A truncated octahedron NaCe(MoO₄)₂ nanostructure: a potential material for blue emission and acetone sensing

Abstract

Scheelite-type NaCe(MoO₄)₂, a promising nanostructure for optoelectronic applications, has been synthesized using a typical hydrothermal technique, and its structural and microstructural properties have been characterized using several microscopic and spectroscopic techniques. We observed intense blue emission from the 5d–4f transitions of Ce³⁺ within CeO₈, while the presence of mono-(CeO₇) and divacancies (CeO₆) of oxygen cause relatively weak emissions at 488 nm, 462 nm and 531 nm. Our study reveals that the use of trisodium citrate during synthesis plays a significant role to tune the formation of CeO₇ and CeO₆, which in consequence also modify the band edges of the system. The Commission Internationale de l'Eclairage (CIE) coordinates are found to be within the blue region with a correlated color temperature (CCT) of ∼7854 K, indicating the potential of NaCe(MoO₄)₂ nanostructures for cold solid-state lighting applications. Moreover, the oxygen deficiencies are found to act as active sites for the selective adsorption of acetone, leading to acetone sensing. Hence, NaCe(MoO₄)₂ nanostructures may be potential smart materials for blue-lighting and acetone sensors. Ab initio calculations have been carried out to obtain theoretical insights into the electronic structure of the bare and oxygen-deficient NaCe(MoO₄)₂ and to understand the guiding parameters for acetone sensing. Our calculations also demonstrate that acetone adsorption occurs mostly through the (112) plane.

---

1. Introduction

Recently, rare-earth (RE)-based scheelite micro/nanostructures have attracted considerable interest from researchers for various optoelectronic applications, including solid-state lasers, fluorescent lamps, optical sensors, scintillators, light-emitting diodes, fiber optic communications, and biological labeling due to their unique features of tuneable emission, narrow bandgap, high quantum yield and high opto-electric conversion, and excellent thermal and chemical stability.^1–3 Among them are the alkali RE binary molybdates with the generic formula A^ILn^III(MoO₄)₂ (A = alkali metal, Ln = RE lanthanide), in which the Ln³⁺ embedded in the A⁺ lattice exhibits a strong 5d–4f transition depending on the symmetry of the unit cell.^4,5 Extensive research work has been carried out by various researchers, including us, to understand the parameters influencing the emission. As an example, Cheng et al. have investigated the influence of Li/Ag on the luminescence properties of Li_1−xAg_xLu(MoO₄)₂:Eu³⁺,⁶ while Song et al. have explored red emission from Eu³⁺-doped AgGd(MoO₄)₂.⁷ The strong integrated emission and higher color purity observed for LiEu(MoO₄)₂ among the AEu(MoO₄)₂ materials (A = Li⁺, Na⁺ and K⁺) have been ascribed to its higher, faster charge transfer due to the shortened and more covalent Li–O bond in comparison with those of its Na or K counterpart.⁸ Most of the conventional techniques (e.g., the Czochralski method, solid-state reaction, etc.) lead to agglomeration with high grain size and irregular morphology;⁹ however, recent studies reveal a remarkable correlation between the morphology and emission as the selection rules of optical transition significantly depend on shape, size, exposure facets, dimensionality, hierarchy, etc. To date, abundant efforts have been devoted to generating various scheelite micro/nanostructures using hydrothermal and other techniques to gain an intuitive understanding of optical emissions. As a reference, Liu et al. reported the preparation of uniform tetragonal microspindles and nanoplates using the surfactant ethylenediaminetetraacetic acid (EDTA-2Na),¹⁰ while microrods, microcuboids and microflowers have been obtained via a synthesis technique based on the ligand polyvinylpyrrolidone.¹¹ Zheng et al. synthesized a cotton-shaped 3D flower-like morphology using a trisodium citrate assisted method,¹² whilst uniform spherical nanoparticles have been achieved via a cetyltrimethyl ammonium bromide (CTAB) assisted hydrothermal method.¹³ A triclinic → tetragonal phase transition was observed by our group during the CTAB-assisted hydrothermal synthesis of NaCe(WO₄)₂.^14,15

In this manuscript, we report for the first time a NaCe(MoO₄)₂ truncated octahedral nanostructure with inherent oxygen mono- and divacancy (V_O and 2V_O) defects and describe the role of these defects in its photoluminescence. In addition, V_O and 2V_O often accumulate at the surface of nanostructures and trap electrons that facilitate gas-sensing properties.^16–19 Although optical studies have been reported in the literature, reports on the gas-sensing of scheelite nano/microstructures are very rare. As an example, Lin et al. reported the V_O-induced gas sensing activity of NaBi(MoO₄)₂ nanomaterials,²⁰ while our recent study demonstrated the NH₃-sensing ability of NaCe(WO₄)₂.¹⁴ We herein also identify acetone-sensing ability in NaCe(MoO₄)₂ nanostructures, with V_O and 2V_O playing crucial roles in determining the sensitivity. To the best of our knowledge, this is the first report on the acetone-sensing ability of NaCe(MoO₄)₂. Additionally, as there have been no in-depth theoretical calculations on the electronic structure of NaCe(MoO₄)₂, to achieve a better understanding, we have also calculated the ab initio band structure of NaCe(MoO₄)₂. The calculations reveal that the Ce 5d_z², d_x²–y², and d_yz orbitals result in blue and green emissions in the presence of V_O and 2V_O, respectively. We have demonstrated that the electronic excitation involved the MoO₄ tetrahedra and that emission takes place within the CeO₈ polyhedra, while MoO₄ → CeO₈ charge transfer is mediated through the ν₄(F₂) phonon mode. Overall, it may be inferred that the truncated octahedron NaCe(MoO₄)₂ nanostructures have strong potential in violet emission and sensing applications.

