mol−1 cm−1 respectively due to the π-π* transitions from diphenylphosphoryl and methylquinolinone. As seen in Fig. 1a, there is no significant change in the position of peak observed at 280 nm when interacted with alkali, alkane earth and transition metal ions. Upon addition of Lu 3+ to L, the absorption peak at 340 nm did not change while peak at 280 nm experience weak blue shift with hyperchromism and molar absorptivity increased from 9740 L·mol −1 cm −1 to 13,100 L·mol −1 cm −1. However, for other lanthanides ions like Sm 3+, Pr 3+ , Dy3+, Er 3+, Nd 3+ and Ho3+ on interacting with L, the wavelength maximum (λ max ) experience red shift to 290 nm with hypochromic effect. These studies revealed that using L, quantitative estimation of Lu3+ by this method is difficult due to serious interference from other lanthanides. Under similar experimental conditions, the cation sensing ability of L (50 μM) was next examined by fluorescence spectroscopy. The neat ligand L exhibits an emission band centred at 400 nm on excitation with 280 nm. The fluorescence of L was enhanced significantly and slightly blue-shifted to 394 nm in the presence of Lu3+. Also, under UV-light irradiation, the change in fluorescence property of L in the presence of Lu3+ can be seen by the naked-eye while other tested metal ions (Fig. 1b) and anions (Fig. 11S) did not show any fluorescence behaviour under similar experimental conditions. The enhancement of emission at 400 nm of L along with slight blue-shift to 394 nm upon addition of Lu3+ can be attributed to the selective complexation between the ligand L and Lu3+. The complexation increases the rigidity of the molecular assembly by restricting the free rotation of the substituents like phosphoryl, hydroxyl and methyl quinolinone moieties and
Scheme 1. Synthesis of L.
34
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
Fig. 1. (a) UV–Vis and (b) fluorescence (λex = 280 nm) spectral changes of L (50 μM) with various metal ions (10 eq) in CH3OH-H2O (1:1, v/v).
promotes the CHEF effect to cause significant enhancement of the emission intensity [29]. Also, Lu3+ cannot significantly change the emission intensity of L because it has no low-energy f-f states for energy transfer [30] and therefore the enhancement is related to the changes in the optical properties of L occurred upon complexation with Lu3+. The effects of solvents ratio between water and methanol on the fluorescence of L and LuL2 were examined at different methanol water ratio. Results show that both L and LuL2 exhibits similar trend (Figs. 12S and 13S), with increase in water content (in methanol: water mixture) resulting in the decrease in the emission intensity due to molecular rigidity attainment by molecular packing modes and conformations in the aggregates states which is confirmed by SEM analyses (Fig. 2). Further, SEM image supports the formation of 1D micro rod; on the surface of these rods exhibits rough surface texture. In the case of LuL2, the emission behaviour follows similar to L but the emission intensity increased by two-fold compare to L. This increase in emission intensity is due to the formation of stable complex with Lu3+; thus all the molecular rotations of LuL2 are arrested. Besides, there is a formation of aggregates which is responsible for the enhanced emission [31]. Further, the formation of LuL2 aggregates (288 nm) was confirmed in dynamic light scattering (DLS) study (Fig. 14S). Therefore, the restriction of intramolecular rotation of the diphenyl phosphoryl and methyl quinolinone moieties of the ligand L in the LuL2 complex
is the main reason for the enhancement of an emission intensity. Furthermore, the intramolecular rotations of the fluorophores are restricted upon aggregation prevent the non-radiative pathway, resulting in the enhancement of emission. Fluorescence titration was carried out using 50 μM of L by adding different amounts of Lu3+ from 0 to 30 μM in MeOH/water (1:1, v/v). Results shows a steady increase in fluorescence intensity with respect to Lu3+ (Fig. 3a). The binding stoichiometry was verified by Job's plot method [32]. The maximum emission intensity was achieved at 0.66 mol ratio of L, which confirms the formation 2:1 ligand to metal complex (Fig. 3b). Table 1 shows the resulting association constants of several fluorescence titrations used for binding capability of L towards Lns. All binding data were fit to a 1:1 and 2:1 binding model and association constants were determined using BindFit v0.5 program [33]. As shown in Table 1, L exhibit a more binding preference to Lu3+ in 1:1 and 2:1 complex formation compare to other Lns. The binding capacity is almost three times more with Lu3+ compare to other Lns. This increase in binding constant may be due to their characteristic coordination geometries with Lu3+. The lowest detection limit (LOD), 3σ/slope, where slope is obtained from calibration curve and σ standard deviation of emission intensity of L [34] was determined from the fluorescence titration data, and the results shows that Lu3+ can be detected down to 24.8 nM (Fig. 3a).
Fig. 2. SEM image of L and LuL2 aggregates prepared in 1:1 methanol and water.
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
35
Fig. 3. (a) Fluorescence spectral change of L with Lu3+ in MeOH/H2O (λex = 280 nm, Insert figure represent calibration plot) (b) Jobs plot.
Fluorescence quantum yields of L and LuL2 were calculated in CH3OH by employing comparative William's method using anthracene in ethanol as a standard [35,36]. The estimated fluorescence quantum yield (φ) of L and LuL2 were found to be 0.063 and 0.115, respectively. In order to achieve a deeper insight into the photo-physical properties of L and LuL2 using time-resolved fluorescence decay measurement was performed in methanol medium (Fig. 4). The free ligand L shows a lifetime of 1.17 ns whereas LuL2 shows a lifetime of 4.66 ns. The rate constant for the radiative (Kr) and nonradiative (Knr) decay process were estimated from the fluorescence quantum yield and lifetime [37]. Accordingly, the calculated values of Kr and Knr for L are 1 × 107 s−1 and 8.45 × 108 s−1 respectively whereas for LuL2 is 1.72 × 107 s−1 and 1.97 × 108 s−1 respectively. Results show that L is exhibiting a weak fluorescence as Kr is less than Knr. However, LuL2 exhibits increase in Kr and decrease in Knr as compare to L. Hence, this phenomenon is also supporting the strong fluorescence enhancement of LuL2. The influence of pH on fluorescence response of L and LuL2 was carried out between pH 2.0 to 10.0 using appropriate buffer solutions (Fig. 5a). The fluorescence intensity of L alone did not get altered from 3.5 to 10.0 but the intensity decreases below pH 3.5 due to the protonation of hydroxyl functional groups. In the case of complex LuL2, the intensity remains same in the pH range between pH 4.0 to 9.0 and whereas the intensity decreases above pH 9.0 and below pH 4.0. The decrease in intensity in acidic region (2.0–4.0) due to protonation nitrogen atom, which restricts the complexation and at higher pH (˃9) due to the formation of Lu(OH)3 [38].
The specificity of ligand L as a fluorogenic sensor for the detection of Lu3+ ions in the presence of various competing cations was explored in CH3OH/H2O (1:1, v/v) (Fig. 5b). For competitive studies, the ligand L (50 μM) was treated with 0.5 equivalent of Lu3+ and 10 equivalents of interfering metal ions then each solution was excited at 280 nm and the emission intensity was measured at 400 nm. The obtained results indicate that no interference in the detection of Lu3+ ions in the presence of interfering metal ions. Further, the performance of L compared with other previously reported sensors shown in Table 1S, It concludes that L is exhibit highest sensitivity (24.2 nM) and no interference from other metal ions [19,20]. Therefore, the present sensor can be applied for the determination of trace amount of Lu3+ ions in actual samples. To check the dynamic response of the LuL2 interactions, the binding reversibility of LuL2 complex in CH3OH/H2O (1:1 v/v) media was performed. Due to the high stability constant of EDTA–Lu3+ complex, the binding reversibility between LuL2 and EDTA was observed (Fig. 15S). The purpose of this test is to show how quickly L can be recoverable from its complex. 3.2. Thermal stability of L and LuL2 In order to study the thermal stability of L and LuL2, thermal analysis was carried out using TGA and DSC. The TGA analysis of L indicates three
Table 1 Association constants, Ka (M−1), of L towards Lns in CH3OH-H2O (1:1, v/v) solution at 298 K calculated using BindFit v0.5. Ln(III)
logKf(1:1)
logKf(2:1)
La3+ Ce3+ Pr3+ Nd3+ Sm3+ Eu3+ Gd3+ Tb3+ Dy3+ Ho3+ Er3+ Tm3+ Yb3+ Lu3+
1.18 1.14 1.96 1.77 1.84 1.26 1.77 1.25 1.47 1.25 1.45 1.15 1.08 3.14
2.64 2.85 2.63 2.97 2.55 2.62 2.63 2.66 2.72 2.66 2.68 2.44 2.44 5.71
Fig. 4. Time resolved fluorescence decay of L and LuL2 in CH3OH by exciting at 295 nm.
