Exploring the milarite minerals, Na₂Mg_5−xM_xSi₁₂O₃₀ (M = Co²⁺, Zn²⁺, Ni²⁺, Cu²⁺) and Na₂Mg_2.5Ca_0.5−xLn_xZn₂Si₁₂O₃₀ (Ln = Eu³⁺, Tm³⁺, Tb³⁺), towards new colored compounds, white light emission, and oxygen evolution reaction (OER) properties

Abstract

New transition metal substituted milarite silicates of the general formula A₂B₂C[T(2)₃T(1)₁₂O₃₀] were prepared by a conventional solid state technique and their structures determined by powder X-ray diffraction (PXRD) methods. Raman spectroscopic studies indicated expected Raman bands. The oxidation states of the transition elements were confirmed by XPS studies. The optical absorption bands were rationalized based on the ligand-centered emission and allowed d–d transitions. The white compounds exhibited good deep UV cut off values, suggesting their possible use as optical filters. The white compounds also showed good near-infrared (NIR) reflectivity comparable to that of the commercial TiO₂. The rare earth substituted compounds, Na₂Mg_2.5Ca_0.5−xLn_xZn₂Si₁₂O₃₀, exhibited intense red (Eu³⁺), green (Tb³⁺) and blue (Tm³⁺) emissions. An optimal composition having all the three lanthanide ion substitutions resulted in white light emissions in Na₂Mg_2.5Ca_0.48Tm_0.01Tb_0.02Eu_0.01Zn₂Si₁₂O₃₀. The presence of Co²⁺ in the Na₂Mg₃Co₂Si₁₂O₃₀ compound facilitated the study of the oxygen evolution reaction (OER) with good values that are comparable to those of RuO₂. The white compounds have reasonably good dielectric constant values with low dielectric loss, which indicates their possible use in communication devices. The milarite based compounds exhibited considerable potential towards new colored compounds, white light emission and OER properties. The present study suggests that it is profitable to explore mineral structures towards new and known material properties.

---

Introduction

In the realm of inorganic materials chemistry, oxides are the dominant class of compounds.^1–3 The studies carried out over the years established that the compounds belonging to the perovskite (ABO₃), pyrochlore (A₂B₂O₇), spinel (AB₂O₄), garnet (A₃B₅O₁₂) etc. structures are highly adaptable.^4,5 These structures accommodate a number of ions with varying sizes and valence states, which are manifested in their different properties and applications in materials science.^6–9

Another important class of compounds belong to the silicates. The ease of formation of polymeric structures from [SiO₄] tetrahedra results in a large variety of interesting structures.^10–12 Many silicates occur naturally and a number of synthetic ones are also known, especially those of zeolites.^13–15 The structural variety and diversity of the silicates, notably the porous ones, find many applications in the area of catalysis, gas sorption and separation processes.

Some of the naturally occurring silicates have alkali/alkaline earth metal ions as a part of the structure.^16–19 One such naturally occurring silicate belongs to the milarite family of compounds. The general formula of the milarite structure can be written as A₂B₂C[T(2)₃T(1)₁₂O₃₀]·xH₂O,^20,21 where T refers to atoms having tetrahedral coordination, A refers to atoms having octahedral coordination, B refers to atoms having 9-coordination and C refers to atoms having 12-coordination. This diversity in accommodating different coordinations within the structure results in a large variety of structural modifications within the milarite family of compounds.^20,22–27 The variety of ions that can be accommodated within the milarite structure is given in Table S1.²⁸

The structure of the milarite consists of T(1) [SiO₄] tetrahedra that form a six-membered ring, which are connected to form a double six ring of [Si₁₂O₃₀] units. The double six rings are joined by the T(2) tetrahedra forming the extended structure (Fig. S1). The T(2) tetrahedra share edges with A-octahedra, which are also connected with T(1) tetrahedra, strengthening the overall tetrahedral framework. The B-polyhedra occupy positions above and below the A-octahedra. Both the A and B sites occupy a 3-fold axis and exhibit considerable distortions. The C-atoms are in the channels formed by the connections between the A, B, T(1) and T(2) polyhedral units.

We have been interested in the study of many mineral structures recently with a view to explore their physical and chemical properties.^29–36 The milarite structure is appealing to us due to the possibility of exploring different substitutional studies at various positions. Thus, we have explored the milarite structure towards the generation of new colored compounds as a host for luminescence and white light emission and towards oxygen evolution reactions. In this paper, we present the results of the studies on the compounds prepared with the milarite structure.

Experimental section

Synthesis

All the compounds were synthesized employing the conventional solid-state synthetic methods. The starting materials, Na₂CO₃, MgCO₃, MC₂O₄·2H₂O (M = Co²⁺, Ni²⁺), CuO, ZnO, CaCO₃, Ln₂O₃ (Ln = Eu³⁺, Tm³⁺, Tb³⁺), and SiO₂, were taken in a stoichiometric ratio, ground thoroughly for 30 minutes and kept in a muffle furnace at 900–1100 °C for 12–24 Hours. The purity of the prepared compounds was established employing powder X-ray diffraction (PXRD) (Fig. 1 and Fig. S2). The compound, Na₂Mg₃Co₂Si₁₂O₃₀, was also prepared employing the hydrothermal method. The compounds NaNO₃ and Mg(NO₃)₂·6H₂O, Co(NO₃)₂·6H₂O were mixed in a beaker (Solution A). SiO₂ was dissolved in an alkali solution by stirring for 1 hour at 70–80 °C. The SiO₂ solution was added to solution A dropwise under continuous stirring. The resultant solution was transferred to a Teflon lined autoclave (fill factor = 70–80%) and kept at 180 °C for 3 days. The PXRD investigation confirmed the phase purity of the compound.

Fig. 1 Experimental PXRD patterns of the prepared compounds, Na₂Mg₅Si₁₂O₃₀, Na₂Mg₃Zn₂Si₁₂O₃₀, Na₂Mg₃Co₂Si₁₂O₃₀, and Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀.

The PXRD patterns were recorded employing a Philips X'pert diffractometer (Ni filtered Cu Kα radiation, λ = 1.5418 Å) (Fig. 1 and Fig. S2). For Rietveld refinement of Na₂Mg₃Co₂Si₁₂O₃₀ and Na₂Mg₃ZnCuSi₁₂O₃₀, the PXRD data were collected at room temperature in the 2Θ range of 8–120° with a step size of 0.02° and a step duration of 50 s. The PXRD pattern was refined using GSAS-EXPGUI. The lattice parameters, scale factor, background (Fourier polynomial background function), pseudo-Voigt (U, V, W, and X), and isothermal parameter (U_iso) were refined (Fig. 2, Table 1, Fig. S3 and Table S2). Thermal parameters were constrained to be the same for different atoms occupying the same site. The optical studies were carried out on the powdered samples (Perkin–Elmer Lambda 950 UV-Vis double-beam spectrometer) in the spectral region of 200–2500 nm. The diffuse reflectance data were converted to the Kubelka–Munk unit:

F(R) = (1 − R)²/2R = α/S

in which R is the reflectance, α is the absorptivity, and S is the scattering factor. The optical band gap of Na₂Mg₃Co₂Si₁₂O₃₀ was calculated using the Tauc plot, with the function

αhν = A[hν − E_g]ⁿ

where α is the absorption coefficient, A is known as the band tailing parameter and n is the power factor of the transition mode. The plot of (αhν)^1/nversus the photon energy (hν) was extrapolated to give the value of the band gap (E_g). The X-ray photoelectron spectroscopy (XPS) data (Thermo Fisher Scientific K-alpha with Al K(alpha) as an X-ray source) were collected at room temperature.