2. Experimental section

2.1 Materials and synthesis

To prepare NaCe(MoO₄)₂ nanomaterials through the typical hydrothermal method, 1.00 mmol (0.434 g) of cerium nitrate [Ce(NO₃)₃·6H₂O, Merck, Germany] and trisodium citrate [Na₃Cit·2H₂O, Merck, Germany] were mixed with 60 ml DI water, followed by stirring for 2 h at room temperature. Another aqueous solution of 2.00 mmol (0.484 g) sodium molybdate [Na₂MoO₄·H₂O, Merck, Germany] in 20 ml DI water was added dropwise to the above solution. After 30 minutes of vigorous stirring, the resulting solution turned a pale yellow colour; the solution was then transferred to a Teflon autoclave for hydrothermal reaction at 180 °C for 24 h. The final product was collected by centrifugation, followed by simultaneous washing with DI water and ethanol, overnight drying at 70 °C and calcination at 800 °C for 5 h in a muffle furnace. The synthesis was repeated using different concentrations of Na₃Cit·2H₂O (0.00, 0.50, 0.78, and 1.00 mmol, i.e., 0.00, 0.147, 0.229, and 0.294 g) to examine its effect on the final product, and the resulting samples were labelled as NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00, respectively. The reaction procedure is presented schematically in Fig. S1 of the ESI.†

2.2 Characterization and gas-sensing properties

X-Ray diffraction (XRD) patterns were obtained using a Rigaku Ultima III powder diffractometer equipped with CuK_α radiation (λ = 1.54056 Å) and utilized to examine the phase purity and other crystallographic information through the Rietveld refinement method using the FullProf program suite. Experimental profiles were fitted with a suitable pseudo-Voigt analytical asymmetric function, while the background was fitted with a fourth-order polynomial function. Field emission scanning electron microscopy (FESEM: Hitachi S – 4800, operated at 5 kV) was used to inspect the morphology of the samples, while a transmission electron microscope (TEM-2100 Plus Electronmicroscope, 200 kV) was used to obtain high-resolution microscopic images and selected area electron diffraction (SAED). Raman spectra obtained using an alpha 300 Witec with a 530 nm laser (power: 3 mW; spot size: 2 μm) and Fourier transform infrared (IR) spectra recorded using an IR Prestige were utilized to examine the short-range structural distortion. X-Ray photoelectron spectra (XPS) were collected using a PHI Versa Probe III Scanning XPS Microprobe with an Al K source. The optical properties of the as-prepared samples were investigated at room temperature using UV-Vis spectroscopy (JASCO V650) and photoluminescence spectroscopy (FP-8300, JASCO, 100 W Xe lamp).

Gas-sensing properties were examined using Taguchi-type sensor module, in which a hollow cylindrical Al₂O₃ substrate having Pt electrodes was coated with a thick slurry of the nanostructured NCMO materials through the drop-casting method. The slurry was prepared by mixing the as-prepared samples with isopropyl alcohol as a binder, and the coated substrates were heated at 100 °C for 12 h to remove residual solvents. Ni–Cr wire was inserted through the hole of the substrate as a heating element to provide the required temperature by adjusting the applied voltage across the wire. Gases supplied from cylinders, balanced with air, were used during measurement of the sensing performance with an Agilent 34461A digital multimeter interfaced with data logger software.

2.3 Calculation of band structure and investigation of gas-sensing mechanism using ab initio density functional theory

It is well-known that the electronic band structure provides theoretical insight into the opto-electronic properties of any material; hence, we computed the spin-polarized band structure, density of states (DOS), partial DOS (pDOS), total DOS (TDOS) and projected DOS (PDOS) using VASP.²¹ Here, the plane-wave pseudo-potential (PAW) was used, while the Perdew–Burke–Ernzerhof (PBE) exchange correlation and ultra-soft potential as basis set were considered.²² We employed the valence electrons of the respective atoms as follows: Na atom (1s² 2s² 2p⁶ 3s¹), Ce atom (5s² 5p⁶ 4f¹ 5d¹ 6s²), Mo atom (4p⁶ 5s² 4d¹⁰), and O atom (2s² 2p⁴). For this calculation, 5 × 5 × 7 Monkhorst–Pack k points were used, as this allowed an accurate representation of the Brillouin zone and the electronic properties of the material. Prior to calculating the band structure along Γ → X → H₁ → C → H → Y → Γ, the structure was optimized with the lowest single-point ground state energy by fixing the cut-off energy at 520 eV, while the convergence was tested between 220 and 620 eV. A maximum atomic displacement of ∼5 × 10⁻⁴ Å and stress of ∼0.02 GPa were considered for this calculation. Each atom was exposed to a 0.01 eV Hellmann–Feynman force to ensure stable convergence, while the EDIFF and force EDIFG parameters were adjusted to 10⁻⁶ eV and 10⁻³ eV, respectively, to ensure high-precision computational convergence.¹⁴

In order obtain theoretical insight into the influence of the defects on the gas-sensing performance, we calculated the adsorption energies of acetone on the NCMO surface as shown in Table S2 (ESI†). Herein, we have modelled the tetragonal NCMO (112) plane with supercell dimension of 7.573 Å × 17.823 Å × 32.422 Å. The k-point mesh for this calculation was optimized at 3 × 3 × 1, while the thickness of the slab and the vacuum level were ∼3 nm and 15 Å, respectively. In these calculations, the bottom half of the layers were frozen, whereas the top half of the layers with acetone as the adsorbent were set to relax. The adsorption energy was then calculated using eqn (1):

E_ads = E_NCMO+acetone − E_NCMO − E_acetone (1)

where E_NCMO+acetone, E_NCMO and E_acetone represent the total ground state energies of NCMO in combination with the acetone molecule, bare NCMO and acetone gas, respectively. In order to examine the influence of defects on the gas-sensing activity, we carried out the above calculations by removing one or two oxygen atoms from the (112) plane of NCMO, as shown in Fig. S2 (ESI†).

3. Results and discussion

3.1 Structural and morphological studies using XRD, FESEM and TEM

The XRD patterns of all the samples (Fig. 1) closely match with that of scheelite-type tetragonal NCMO with the space group I4₁/a (C_4h⁶, ICDD Database No. 04-007-5489). The absence of any peak other than those of NCMO confirms the phase purity of the synthesized nanostructures, while the sharp diffraction peaks indicate good crystallinity. The unit cell is represented in Fig. S3, and Table S1 in the ESI† summarizes the structural parameters as obtained from Rietveld refinement, which was carried out until satisfactory convergence of χ², R_p and R_wp were achieved.²³ As shown in Fig. S3 (ESI†), the unit cell of NCMO comprises Na and Ce atoms at the 4b sites and is coordinated to eight O atoms at 16f site via two different bond lengths to produce (Na/Ce)O₈ polyhedra; in turn, these polyhedrons are connected to MoO₄ tetrahedra with Mo at the 4a site. The unit cell volume, c value, c/a ratio, Na/Ce–O bond lengths, and the Na/Ce–O–Mo and O–Mo–O bond angles (but not the Mo–O bond lengths) decrease monotonically from NCMO_0.00 to NCMO_1.00, indicating an enhancement in the distortion of the unit cells. The (Na/Ce)O₈ polyhedral (N) and MoO₄ tetrahedral (K) distortions were calculated using the expressions ,²⁴ where α_{1(O–Mo–O)} and α_{2(O–Mo–O)} denote the two different bond angles, and , where d_1(Na,Ce–O) and d_2(Na,Ce–O) represent the two different bond lengths and were found to be N = 1.011, 1.029, 1.043, 1.052 and K = 1.0587, 1.0827, 1.0838, 1.0842 for NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00, respectively.^25,26 According to previous studies, a decrease in the c/a ratio is associated with the partial substitution of Na⁺ by Ce³⁺,^27,28 while the perquisite charge neutrality is balanced by O vacancies (V_O), causing increases in K and N.