36
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
Fig. 5. (a) Effect of pH on the fluorescence of L and LuL2 and (b) Bar diagram of LuL2 fluorescence in the presence of various competing metal ions.
steps of weight loss (Fig. 16S). The first weight change corresponding to loss of water molecules observed at 100 °C. The second weight loss was from 240 °C to 500 °C and may be due to the removal of aromatic units and the final weight loss was due to the conversion to final product to P2O5 residue. Similarly, the LuL2 complex also shows three steps weight loss. The first weight loss occurred at about 140 °C corresponding to the removal of water molecule present in the system. The second weight loss between 150 and 600 °C is corresponding to the removal of organic molecules and the final weight loss can be attributed to the decomposition to LuPO4 and P2O5 [39]. On the other hand DSC analysis reveals that the ligand L shows (Fig. 17S) an endothermic signal at 236 °C with the transition enthalpy of 19.95 J/kg, corresponding to the melting point and relative transition energy involves in the process. The DSC analysis indicates two endothermic signals at 77 °C and 210 °C, which also confirms the presence of water molecule in the complex which melts at 77 °C with transition enthalpy of 116.1 J/kg; the LuL2 complex starts to melt at 210 °C with transition enthalpy of 177.1 J/kg. The comparison of L with the LuL2 complex shows lower melting point with higher transition enthalpy [40]. 3.3. Binding mechanism In order to support the binding mechanism between L and Lu3+, we have studied FTIR, 1H NMR, 13C NMR, 31P NMR and ESI mass analysis. As seen from Fig. 6a, the FT-IR of free ligand L shows vibrational bands at 3224 cm−1 assigned to vibrations of \\NH and \\OH stretch,
1645 cm−1 assigned to C_O stretch and 1180 cm−1 assigned to P_O stretch. The vibrational bands at 694 cm−1 and 515 cm−1 are due to the bending vibrations of P_O group. After complexing with Lu3+, a broad signal appeared at 3140 cm−1 and it corresponds to\\OH stretch of water molecules which is present in the complex. The carbonyl group splits into two sharp ends and appeared at 1637 cm−1 and 1614 cm−1; these are due to non-symmetrical vibrations in LuL2. The P_O symmetrical vibrational band shift to 1140 cm−1 and bending vibrational bands are also shifted to 690 cm−1 and 510 cm−1. The shifting of bands to lower wavenumber indicates the formation of LuL2 complex. The formation of LuL2 complex was further confirmed by 1H, 13C and 31 P NMR analysis in DMSO‑d6. In case of 1H NMR, L alone exhibits characteristics peak at 11.76 ppm, 6.48 ppm and 6.0 ppm corresponding to\\NH, \\OH and \\CH (chiral) protons, respectively (Fig. 6b). Upon adding Lu3+ from 0.1 to 0.5 eq to L, the \\NH peak experience a deshielding effect due to the involvement ˃C_O group in bonding formation with Lu3+ thereby the electron density present at hydrogen (\\NH) is attracted by ˃C_O group. The \\OH peak appeared at 6.48 ppm begins to decrease in its intensity as Lu3+ contents increased and at 0.5 eq, the\\OH peak completely disappeared due to the formation of 2:1 stoichiometry complexation. The hydrogen attached to chiral carbon is suffering from weak shielding effect due to the formation LuL2. In case of 13C, the peak appeared at 161.57 ppm corresponds to carbonyl carbon, upon complexation with Lu3+ the carbonyl carbon peak undergoes weak deshielding effect (161.08 ppm) confirming involvement of ˃C_O group in the complexation (Fig. 8S). In the case
Fig. 6. (a) FT-IR spectra for L and LuL2 and (b) The 1H NMR titration of L upon incremental addition of Lu3+.
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
of 31P NMR, the free P_O group at 28.41 ppm after complexing with Lu3 + , the P_O group resonates at 29.65 ppm with significant peak broadening which indicates the P_O group involved in the complexation (Fig. 18S). Further, the ESI-TOF mass analysis of the complex exhibits two major peaks (m/z) appeared at 950.9247 and 1013.8861 (Fig. 10S). The mass 950.9247 corresponds to the involvement of 2L and 1Lu3+ which is agreeing with the calculated mass of 951.1604 while the mass corresponding to 1013.8961 is due to the involvement of 2L, 1Lu and 2 methanol. Both the mass peaks confirm the formation of complex in 2:1 binding stoichiometry between L and Lu3+. 3.4. Theoretical studies The possible 3D structure of the receptor L and LuL2 complex was proposed through quantum mechanical calculations by applying the B3LYP exchange-correlation functional and the basis sets 6-31G** for the C, H, N, O, P atoms whereas SDD for the Lu atom. The optimized structure of L shown in Fig. 7a preferred an enol-form, where the carbonyl group of quinolone-O and hydroxyl group are oriented in the same direction and form an intramolecular hydrogen bond of length 1.558 Å. Also, the computed L structure indicates that the complexation of Lu3+ with L through the oxygen atoms of phosphine oxide, hydroxyl and quinolone-O need apparent conformational changes in the receptor. Considering the experimental evidences, different possible coordination modes were calculated for the 1:2 binding ratio between the Lu and L, where the receptor L was considered either as bidentate or a tridentate ligand (Fig. 7b–d). When the bidentate mode was considered for the LuL2 complex, the optimization resulted distorted tetrahedral shaped complex and the interaction energy (Eint = ELuL2-2EL-ELu) was decreased by −209.58 kcal/mol and −220.74 kcal/mol for the binding mode-I and mode-II, respectively. The calculated Eint indicates the formation of the stable LuL2 complex where the participation of hydroxyl and quinoline-O group (i.e. binding mode-I) is more favourable due to
37
the conformational adjustment required in mode-II. When the ligand was considered as a tridentate ligand, the distorted octahedral shaped binding mode-III was obtained for the LuL2 complex (Fig. 7d). In this binding mode-III, the Eint was estimated as −247.38 kcal/mol, which is significantly lowered in compared to the mode-I and mode-II where the L acts as a bidentate ligand. The significant lowering of interaction energy in compared to modes-I/II indicate the formation of the more stable complex may be due to the higher coordination number preferred by Lu atom and preferential binding with hard donor atoms. Similar studies were conducted with selected Ln's such as La3+, Ce3+, Pm3+, Gd3+. Er3+ and the results are depicted in Fig. 19S. On comparison of binding energy data, LuL2 complex has lowest possible energy and it confirms L is having high selectivity with Lu3+. In order to obtain insights into the electronic excitation of L in the presence and absence of Lu3+, Time dependent density functional theory (TD-DFT) was carried out. The frontier molecular orbital involved in the UV–vis absorption of L and LuL2 shown in Fig. 20S and corresponding excitation energy and oscillation strength were tabulated in Table 2S. The theoretical absorption spectra are well match with experimental data. HOMO and LUMO of L are mainly located in the πelectrons of the quinolone ring, which indicates the π-π* transitions are mainly contributed for the absorption spectra. Whereas in LuL2 the HOMO orbital are mainly located on P_O group and LUMO orbitals are located on Lu atom. The energies of HOMO and LUMO of L and LuL2 were shown in Fig. 8 (a) and (b), the free ligand L exhibits the band gap of 4.08 eV and LuL2 complex exhibits the band gap of 0.56 eV. When Lu3+ was added, it showed a significant change in the distribution of π-electrons on quinolone ring system. More significantly, the HOMO and LUMO densities in free ligand L with Lu3+ were delocalized over the entire molecule except for extended the conjugation throughout both side quinolone rings which results the enhancing in the fluorescence emission with significantly higher energy than that of L.
Fig. 7. DFT computed structure of L. (a) Various possible complexes of L with Lu3+ (b–d).
38
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
Fig. 8. Frontier molecular orbital for L (a) and LuL2 (b) by DFT/B3LYP method.