Fig. 2 Rietveld refinement of Na₂Mg₃Co₂Si₁₂O₃₀. The observed (O), the calculated (red line), and the difference (bottom blue line) are shown. The vertical bars (|) indicate Bragg's reflections. The simulated pattern was generated from Na₂Mg₅Si₁₂O₃₀.²⁶

Table 1 Crystallographic data of Na₂Mg₃Co₂Si₁₂O₃₀ from the Rietveld refinement data^a,b

Elements	x/a	y/b	z/c	Occ.	U iso	Site
a Space group P6/mcc; a = b = 10.1537(1) Å; c = 14.2536(1) Å; V = 1272.65(5) Å^3; α = β = 90°; γ = 120°; reliability factors: RP = 1.2%; Rwp = 1.9%; χ^2 = 4.80; average bond lengths [Å]: Mg2/Co2–O = 2.0410(1); Si–O = 1.5854(1); Mg1–O = 2.1324(1). b The refinement structural model was taken from Na2Mg5Si12O30.^26
Na1	0	0	0.2500	0.670	0.084(10)	2a
Na2	0.3333	0.6666	0	0.670	0.023(4)	4d
Mg1	0.3333	0.6666	0.2500	1	0.012(3)	4c
Mg2	0.5000	0	0.2500	0.363	0.037(2)	6f
Co2	0.5000	0	0.2500	0.637	0.037(2)	6f
Si1	0.7641(3)	0.1117(3)	0.1077(2)	1	0.018(1)	24m
O1	0.7385(6)	0.1370(7)	0	1	0.024(3)	12l
O2	0.9380(6)	0.2145(5)	0.1301(3)	1	0.019(2)	24m
O3	0.6674(5)	0.1593(4)	0.1667(2)	1	0.017(2)	24m

---

The dielectric measurements (Novocontrol impedance analyzer (Alpha-A)) were carried out in the frequency range of 0.1 Hz–1 MHz. Micro-Raman spectral studies (HORIBA LabRAM HR Evolution) were performed at room temperature using a triple monochromator DXR microscope Raman spectrometer, equipped with 1800 groove per mm diffraction grating and a liquid nitrogen-cooled CCD detector. The 10 mW output of the 532 nm line of Nd:YAG (10 mW) was used as the excitation source.

Magnetic measurements were carried out in the temperature range 2–300 K employing a SQUID magnetometer (Quantum Design, USA) under an applied field of 1000 Oe. The photoluminescence and lifetime studies (Fluoromax plus spectrofluorometer and Edinburgh spectrofluorometer) were performed at room temperature.

The electrocatalytic performance of Na₂Mg₃Co₂Si₁₂O₃₀ (PARSTAT analytical electrochemical workstation) was studied employing a three-electrode system with Hg/HgO (0.5 M KOH) as the reference electrode, a graphite rod as the counter electrode, and glassy carbon coated with Na₂Mg₃Co₂Si₁₂O₃₀ ink as the working electrode in a 0.5 M KOH electrolyte.

Results and discussion

The prepared compounds were characterized by PXRD which indicated the formation of single-phase compounds. The compounds, Na₂Mg₅Si₁₂O₃₀ and Na₂Mg₃Zn₂Si₁₂O₃₀, along with their transition metal-substituted analogues were found to be phase pure (Fig. 1 and Fig. S2). As part of the study, we explored the possibility of replacing the Mg²⁺ ions by Ca²⁺ ions, which resulted in Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀.

We made attempts to refine the Co²⁺ ion substituted compound, Na₂Mg₃Co₂Si₁₂O₃₀, by employing the Na₂Mg₅Si₁₂O₃₀ compound as the model using the GSAS-EXPGUI powder structure refinement program (Fig. 2). The results of the refinement indicate a good fit with the reported structure, Na₂Mg₅Si₁₂O₃₀.²⁶ A minor peak observed around ∼22° appears to be due to a SiO₂ phase, which has been previously reported in the synthesis of milarites.²⁶ This small peak was excluded during the refinement. The refinement suggested that the Co²⁺ ions are substituted at the tetrahedral sites (T(2)) and resulted in a small reduction in the overall volume of the unit cell (1275.03(5) Å to 1272.65(5) Å). This reduction in the overall volume appears to be consistent with the contraction of the c-axis (14.29(3) Å to 14.25(5) Å), reflecting the stronger polarizing power exhibited by the Co²⁺ ions compared to the Mg²⁺ ions in the tetrahedral coordination (Fig. 2, 3 and Table 1). We carried out the bond valence sum (BVS) calculation on the compounds. We observed that the BVS value for the Co²⁺ ions at the tetrahedral T(2) site was ∼1.56, which suggests that the tetrahedral site may be slightly under-bonded. In the present compound, Na₂Mg₃Co₂Si₁₂O₃₀, the BVS calculations reveal that the octahedral A site (Mg²⁺ ion) gives a value of ∼1.85 and the T(2) site (Mg²⁺ ions) was found to have a value of ∼1.548. This indicates that the T(2) tetrahedral positions in the milarite structure are inherently under-bonded. The BVS calculations for Co²⁺ ions showed that they were under-bonded, and it is likely that the Co²⁺ ion occupies the T(2) site rather than the A site.

Fig. 3 Crystal structure of Na₂Mg₃Co₂Si₁₂O₃₀, (a) viewed down the ‘ca’ plane showing the tetrahedral SiO₄ (blue color) and Mg2/Co2–O₄ tetrahedral (green color) layers and incorporation of Mg1 octahedra, and Na polyhedra, (b) the view along the ‘ab’ plane, showing the tunnels where Na(1) is present, and (c) the view along the ‘ac’ plane.

Raman spectroscopic studies

Raman spectroscopic studies were carried out on Na₂Mg₅Si₁₂O₃₀ and other substituted compounds (Fig. 4 and Fig. S5). The main Raman bands are observed in the region of 60–1500 cm⁻¹. The centrosymmetricity (P6/mcc: no. 192) of the space group suggests the presence of mutually exclusive IR-active and Raman-active bands. The structure has 98 atoms in the unit cell, which results in a total of 288 vibrational modes (Table S3).^37–39

Fig. 4 Raman spectra of the synthesized compounds, Na₂Mg_5−xCo_xSi₁₂O₃₀ (x = 0–2).

Of these, 3 vibrational modes (A_2u + E_1u) are acoustic in nature, and 67 modes (13A_2u + 27E_1u) are IR-active. Based on the nuclear site analysis, the following Raman active modes were to be expected.