Fig. 1 XRD patterns of the NaCe(MoO₄)₂ samples (a) NCMO_0.00, (b) NCMO_0.50, (c) NCMO_0.78 and (d) NCMO_1.00, respectively.

It can be noted from the FESEM image that NCMO_0.00 possesses an irregular shape (Fig. 2(a)), while a gradual transformation from an irregular to a truncated octahedral shape was observed for NCMO_0.50, NCMO_0.78 and NCMO_1.00 (Fig. 2(b)–(d)) suggesting that Na₃Cit has significant role during hydrothermal reaction to tune morphology. Mapping of the samples (Fig. S4, ESI†) clearly reveals even distribution of elements. For a better understanding of the shape of these microstructures, transmission electron microscopy (TEM) was further employed. Fig. 2(e) depicts a typical TEM image of NCMO_1.00, which corroborates the truncated octahedral shape, while the selected area diffraction pattern (Fig. 2(f)) consists of five diffraction spots corresponding to the (112), (211), (123), (222) and (224) planes of tetragonal NCMO, denoting its crystalline character.

Fig. 2 FESEM images of NCMO samples (a) NCMO_0.00, (b) NCMO_0.50, (c) NCMO_0.78 and (d) NCMO_1.00, (e) and (f) TEM image and SAED pattern of NCMO_1.00.

3.2 Investigations of local structure using FTIR, Raman and X-ray photoelectron spectroscopies

To understand the influence of V_O on the short-range structural distortions of (Na/Ce)O₈ polyhedra and MoO₄ tetrahedra, we obtained the FTIR and Raman spectra of our synthesized samples. The various peaks in the absorption band (shown in Fig. S5 of the (ESI†)) between 500 and 1000 cm⁻¹ are attributed to the various active internal vibrations of MoO₄ tetrahedra. Commonly, MoO₄ tetrahedra crystallize in either regular (Mo^RO₄) or distorted form (Mo^DO₄) due to V_O, while the vibration (Γ_Td) of the isolated Mo^RO₄ is represented by:²⁹

Γ_Td = A₁ + E + 2F₂ (2)

However, the symmetry of the tetrahedra changes to S₄ in NCMO, for which the IR active modes include ν₁(A₁), ν₂(E₁), ν₃(F₂), ν₄(F₂), one free rotation ν_f.r.(F₁) and one translation F₂. In contrast to Mo^RO₄, Mo^DO₄, which has the C_2h⁶ space group, exhibits four formula units per crystallographic unit cell with 2Mo(C₂), 4Mo(C₁), 4Ce(C₁) and 24O(C₁) symmetries of the atoms, indicating splitting of the vibrational modes (i.e., O → Mo(C₁) → O and O → Mo(C₂) → O) within Mo^DO₄ due to lower site symmetry.³⁰ The two bands at 650–749 and 750–1000 cm⁻¹ are ascribed to ν₃(F₂) and ν₄(F₂), while the peak at 670 cm⁻¹ is assigned to ν₂(E₁). Careful deconvolution (Fig. S6, ESI†) of the ν₃(F₂) and ν₄(F₂) modes yields three peaks (i.e., 693, 705, 722 cm⁻¹ and 813, 858, 909 cm⁻¹). As per the bond length–stretching frequency correlation described by Hardcastle et al., an increase in the bond length decreases the vibrational frequency.³¹ Hence, the vibrational mode with lower energy corresponds to Mo^DO₄, while the high-energy peaks are ascribed to Mo^RO₄. Therefore, the peaks at 722 and 909 cm⁻¹ are attributed to the ν₃(F₂) and ν₄(F₂) modes of Mo^RO₄, respectively, while the peaks at 693/705 cm⁻¹ and 813/858 cm⁻¹ correspond to the ν₃(F₂) and ν₄(F₂) mode of vibration within Mo^DO₄.³⁰ The Mo^DO₄/Mo^RO₄ ratio, which was calculated from the areas under the curve, was found to be 0.8, 1.0, 1.6 and 2.9 for NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00, respectively, indicating an increase in tetrahedral distortion with increasing Na₃Cit concentration, which corroborates the previous structural studies using XRD.

The Raman spectra, which were obtained up to the range of 1000 cm⁻¹ at ambient temperature (Fig. S7, ESI†), consist of a peak at 463 cm⁻¹ and three bands at 64–231, 255–425 and 681–966 cm⁻¹. In general, Raman spectra of scheelite materials at the Γ-point is expressed as:

Γ = 3A_g + 5B_g + 5E_g (3)

where the non-degenerate A_g and B_g modes and the doubly degenerate E_g mode correspond to the various external and internal vibrations of MoO₄ and CeO₈.²⁹ The peak at 64–231 cm⁻¹ corresponds to the rotation and translation of the MoO₄ tetrahedra and Na⁺ and Ce³⁺, respectively. The sharp peak is attributed to the external vibration of CeO₈, which is in good agreement with previous results,³² while the bands are ascribed to the internal bending of the MoO₄ tetrahedra. After thorough examination, the two peaks at 318 and 380 cm⁻¹ were assigned to the symmetric A_g bending and antisymmetric B_g bending vibrational modes of MoO₄, respectively.³³ While the peak at 463 cm⁻¹ is assigned to antisymmetric A_g bending mode of MoO₄. The other two Raman peaks at 823 and 889 cm⁻¹ are assigned to (O → Mo → O) antisymmetric stretching B_g and (O ← Mo → O) symmetric A_g mode of vibration.^29,34,35 The decrease in Raman intensity with increasing Na₃Cit concentration is believed to be related to the increase of V_O and antisite defects, in agreement with the XRD and FTIR studies.^36,37