3.5. Applications Analytical applications of L were tested for the quantification of Lu3+ in three different real and spiked samples such as demineralised water, tap water and river water. The test for each sample was carried out in three measurements and the results show that there is a good match with standard addition and recovery of the Lu3+ ions. Also, the novelty and analytical parameters of sensor L was compared with the reported Lu3+ sensors (Table 3S). This sensor does not suffer from any interference from any of the selected metal ions. Hence, L can apply as a florigenic sensor for the estimation of Lu3+ in actual samples containing Lu3+ contents. 4. Conclusion In summary, we have developed an easy-to-synthesize phosphoryl quinolone-based ligand (L) for the selective detection of Lu3+ by the spectrofluorometric method. The 2:1 binding stoichiometry for the LuL2 complex was proposed by Job's plot and mass analysis. The coordination modes of L was established by FT-IR and NMR data, followed by the 3D structure of L and its complex with Lu3+ was proposed by DFT calculations. L exhibits highly selective CHEF with Lu3+. Based on the spectral data, the LOD for the analysis of Lu3+ was calculated down to be 24.2 nM and the formation of LuL2 complex was reversible with EDTA. The sensor L could work in wide pH range from 4.0 to 9.0. The performance of L was successfully used for the analysis of Lu3+ contents present in different water samples. Acknowledgements Authors thank Board of Research in Nuclear Sciences (BRNS), GOI for supporting the work through the project grant (34/14/11/2015/BRNS).
Authors are also thankful to DST-VIT-FIST for NMR and Vellore Institute of Technology, Vellore for other research facilities. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.saa.2019.02.003.
References [1] S. Kantevari, M. Matsuzaki, Y. Kanemoto, H. Kasai, G.C.R. Ellis-Davies, Two-color, two-photon uncaging of glutamate and GABA, Nat. Methods 7 (2010) 123–125. [2] G. Jones, W.R. Jackson, C.Y. Choi, W.R. Bergmark, Solvent effects on emission yield and lifetime for coumarin laser dyes. Requirements for a rotatory decay mechanism, J. Phys. Chem. 89 (1985) 294–300. [3] S.K. Asthana, A. Kumar, S.K. Hira, P.P. Manna, K.K. Upadhyay, Brightening quinolineimines by Al3+ and subsequent quenching by PPi/PA in aqueous medium: synthesis, crystal structures, binding behavior, theoretical and cell imaging studies, Inorg. Chem. 56 (2017) 3315–3323. [4] K. Zhang, Z.Y. Yang, B.D. Wang, S.B. Sun, Y.D. Li, T.R. Li, Z.C. Liu, J.M. An, A highly selective chemosensor for Al3+ based on 2-oxo-quinoline-3-carbaldehyde Schiff-base, Spectrochim. Acta A Mol. Biomol. Spectrosc. 124 (2014) 59–63. [5] Y. Laras, V. Hugues, Y. Chandrasekaran, M. Blanchard-Desce, F.C. Acher, N. Pietrancosta, Synthesis of quinoline dicarboxylic esters as biocompatible fluorescent tags, J. Org. Chem. 77 (2012) 8294–8302. [6] C. McDonagh, C.S. Burke, B.D. MacCraith, Optical chemical sensors, Chem. Rev. 108 (2008) 400–422. [7] M. Dutta, D. Das, Recent developments in fluorescent sensors for trace-level determination of toxic-metal ions, TrAC Trends Anal. Chem. 32 (2012) 113–132. [8] Z. Xu, J. Yoon, D.R. Spring, Fluorescent chemosensors for Zn2+, Chem. Soc. Rev. 39 (2010) 1996–2006. [9] L. Feng, W. Shi, J. Ma, Y. Chen, F. Kui, Y. Hui, Z. Xie, A novel thiosemicarbazone Schiff base derivative with aggregation-induced emission enhancement characteristics and its application in Hg2+ detection, Sensors Actuators B Chem. 237 (2016) 563–569. [10] M.R. Ganjali, V.K. Gupta, F. Faridbod, P. Norouzi, Lanthanides Series Determination by Various Analytical Methods, Elsevier, 2016.
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39 [11] C. Bariáin, I.R. Matías, C. Fernández-Valdivielso, F.J. Arregui, M.L. Rodríguez-Méndez, J.A. De Saja, Optical fiber sensor based on lutetium bisphthalocyanine for the detection of gases using standard telecommunication wavelengths, Sensors Actuators B Chem. 93 (2003) 153–158. [12] D.L. Massart, J. Hoste, Activation analysis of rare earths: part II. Determination of lutetium in gadolinite, Anal. Chim. Acta 42 (1968) 15–20. [13] V.N. Azarova, A.I. Kirillov, Spectrophotometric determination of lanthanum, lutetium, and yttrium with triphenylmethane reagents and cetylpyridinium bromide, J. Anal. Chem. USSR 41 (1986) 1408–1412. [14] T.B. Krustev, S.T. Mincheva, D.A. Angelov, E.P. Vidolova-Angelova, Determination of traces of lutetium in geological samples by resonance ionization spectroscopy, J. Anal. At. Spectrom. 8 (1993) 1029–1031. [15] C. Peng, W. Li, P. Yuan, W. Feng, X. Wang, Determination of rare-earth impurities in ultra-pure thulium oxide, ytterbium oxide, lutetium oxide by extraction chromatography-atomic emission spectrometry, Chin. J. Anal. Chem. 25 (1997) 377–381. [16] J.C. Van Loon, J.H. Galbraith, H.M. Aarden, The determination of yttrium, europium, terbium, dysprosium, holmium, erbium, thulium, ytterbium and lutetium in minerals by atomic-absorption spectrophotometry, Analyst 96 (1971) 47–50. [17] C. Karadaş, D. Kara, Preconcentration of rare earth elements using amberlite XAD-4 modified with 2, 6-pyridinedicarboxaldehyde and their determination by inductively coupled plasma optical emission spectrometry, Water Air Soil Pollut. 225 (2014) 1972–1982. [18] F.G. Sanchez, M. Gracia, L. Hernandez, J.C. Marquez, A graphical derivative approach to the photometric determination of lutetium and praseodymium in mixtures, Talanta 34 (1987) 639–644. [19] M. Hosseini, M.R. Ganjali, F. Aboufazeli, F. Faridbod, H. Goldooz, A. Badiei, P. Norouzi, A selective fluorescent bulk sensor for lutetium based on hexagonal mesoporous structures, Sensors Actuators B Chem. 184 (2013) 93–99. [20] M. Hosseini, M.R. Ganjali, Z. Rafiei-Sarmazdeh, F. Faridbod, H. Goldooz, A. Badiei, P. Nourozi, G.A. Mohammadi Ziarani, A novel Lu3+ fluorescent nano-chemosensor using new functionalized mesoporous structures, Anal. Chim. Acta 771 (2013) 95–101. [21] F. Faridbod, M. Sedaghat, M. Hosseini, M.R. Ganjali, M. Khoobi, A. Shafiee, P. Norouzi, Turn-on fluorescent chemosensor for determination of lutetium ion, Spectrochim. Acta A Mol. Biomol. Spectrosc. 137 (2015) 1231–1234. [22] L. Wang, T. Jang, N. Sui, X. Yang, L. Zhang, X. Yao, Q. Zhou, H. Xiao, S. Kuang, W.Y. William, A highly selective Schiff base fluorescent chemosensor for Lu3+, Anal. Methods 44 (2017) 6254–6260. [23] M.J. Frisch, et al., Gaussian 09W, Revision A, Gaussian Inc., Wallingford, CT, 2009. [24] Y. Luo, J. Baldamus, O. Tardif, Z. Hou, DFT study of the tetranuclear lutetium and yttrium polyhydride cluster complexes [(C5Me4SiMe3) 4Ln4H8](Ln = Lu, Y) that contain a four-coordinate hydrogen atom, Organometallics 24 (2005) 4362–4366.