Γ_Raman = 11A_1g + 26E_2g + 24E_1g

We observed a lower number due to the doubly degenerate Raman modes and overlap of bands. A weak band at ∼567 cm⁻¹ could be due to the bending vibration of Si–O bonds within the SiO₄ tetrahedra.⁴⁰ A weak broad band observed at ∼823 cm⁻¹ might be due to the Si–O–Si vibrations, and is commonly observed in many silicates.²⁶ The bands at ∼1050 and ∼1121 cm⁻¹ (A_1g) could be due to the stretching vibration of the Si₁₂O₃₀ cluster.^24,25

The Mg(1)O₆ bending vibration was observed at ∼241 cm⁻¹.⁴¹ The substitution of Co²⁺ ions in place of Mg²⁺ ions results in a Raman splitting of the band centred around 240–275 cm⁻¹ and transforms into a doublet. In addition, we also note that the Raman band centred at ∼176 cm⁻¹ appeared to gain intensity due to the substitution of Co²⁺ ions. This suggests a stronger bonding and more polarizability in the Co-substituted compound. The medium Raman bands centred at ∼511 cm⁻¹ (E_2g) and ∼567 cm⁻¹ could be due to the Mg(2)–O₄ stretching vibrations. The substitution of Co²⁺ ions in place of the tetrahedral Mg²⁺ ions (Mg(2)) leads to the following changes in the Raman spectra: (i) the band centred at ∼511 cm⁻¹ increases in intensity as the amount of Co²⁺ ions is increased and (ii) the band centred at ∼567 cm⁻¹ shifts to a lower wavenumber with the increase in the substitution of Co²⁺ ions. The strong Raman band observed at ∼467 cm⁻¹ could be due to the stretching vibration of Mg–O–Si bonds (Si₁₂O₃₀ units binding with Mg), which appears to be a characteristic marker of the milarite family of compounds.²⁶

Optical studies

The optical properties of the transition metal-substituted compounds were examined employing ultraviolet–visible (UV-Vis) and near-infrared spectroscopy (NIR). The transition metal–substituted compounds exhibited vibrant colors under daylight, which is characteristic of the transition metal ions occupying octahedral and tetrahedral coordination environments (Fig. 5 and Fig. S6).^42–44

Fig. 5 Optical absorption spectra of the synthesised compounds, Na₂Mg_5−xCo_xSi₁₂O₃₀ (x = 0–2).

The Co²⁺ ions (3d⁷) possess ⁴F and ⁴P states as the free ion state. On application of the octahedral or tetrahedral ligand field, these terms split into multiple levels, which results in the characteristic d–d transitions.⁴⁵ Normally, the Co²⁺ ions in a tetrahedral coordination environment have three spin-allowed transitions: ⁴A₂ → ⁴T₂ (⁴F), ⁴A₂ → ⁴T₁ (⁴F), and ⁴A₂ → ⁴T₁ (⁴P) (Fig. S6a). Out of these, the first two transitions are observed in the infrared (IR) region and the third transition, ⁴A₂ → ⁴T₁ (⁴P), is observed in the visible region. In many compounds, this allowed transition is often found overlapped with the spin-forbidden transitions such as ⁴A₂(⁴F) → ²E(²G).^46–48 In the UV-Vis spectra, the compound, Na₂Mg_5−xCo_xSi₁₂O₃₀ (x = 0.25–2), exhibits two maximas that can be separated based on the Co²⁺ ion concentration in the compounds. Thus at the low substitution region (x = 0.25–0.5) the maxima is observed at ∼2.03 eV (∼610 nm) and ∼2.48 eV (∼500 nm) with a shoulder at 2.13 eV (∼580 nm). These transitions correspond to the ⁴A₂ → ⁴T₁ (⁴P) transition of the Co²⁺ ions in tetrahedral coordination.⁴⁹ In addition we also observed the spin forbidden transitions at ⁴A₂(⁴F) → ²E(²G) (a doublet at ∼2.98 eV (∼416 nm) and ∼3.02 eV (∼410 nm)) and ⁴A₂(⁴F) → ²T₁(²G) (∼2.31 eV (∼536 nm)). Similar earlier observations were rationalised based on the highly distorted Co²⁺ ions in tetrahedral coordination along with strong L–S coupling.⁵⁰ The absorptions for the Co²⁺ ions were observed in the red and green-blue regions, which would result in a lavender color. When the concentration of the Co²⁺ ions increases (x = 1–2), the strong absorption band observed at ∼2.03 eV reduces in intensity along with an increase and broadening of the absorption at ∼2.48 eV. This results in the deepening of the lavender color.⁵¹ In addition, two valleys of low absorption around ∼2.3 eV and ∼2.88 eV, corresponding to the yellow-green and deep blue regions, were also observed. A combination of the absorption bands along with valleys of low absorption would result in the observed color of the compounds containing higher concentration of Co²⁺ ions.⁵¹ The observation of a triplet broad band in the region between 0.88 and 1.02 eV, which corresponds to the ⁴A₂ → ⁴T₁ (⁴F) transition, suggests the presence of Co²⁺ ions in a tetrahedral coordination environment (Fig. 5).^52,53 We also observed similar absorption bands for the Co²⁺ ion substituted Na₂Mg₃ZnCoSi₁₂O₃₀ compound.

The Cu²⁺ ion substituted Na₂Mg_4−xCu_xSi₁₂O₃₀ (x: 0.5, 1) and Na₂Mg₃Z_2−xCu_xSi₁₂O₃₀ (x: 0.5, 1) compounds exhibit colors that have different shades of blue under the daylight (Fig. S6). The UV-visible spectra were featureless. We observed a broad transition in the near infrared (IR) region (1.34 eV–1.51 eV) which could be due to the ²T₂ → ²E₂ transition. Similar optical absorption features have been observed for the Cu²⁺ ions in oxide hosts.^33,36

The Ni²⁺ ion substituted compounds, Na₂Mg_4.5Ni_0.5Si₁₂O₃₀ and Na₂Mg₃Zn_1.5Ni_0.5Si₁₂O₃₀, exhibited a greyish green color under the daylight (Fig. S6). We observed a broad absorption band with a maximum centred at ∼1.96 eV (∼632 nm) and ∼2.44 eV (∼508 nm) in the UV-Visible spectrum. Normally, Ni²⁺ ions prefer either a square-planar or an octahedral coordination. The Ni²⁺ ions substituted in tetrahedral coordination are also known and the colors exhibited by these compounds change from pink to grey color.^{31–33,35,51,54} In the present compound, we successfully substituted a small concentration of Ni²⁺ ions in the tetrahedral coordination. The Ni²⁺ (d⁸) ions in a tetrahedral environment are expected to exhibit three transitions, ³T₁ (F) → ³T₂ (F), ³T₁ (F) → ³A₂ (F), and ³T₁ (F) → ³T₁ (P). Of these, the ³T₁ (F) → ³T₁ (P) transition corresponds to the observed absorption band.^33,35,36

As part of the study, we also carried out the NIR reflectance studies on the white-colored compounds (Fig. S7). All the white compounds exhibited good near-IR reflectivity behavior with values in the range of 70–85%, which is comparable to that of the industrial standard NIR compound, TiO₂ (∼70%) (Fig. S7). The high near-IR reflectivity values suggest that these compounds may find use as NIR reflective coatings.³⁵

There has been intensive research towards developing materials that exhibit deep ultraviolet (UV) cut-off characteristics, which are useful for advanced optical applications such as in UV filters, protective coatings, optical components in lithography and sterilization technologies etc. It has been shown that many borates, phosphates, and fluorides exhibit a short UV cutoff edge.^55–57 We explored the compounds, Na₂Mg₅Si₁₂O₃₀ and Na₂Mg₃Zn₂Si₁₂O₃₀, towards the deep UV-cutoff (Fig. S8). The investigations indicate that the compounds exhibit ∼72% of the UV cut-off at 204 nm. This value appears to be promising towards the development of optical components requiring a deep UV-cut off.

