Hydrothermal synthesis and investigation of optical properties of Nb5+-doped lithium silicate nanostructures

Synthesis of niobium-doped lithium metasilicate (Li2−2xNb2xSiO3+δ) (x = 0.0025, 0.005)

Appropriate molar amounts of LiNO ₃ (MW = 68.95 g mol ⁻¹ ) (10 and 11.9 mol, respectively), SiO ₂ ·H ₂ O (MW = 96.11 g mol ⁻¹ ) (20 and 23.92 mol, respectively) and Nb ₂ O ₅ (MW = 265.815 g mol ⁻¹ ) (0.0263 and 0.06 mol, respectively) were dissolved in 60 mL of hot NaOH solution (0.67 and 0.80 M solution, respectively) under magnetic stirring at 80 °C. The resultant solution was transferred and sealed in a Teflon-lined stainless steel autoclave of 100 mL capacity, under autogenous pressure and heated to 180 °C for 96 h. The autoclave was then allowed to cool naturally to room temperature and the resulting white precipitate was recovered.

Synthesis of niobium-doped lithium disilicate (Li2−2xNb2xSi2O5+δ) (x = 0.005, 0.0075 and 0.01)

Appropriate molar amounts of LiNO ₃ (MW = 68.95 g mol ⁻¹ ) (11.9, 10 or 9.9 mol, respectively), SiO ₂ ·H ₂ O (MW = 96.11 g mol ⁻¹ ) (35.9, 30.22 or 30 mol, respectively) and Nb ₂ O ₅ (MW = 265.815 g mol ⁻¹ ) (0.06, 0.073 or 0.1 mol, respectively) were dissolved in 60 mL of hot NaOH solution (1.20, 1.0 and 1.0 M solution, respectively) under magnetic stirring at 80 °C. The resultant solution was transferred and sealed in a Teflon-lined stainless steel autoclave of 100 mL capacity, under autogenous pressure and heated to 180 °C for 96 h. The autoclave was then allowed to cool naturally to room temperature and the resulting white precipitate was recovered.

Powder X-ray diffraction analysis

Phase identifications were performed on a powder X-Ray diffractometer Siemens D5000 using Cu-K _α radiation. The morphology of the obtained materials was examined with a Philips XL30 Scanning Electron Microscope equipped with energy-dispersive X-ray (EDX) spectrometer. Absorption and photoluminescence spectra were recorded on a Jena Analytik Specord 40 and a Perkin Elmer LF-5 spectrometer, respectively.

Figure  ¹ a, b, respectively, shows the EDX spectra of the synthesized Nb ⁵⁺ -doped lithium metasilicate and lithium disilicate nanomaterials, which verify the doping and the compositional analysis of Nb ⁵⁺ in the nanoparticles of lithium silicates.

Fig. 1

The crystal phases of the synthesized materials were examined by powder X-ray diffraction technique. Figures  ² and ³ show the powder XRD patterns of the Nb ⁵⁺ -doped lithium metasilicate and lithium disilicate, respectively. The measured powder XRD data are in good agreement with those of corresponding undoped lithium metasilicate or lithium disilicate nanomaterials [ ³¹ ] and the obtained stable phases are, respectively, isostructural with Li ₂ SiO ₃ (space group Cmc21) [ ³¹ – ⁴¹ ] and Li ₂ Si ₂ O ₅ (space group Ccc2) [ ³¹ , ⁴² – ⁴⁴ ]. The measured data are in agreement with the respective Joint Committee on Powder Diffraction Standards (JCPDS) card for Li ₂ SiO ₃ (JCPDS 29-0829) ( a  = 9.3808 Å, b  = 5.3975 Å and c  = 4.6615 Å) and for Li ₂ Si ₂ O ₅ (JCPDS 15-0637) ( a  = 5.825 Å, b  = 14.56 Å and c  = 4.796 Å). The standard crystallographic data for lithium metasilicate (JCPDS 29-0829) and lithium disilicate (JCPDS 15-0637) and the powder XRD data for respective hydrothermally synthesized undoped nanomaterials [ ³¹ ] are summarized in Tables  ¹ and ² , respectively. Also, the powder XRD data for respective hydrothermally synthesized Nb-doped lithium metasilicate and Nb-doped lithium disilicate are summarized in Tables  ³ and ⁴ for comparisons. Moreover, the intense sharp diffraction patterns suggest that the as-synthesized products are well crystallized.

Fig. 2

Fig. 3

The doping limitations are 0–0.25 and 0–0.75 mol% of Nb ⁵⁺ for lithium metasilicate and lithium disilicate, respectively. Excess mol% concentration of the dopant agent in the reaction mixture, as shown in Figs.  ² and ³ , results in impurity peaks in the XRD patterns. The diffraction line at 2 θ  ≈ 49° is assigned by its peak position to the excess Nb ₂ O ₅ [ ⁴³ ]. Moreover, the formation of other phases of lithium silicates and raw materials was already detected for higher mol% concentration of the dopant agent in the reaction mixture (Figs.  ² , ³ ) [ ³¹ , ⁴¹ , ⁴² , ⁴⁸ ].

Compared to those of the nanomaterials of undoped lithium silicates, the diffraction lines in the powder XRD patterns of the Nb ⁵⁺ -doped lithium silicates nanomaterials shift to lower 2 θ values and, therefore, to larger d values. For the most intensive diffraction line (200) a diffraction line shift of ∆2 θ  = 18.881° (pure)−18.80° (doped) = 0.081° (∆ d  = 4.7206 Å (doped)−4.7005 Å (pure) = 0.0201 Å) for Nb ⁵⁺ -doped lithium metasilicate and for the most intensive diffraction line (040) a diffraction line shift of ∆2 θ  = 24.78° (pure)−24.70° (doped) = 0.08° (∆ d  = 3.600Å (doped)−3.589 Å (pure) = 0.011 Å) for Nb ⁵⁺ -doped lithium disilicate are calculated via Bragg’s law. Tables  ⁵ and ⁶ show the crystal sizes of the Nb-doped materials in different dopant amounts via Debye–Scherrer equation.