39
[25] C. Adamo, D. Jacquemin, The calculations of excited-state properties with timedependent density functional theory, Chem. Soc. Rev. 42 (2013) 845–856. [26] M.K. Singh, A. Chandra, B. Singh, R.M. Singh, Synthesis of diastereomeric 2, 4disubstituted pyrano [2, 3-b] quinolines from 3-formyl-2-quinolones through O–C bond formation via intramolecular electrophilic cyclization, Tetrahedron Lett. 48 (2007) 5987–5990. [27] D. Das, G. Gopakumar, C.V.S. Brahmmananda Rao, N. Sivaraman, A. Sivaramakrishna, K. Vijayakrishna, Extraction and coordination behavior of diphenyl hydrogen phosphine oxide towards actinides, J. Coord. Chem. 70 (2017) 3338–3352. [28] R.G. Pearson, Hard and soft acids and bases, J. Am. Chem. Soc. 85 (1963) 3533–3539. [29] A.W. Czarnik, Fluorescent chemosensors for ion and molecule recognition, ACS Symp. Ser. (1993) 538. [30] W.S. Xia, R.H. Schmehl, C.J. Li, A fluorescent 18-crown-6 based luminescence sensor for lanthanide ions, Tetrahedron 56 (2000) 7045–7049. [31] Y. Hong, J.W. Lam, B.Z. Tang, Aggregation-induced emission, Chem. Soc. Rev. 40 (2011) 5361–5388. [32] P. Job, Formation and stability of inorganic complexes in solution, Ann. Chim. Appl. 9 (1928) 113–203. [33] P. Thordarson, Determining association constants from titration experiments in supramolecular chemistry, Chem. Soc. Rev. 40 (2011) 1305–1323. [34] G.L. Long, J.D. Winefordner, Limit of detection. A closer look at the IUPAC definition, Anal. Chem. 55 (1983) 712A–724A. [35] A.T.R. Williams, S.A. Winfield, J.N. Miller, Relative fluorescence quantum yields using a computer-controlled luminescence spectrometer, Analyst 108 (1983) 1067–1071. [36] W.H. Melhuish, Quantum efficiencies of fluorescence of organic substances: effect of solvent and concentration of the fluorescent solute1, J. Phys. Chem. 65 (1961) 229–235. [37] R. Selva kumar, S.K. Ashok Kumar, K. Vijayakrishna, A. Sivaramakrishna, P. Paira, C.V.S. Brahmmananda Rao, N. Sivaraman, S.K. Sahoo, Bipyridine bisphosphonatebased fluorescent optical sensor and optode for selective detection of Zn2+ ions and its applications, New J. Chem. 42 (2018) 8494–8502. [38] H. López-González, M. Jiménez-Reyes, M. Solache-Ríos, A. Rojas-Hernández, Solubility and hydrolysis of lutetium at different [Lu3+]initial, J. Radioanal. Nucl. Chem. 274 (2007) 103–108. [39] K.S. Gavrichev, N.N. Smirnova, V.M. Gurevich, V.P. Danilov, A.V. Tyurin, M.A. Ryumin, L.N. Komissarova, Heat capacity and thermodynamic functions of LuPO4 in the range 0–320 K, Thermochim. Acta 448 (2006) 63–65. [40] P. Gabbott (Ed.), Principles and Applications of Thermal Analysis, John Wiley & Sons, 2008.
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Highly selective CHEF-type chemosensor for lutetium (III) recognition in semi-aqueous media R. Selva Kumar a, S.K. Ashok Kumar a,⁎, Kari Vijayakrishna a, Akell Sivaramakrishna a, C.V.S. Brahmnanda Rao b, N. Sivaraman b, Suban K. Sahoo c a b c
Department of Chemistry, School of Advanced Sciences, Vellore Institute of Technology, Vellore, 632014, Tamil Nadu, India Indira Gandhi Centre for Atomic Research, Homi Bhabha National Institute (HNBI), Kalpakkam 603102, Tamil Nadu, India Department of Applied Chemistry, S. V. National Institute Technology, Surat 395007, Gujarat, India
a r t i c l e
i n f o
Article history: Received 18 November 2018 Received in revised form 29 January 2019 Accepted 2 February 2019 Available online 5 February 2019 Keywords: Fluorescent chemosensors Lu3+ CHEF AIE DFT
a b s t r a c t A simple phosphoryl quinolone (L) based sensor has been synthesized for the selective recognition of Lu3+ by spectrofluorimetric method. In methanol-water (1:1, v/v), the ligand L exhibits a weak emission peak at 400 nm upon excitation at 280 nm. Upon interaction with various f-metal and other selected metals from s, p, and d-block elements, the fluorescence of L is selectively enhanced in the presence of Lu3+ due to the chelation enhanced fluorescence (CHEF) effects. The quantum yield (φ) of L (φ = 0.063) is enhanced to φ = 0.118 upon chelation with Lu3+ ion. From the titration experiment, the limit of detection (LOD) of sensor L to recognize Lu3+ is estimated down to 24.2 nM, which is much lower than the WHO guidelines (76 μM) in drinking water. The formation of host-guest complexation between L and Lu3+ in 2:1 binding stoichiometry is studied by Job's method and the binding constant is estimated by band fit analysis (logKf = 5.1). Further, the coordination behaviour between L and Lu3+ is well supported by FT-IR, 1H NMR, 13C NMR, 31P NMR, ESI mass spectral data and the theoretical results. © 2019 Published by Elsevier B.V.
1. Introduction Quinoline based fluorescent derivatives are widely applied for the sensing of metal ions, where the quinolone unit is used as a central fluorescent signalling unit [1–4]. Quinoline derivatives are quite attractive due to their relative synthetic versatility and the facile introduction of substituent to quinoline heterocyclic scaffolds which allows subsequent tuning of photophysical properties of probes [5–8]. The development of probes undergoing fluorescence enhancement in the presence of metal ions has always been attractive as it is easier and more efficient mode of detection [9]. In this view, the design and synthesis of fluorescence turn-on sensor with high selectivity, still remain an important topic of research. Due to similarity in chemical and physical properties of the lanthanides, their analysis in a mixture of several members of this group is extremely time-consuming and complicated [10]. Lutetium is an interesting metal in materials applications and it can be found in household items such as colour television, fluorescent bulbs, glasses, magnetic resonance image contrast agents and catalyst in petroleum refinery units. An increase in lutetium content in the environment is due to the improper disposal of such exhausted products. Eventually, this leads ⁎ Corresponding author. E-mail address: [email protected] (S.K. Ashok Kumar).
https://doi.org/10.1016/j.saa.2019.02.003 1386-1425/© 2019 Published by Elsevier B.V.