Photoluminescence studies

The milarite structure has multiple coordination sites and it occurred to us that the compound may have the potential to act as a host towards the lanthanide emission. To this end, we have taken the Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ compound as a host and attempted the substitution of different lanthanide ions such as Eu³⁺, Tb³⁺ and Tm³⁺ ions. The idea is not only to investigate the emission behavior, but also to generate white light emission by judicious combination of the above elements. It is well known that a combination of the red (Eu³⁺), green (Tb³⁺), and blue (Tm³⁺) emissions can result in a white light.^8,58–60

All the lanthanide substituted compounds were found to be single phasic by PXRD (Fig. S2). The room temperature photoluminescence of all the compounds exhibited the expected emission consistent with the lanthanide ions (Fig. 6 and Fig. S9). In all the compounds, the lanthanide concentrations were varied to establish maximum emission intensity. The studies clearly establish that for Eu³⁺ ions, the maximum intensity was observed at 6%; for Tb³⁺ ions it was at 9% and for Tm³⁺ it was at 4% (Fig. S9).

Fig. 6 PL and PLE spectra of Na₂Mg_2.5Ca_0.5−xLn_xZn₂Si₁₂O₃₀: (a) Eu³⁺ (6%), (b) Tb³⁺ (9%), and (c) Tm³⁺ (4%).

The photoluminescence excitation (PLE) spectra of the 6% Eu³⁺ ion substituted compound, Na₂Mg_2.5Ca_0.44Eu_0.06Zn₂Si₁₂O₃₀, were monitored at 612 nm (Fig. 6a). This exhibited excitation peaks at 361 nm, 381 nm, 394 nm, 412 nm, and 464 nm, which correspond to the f–f transitions, originating from the ground state ⁷F₀ to ⁵D₄, ⁵L₇, ⁵L₆, ⁵D₃, and ⁵D₂ excited states, respectively.^30,61 Additionally, we also observed an excitation peak at 531 nm, which may correspond to the ⁷F₁ → ⁵D₁ transition.²⁹ The emission spectra (λ_ex = 394 nm) indicated a maximum intensity at 612 nm wavelength, which could be due to the ⁵D₀ → ⁷F₂ transition. The other emission peaks were observed at 588 nm, 649 nm and 702 nm in the photoluminescence emission (PL) spectra, which correspond to the ⁵D₀ → ⁷F₁, ⁵D₀ → ⁷F₃, and ⁵D₀ → ⁷F₄ transitions, respectively. It is known that the emission of Eu³⁺ ions in a host can provide clues to the local environment, as the emission characteristics are sensitive to the site symmetry and the coordination. In the present compounds the maximum intensity was observed at 612 nm, assigned to the ⁵D₀ → ⁷F₂ transition, which is the electric dipole transition. This transition exhibits enhanced intensity when the Eu³⁺ ions occupy non-centrosymmetric or low-symmetry sites. The magnetic dipole transition, ⁵D₀ → ⁷F₁, on the other hand, is parity-allowed and relatively insensitive to the changes in the crystal field. The observation of maximum intensity for the ⁵D₀ → ⁷F₂ transition (612 nm) suggests that the Eu³⁺ ions are located at sites lacking the inversion symmetry and corroborates the low-symmetry environment.⁶¹

The PLE and PL spectra of the 9% Tb³⁺ ion substituted compound, Na₂Mg_2.5Ca_0.41Tb_0.09Zn₂Si₁₂O₃₀, were monitored at 541 nm and 377 nm, respectively (Fig. 6b). The PLE spectra of the Tb³⁺ ion substituted compound exhibited excitation peaks at 319 nm, 333 nm, 351 nm, 367 nm, 377 nm, and 483 nm, which correspond to the excitation from the ground state, ⁷F₆, to the excited states: ⁶D₁, ⁵L_7,8, ⁵L₁₀, ⁵G₆, ⁵D₃ and ⁵D₄, respectively.^62,63 The PL spectrum of the compound (λ_ex = 377 nm) exhibits emission bands in the blue region at 436 and 486 nm, the green region at 541 and 548 nm, and the orange-red region at 583 and 619 nm.^64,65 Of these, the most intense emission was found to be at 541 nm, which could be due to the ⁵D₄ → ⁷F₅ transition.⁶⁶ The other emission bands observed in the PL spectra could be assigned to ⁵D₄ → ⁷F₆ (486 nm), ⁵D₄ → ⁷F₄ (583 nm), and ⁵D₄ → ⁷F₃ (619 nm) transitions. The weak emission observed at ∼436 nm could be due to the ⁵D₃ → ⁷F₄ transition.^64–67

The PL and PLE spectra of the compound, Na₂Mg_2.5Ca_0.46Tm_0.04Zn₂Si₁₂O₃₀, doped with 4% Tm³⁺ ions were recorded by monitoring emissions at 357 nm and 457 nm, respectively (Fig. 6c). It is known that the Tm³⁺ ions exhibit many transitions due to the strong Russell–Saunders (L–S) coupling of the 4f electrons.⁶⁸ The PLE spectra exhibited a strong excitation peak at 357 nm, which could be due to the ³H₆ → ¹D₂ transition. The Tm³⁺ substituted compound exhibits an intense blue emission at 457 nm (λ_ex = 357 nm), which can be assigned to the ¹D₂ → ³F₄ transition.^58,61,68,69

The Commission Internationale de L'Eclairage (CIE) chromaticity coordinates for the substituted compounds exhibit intense blue (4% Tm³⁺), green (9% Tb³⁺) and red (6% of Eu³⁺) emission (Fig. S10). The corresponding color correlating temperature (CCT) values are given in Table S4.