X-Ray photoelectron spectroscopy (XPS), a well-known technique to investigate the chemical state of elements, was adopted here to examine V_O-associated changes in the valence states of Na, Ce, Mo and O, as they are highly sensitive to the local electronic environment. The binding energy data of all elements were adjusted with respect to the binding energy (∼284.6 eV) of surface-adsorbed atmospheric C 1s.¹⁴ The survey scans for all the synthesized samples are presented in Fig. S8 of the ESI,† while the high-resolution spectra of Mo, Ce, and O are illustrated in Fig. S9–S11 of the ESI.† Two peaks (i.e., 232.7 and 236.0 eV) of Mo can be readily ascribed to the spin–orbit splitting of the 3d_5/2 and 3d_3/2 orbitals of Mo⁶⁺, while careful deconvolution (Fig. S9(a)–(d), ESI†) reveals the presence of two peaks for the two orbitals (i.e., 232.5 and 232.9 eV for 3d_5/2 and 235.7 and 236.2 eV for 3d_3/2), which were assigned as Mo3d_5/2(I), Mo3d_5/2(II) and Mo3d_3/2(I), Mo3d_3/2(II), respectively. Accordingly, we ascribed Mo3d_5/2(II) and Mo3d_3/2(II) to Mo^RO₄, while Mo3d_5/2(I) and Mo3d_3/2(I) corresponded to Mo^DO₄; their difference can be assigned to the V_O-induced changes in electronic repulsion. In brief, electrons are located on O due to its strong electronegativity, which makes the Mo–O bond significantly ionic, while V_O increases the effective charge on Mo. Hence, the electron–electron repulsion on Mo reduces the binding energies of the Mo 3d_5/2 and 3d_3/2 orbitals in Mo^DO₄. In this context, the smaller energy difference between 3d_5/2 and 3d_3/2 in Mo^DO₄ with respect to Mo^RO₄ (i.e., 3.23 and 3.19 eV) again indicates the influence of the local field generated from V_O on spin–orbit splitting. The weighted percentage of Mo^DO₄ calculated from the area under the curve was found to be ∼26, 32, 35 and 42% for NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00, respectively, which agrees well with the previous finding of increasing Mo^DO₄ content. As illustrated in Fig. S10(a)–(d) (ESI†), the two asymmetric peaks observed at 877–891 and 894–912 eV are attributed to the spin–orbit splitting of the 3d_5/2 and 3d_3/2 orbitals of Ce³⁺. Similar to the Mo orbitals, careful deconvolution of the asymmetric d_5/2 peak yields three peaks at 882.0, 884.6, and 886.9 eV, which were designated as Ce3d_5/2(I), Ce3d_5/2(II) and Ce3d_5/2(III), indicating the presence of Ce with three different oxidation states. Ce3d_5/2(III) was assigned to the Ce of the regular CeO₈ polyhedra, while Ce3d_5/2(II) and Ce3d_5/2(I) correspond to Ce belonging to the CeO₈ polyhedra with one or two oxygen vacancies (ca. V_O and 2V_O) and with the nomenclature CeO₇ and CeO₆, respectively. In this context, it may be stated that CeO₇ and CeO₆ represent deformed CeO₈ polyhedra with one or two nearby Mo^DO₄ and Mo^DO₄ subunits (shown schematically in Fig. 3). Due to smaller difference in electronegativity between Ce and O, the Ce–O bond is supposed to be more covalent; therefore, V_O plays the predominant role in terms of this covalence and the binding energy. We observed from the Rietveld analysis that V_O shortens Ce–O bond lengths, which in consequence increases electron–electron repulsion at Ce site; thus, the binding energy of the 3d orbitals is decreased in the distorted polyhedra.³⁸ The analysis showed that the weighted percentage of Ce_a increases (∼19, 23, 37, 41%), while there is a significant decrease in the weighted percentage of Ce_b (∼48, 46, 37, 26%) from NCMO_0.00 to NCMO_1.00. On the other hand, Ce_c remains almost unchanged for all samples (∼30%). Therefore, the analysis indicates greater generation of 2V_O at the cost of V_O; hence, the tendency for V_O cluster formation increases. The XPS of O 1s displays asymmetry, indicating the presence of two different O species. A careful analysis reveals two O 1s peaks (Fig. S11, ESI†) at 530.6 and 532.4 eV, corresponding to lattice O and the O atom of the Mo^DO₄ tetrahedra.³⁹ The increasing ratio of the area under the fitted curves illustrates the monotonic increase in Mo^DO₄ tetrahedra from NCMO_0.00 to NCMO_1.00.

Fig. 3 Schematic diagram of oxygen vacancies of the CeO₈ polyhedral unit.

3.3 Optical properties of the as-prepared NCMO samples using UV-Vis and photoluminescence spectroscopy

The optical band gap (E_g) was calculated to be ∼3.07, 3.05, 3.02 and 2.99 eV for NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00 based on the UV-Vis absorption spectra (shown in Fig. S12, ESI†). This variation can be understood as follows: our band structure calculations (discussed later) reveal that the valence band (VB) is made up of O 2p–Mo 4d hybridization, while the conduction band (CB) comprises Mo 4d, Ce 4f and a minute contribution from O 2p orbitals. The decrease of Mo–O bond length in the presence of V_O and 2V_O reduces the O 2p–Mo 4d overlap, which causes a blue-shift of the VB, and subsequently, E_g is reduced. All the samples exhibit a broad peak at 380 nm in their photoluminescence excitation (PLE) spectra, which were obtained using an emission wavelength λ_em of 531 nm (Fig. S13, ESI†), while the emission spectra recorded in the visible region using an excitation wavelength of λ_ex = 380 nm are depicted in Fig. 4(a)–(d). The PLE peak was assigned to O 2p–Mo 4d charge transfer absorption within MoO₄ tetrahedra.^40,41 The emission spectra have four distinct peaks in the blue (i.e. blue I, blue II, blue III) and green regions, with an intense blue I peak being observed at 440 nm (22 727 cm⁻¹), while relatively weaker blue II, blue III peaks and a prominent green peak are observed at 462 nm (21 645 cm⁻¹), 488 nm (20 492 cm⁻¹) and 531 nm (18 832 cm⁻¹), respectively. The Stokes shift of the excitation and emission spectra were noted to be ∼3588 cm⁻¹, suggesting the absence of reabsorption and interference among the various emissions.⁴² Although understanding the emission spectra of lanthanides is difficult for several reasons, including symmetry, the crystal field surrounding the host crystal, anion polarizability, and covalence of the host crystal, the present emissions are believed to be 5d–4f transitions within Ce^3+ 43 and can be validated as follows. The emission wavelength (λ in nm) for intra-5d–4f transitions can be calculated using eqn (4):^44,45where, Q*, V, ‘n’, E_a and ‘r’ represent the energy of the 5d band edge of a free Ce³⁺ ion (= 50 000 cm⁻¹), the valence of the Ce³⁺, the number of anions in the immediate shell around the Ce³⁺, the electron affinity of the atoms forming the anions (∼2.16 eV) and the difference between the average bond length and radius of Ce³⁺ within the CeO₈ polyhedra (1.03 Å). Very careful calculation yields λ ∼ 449 nm, which corresponds to blue I emission. Previous studies indicate that V_O and 2V_O can significantly modify the 5d–4f transitions of lanthanide atoms, resulting in different emissions. As an example, Sokolenko et al.⁴⁶ assigned green–red emission to oxygen-deficient complexes, while Korzhik et al. ascribed green emission to WO₃ centres.⁴⁷ Considering CeO₇ and CeO₆ to be V_O- and 2V_O-containing distorted CeO₈ polyhedra, we have calculated λ values of 487 and 537 nm for intra-Ce³⁺ 5d–4f transitions using eqn (4), which are in good agreement with the blue III and green emissions (schematically represented in Fig. S14, ESI†).