to the contamination of soil and water with significant concentration of Lu3+, and further in human and animal philological systems. It can cause the cell membrane damage, creating a negative influence on the reproduction and nervous system function [11]. Therefore, it extremely desirable to find highly selective and sensitive analytical methods for the trace level quantification of Lu3+. The conventional method for the determination of Lu3+ includes neutron activation analysis [12], spectrophotometric determinations [13], resonance ionization spectroscopy determination [14], extractions chromatography-atomic emission spectrometric [15], atomic spectroscopy [16], inductive coupled plasma-atomic spectroscopy [17], and ion-selective electrodes [18]. Most of these methods suffer from a high limit of detection, narrow working concentration range, serious interferences from various metal ions and the cost involved. Therefore, there is a continuous growth in the development of highly selective and sensitive florigenic sensors for quantitative detection of lanthanides. But a very few reports are available for Lu3+ determination using fluorescence method such as 8-hydroxyquinoline functionalized mesoporous silica [19,20], dipicolinamide based chemosensor [21] and salicyloylhydrazone base Schiff base [22]. These reported Lu3+ selective probes suffer from serious interference from other lanthanides (Table 1S). In order to overcome the drawbacks discussed above in the detection of Lu3+, herein, we have designed a new receptor (3-((diphenylphosphoryl)(hydroxy) methyl)-6-methylquinolin-2(1H)-one) (L) for the selective detection
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
and quantification of Lu3+ without any interferences from other lanthanides. 2. Experimental section 2.1. Materials and instrumentations The chloride salts of lanthanides and the nitrate salts of other metals were procured from Sigma Aldrich and SD Fine chemicals, respectively. All other solvents used in various experiments were procured from Merck chemicals (AR grade) and were used without further purification. Fluorescence measurements were carried out in a Hitachi F7000 Fluorescence spectrophotometer equipped with a xenon lamp. Instrumental parameters were controlled with FL Winlan software. The absorption spectra were recorded in a JASCO V730 UV–visible spectrometer using 1 cm quartz cell. Fluorescence lifetime analysis was done with JOBIN-VYON with pulsed-diode excitation sources of 295 nm LED. Infrared spectra were recorded in Shimadzu Affinity 1 FTIR spectrometer in the range of 4000–400 cm−1. NMR spectra were recorded using a 400 MHz Advance Bruker NMR spectrometer. 1H, 13C, and DEPT chemical shift were recorded in ppm downfield from TMS. 31P NMR chemical shift was recorded in ppm using 30% H3PO4 as an internal standard. All theoretical calculations were done by using Gaussian G09W [23,24]. The ground state geometry was optimized by employing DFT at the B3LYP level with the basis set, 6–31(d,p) for L and the SDD effective core potential for Lu. All the structures corresponding to true minima of the potential energy surface were confirmed by the vibrational frequency calculations. The excited state parameters were calculated using the TDDFT method [25]. 2.2. Synthesis of L and LuL2 complex Initially, 6-methyl-2-oxo-1,2-dihydroquinoline-3-carbaldehyde and diphenyl phosphine oxide were synthesized as per literature without any modification [26,27]. The toluene solution of 6-methyl-2-oxo-1,2dihydroquinoline-3-carbaldehyde (0.374 g, 2 mmol) was added to an equimolar diphenyl phosphine oxide (0.4 g, 2 mmol) in 10 mL of toluene in the presence of piperidine. The reaction mixture stirred for overnight. The pale white solid precipitate was filtered, washed with toluene and finally dried under a vacuum over anhydrous CaCl2. A pale white solid of L was obtained in 92% yield. The L was characterized through various spectroscopy techniques such as FTIR, 1H, 13C, 31P NMR and ESI-mass (for spectral details, see Figs. 1S–5S in SI). FT-IR (ATR, cm−1): 3224 (N\\H), 3020 (Ar-CH), 2843 (alk-CH2), 1645 (C_O), 1571 (C_C), 1450 (CH2-bending), 1180 (P_O), 694 (N\\H bending), 515 (P\\C). 1H NMR (400 MHz, DMSO‑d6): 2.31 (s, 3H), 5.98–6.00 (t,1H), 6.40–6.45 (dd,1H), 7.16 (d,1H), 7.28 (s,1H), 7.44 (t,2H), 7.49–7.55 (m,3H), 7.56 (d, 1H), 7.70 (d,1H), 7.73–7.77 (t,2H), 7.80–7.85 (t, 2H), 11.71 (s,1H). 13C NMR (100 MHz, DMSO‑d6): 20.80, 64.61, 65.48, 115.28, 119.21, 127.70, 128.68, 128.78, 129.37, 130.52, 131.30, 131.42, 131.45, 131.50, 132.14, 132.26, 132.34, 132.50, 133.43, 136.46, 138.57, 138.62, 161.07. DEPT135 (100 MHz, DMSO‑d6): 20.80, 64.61, 65.47, 115.21, 127.70, 128.68, 128.79, 131.42, 131.50, 132.09, 132.15, 132.26, 132.34, 138.57, 138.63. 31 P NMR (162 MHz, DMSO‑d6): 28.46. HR-MS: 389.1210 and calculated 389.1200. Complex of LuL2 was synthesized by adding a 10 mL of methanolic solution of LuCl3·6H2O (0.25 g, 0.64 mmol) slowly to the 10 mL methanolic solution of L (0.5 g, 1.28 mmol). Subsequently, the mixture was further stirred at room temperature for an overnight. The solvent was removed under reduced pressure and washed several times with diethyl ether. The colour of the complex is yellow with a yield of 95%. The complex was characterized through various spectroscopic and analytical techniques including; 1H, 13C, 31P NMR FTIR and ESI-mass (for spectral details, see Figs. 6S–10S). FTIR (ATR, cm−1): 3140 (OH, H2O), 3055 (Ar-CH), 2964 (alk-CH2), 1637, 1614 (C_N), 1562 (C_C),
33
1435 (CH2 bending), 1143 (P_O). 1H NMR (400 MHz, DMSO-d6): 2.29 (s, 3H), 6.11 (d, 1H), 7.34 (d, 2H), 7.40 (s, 3H), 7.46 (m, 2H), 7.65 (s, 3H), 7.75–7.78 (m, 4H). 13C NMR (100 MHz, DMSO-d6): 161.12, 138.63, 136.42, 132.37, 132.29, 131.56, 131.48, 130.45, 128.81, 128.70, 127.68, 119.27, 115.37, 65.55, 6470, 20.82. 31P NMR (162 MHz, DMSO-d6): 32.88. ESI-TOF mass: 950.9247, calculated mass: 951.7302. 3. Results and discussion The synthesis of desired ligand L was accomplished in high yield by reacting 6-methyl-2-oxo-1,2-dihydroquinoline-3-carbaldehyde with diphenyl phosphine oxide (Scheme 1). The compound was systematically characterized by 1H NMR, 13C NMR, 31P NMR, FT-IR and HR-mass analysis. 3.1. Absorption and emission studies The ligand L contains several oxygen donor atoms such as phosphoryl, hydroxyl, and carbonyl groups [28], which can participate in coordinate bonding with metal ions while the methyl quinolinone unit can provides analytically useful optical response. The sensing ability of L (50 μM) was investigated in the presence of various metal ions such as Na+, Cs+, Mg2+, Ca2+, Mn2+, Fe3+, Pb2+, Cd2+, Hg2+, Zn2+, Cr3+, Co2+, Cu2+, Al3+, La3+, Ce3+, Ce4+, Pr3+, Nd3+, Sm3+, Eu3+, Gd3+, Tb3+, Dy3+, Ho3+, Er3+, Tm3+, Yb3+, Lu3+, Th4+, and UO22+ (50 μM) by recording the UV–visible spectra in CH3OH-H2O (1:1, v/v). The neat ligand L exhibits two electronic bands appeared at 280 nm and 340 nm with a molar absorptivity of 9740 L·mol−1 cm−1 and 7380 L·mol−1 cm−1 respectively due to the π-π* transitions from diphenylphosphoryl and methylquinolinone. As seen in Fig. 1a, there is no significant change in the position of peak observed at 280 nm when interacted with alkali, alkane earth and transition metal ions. Upon addition of Lu 3+ to L, the absorption peak at 340 nm did not change while peak at 280 nm experience weak blue shift with hyperchromism and molar absorptivity increased from 9740 L·mol −1 cm −1 to 13,100 L·mol −1 cm −1. However, for other lanthanides ions like Sm 3+, Pr 3+ , Dy3+, Er 3+, Nd 3+ and Ho3+ on interacting with L, the wavelength maximum (λ max ) experience red shift to 290 nm with hypochromic effect. These studies revealed that using L, quantitative estimation of Lu3+ by this method is difficult due to serious interference from other lanthanides. Under similar experimental conditions, the cation sensing ability of L (50 μM) was next examined by fluorescence spectroscopy. The neat ligand L exhibits an emission band centred at 400 nm on excitation with 280 nm. The fluorescence of L was enhanced significantly and slightly blue-shifted to 394 nm in the presence of Lu3+. Also, under UV-light irradiation, the change in fluorescence property of L in the presence of Lu3+ can be seen by the naked-eye while other tested metal ions (Fig. 1b) and anions (Fig. 11S) did not show any fluorescence behaviour under similar experimental conditions. The enhancement of emission at 400 nm of L along with slight blue-shift to 394 nm upon addition of Lu3+ can be attributed to the selective complexation between the ligand L and Lu3+. The complexation increases the rigidity of the molecular assembly by restricting the free rotation of the substituents like phosphoryl, hydroxyl and methyl quinolinone moieties and
Scheme 1. Synthesis of L.