The observation of RGB emission in the Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ compound suggested the possibility of obtaining white light by careful choice of the substitution of the three rare-earth ions (Table S5a). The CIE coordinates and the CCT values were also found to be comparable to those of some of the known silicate compounds (Table S6). It may be noted that the ratio of the Tm³⁺ : Tb³⁺ : Eu³⁺ ions in the host exhibiting maximum intensity was found to be approximately 1 : 2 : 1. To this end, we synthesized a series of compounds by fixing the concentrations of Tm³⁺ (1%) and Tb³⁺ (2%), along with systematically varying the Eu³⁺ ion concentration (x = 0, 0.5, 1, 1.5, 2 mol%) (Fig. 7). The photoluminescence (PL) spectra of the different samples were recorded by employing an excitation wavelength of 358 nm (Fig. 7). The variation of Eu³⁺ ion concentration in the host Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and x% Eu³⁺ with x = 0–2%) changes the color of the emission from the blue green region to the white region (Fig. 8). In addition, we also noted that on increasing the Eu³⁺ concentration from 0 to 2 mol%, there was a progressive quenching of the emission intensities of the Tm³⁺ and the Tb³⁺ ions, accompanied by an enhancement of the intensity of the Eu³⁺ emission. This indicates possible energy transfer from the Tm³⁺ and Tb³⁺ ions to the Eu³⁺ ions. Similar behavior has been observed before.^30,61 The PL Tm³⁺ emission at ∼457 nm (blue) and the Tb³⁺ emission at ∼541 nm (green) are progressively quenched as the concentration of Eu³⁺ substitution increases (Fig. 7). The change in the intensities of the individual emission indicates that there is a possibility of cascade energy transfer (Tm³⁺ → Tb³⁺ → Eu³⁺).

Fig. 7 PL emission spectra (λ_ex = 358 nm) of Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and x% Eu³⁺), x = 0 to x = 2%. Please note the variation in the different intensities as a function of Eu³⁺ concentration (see the text).

Fig. 8 CIE diagram for Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and x% Eu³⁺) with the concentration of Eu³⁺ varying from x = 0 to x = 4%. Note the white light region for x = 1.

We have carried out a temperature-dependent photoluminescence on a freshly prepared and characterized white light emitting sample (Fig. S9d). The CIE coordinates obtained at 20 K indicate that the values are marginally different from those obtained at room temperature (Table S5b).

Energy transfer studies

It has been well established that the excited state levels of Tb³⁺ ions generally have higher energy compared to that of the Eu³⁺ ions. Thus, in many compounds, facile transfer of energy from the Tb³⁺ ions to the Eu³⁺ ions has been reported.^70–73 We wanted to explore the energy transfer from Tb³⁺ ions to Eu³⁺ ions in the compound, Na₂Mg_2.5Ca_{0.5−0.08−x}Tb_0.08Eu_xZn₂Si₁₂O₃₀ (x = 0.02–0.06). The photoluminescence emission spectra of Na₂Mg_2.5Ca_{0.5−0.08−x}Tb_0.08Eu_xZn₂Si₁₂O₃₀ (x = 0.02–0.06) (λ_ex = 377 nm) indicated Tb³⁺ ion emission at ∼541 nm (⁵D₄ → ⁷F₅) along with the Eu³⁺ ion emission at ∼612 nm (⁵D₀ → ⁷F₂) (Fig. 9). As the Eu³⁺ ion concentration increases, the intensity of the Tb³⁺ emission band is found to decrease progressively along with an increase in the intensity of the emission due to Eu³⁺ ions. This behavior indicates possible transfer of energy from the Tb³⁺ ions (donor) to the Eu³⁺ ions (acceptor) in the host. Further increase of Eu³⁺ results in the red emission suggestive of quenching of the Tb³⁺ emission.^74,75 The energy transfer mechanisms involving the sensitizer and the activator ions can be classified as either exchange interactions or electric multipolar interactions.⁷⁶ The distinction between the two types of mechanisms is determined by evaluating the critical distance (R_c) between the two interacting ions. When the value of R_c is below 5 Å, the exchange interactions could be more feasible and for higher values of R_c, the energy transfer could proceed through electrical multipolar interactions. The critical distance between the Tb³⁺ ions (sensitizer) and the Eu³⁺ ions (activator) was calculated following the approach outlined by Blasse,⁷⁷
where N is the number of atoms in the unit cell (N = 2), V is the volume of the unit cell (V = 1271 ų), and X_c is the optimal concentration of Tb³⁺ and Eu³⁺ ions (combined concentration of Tb³⁺ and Eu³⁺ at which the Tb³⁺ emission intensity is reduced to half of its value in the absence of Eu³⁺) (X_c = 0.12).^73,78

Fig. 9 PL emission spectra (λ_ex = 377 nm) of Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (8% Tb³⁺ and x% Eu³⁺), x = 2 to x = 6%.

Thus, the critical distance (R_c) for the Tb–Eu pair in the present structure was calculated to be ∼21.6 Å. This value is considerably larger than the 5 Å required for the exchange interactions and suggests that the energy transfer occurs through a long-range interaction (multi-ploar interaction). It is likely that the energy transfer between the Tb³⁺ ions and the Eu³⁺ ions could be through the resonant electric dipole–dipole coupling pathway. This is also consistent with the observation that the Tb³⁺ ion emission intensity is substantially reduced in the presence of a higher concentration of Eu³⁺ ions. It may be noted that the Eu³⁺ (⁵D₀ → ⁷F₂) emission intensity is increased which indicates that the Tb³⁺ ions act as a good sensitizer for the Eu³⁺ ions through long-range dipole–dipole interactions resulting in the energy transfer (Fig. 10).^{74,75,79–81}

Fig. 10 Possible energy transfer pathway in the compound, Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and x% Eu³⁺) with x = 0 to x = 2%.

Fluorescence decay average lifetime

In order to explore the energy transfer process further between the Tb³⁺ ions and the Eu³⁺ ions, we measured the lifetime of the compound, Na₂Mg_2.5Ca_{0.5−0.08−x}Tb_0.08Eu_xZn₂Si₁₂O₃₀ (x = 0.02–0.06). The fluorescence decay of 8% Tb³⁺ and x% Eu³⁺ (x = 0.02–0.06) was measured employing 377 nm excitation while monitoring the emission wavelength at 541 nm (Fig. 11). The decay curves were fitted using a tri-exponential function:
and the average lifetime was calculated employing

τ_avg = (B₁τ₁² + B₂τ₂² + B₃τ₃²)/(B₁τ₁ + B₂τ₂ + B₃τ₃)

where I and I₀ are the intensities at t and at t = 0, B₁, B₂ and B₂ are the amplitude of the decay process and τ₁, τ₂ and τ₃ are the fluorescence decay times.^75,76,82,83

Fig. 11 Fluorescence lifetime decay curves (λ_ex = 377 nm; λ_em = 541 nm) of Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (8% Tb³⁺, and x% Eu³⁺) with the concentration of Eu³⁺ varying from x = 2 to x = 6%.

The fluorescence average lifetime decay was also estimated for a single substituent having maximum emission intensity: 6% Eu³⁺, 9% Tb³⁺, and 4% Tm³⁺ in the compound Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (Fig. S11).

Similarly, the fluorescence decay measurements were also conducted on Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and x% Eu³⁺) with the concentration of Eu³⁺ varying from x = 0 to x = 2%. All the samples were excited at 358 nm, and the emission was monitored at 541 nm, corresponding to the ⁵D₄ → ⁷F₅ transition of Tb³⁺ ions (Fig. 12).