Fig. 4 Photoluminescence spectra of (a) NCMO_0.00, (b) NCMO_0.50, (c) NCMO_0.78 and (d) NCMO_1.00. (e) Normalized PL intensity ratio of emission wavelengths. (f) CIE chromaticity diagram of the NCMO phosphors.

The shift of the centroids of the d-orbitals (in eV) of Ce³⁺, ε_c(1,3+,Ce³⁺) in CeO₇ and CeO₆ are believed to tune the 5d–4f transition, giving various emissions (see discussion in the ESI†). Careful monitoring of the green emission reveals the presence of a phonon side band (PSB) at 516 nm with a phonon energy of 930 cm⁻¹, which exactly matches the energy of the ν₄(F₂) phonon mode, indicating that MoO₄ → CeO₆ charge transfer is mediated by the ν₄(F₂) phonon mode. In this context, the Huang–Rhys (S) factor, an indicator of electron–phonon coupling, was calculated using eqn (5) and (6):^48,49

where, I_PSB, I_ZP and I_1P are the integrated intensities of the PSB, zero-phonon line and one-phonon line. As this emission includes a single phonon (p ≥ 0 in eqn (6)), and considering the ⁵D₀–²F_5/2 transition to be purely dipolar, we may write I_PSB = I_1P. In addition, 〈1 + m〉 can also be considered to be 1 due to the higher phonon energy at room temperature. The decrease in the S-factor (0.098, 0.077, 0.032 and 0.039 for the four respective samples) is ascribed to increasing distortion in the lattice structure due to V_O and 2V_O.

The Commission International De l'Eclairage (CIE) coordinates (Fig. 4(f)), which were calculated to be (0.180, 0.149), (0.180, 0.147), (0.170, 158) and (0.172, 0.148) for the NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00 samples, are well spread throughout the blue region of the visible spectrum, indicating the color purity (CP) of the NCMO samples.⁵⁰ We also validated the CP using eqn (7):^51,52

where (x, y) denotes the color coordinates of the phosphor; (x_i, y_i) is the 1931 CIE Standard Source's illuminant point, with color coordinates of (0.3101, 0.3162); and (x_d, y_d) is the color coordinates of the dominant wavelength. Outstanding CP values of ∼90–94% were observed for the samples, indicating that NCMO could be a good candidate for blue-light-emitting applications. The correlated colour temperature (CCT) was calculated using McCamy's relation:⁵³

CCT = 449n³ + 3525n² + 6823.3n + 5520.3 (8)

where and (x, y) represent the chromaticity co-ordinates. The CCT values were ∼7819 K, 7630 K, 7715 K and 7854 K for the samples NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00, respectively. The high CCT indicates that NCMO can be used for cold blue lighting.

3.4 Investigations of gas-sensing properties of NCMO nanostructures

The gas-sensing abilities of metal oxide nanostructures involve a solid–gas interfacial reaction mechanism in which the adsorption/desorption of the targeted gas molecules on the surface of the nanostructure leads to significant changes in resistance. A careful literature survey indicates that the following redox reactions between the nanostructure and acetone (CH₃COCH₃), ethanol (C₂H₅OH), ammonia (NH₃), methanol (CH₃OH), formaldehyde (HCHO) take place in the presence of O₂⁻(ads).

CH₃COCH₃(gas) + 8O₂⁻(ads) ↔ 3H₂O(gas) + 3CO₂(gas) + 8e⁻ (9)

CH₃CH₂OH(gas) + O₂⁻(ads) ↔ 2H₂O(gas) + C₂H₂O + 2e⁻ (10)

4NH₃(gas) + 3O₂⁻(ads) ↔ 6H₂O(gas) + 2N₂(gas) + 3e⁻ (11)

CH₃OH(gas) + O₂⁻(ads) ↔ 2H₂O(gas) + CO(gas) + 2e⁻ (12)

HCHO(gas) + O₂⁻(ads) ↔ H₂O(gas) + CO₂(gas) + 2e⁻ (13)

Our previous investigations indicate that V_O and 2V_O defects are often ionized (e.g., , ) and act as electron donors ([CeO₈]′ and [CeO₈]′′), which can be understood from Kröger–Vink notation:

These [CeO₈]′ and [CeO₈]′′ facilitate electron transfer from the host materials to the chemisorbed oxygen molecule to generate O₂⁻(ads), as described by the reaction: O₂(gas) + e⁻ ↔ O₂⁻(ads).^54–57 In order to validate the gas-sensing ability of our synthesized NCMO nanostructures as a proof of concept, we carried out a series of sensing measurements using NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00 as a resistive sensor probe against NH₃, CH₃COCH₃, C₂H₅OH, CH₃OH and HCHO at different operating temperatures (200–450 °C) at intervals of 50 °C. An optimum operating temperature of ∼300 °C was observed for all the samples (shown in Fig. 5(a)), and they are highly selective toward CH₃COCH₃ (Fig. 5(b), 5.0 ppm concentration) in comparison with C₂H₅OH, NH₃, CH₃OH, and HCHO . The sensitivity monotonically increases from NCMO_0.00 to NCMO_1.00, with the highest sensitivity being observed for NCMO_1.00. To understand the role of V_O, 2V_O in CH₃COCH₃ adsorption/desorption, we fitted the response data using the Freundlich adsorption isotherm as given in eqn (16) and (17).