34
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
Fig. 1. (a) UV–Vis and (b) fluorescence (λex = 280 nm) spectral changes of L (50 μM) with various metal ions (10 eq) in CH3OH-H2O (1:1, v/v).
promotes the CHEF effect to cause significant enhancement of the emission intensity [29]. Also, Lu3+ cannot significantly change the emission intensity of L because it has no low-energy f-f states for energy transfer [30] and therefore the enhancement is related to the changes in the optical properties of L occurred upon complexation with Lu3+. The effects of solvents ratio between water and methanol on the fluorescence of L and LuL2 were examined at different methanol water ratio. Results show that both L and LuL2 exhibits similar trend (Figs. 12S and 13S), with increase in water content (in methanol: water mixture) resulting in the decrease in the emission intensity due to molecular rigidity attainment by molecular packing modes and conformations in the aggregates states which is confirmed by SEM analyses (Fig. 2). Further, SEM image supports the formation of 1D micro rod; on the surface of these rods exhibits rough surface texture. In the case of LuL2, the emission behaviour follows similar to L but the emission intensity increased by two-fold compare to L. This increase in emission intensity is due to the formation of stable complex with Lu3+; thus all the molecular rotations of LuL2 are arrested. Besides, there is a formation of aggregates which is responsible for the enhanced emission [31]. Further, the formation of LuL2 aggregates (288 nm) was confirmed in dynamic light scattering (DLS) study (Fig. 14S). Therefore, the restriction of intramolecular rotation of the diphenyl phosphoryl and methyl quinolinone moieties of the ligand L in the LuL2 complex
is the main reason for the enhancement of an emission intensity. Furthermore, the intramolecular rotations of the fluorophores are restricted upon aggregation prevent the non-radiative pathway, resulting in the enhancement of emission. Fluorescence titration was carried out using 50 μM of L by adding different amounts of Lu3+ from 0 to 30 μM in MeOH/water (1:1, v/v). Results shows a steady increase in fluorescence intensity with respect to Lu3+ (Fig. 3a). The binding stoichiometry was verified by Job's plot method [32]. The maximum emission intensity was achieved at 0.66 mol ratio of L, which confirms the formation 2:1 ligand to metal complex (Fig. 3b). Table 1 shows the resulting association constants of several fluorescence titrations used for binding capability of L towards Lns. All binding data were fit to a 1:1 and 2:1 binding model and association constants were determined using BindFit v0.5 program [33]. As shown in Table 1, L exhibit a more binding preference to Lu3+ in 1:1 and 2:1 complex formation compare to other Lns. The binding capacity is almost three times more with Lu3+ compare to other Lns. This increase in binding constant may be due to their characteristic coordination geometries with Lu3+. The lowest detection limit (LOD), 3σ/slope, where slope is obtained from calibration curve and σ standard deviation of emission intensity of L [34] was determined from the fluorescence titration data, and the results shows that Lu3+ can be detected down to 24.8 nM (Fig. 3a).
Fig. 2. SEM image of L and LuL2 aggregates prepared in 1:1 methanol and water.
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
35
Fig. 3. (a) Fluorescence spectral change of L with Lu3+ in MeOH/H2O (λex = 280 nm, Insert figure represent calibration plot) (b) Jobs plot.
Fluorescence quantum yields of L and LuL2 were calculated in CH3OH by employing comparative William's method using anthracene in ethanol as a standard [35,36]. The estimated fluorescence quantum yield (φ) of L and LuL2 were found to be 0.063 and 0.115, respectively. In order to achieve a deeper insight into the photo-physical properties of L and LuL2 using time-resolved fluorescence decay measurement was performed in methanol medium (Fig. 4). The free ligand L shows a lifetime of 1.17 ns whereas LuL2 shows a lifetime of 4.66 ns. The rate constant for the radiative (Kr) and nonradiative (Knr) decay process were estimated from the fluorescence quantum yield and lifetime [37]. Accordingly, the calculated values of Kr and Knr for L are 1 × 107 s−1 and 8.45 × 108 s−1 respectively whereas for LuL2 is 1.72 × 107 s−1 and 1.97 × 108 s−1 respectively. Results show that L is exhibiting a weak fluorescence as Kr is less than Knr. However, LuL2 exhibits increase in Kr and decrease in Knr as compare to L. Hence, this phenomenon is also supporting the strong fluorescence enhancement of LuL2. The influence of pH on fluorescence response of L and LuL2 was carried out between pH 2.0 to 10.0 using appropriate buffer solutions (Fig. 5a). The fluorescence intensity of L alone did not get altered from 3.5 to 10.0 but the intensity decreases below pH 3.5 due to the protonation of hydroxyl functional groups. In the case of complex LuL2, the intensity remains same in the pH range between pH 4.0 to 9.0 and whereas the intensity decreases above pH 9.0 and below pH 4.0. The decrease in intensity in acidic region (2.0–4.0) due to protonation nitrogen atom, which restricts the complexation and at higher pH (˃9) due to the formation of Lu(OH)3 [38].
The specificity of ligand L as a fluorogenic sensor for the detection of Lu3+ ions in the presence of various competing cations was explored in CH3OH/H2O (1:1, v/v) (Fig. 5b). For competitive studies, the ligand L (50 μM) was treated with 0.5 equivalent of Lu3+ and 10 equivalents of interfering metal ions then each solution was excited at 280 nm and the emission intensity was measured at 400 nm. The obtained results indicate that no interference in the detection of Lu3+ ions in the presence of interfering metal ions. Further, the performance of L compared with other previously reported sensors shown in Table 1S, It concludes that L is exhibit highest sensitivity (24.2 nM) and no interference from other metal ions [19,20]. Therefore, the present sensor can be applied for the determination of trace amount of Lu3+ ions in actual samples. To check the dynamic response of the LuL2 interactions, the binding reversibility of LuL2 complex in CH3OH/H2O (1:1 v/v) media was performed. Due to the high stability constant of EDTA–Lu3+ complex, the binding reversibility between LuL2 and EDTA was observed (Fig. 15S). The purpose of this test is to show how quickly L can be recoverable from its complex. 3.2. Thermal stability of L and LuL2 In order to study the thermal stability of L and LuL2, thermal analysis was carried out using TGA and DSC. The TGA analysis of L indicates three
Table 1 Association constants, Ka (M−1), of L towards Lns in CH3OH-H2O (1:1, v/v) solution at 298 K calculated using BindFit v0.5. Ln(III)
logKf(1:1)
logKf(2:1)
La3+ Ce3+ Pr3+ Nd3+ Sm3+ Eu3+ Gd3+ Tb3+ Dy3+ Ho3+ Er3+ Tm3+ Yb3+ Lu3+
1.18 1.14 1.96 1.77 1.84 1.26 1.77 1.25 1.47 1.25 1.45 1.15 1.08 3.14
2.64 2.85 2.63 2.97 2.55 2.62 2.63 2.66 2.72 2.66 2.68 2.44 2.44 5.71
Fig. 4. Time resolved fluorescence decay of L and LuL2 in CH3OH by exciting at 295 nm.
36
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
Fig. 5. (a) Effect of pH on the fluorescence of L and LuL2 and (b) Bar diagram of LuL2 fluorescence in the presence of various competing metal ions.
steps of weight loss (Fig. 16S). The first weight change corresponding to loss of water molecules observed at 100 °C. The second weight loss was from 240 °C to 500 °C and may be due to the removal of aromatic units and the final weight loss was due to the conversion to final product to P2O5 residue. Similarly, the LuL2 complex also shows three steps weight loss. The first weight loss occurred at about 140 °C corresponding to the removal of water molecule present in the system. The second weight loss between 150 and 600 °C is corresponding to the removal of organic molecules and the final weight loss can be attributed to the decomposition to LuPO4 and P2O5 [39]. On the other hand DSC analysis reveals that the ligand L shows (Fig. 17S) an endothermic signal at 236 °C with the transition enthalpy of 19.95 J/kg, corresponding to the melting point and relative transition energy involves in the process. The DSC analysis indicates two endothermic signals at 77 °C and 210 °C, which also confirms the presence of water molecule in the complex which melts at 77 °C with transition enthalpy of 116.1 J/kg; the LuL2 complex starts to melt at 210 °C with transition enthalpy of 177.1 J/kg. The comparison of L with the LuL2 complex shows lower melting point with higher transition enthalpy [40]. 3.3. Binding mechanism In order to support the binding mechanism between L and Lu3+, we have studied FTIR, 1H NMR, 13C NMR, 31P NMR and ESI mass analysis. As seen from Fig. 6a, the FT-IR of free ligand L shows vibrational bands at 3224 cm−1 assigned to vibrations of \\NH and \\OH stretch,
1645 cm−1 assigned to C_O stretch and 1180 cm−1 assigned to P_O stretch. The vibrational bands at 694 cm−1 and 515 cm−1 are due to the bending vibrations of P_O group. After complexing with Lu3+, a broad signal appeared at 3140 cm−1 and it corresponds to\\OH stretch of water molecules which is present in the complex. The carbonyl group splits into two sharp ends and appeared at 1637 cm−1 and 1614 cm−1; these are due to non-symmetrical vibrations in LuL2. The P_O symmetrical vibrational band shift to 1140 cm−1 and bending vibrational bands are also shifted to 690 cm−1 and 510 cm−1. The shifting of bands to lower wavenumber indicates the formation of LuL2 complex. The formation of LuL2 complex was further confirmed by 1H, 13C and 31 P NMR analysis in DMSO‑d6. In case of 1H NMR, L alone exhibits characteristics peak at 11.76 ppm, 6.48 ppm and 6.0 ppm corresponding to\\NH, \\OH and \\CH (chiral) protons, respectively (Fig. 6b). Upon adding Lu3+ from 0.1 to 0.5 eq to L, the \\NH peak experience a deshielding effect due to the involvement ˃C_O group in bonding formation with Lu3+ thereby the electron density present at hydrogen (\\NH) is attracted by ˃C_O group. The \\OH peak appeared at 6.48 ppm begins to decrease in its intensity as Lu3+ contents increased and at 0.5 eq, the\\OH peak completely disappeared due to the formation of 2:1 stoichiometry complexation. The hydrogen attached to chiral carbon is suffering from weak shielding effect due to the formation LuL2. In case of 13C, the peak appeared at 161.57 ppm corresponds to carbonyl carbon, upon complexation with Lu3+ the carbonyl carbon peak undergoes weak deshielding effect (161.08 ppm) confirming involvement of ˃C_O group in the complexation (Fig. 8S). In the case
Fig. 6. (a) FT-IR spectra for L and LuL2 and (b) The 1H NMR titration of L upon incremental addition of Lu3+.