Fig. 12 Fluorescence lifetime decay curves (λ_ex = 358 nm; λ_em = 541 nm) of the Ln³⁺ doped host Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and x% Eu³⁺) with the concentration of Eu³⁺ varying from x = 0 to x = 2%.

The fluorescence decay profile was fitted employing the tri-exponential function and the average lifetime was calculated (Table S7). The decreasing lifetime decay agrees with the energy transfer from Tm³⁺ to Tb³⁺ to Eu³⁺ ions, where the excited-state energy of Tb³⁺ is non-radiatively transferred to the Eu³⁺ ions. We obtained the white light emission at x = 1 for Eu³⁺ ions in the host Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and x% Eu³⁺), which exhibited a fluorescence average lifetime decay of 485 μs. The fluorescence decay studies on the Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and 1% Eu³⁺) compound at 20 K gave a lifetime value of 783 μs (Fig. S11d).

To evaluate the performance of the phosphor, we also measured photoluminescence quantum yield (PLQY) for the maximum La³⁺ substituted, Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀, and for the white light compound, Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and 1% Eu³⁺). The Eu³⁺ substituted compound, Na₂Mg_2.5Ca_0.44Eu_0.06Zn₂Si₁₂, exhibited a PLQY of 55% (λ_ex = 394 nm). The Tb³⁺ substituted compound, Na₂Mg_2.5Ca_0.41Tb_0.09Zn₂Si₁₂, and Tm³⁺ substituted compound, Na₂Mg_2.5Ca_0.46Tm_0.04Zn₂Si₁₂, exhibited lower PLQY values of 16% (λ_ex = 377 nm) and 1% (λ_ex = 358 nm), respectively. The white light emitting phosphor, Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (1% Tm³⁺, 2% Tb³⁺, and 1% Eu³⁺), yielded a PLQY value of 5%. Similar values have been reported in the literature.³⁰

Magnetic studies

We substituted the Co²⁺ ions in Na₂Mg₃Co₂Si₁₂O₃₀ compounds and investigated the magnetic behavior in the temperature range of 2 K to 300 K (Fig. S12).

The molar magnetic susceptibility (χ_m) increases with decreasing temperature, exhibiting a sharp rise below 50 K. The room temperature molar magnetic susceptibility was found to be 0.23 emu mol⁻¹, which increases to a value of 1.38 emu mol⁻¹ at 4 K. The inverse susceptibility (1/χ_m) data were fitted in two temperature ranges: (i) in the range of 50–150 K and (ii) in the range of 150–300 K, employing the Curie–Weiss law.⁸⁴

The fitting yielded a Curie constant (C) of 18.03 emu mol⁻¹ K and a Curie–Weiss temperature (θ) of −140.9 K in the temperature range of 150–300 K. The fit in the temperature range of 50–150 K gave a Curie constant (C) value of 11.03 emu mol⁻¹ K⁻¹ and a Curie–Weiss temperature (θ) of −26.8 K. The negative value of θ at both the temperature ranges suggests the presence of antiferromagnetic interactions between the Co²⁺ ions in the compound.

The effective magnetic moment (μ_eff) values were also calculated from the Curie constant using the relationship, . The μ_eff values obtained are 5.98μ_B in the range 150–300 K and 4.685μ_B in the range 50–150 K, which are higher than the spin-only magnetic moment values of 3.87μ_B for the high-spin Co²⁺ ions (S = 3/2). The observed magnetic moment value appears to be marginally larger (150 K–300 K) than the spin-only values of the tetrahedral Co²⁺ ions. These larger magnetic moment values observed may be due to some orbital contribution, which is known for Co²⁺ ions (Fig. S12c).⁸⁴

Electrocatalytic studies

Cobalt(II)-based oxides have emerged as a promising class of electrocatalysts, especially towards the oxygen evolution reaction (OER).^85–89 In many of the studies, the Co²⁺ ions were found to have an octahedral, square-pyramidal or trigonal–bipyramidal geometry.^85–89 Octahedral Co²⁺ ions are generally more active in OER catalysis due to their ease of transition to higher oxidation states (Co³⁺/Co⁴⁺), which is needed for oxygen evolution.^90,91 The Co²⁺ ions in tetrahedral coordination were also explored towards the OER, but the number of such reports is limited.⁹² In the present study the compound Na₂Mg₃Co₂Si₁₂O₃₀ was explored towards the possible electrocatalysis in the OER under alkaline conditions. We estimated the band gap from the UV-Vis studies, which gave a value of 1.28 eV for Na₂Mg₃Co₂Si₁₂O₃₀ (Fig. S13). This value suggested that the compound can be a potential candidate towards the OER. The catalyst ink for the electrocatalytic studies was prepared by mixing the Na₂Mg₃Co₂Si₁₂O₃₀ compound with acetylene black in a 4 : 1 ratio. 5% Nafion was also added to the catalyst. The entire mixture was taken in an ethanol : water (2 : 3) solution and sonicated for 60 minutes. This solution was dropcast onto the glassy carbon electrode in small increments to ensure a uniform and homogeneous coating of the electrode surface. The electrode was dried under an infrared lamp to remove any residual solvent.

The reaction was carried out in 0.5 M KOH solution. The entire solution was bubbled with Ar gas for 30 min at room temperature to remove any dissolved oxygen before starting the reaction. The electrocatalytic reaction was carried out employing a three-electrode set up. The glassy carbon electrode, containing the catalyst, was the working electrode, mercury/mercuric oxide and graphite rod were employed as the reference and counter electrodes. The electrocatalytic studies were screened by linear scan voltammetry (LSV) with a scan rate of 10 mV s⁻¹. The LSV indicated that the oxygen evolution was observed at a overpotential of 1.54 V vs. RHE. The LSV studies were also compared with those with RuO₂, which exhibited a overpotential of 1.43 V vs. RHE (Fig. S14). The kinetics of the electrocatalytic behavior was obtained from the Tafel plot (Fig. 13).

Fig. 13 (a) Linear sweep voltammogram (LSV), (b) Tafel plot, (c) Nyquist plot, and (d) chronoamperometric studies on the Na₂Mg₃Co₂Si₁₂O₃₀ compound.

The slope for the Na₂Mg₃Co₂Si₁₂O₃₀ compound was found to be 92.86 mV dec⁻¹, which is marginally higher than that observed for RuO₂ (90.37 mV dec⁻¹) (Fig. 13b). These values suggest that the charge transfer kinetics of the Na₂Mg₃Co₂Si₁₂O₃₀ compound is comparable to that of RuO₂.

Electrochemical impedance spectroscopy (EIS) studies were carried out to understand the interfacial charge transfer behavior of Na₂Mg₃Co₂Si₁₂O₃₀ during the oxygen evolution reaction (OER).⁹³ The studies were carried out in the AC field with frequencies in the range of 0.1 Hz–1 MHz (Fig. 13c). The Nyquist plot exhibits typical behavior where the electrode processes involve both the charge transfer and double-layer capacitance effects.^94,95 The experimental data were fitted employing an equivalent circuit model, the Randles’ circuit, comprising R₁, R₂, and CPE₁. In this model, R₁ corresponds to the uncompensated solution resistance, while R₂ represents the charge transfer resistance at the electrode–electrolyte interface. The constant phase elements, CPE₁, account for the non-ideal capacitive behavior (Table S8). The low value of R₂ indicates reasonable charge transfer kinetics, which appears to be consistent with the electrocatalytic activity observed both in LSV and Tafel analysis. The reasonable agreement between the fitted curve and the experimental data suggests that the employed model is good (Fig. 13c and Fig. S15).