S_g = 1 + aC_g^b (16)

i.e.,

log(S_g − 1) = a + b log(C_g) (17)

where S_g and C_g represent sensitivity and concentration respectively, while ‘a’ and ‘b’ are constants that depend on the electrical charge of the species and stoichiometry of the reactions on the surface of the sensing probe. In the literature, ‘b’ is noted to be dependent on the adsorbed O species, with b ≤ 0.5 for O²⁻ adsorption and ≈1.0 for O⁻ adsorption.⁵⁸ From slope of the linear regressions (Fig. 5(c)) curves obtained using eqn (17), we calculated b values of ∼0.423, 0.279, 0.088, 0.053 for NCMO_1.00, NCMO_0.78, NCMO_0.50 and NCMO_0.00, respectively, suggesting that 2V_O acts as a more-active site for O²⁻ generation, facilitating CH₃COCH₃ adsorption. The dynamic sensing responses ofNCMO_1.00 (Fig. 5(d)) measured at 300 °C demonstrate sensitivity values of ∼1.86, 2.81, 4.95, 6.21 and 6.89 in the presence of 1.0, 5.0, 10.0, 50.0 and 100.0 ppm of CH₃COCH₃, while the response and recovery (at 5.0 ppm of CH₃COCH₃, see Fig. S15(b) of the ESI†) remain the same for six consecutive cycles (error ≤ 2%), indicating the excellent reproducibility and reliability of NCMO_1.00 as a sensor probe in comparison with the other samples (Fig. S15(c–e), ESI†). In addition, the very low response and recovery time of ∼1.5 s/33.8 s (Fig. 5(e), measured at 10.0 ppm) indicate the fast response of NCMO_1.00, while its long-term stability was checked and verified over 90 days at intervals of 10 days (Fig. 5(f)).

Fig. 5 (a) Temperature calibration curves of NCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00, respectively. (b) Selectivity ofNCMO_0.00, NCMO_0.50, NCMO_0.78 and NCMO_1.00 sensors measured at an operating temperature of 300 °C. (c) Linear fitting of the logarithm of sensitivity and logarithm of acetone concentration data of all samples. (d) Dynamic acetone response curves of the NCMO_1.00 sensor to acetone concentrations from 1–100 ppm at 300 °C. (e) Response and recovery time of the NCMO_1.00-based sensor for 1 ppm acetone at 300 °C. (f) Long-term stability of the NCMO_1.00-based sensor in sensing 1 ppm acetone measured over 90 days.

3.5 Density functional theory (DFT) calculations of NCMO

In order to validate the V_O- and 2V_O-associated optical emissions and sensing activity, we calculated the electronic band structure, total density of states (TDOS), and angular momentum projected partial density of states (PDOS) using ab initio density functional theory (DFT), which provides valuable insights to explain the luminescence and gas sensing behaviour. For this, we calculated band structure along several high symmetry k-points within the Brillouin zone. In this calculation, we removed one or two O atoms from the NCMO unit cell to mimic CeO₇ (V_O) and CeO₆ (2V_O). Prior to calculating the electronic band structure, we optimized the unit cell of pure NCMO and obtained lattice parameters a = b = 5.355 and c = 11.611 Å and Na/Ce–O and Mo–O lengths of ∼2.453 and 1.806 Å, respectively. These were in good agreement with the experimental results, proving the accuracy of our calculations, specifically, the choice of exchange correlation and pseudopotential. The top of the valence band maxima (VBM) was set to zero as a reference for other calculations. As shown in Fig. 6(a), NCMO appears to be an indirect bandgap material (E_g ∼ 3.23 eV) with a spin-unpolarized VBM and spin-polarized conduction band minima (CBM) at Γ- and X-points, respectively. The curvature of the VBM is observed to be low, indicating a comparatively high effective hole mass, while the CBM has a high curvature suggesting a low effective electron mass. The TDOS (shown in Fig. 6(b)) and PDOS contributions of O, Mo, Ce and Na (shown in Fig. S16(c)–(f), ESI†) demonstrate that the upper part of the VB is primarily composed of the O 2p_x, 2p_y, and 2p_z orbitals with very small contributions from Mo 2p_x, 2p_y, and 2p_z, giving a low curvature. The lower part of the VB originates from predominant hybridization between Mo 4d_xy, 4d_xz, 4d_z² and O 2p_x, 2p_y, and 2p_z orbitals, whereas 4d_x² and 4d_yz make insignificant contributions. We noted almost no contribution to the VB from the Ce 4f and Na 3s orbitals. The CBM comprises spin-polarized (i.e., up-spin) Ce 4f_z³, 4f_xz² and 4f_y³x² orbitals, among which the most predominant contribution comes from 4f_xz², hybridized with O 2p_x, 2p_y, and 2p_z orbitals, giving a higher curvature in comparison with the VBM, while the upper portion of the CB is made up of Mo–4d_z² and 4d_x². Thus, it may be stated that the MoO₄ tetrahedra mostly construct the VB, while the CB is predominantly contributed by the CeO₈ polyhedra, and the band-to-band transition includes a Mo ↔ Ce charge transfer process.

Fig. 6 (a) Band structure and (b) spin-polarised TDOS of NCMO.

The band structures, TDOS and PDOS of the - and -containing NCMO samples, designated as and , are represented in Fig. S17 and S18 (ESI†), respectively. Notably, both the VBM and CBM (shown in Fig. S17(a), ESI†) of remain unchanged, while in case of , the CBM band has shifted into the Γ-point (shown in Fig. S17(a), ESI†), indicating the transformation from an indirect to a direct E_g nature for NCMO. We observed an E_g of 3.01 and 2.95 eV for and , suggesting a decrease in E_g due to . Thus, the monotonic decrease in the experimentally measured E_g from NCMO_0.00 to NCMO_1.00 is believed to be due to increasing V_O and 2V_O, corroborating previous studies. The reduction of E_g can be understood as follows: TDOS (Fig. S17(b) and S18(b), ESI†) and PDOS (Fig. S17(c)–(f) and S18(c)–(f), ESI†) calculations show that the VB width decreases in due to the greater localization of the O 2p orbitals, while the curvature increases in and , which is attributed to the enhanced overlap between the Ce 4f_zx², 4f_xz², 4f_yz, and 4f_xyz and O 2p_x, 2p_y, and 2p_z orbitals due to the decrease in the Ce–O bond length in the CeO₈ polyhedra, indicating higher carrier mobility. Further analysis of the TDOS and PDOS reveals that the contributions from Ce 5d_xy and 5d_xz to the CB increase in and . The defect state at 2.44 eV above the valence band in was assigned to Ce 5d_z², whilst these defects are found at 2.26 and 2.40 eV above in due to Ce 5d_z² and Ce 5d_x², respectively. Therefore, it can be stated that the blue III emission originates from Ce 5d_z² → 4f transitions in the CeO₇ polyhedra, while the blue II and green emissions are attributed to Ce 5d_z² → 4f and Ce 5d_x² → 4f transitions in the CeO₆ polyhedra, respectively, as schematically presented in Fig. 7.