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
of 31P NMR, the free P_O group at 28.41 ppm after complexing with Lu3 + , the P_O group resonates at 29.65 ppm with significant peak broadening which indicates the P_O group involved in the complexation (Fig. 18S). Further, the ESI-TOF mass analysis of the complex exhibits two major peaks (m/z) appeared at 950.9247 and 1013.8861 (Fig. 10S). The mass 950.9247 corresponds to the involvement of 2L and 1Lu3+ which is agreeing with the calculated mass of 951.1604 while the mass corresponding to 1013.8961 is due to the involvement of 2L, 1Lu and 2 methanol. Both the mass peaks confirm the formation of complex in 2:1 binding stoichiometry between L and Lu3+. 3.4. Theoretical studies The possible 3D structure of the receptor L and LuL2 complex was proposed through quantum mechanical calculations by applying the B3LYP exchange-correlation functional and the basis sets 6-31G** for the C, H, N, O, P atoms whereas SDD for the Lu atom. The optimized structure of L shown in Fig. 7a preferred an enol-form, where the carbonyl group of quinolone-O and hydroxyl group are oriented in the same direction and form an intramolecular hydrogen bond of length 1.558 Å. Also, the computed L structure indicates that the complexation of Lu3+ with L through the oxygen atoms of phosphine oxide, hydroxyl and quinolone-O need apparent conformational changes in the receptor. Considering the experimental evidences, different possible coordination modes were calculated for the 1:2 binding ratio between the Lu and L, where the receptor L was considered either as bidentate or a tridentate ligand (Fig. 7b–d). When the bidentate mode was considered for the LuL2 complex, the optimization resulted distorted tetrahedral shaped complex and the interaction energy (Eint = ELuL2-2EL-ELu) was decreased by −209.58 kcal/mol and −220.74 kcal/mol for the binding mode-I and mode-II, respectively. The calculated Eint indicates the formation of the stable LuL2 complex where the participation of hydroxyl and quinoline-O group (i.e. binding mode-I) is more favourable due to
37
the conformational adjustment required in mode-II. When the ligand was considered as a tridentate ligand, the distorted octahedral shaped binding mode-III was obtained for the LuL2 complex (Fig. 7d). In this binding mode-III, the Eint was estimated as −247.38 kcal/mol, which is significantly lowered in compared to the mode-I and mode-II where the L acts as a bidentate ligand. The significant lowering of interaction energy in compared to modes-I/II indicate the formation of the more stable complex may be due to the higher coordination number preferred by Lu atom and preferential binding with hard donor atoms. Similar studies were conducted with selected Ln's such as La3+, Ce3+, Pm3+, Gd3+. Er3+ and the results are depicted in Fig. 19S. On comparison of binding energy data, LuL2 complex has lowest possible energy and it confirms L is having high selectivity with Lu3+. In order to obtain insights into the electronic excitation of L in the presence and absence of Lu3+, Time dependent density functional theory (TD-DFT) was carried out. The frontier molecular orbital involved in the UV–vis absorption of L and LuL2 shown in Fig. 20S and corresponding excitation energy and oscillation strength were tabulated in Table 2S. The theoretical absorption spectra are well match with experimental data. HOMO and LUMO of L are mainly located in the πelectrons of the quinolone ring, which indicates the π-π* transitions are mainly contributed for the absorption spectra. Whereas in LuL2 the HOMO orbital are mainly located on P_O group and LUMO orbitals are located on Lu atom. The energies of HOMO and LUMO of L and LuL2 were shown in Fig. 8 (a) and (b), the free ligand L exhibits the band gap of 4.08 eV and LuL2 complex exhibits the band gap of 0.56 eV. When Lu3+ was added, it showed a significant change in the distribution of π-electrons on quinolone ring system. More significantly, the HOMO and LUMO densities in free ligand L with Lu3+ were delocalized over the entire molecule except for extended the conjugation throughout both side quinolone rings which results the enhancing in the fluorescence emission with significantly higher energy than that of L.
Fig. 7. DFT computed structure of L. (a) Various possible complexes of L with Lu3+ (b–d).
38
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39
Fig. 8. Frontier molecular orbital for L (a) and LuL2 (b) by DFT/B3LYP method.
3.5. Applications Analytical applications of L were tested for the quantification of Lu3+ in three different real and spiked samples such as demineralised water, tap water and river water. The test for each sample was carried out in three measurements and the results show that there is a good match with standard addition and recovery of the Lu3+ ions. Also, the novelty and analytical parameters of sensor L was compared with the reported Lu3+ sensors (Table 3S). This sensor does not suffer from any interference from any of the selected metal ions. Hence, L can apply as a florigenic sensor for the estimation of Lu3+ in actual samples containing Lu3+ contents. 4. Conclusion In summary, we have developed an easy-to-synthesize phosphoryl quinolone-based ligand (L) for the selective detection of Lu3+ by the spectrofluorometric method. The 2:1 binding stoichiometry for the LuL2 complex was proposed by Job's plot and mass analysis. The coordination modes of L was established by FT-IR and NMR data, followed by the 3D structure of L and its complex with Lu3+ was proposed by DFT calculations. L exhibits highly selective CHEF with Lu3+. Based on the spectral data, the LOD for the analysis of Lu3+ was calculated down to be 24.2 nM and the formation of LuL2 complex was reversible with EDTA. The sensor L could work in wide pH range from 4.0 to 9.0. The performance of L was successfully used for the analysis of Lu3+ contents present in different water samples. Acknowledgements Authors thank Board of Research in Nuclear Sciences (BRNS), GOI for supporting the work through the project grant (34/14/11/2015/BRNS).
Authors are also thankful to DST-VIT-FIST for NMR and Vellore Institute of Technology, Vellore for other research facilities. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.saa.2019.02.003.
References [1] S. Kantevari, M. Matsuzaki, Y. Kanemoto, H. Kasai, G.C.R. Ellis-Davies, Two-color, two-photon uncaging of glutamate and GABA, Nat. Methods 7 (2010) 123–125. [2] G. Jones, W.R. Jackson, C.Y. Choi, W.R. Bergmark, Solvent effects on emission yield and lifetime for coumarin laser dyes. Requirements for a rotatory decay mechanism, J. Phys. Chem. 89 (1985) 294–300. [3] S.K. Asthana, A. Kumar, S.K. Hira, P.P. Manna, K.K. Upadhyay, Brightening quinolineimines by Al3+ and subsequent quenching by PPi/PA in aqueous medium: synthesis, crystal structures, binding behavior, theoretical and cell imaging studies, Inorg. Chem. 56 (2017) 3315–3323. [4] K. Zhang, Z.Y. Yang, B.D. Wang, S.B. Sun, Y.D. Li, T.R. Li, Z.C. Liu, J.M. An, A highly selective chemosensor for Al3+ based on 2-oxo-quinoline-3-carbaldehyde Schiff-base, Spectrochim. Acta A Mol. Biomol. Spectrosc. 124 (2014) 59–63. [5] Y. Laras, V. Hugues, Y. Chandrasekaran, M. Blanchard-Desce, F.C. Acher, N. Pietrancosta, Synthesis of quinoline dicarboxylic esters as biocompatible fluorescent tags, J. Org. Chem. 77 (2012) 8294–8302. [6] C. McDonagh, C.S. Burke, B.D. MacCraith, Optical chemical sensors, Chem. Rev. 108 (2008) 400–422. [7] M. Dutta, D. Das, Recent developments in fluorescent sensors for trace-level determination of toxic-metal ions, TrAC Trends Anal. Chem. 32 (2012) 113–132. [8] Z. Xu, J. Yoon, D.R. Spring, Fluorescent chemosensors for Zn2+, Chem. Soc. Rev. 39 (2010) 1996–2006. [9] L. Feng, W. Shi, J. Ma, Y. Chen, F. Kui, Y. Hui, Z. Xie, A novel thiosemicarbazone Schiff base derivative with aggregation-induced emission enhancement characteristics and its application in Hg2+ detection, Sensors Actuators B Chem. 237 (2016) 563–569. [10] M.R. Ganjali, V.K. Gupta, F. Faridbod, P. Norouzi, Lanthanides Series Determination by Various Analytical Methods, Elsevier, 2016.