The long-term electrochemical stability was evaluated employing chronoamperometric measurements at a fixed applied potential (1.647 V vs. RHE). The compound exhibited good OER activity without a drop in current for up to 12 h (Fig. 13d). The LSV curves recorded before and after the 12-hour chronoamperometry test show minimal changes, indicating that the catalyst maintains its electrocatalytic activity over time. This consistency suggests that the material remains stable and effective for oxygen evolution reaction (OER) applications under prolonged operational conditions (Fig. 13d).

The XPS studies were carried out on the Na₂Mg₃Co₂Si₁₂O₃₀ compound to learn about the oxidation state of Co²⁺ ions. The XPS spectra exhibited two peaks at ∼782.5 eV and ∼798.2 eV, which may be assigned to the 2p_3/2 and 2p_1/2 binding energies of Co²⁺ ions (Fig. S16a).^29,54,96 As can be observed, in addition to the main peaks, a prominent satellite was also observed at ∼787.7 eV (∼5.2 eV higher than 2p_3/2) for Co-2p_3/2. This is suggestive of Co²⁺ ions in a high spin state.^97–99 The separation between the 2p_3/2 and 2p_1/2 binding energies was observed to be ∼15.7 eV which also suggests the presence of Co²⁺ ions in the tetrahedral coordination.¹⁰⁰

The XPS spectra of the Na₂Mg₃Co₂Si₁₂O₃₀ compound was also recorded after 1000 cycles of cyclovoltammetry. We observed a similar behavior for the Co²⁺ ions with 2p_3/2 and 2p_1/2 peaks at ∼782.2 eV and ∼797.8 eV, respectively. As can be noted, the separation between the two XPS peaks was found to be ∼15.6 eV. This also indicates that there is no change in the coordination environment of the Co²⁺ ions in Na₂Mg₃Co₂Si₁₂O₃₀. We observed a peak at ∼781 eV and at ∼796.4 eV, which is suggestive of the binding energies of 2p_3/2 and 2p_1/2 states of Co³⁺ ions. It is likely that the Co²⁺ ions undergo oxidation during the electrocatalytic reaction. The weak satellite peak observed at ∼791 eV is also suggestive of Co³⁺ (3d⁶). The ratio of Co²⁺/Co³⁺ can be calculated from the 2p_3/2 peak, which indicates a value of ∼6.5 : 1 (Fig. S16b).

The electrochemically active surface area (ECSA) of Na₂Mg₃Co₂Si₁₂O₃₀ was estimated by analyzing the capacitive current response in the non-faradaic region using cyclic voltammetry at varying scan rates (0.01–0.1 V s⁻¹). The capacitive current density vs. the scan rate (V s⁻¹), gives a linear relationship (Fig. S17). The slope of the linear fit corresponds to the double-layer capacitance (C_dl), which was found to be 0.702 mF cm⁻². Based on the standard specific capacitance value of 40 μF cm⁻² for a flat surface in 0.5 M KOH, the ECSA was calculated using the relationship, ECSA = C_dl/C_s, resulting in a value of 17.6. The large surface area suggests that there are large numbers of electrochemically accessible sites at the interface.^32,101

The turnover frequency (TOF) for Na₂Mg₃Co₂Si₁₂O₃₀ was determined using linear scan voltammetry (LSV) performed at a scan rate of 10 mV s⁻¹ in 0.5 M KOH (Fig. S18). The TOF quantifies the rate at which catalytic sites facilitate the conversion of the reactants to the products. Based on the measured current densities and the calculated number of active centers from the redox peak integration, the TOF at the relevant potential was determined to be 6.01 s⁻¹ for a 10 mA cm⁻² current density.³² Similar values have been reported before.^102–104

XPS studies

We carried out XPS studies on all the highest transition metal substituted milarite compounds to understand the oxidation states of the substituted ions and to obtain possible clues about the chemical environment present in the as-prepared samples. We examined Na₂Mg₃Co₂Si₁₂O₃₀, Na₂Mg₃CuZnSi₁₂O₃₀, and Na₂Mg_4.5Ni_0.5Si₁₂O₃₀ compounds by XPS studies (Fig. S4 and S16).

The Cu²⁺ ions in the compound, Na₂Mg₃CuZnSi₁₂O₃₀, exhibited two main peaks at ∼933.3 eV and ∼952.7 eV, corresponding to the 2p_3/2 and 2p_1/2 binding energies of Cu²⁺ ions.^105–107 Cu²⁺ ions exhibit strong satellite features in the XPS spectra, which are not generally observed for Cu⁺ ions.¹⁰⁸ We observed strong satellite features at ∼9 eV higher for 2p_3/2 and 2p_1/2 at ∼942.2 eV and ∼961.6 eV, respectively. The difference between the two binding energies both for the main peaks and the satellite peaks was observed to be ∼19.3 eV, which is suggestive of Cu²⁺ ions in tetrahedral coordination.¹⁰⁹

The XPS spectra of Ni-2p in Na₂Mg_4.5Ni_0.5Si₁₂O₃₀ exhibited two main peaks at ∼856.3 eV and at ∼873.9 eV which correspond to the binding energies of 2p_3/2 and 2p_1/2 for Ni²⁺ ions, respectively. In addition, we also observed satellite peaks at ∼861.9 eV and ∼879.9 eV, which is also suggestive of the presence of Ni²⁺ ions. The separation between the two binding energies (2p_3/2 and 2p_1/2) was found to be ∼17.6 eV, which indicates the Ni²⁺ ions may be present in a tetrahedral coordination environment.^54,110

Dielectric studies

Room temperature dielectric behavior was examined on the white compounds: Na₂Mg₅Si₁₂O₃₀, Na₂Mg₄ZnSi₁₂O₃₀, Na₂Mg₃Zn₂Si₁₂O₃₀, and Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ (Fig. 14). All the compounds exhibited high dielectric constant values, with Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀ exhibiting the highest dielectric constant. In all the cases, we observed a decrease in the dielectric constant values as the applied frequency was increased (Fig. 14a). This dielectric behavior has been commonly observed in many ceramic oxides. The main factors responsible for this behavior could be the density, porosity, grain boundaries, and the defects present in the compound along with the changes that occur due to the different types of polarizations (ionic, dipolar, electronic, and interfacial), during the increase in the applied field.^111–113 The values for the dielectric constant, at a low field, were found to be larger in Na₂Mg_2.5Ca_0.5Zn₂Si₁₂O₃₀, which may be due to the higher ionic polarizability of Ca²⁺ ions compared to that of the Mg²⁺ ions. This can give rise to additional lattice distortion, with enhanced displacement polarization.¹¹⁴ One of the notable points is that the compounds consistently exhibit a lower dielectric loss (Fig. 14b). This suggests that these compounds could have the potential to be employed in electronic communication related purposes.