Fig. 7 Schematic picture of luminescent transitions within different CeO₈ polyhedra involving different orbitals.

From the calculation of the adsorption energies on the (112) plane with the lowest energy for the NCMO, and systems using eqn (1), we found the energies listed in Table S2 (ESI†). The E_ads (eV) values represent the adsorption energies of the pristine NCMO, , systems, and it was found that E_ads < 0, indicating that the adsorption processes involved in these systems are thermodynamically stable and exothermic, meaning that they release energy during the adsorption process. Additionally, the increasingly negative values of and suggest the presence of active sites (oxygen vacancies) on the NCMO surface, which create localized electron-rich sites enabling strong interactions to occur between the CH₃COCH₃ and NCMO surfaces.⁵⁹ Hence, the highest sensitivity of NCMO_1.00 (Fig. 5) can purely be assigned to the presence of the highest 2V_O content, which is in good agreement with the experimental results and theoretical calculations.

4. Conclusion

In summary, this study reports the synthesis and characterization of oxygen-deficient NCMO-truncated octahedral nanostructures, which exhibit intense blue I and green emissions and comparatively low-intensity blue II and blue III emissions. Excitation involves MoO₄ tetrahedra, while emissions are attributed to the 5d → 4f transitions of Ce in different CeO₈ polyhedral configurations. This study also identified the acetone-sensing ability of NCMO, which has been attributed to CeO₆. The valence and conduction bands of NCMO comprise O 2p and O 2p and Ce 5d orbitals, respectively; hence, they are highly sensitive to oxygen vacancies and have a potential role in luminescence and gas sensing. Our study suggests that NCMO may be a promising material for blue-light-emitting and acetone-sensing applications.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

One of the authors (NH) thanks UGC, Govt. of India, and TM thanks CSIR, Govt. of India, for financial support during execution of their work. NH and TM made equal contributions to this manuscript. The authors would also like to acknowledge Researchers Supporting Project (RSP2024R373), King Saud University, Riyadh, Saudi Arabia.