Since the ionic radius of the Nb ⁵⁺ (0.64Å [ ⁴⁶ ]) is closer to the ionic radius of Li ⁺ (0.59Å [ ⁴⁶ ]) rather than the Si ⁴⁺ (0.26Å [ ⁴⁶ ]), in the Nb ⁵⁺ -doped lithium metasilicate and lithium disilicate, it may be expected that the dopant ion will replace with Li ⁺ ions in the structure. The larger radius of the dopant ion, compared to the Li ⁺ , may cause an expansion of the lattice parameter in the Nb ⁵⁺ -doped lithium silicate nanomaterials. Since both ionic radii and charges are not the same for the dopant and Li ⁺ ions, it is also possible that the dopant ion takes an interstitial position in lattice rather than replacing any Li ⁺ ions, where additional patterns will be observed in XRD pattern [ ⁴⁷ ]. However, here, the powder XRD data measured for the doped samples are in accordance with those of the undoped materials without any residual or impurity phase formation. The powder XRD patterns of the doped samples, therefore, suggest the fact that the dopant ions are indeed going to lattice positions rather than interstitial positions.

Moreover, on replacing Li ⁺ ions, the dopant ions are bound to create some oxygen-related defect centers or Li ⁺ vacancies for charge compensation. Therefore, it is believed that the dopant ions will be in a structurally disordered environment.

Cellref version 3 was used to refine the cell parameters from the measured powder XRD data of the synthesized doped nanomaterials. Compared to the standard crystallographic data for lithium metasilicate (JCPDS 29-0829) and lithium disilicate (JCPDS 15-0637), the refined unit cell parameters of the synthesized Nb-doped lithium metasilicate and lithium disilicate nanomaterials are a  = 9.3702 Å, b  = 5.3994 Å, c  = 4.6643 Å and a  = 5.826 Å, b  = 14.6168 Å, c  = 4.878 Å, respectively.

Microstructure analysis

SEM images of the pure lithium metasilicate and lithium disilicate are present in our previous work [ ³¹ ]. Figure  ⁴ shows typical SEM images of the synthesized Li _1.995 Nb _0.001 SiO _{3 + δ} nanoparticles. The synthesized sample is composed of multi-ply sheets (thickness and length of about 100 nm and 5 μm, respectively) join together to form nano-flowers. Typical SEM images of the synthesized Li _1.99 Nb _0.002 Si ₂ O _{5 + δ} and Li _1.985 Nb _0.003 Si ₂ O _{5 + δ} are given in Figs.  ⁵ and ⁶ , respectively. The synthesized Li _1.99 Nb _0.002 Si ₂ O ₅ nanomaterial is composed of plate-like nanoparticles with homogenous dispersion (Fig.  ⁵ b, c). The length of the nano-plates is approximately 0.7–0.8 μm. As shown in Fig.  ⁶ , with increasing the dopant concentration in the structure to x _Nb  = 0.0075, the resultant nano-plates assemble to each other to form nano-flower-like structures. The length and thickness of the nano-plates are estimated to be 500 and 80–100 nm approximately.

Fig. 4

Fig. 5

Fig. 6

Optical properties

The emission spectra of pure Li ₂ SiO ₃ and Li ₂ Si ₂ O ₅ are shown in Figs.  ⁷ and ⁸ . In the excitation spectrum of the synthesized Li ₂ SiO ₃ and Li ₂ Si ₂ O ₅ nanomaterials, a band is observed with maxima at 360 and 250 nm, respectively. Accordingly, in the emission spectrum of the synthesized Li ₂ SiO ₃ nanomaterials, an intense peak appears at 410.03 nm. In comparison, an intense peak at 291.45 nm is observed in the emission spectrum of the synthesized Li ₂ Si ₂ O ₅ nanomaterials. With increasing in the reaction time, no shift is observed in the emission spectrum of the obtained Li ₂ SiO ₃ and Li ₂ Si ₂ O ₅ nanomaterials. However, increasing band intensities in the emission spectra of both compounds are observed with increasing reaction time. In the emission spectrum of Nb ⁵⁺ -doped lithium metasilicate nano-flowers (Fig.  ⁹ ), under excitation with light at 234 nm, the main emission band is located at 360 nm with shoulders at 310, 340 and 425 nm. The shoulder appeared at 310 nm is assigned to the band edge emission. Also, the broad band with maxima at 360 nm and the shoulder at 340 nm are assigned to the trap state emission of the nanoparticles. Considering that the energy gap of bulk lithium silicates is above 3.3 eV, the purple-blue photoluminescence appeared as a shoulder at 425 nm (approximately 2.92 eV) is probably due to a triplet to ground state transition of a neutral oxygen vacancy defect, as suggested by ab initio molecular orbital calculations for many other well-studied metal oxides. Also, the emission band related to the Nb(V) centers in the structure is expected to be superimposed on the shoulder at 425 nm [ ⁴⁴ ]. In comparison, the synthesized Nb ⁵⁺ -doped lithium disilicate nanoparticles exhibit an intense broad emission band ( λ _ex  = 229 nm) at 420 nm (~2.95 eV) (Fig.  ¹⁰ ) assigned to the oxygen-related defects and Nb ⁵⁺ centers in the structure, which shows an increasing intensity with increasing the dopant concentration in the structure [ ⁴⁵ ].

Fig. 7

Fig. 8

Fig. 9

Fig. 10