R.S. Kumar et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 32–39 [11] C. Bariáin, I.R. Matías, C. Fernández-Valdivielso, F.J. Arregui, M.L. Rodríguez-Méndez, J.A. De Saja, Optical fiber sensor based on lutetium bisphthalocyanine for the detection of gases using standard telecommunication wavelengths, Sensors Actuators B Chem. 93 (2003) 153–158. [12] D.L. Massart, J. Hoste, Activation analysis of rare earths: part II. Determination of lutetium in gadolinite, Anal. Chim. Acta 42 (1968) 15–20. [13] V.N. Azarova, A.I. Kirillov, Spectrophotometric determination of lanthanum, lutetium, and yttrium with triphenylmethane reagents and cetylpyridinium bromide, J. Anal. Chem. USSR 41 (1986) 1408–1412. [14] T.B. Krustev, S.T. Mincheva, D.A. Angelov, E.P. Vidolova-Angelova, Determination of traces of lutetium in geological samples by resonance ionization spectroscopy, J. Anal. At. Spectrom. 8 (1993) 1029–1031. [15] C. Peng, W. Li, P. Yuan, W. Feng, X. Wang, Determination of rare-earth impurities in ultra-pure thulium oxide, ytterbium oxide, lutetium oxide by extraction chromatography-atomic emission spectrometry, Chin. J. Anal. Chem. 25 (1997) 377–381. [16] J.C. Van Loon, J.H. Galbraith, H.M. Aarden, The determination of yttrium, europium, terbium, dysprosium, holmium, erbium, thulium, ytterbium and lutetium in minerals by atomic-absorption spectrophotometry, Analyst 96 (1971) 47–50. [17] C. Karadaş, D. Kara, Preconcentration of rare earth elements using amberlite XAD-4 modified with 2, 6-pyridinedicarboxaldehyde and their determination by inductively coupled plasma optical emission spectrometry, Water Air Soil Pollut. 225 (2014) 1972–1982. [18] F.G. Sanchez, M. Gracia, L. Hernandez, J.C. Marquez, A graphical derivative approach to the photometric determination of lutetium and praseodymium in mixtures, Talanta 34 (1987) 639–644. [19] M. Hosseini, M.R. Ganjali, F. Aboufazeli, F. Faridbod, H. Goldooz, A. Badiei, P. Norouzi, A selective fluorescent bulk sensor for lutetium based on hexagonal mesoporous structures, Sensors Actuators B Chem. 184 (2013) 93–99. [20] M. Hosseini, M.R. Ganjali, Z. Rafiei-Sarmazdeh, F. Faridbod, H. Goldooz, A. Badiei, P. Nourozi, G.A. Mohammadi Ziarani, A novel Lu3+ fluorescent nano-chemosensor using new functionalized mesoporous structures, Anal. Chim. Acta 771 (2013) 95–101. [21] F. Faridbod, M. Sedaghat, M. Hosseini, M.R. Ganjali, M. Khoobi, A. Shafiee, P. Norouzi, Turn-on fluorescent chemosensor for determination of lutetium ion, Spectrochim. Acta A Mol. Biomol. Spectrosc. 137 (2015) 1231–1234. [22] L. Wang, T. Jang, N. Sui, X. Yang, L. Zhang, X. Yao, Q. Zhou, H. Xiao, S. Kuang, W.Y. William, A highly selective Schiff base fluorescent chemosensor for Lu3+, Anal. Methods 44 (2017) 6254–6260. [23] M.J. Frisch, et al., Gaussian 09W, Revision A, Gaussian Inc., Wallingford, CT, 2009. [24] Y. Luo, J. Baldamus, O. Tardif, Z. Hou, DFT study of the tetranuclear lutetium and yttrium polyhydride cluster complexes [(C5Me4SiMe3) 4Ln4H8](Ln = Lu, Y) that contain a four-coordinate hydrogen atom, Organometallics 24 (2005) 4362–4366.
39
[25] C. Adamo, D. Jacquemin, The calculations of excited-state properties with timedependent density functional theory, Chem. Soc. Rev. 42 (2013) 845–856. [26] M.K. Singh, A. Chandra, B. Singh, R.M. Singh, Synthesis of diastereomeric 2, 4disubstituted pyrano [2, 3-b] quinolines from 3-formyl-2-quinolones through O–C bond formation via intramolecular electrophilic cyclization, Tetrahedron Lett. 48 (2007) 5987–5990. [27] D. Das, G. Gopakumar, C.V.S. Brahmmananda Rao, N. Sivaraman, A. Sivaramakrishna, K. Vijayakrishna, Extraction and coordination behavior of diphenyl hydrogen phosphine oxide towards actinides, J. Coord. Chem. 70 (2017) 3338–3352. [28] R.G. Pearson, Hard and soft acids and bases, J. Am. Chem. Soc. 85 (1963) 3533–3539. [29] A.W. Czarnik, Fluorescent chemosensors for ion and molecule recognition, ACS Symp. Ser. (1993) 538. [30] W.S. Xia, R.H. Schmehl, C.J. Li, A fluorescent 18-crown-6 based luminescence sensor for lanthanide ions, Tetrahedron 56 (2000) 7045–7049. [31] Y. Hong, J.W. Lam, B.Z. Tang, Aggregation-induced emission, Chem. Soc. Rev. 40 (2011) 5361–5388. [32] P. Job, Formation and stability of inorganic complexes in solution, Ann. Chim. Appl. 9 (1928) 113–203. [33] P. Thordarson, Determining association constants from titration experiments in supramolecular chemistry, Chem. Soc. Rev. 40 (2011) 1305–1323. [34] G.L. Long, J.D. Winefordner, Limit of detection. A closer look at the IUPAC definition, Anal. Chem. 55 (1983) 712A–724A. [35] A.T.R. Williams, S.A. Winfield, J.N. Miller, Relative fluorescence quantum yields using a computer-controlled luminescence spectrometer, Analyst 108 (1983) 1067–1071. [36] W.H. Melhuish, Quantum efficiencies of fluorescence of organic substances: effect of solvent and concentration of the fluorescent solute1, J. Phys. Chem. 65 (1961) 229–235. [37] R. Selva kumar, S.K. Ashok Kumar, K. Vijayakrishna, A. Sivaramakrishna, P. Paira, C.V.S. Brahmmananda Rao, N. Sivaraman, S.K. Sahoo, Bipyridine bisphosphonatebased fluorescent optical sensor and optode for selective detection of Zn2+ ions and its applications, New J. Chem. 42 (2018) 8494–8502. [38] H. López-González, M. Jiménez-Reyes, M. Solache-Ríos, A. Rojas-Hernández, Solubility and hydrolysis of lutetium at different [Lu3+]initial, J. Radioanal. Nucl. Chem. 274 (2007) 103–108. [39] K.S. Gavrichev, N.N. Smirnova, V.M. Gurevich, V.P. Danilov, A.V. Tyurin, M.A. Ryumin, L.N. Komissarova, Heat capacity and thermodynamic functions of LuPO4 in the range 0–320 K, Thermochim. Acta 448 (2006) 63–65. [40] P. Gabbott (Ed.), Principles and Applications of Thermal Analysis, John Wiley & Sons, 2008.