References

1. B. Glorieux, V. Jubera, A. Apheceixborde and A. Garcia, Sol. State. Sci., 2011, 13, 460–467.
2. H. Yang, F. Peng, Q. Zhang, W. Liu, D. Sun, Y. Zhao and X. Wei, Opt. Mater., 2013, 35, 2338–2342.
3. J. R. de Moraes, S. L. Baldochi, L. D. R. L. Soares, V. L. Mazzocchi, C. B. R. Parente and L. C. Courrol, Mater. Res. Bull., 2012, 47, 744–749.
4. Z. Xu, S. Li, Q. Meng, X. Wang, Q. Zhu and J. G. Li, J. Alloys Compd., 2020, 830, 154676.
5. J. Kim, Inorg. Chem., 2017, 56, 8078–8086.
6. F. Cheng, Z. Xia, M. S. Molokeev and X. Jing, Dalton Trans., 2015, 44, 18078–18089.
7. J. Song, Y. Du, L. Xu, H. Wei, D. He and H. Jiao, J. Am. Ceram. Soc., 2014, 97, 1442–1449.
8. A. Sarapulova, D. Mikhailova, A. Senyshyn and H. Ehrenberg, J. Solid State Chem., 2009, 182, 3262–3268.
9. L. S. Cavalcante, J. C. Sczancoski, R. L. Tranquilin, M. R. Joya, P. S. Pizani, J. A. Varela and E. Longo, J. Phys. Chem. Solids, 2008, 69, 2674–2680.
10. X. Liu, W. Hou, X. Yang and J. Liang, CrystEngComm, 2014, 16, 1268–1276.
11. S. Huang, D. Wang, C. Li, L. Wang, X. Zhang, Y. Wan and P. Yang, CrystEngComm, 2012, 14, 2235–2244.
12. Y. Tian, B. Chen, R. Hua, N. Yu, B. Liu, J. Sun, L. Cheng, H. Zhong, X. Li, J. Zhang and B. Tian, CrystEngComm, 2012, 14, 1760–1769.
13. Z. Wang, Y. Li, Q. Jiang, H. Zeng, Z. Ci and L. Sun, J. Mater. Chem. C, 2014, 2, 4495–4501.
14. N. Haldar, T. Mondal, T. Das, D. Sarkar, M. Pal and C. K. Ghosh, CrystEngComm, 2023, 25, 3514–3527.
15. N. Haldar, T. Mondal, A. Dutta, D. Sarkar, U. K. Ghorai and C. K. Ghosh, Appl. Phys. A: Mater. Sci. Process., 2023, 129, 708.
16. S. Das, U. K. Ghorai, R. Dey, C. K. Ghosh and M. Pal, New J. Chem., 2022, 46, 17585–17595.
17. S. Das, U. K. Ghorai, R. Dey, C. K. Ghosh and M. Pal, Phys. Chem. Chem. Phys., 2017, 19, 22995–23006.
18. B. Ding, H. Qian, C. Han, J. Zhang, S. E. Lindquist, B. Wei and Z. Tang, J. Phys. Chem. C, 2014, 118, 25633–25642.
19. R. Fernández-Climent, S. Giménez and M. García-Tecedor, Sustainable Energy Fuels, 2020, 4, 5916–5926.
20. Z. Lin, M. Xu, Y. Hong, X. Wang and P. Fu, Mater. Lett., 2016, 168, 72–75.
21. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
22. G. Kresse and D. Joubert, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 1758–1775.
23. W. R. Liu, C. W. Yeh, C. H. Huang, C. C. Lin, Y. C. Chiu, Y. T. Yeh and R. S. Liu, J. Mater. Chem., 2011, 21, 3740–3744.
24. L. S. Cavalcante, F. M. C. Batista, M. A. P. Almeida, A. C. Rabelo, I. C. Nogueira, N. C. Batista, J. A. Varela, M. R. M. C. Santos, E. Longo and E. M. S. Li, RSC Adv., 2012, 2, 6438–6454.
25. G. M. Kuzmicheva, A. V. Eremin, V. B. Rybakov, K. A. Subbotin and E. V. Zharikov, Russ. J. Inorg. Chem., 2009, 54, 854–863.
26. R. L. Andrews, A. M. Heyns and P. M. Woodward, Dalton Trans., 2015, 44, 10700–10707.
27. L. Li, Y. Su and G. Li, Appl. Phys. Lett., 2007, 90, 054105.
28. G. M. Kuz'Micheva, I. A. Kaurova, V. B. Rybakov, P. A. Eistrikh-Geller, E. V. Zharikov, D. A. Lis and K. A. Subbotin, CrystEngComm, 2016, 18, 2921–2928.
29. J. V. B. Moura, G. S. Pinheiro, J. V. Silveira, P. T. C. Freire, B. C. Viana and C. Luz-Lima, J. Phys. Chem. Solids, 2017, 111, 258–265.
30. Z. Xu, P. Du, Q. Zhu, X. Li, X. Sun and J. G. Li, Dalton Trans., 2021, 50, 17703–17715.
31. J. A. Horsley, I. E. Wachs, J. M. Brown, G. H. Via and F. D. Hardcastle, J. Phys. Chem., 1987, 91, 4014–4020.
32. B. P. Singh and R. A. Singh, New J. Chem., 2015, 39, 4494–4507.
33. F. F. Wu, D. Zhou, C. Du, S. K. Sun, L. X. Pang, B. B. Jin, Z. M. Qi, J. Varghese, Q. Li and X. Q. Zhang, J. Mater. Chem. C, 2021, 9, 9962–9971.
34. Z. Zhao, Z. Sui, X. Wei, J. Zuo, X. Zhang, R. Dai, Z. Zhang and Z. Ding, CrystEngComm, 2015, 17, 7905–7914.
35. J. Victor-Barbosa-Moura, A. Amison-Gomes-de Souza, P. de Tarso-Cavalcante-Freire, C. da Luz-Lima and T. Mielle-Brito-Ferreira-Oliveira, Int. J. Appl. Ceram. Technol., 2021, 18, 615–621.
36. Y. Xiong, B. Wang, H. Zhuang, X. Jiang, G. Ma, Y. Yi, W. Hu and Y. Zhou, RSC Adv., 2014, 4, 36738–36741.
37. N. Dirany, E. McRae and M. Arab, CrystEngComm, 2017, 19, 5008–5021.
38. S. Mandal, C. K. Ghosh, D. Sarkar, U. N. Maiti and K. K. Chattopadhyay, Solid State Sci., 2010, 12, 1803–1808.
39. T. Thomas, N. Jayababu, J. Shruthi, A. Mathew, A. Cerdán-Pasarán, J. A. Hernández-Magallanes, K. C. Sanal and R. Reshmi, Thin Solid Films, 2021, 722, 138575.
40. A. I. Becerro, M. Allix, M. Laguna, D. González-Mancebo, C. Genevois, A. Caballero, G. Lozano, N. O. Núñez and M. Ocaña, J. Mater. Chem. C, 2018, 6, 12830–12840.
41. L. Xu, G. Liu, H. Xiang, R. Wang, Q. Shan, S. Yuan, B. Cai, Z. Li, W. Li, S. Zhang and H. Zeng, Appl. Phys. Rev., 2022, 9, 021308.
42. R. Xiao, N. Guo, X. Lv, Q. Ma, B. Shao and R. Ouyang, Dalton Trans., 2022, 51, 15484–15495.
43. G. Li, Y. Tian, Y. Zhao and J. Lin, Chem. Soc. Rev., 2015, 44, 8688–8713.
44. Y. Zhu, Y. Liang, M. Zhang, M. Tong, G. Li and S. Wang, RSC Adv., 2015, 5, 98350–98360.
45. L. G. Van Uitert, J. Lumin., 1984, 29, 1–9.
46. E. V. Sokolenko, V. M. Zhukovskii, E. S. Buyanova and Y. A. Krasnobaev, Inorg. Mater., 1998, 34, 499–502.
47. M. V. Korzhik, V. B. Pavlenko, T. N. Timoschenko, V. A. Katchanov, A. V. Singovskii, A. N. Annenkov, V. A. Ligun, I. M. Solskii and J. P. Peigneux, Phys. Status Solidi A, 1996, 154, 779–788.
48. A. Lecointre, A. Bessière, A. J. J. Bos, P. Dorenbos, B. Viana and S. Jacquart, J. Phys. Chem. C, 2011, 115, 4217–4227.
49. J. Raha, N. Haldar and C. K. Ghosh, Appl. Phys. A: Mater. Sci. Process., 2021, 127, 1–8.
50. S. Das, C. K. Ghosh, R. Dey and M. Pal, RSC Adv., 2016, 6, 236–244.
51. Y. Shi, Y. Wen, M. Que, G. Zhu and Y. Wang, Dalton Trans., 2014, 43, 2418–2423.
52. Q. Wang, G. Zhu, Y. Li and Y. Wang, Opt. Mater., 2015, 42, 385–389.
53. M. Rai, G. Kaur, S. K. Singh and S. B. Rai, Dalton Trans., 2015, 44, 6184.
54. Z. Wang, X. Qiu, W. Xi, M. Tang, J. Liu, H. Jiang and L. Sun, Sens. Actuators, B, 2021, 345, 130417.
55. Z. Wang, X. Qiu, W. Xi, M. Tang, J. Liu, H. Jiang and L. Sun, Sens. Actuators, B, 2023, 376, 132936.
56. T. Das, S. Das, M. Karmakar, S. Chakraborty, D. Saha and M. Pal, Sens. Actuators, B, 2020, 325, 128765.
57. T. Das, S. Mojumder, S. Chakraborty, D. Saha and M. Pal, Appl. Surf. Sci., 2022, 602, 154340.
58. Z. Wang, J. Xue, D. Han and F. Gu, ACS Appl. Mater. Interfaces, 2015, 7, 308–317.
59. M. Ali, N. Amrane and N. Tit, Results Phys., 2020, 16, 102907.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ma00306c

‡ At present: University of Engineering and Management, New Town, Kolkata-700160, India.

---