Fig. 14 (a) The variation of the dielectric constant as a function of frequency and (b) the dielectric loss as a function of frequency. Note the small loss for all the compounds.

Conclusions

In this study, the synthesis and structural characterization of a number of new analogues of the milarite structure have been accomplished. The substitution of transition elements in the milarite structure results in compounds with interesting colors under daylight. XPS and Raman spectroscopic studies confirm the presence of transition elements with the expected oxidation states in the compounds. The white compounds exhibited (i) good dielectric constant values with low dielectric loss; (ii) NIR reflectivity values comparable to that of TiO₂ and (iii) deep UV-cut off values. These indicate the utility of milarite compounds in electronic communication, UV-blocking agents and thermal shielding. The lanthanide doped compounds gave good luminescence behavior in the red (Eu³⁺), green (Tb³⁺) and blue (Tm³⁺) regions. A suitable combination of these elements resulted in white light emission. The Co²⁺-substituted Na₂Mg₃Co₂Si₁₂O₃₀ compound also was found to be a good electrocatalyst towards the oxygen evolution reaction (OER). The present study illustrates the utility of exploring the mineral structures towards different physical and chemical properties, which opens a newer avenue in material research.

Author contributions

The synthesis of the compounds, characterisation of the compounds (PXRD, UV-visible optical studies, Raman studies, and dielectric studies), photoluminescence studies, and electrochemical studies were carried out by DM. The overall scientific problem was conceptualized by SN.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as a part of the supplementary information (SI). Supplementary information: S1–S3: PXRD, Rietveld refinement and crystal structures of the as-synthesized compounds; S4 and S5: XPS and Raman spectra; S6–S8: Optical studies; S9–S11: Photoluminescence studies; S12: Magnetic studies; and S13–S18: Electrochemical studies of compounds. See DOI: https://doi.org/10.1039/d5dt01850a.

Acknowledgements

S. Natarajan thanks the Science and Engineering Research Board (SERB), Govt. of India, for a research grant and for a JC Bose national fellowship.

References

1. C. N. R. Rao and J. Gopalakrishnan, New Directions in Solid State Chemistry, Cambridge University Press, 1997.
2. J. M. Rondinelli, K. R. Poeppelmeier and A. Zunger, APL Mater., 2015, 3(8), 080701.
3. J. M. Rondinelli and E. Kioupakis, Annu. Rev. Mater. Res., 2015, 45(1), 491–518.
4. J. P. Smit, P. C. Stair and K. R. Poeppelmeier, Chem. – Eur. J., 2006, 12(23), 5944–5953.
5. Z. Xia and K. R. Poeppelmeier, Acc. Chem. Res., 2017, 50(5), 1222–1230.
6. R. E. Schaak and T. E. Mallouk, Chem. Mater., 2002, 14(4), 1455–1471.
7. A. G. Slater and A. I. Cooper, Science, 2015, 348(6238), aaa8075.
8. Y.-C. Lin, M. Karlsson and M. Bettinelli, Photoluminescent Mater. Electroluminescent Devices, 2017, pp. 309–355.
9. S. M. Woodley and R. Catlow, Nat. Mater., 2008, 7(12), 937–946.
10. F. Liebau, Struct. Chem. Silicates: Struct. Bonding Classif, 1985, pp. 14–51.
11. H. Bennett, Analyst, 1977, 102(1212), 153–179.
12. D. Y. Pushcharovsky, N. V. Zubkova and I. V. Pekov, Struct. Chem., 2016, 27, 1593–1603.
13. E. Pérez-Botella, S. Valencia and F. Rey, Chem. Rev., 2022, 122(24), 17647–17695.
14. J. Weitkamp, Solid State Ionics, 2000, 131(1–2), 175–188.
15. Y. Li and J. Yu, Nat. Rev. Mater., 2021, 6(12), 1156–1174.
16. M. C. Day and F. C. Hawthorne, Mineral. Mag., 2020, 84(2), 165–244.
17. N. V. Zubkova, I. V. Pekov and D. Y. Pushcharovsky, Mineral. Mag., 2014, 78(2), 253–265.
18. F. C. Hawthorne, Y. A. Uvarova and E. Sokolova, Mineral. Mag., 2019, 83(1), 3–55.
19. J. J. Papike, Rev. Geophys., 1988, 26(3), 407–444.
20. W. C. Forbes, W. H. Baur and A. A. Khan, Am. Mineral., 1972, 57(3–4, Part 1), 463–472.
21. P. Cerny, F. C. Hawthorne and E. Jarosewich, Can. Mineral., 1980, 18(1), 41–57.
22. O. C. Gagné and F. C. Hawthorne, Can. Mineral., 2016, 54(5), 1229–1247.
23. M. A. Cooper, F. C. Hawthorne and E. S. Grew, Am. Mineral., 1999, 84(5–6), 764–768.
24. C. Ferraris, I. Pignatelli, F. Camara, G. Parodi, S. Pont, M. Schreyer and F. Wei, Eur. J. Mineral., 2020, 32(3), 355–365.
25. C. L. Lengauer, N. Hrauda, U. Kolitsch, R. Krickl and E. Tillmanns, Mineral. Petrol., 2009, 96, 221–232.
26. (a) N. Nguyen, J. Choisnet and B. Raveau, J. Solid State Chem., 1980, 34(1), 1–9; (b) L. Qin, C. Chen, J. Wang, S. Bi, Y. Huang and H. J. Seo, Mater. Res. Bull., 2019, 118, 110494; (c) G. V. Kanmani, V. Ponnusamy, G. Rajkumar and M. T. Jose, Opt. Mater., 2019, 96, 109350.
27. W. Postl, F. Walter, K. Ettinger, C. Hauzenberger and H.-P. Bojar, Eur. J. Mineral., 2004, 16(2), 375–380.
28. B. Röska, I. Akter, M. Hoelzel and S.-H. Park, J. Solid State Chem., 2018, 264, 98–107.
29. D. Malik and S. Natarajan, Chem. – Asian J., 2025, 20(6), e202401230.
30. A. Bhim, A. Mohanty and S. Natarajan, Dalton Trans., 2020, 49(48), 17649–17657.
31. A. Bhim, J. Gopalakrishnan and S. Natarajan, Eur. J. Inorg. Chem., 2018, 2018(20–21), 2277–2284.
32. I. G. Shanmugapriya, S. Sa and S. Natarajan, Dalton Trans., 2025, 54(7), 2753–2764.
33. A. Bhim, S. Laha, J. Gopalakrishnan and S. Natarajan, Chem. – Asian J., 2017, 12(20), 2734–2743.
34. S. Gurursamy, A. T. M. Munthasir, P. Thilagar and S. Natarajan, Chem. Mater., 2024, 36(11), 5356–5369.
35. A. Bhim, W. Zhang, P. S. Halasyamani, J. Gopalakrishnan and S. Natarajan, Inorg. Chem., 2019, 58(13), 8560–8569.
36. S. Gurusamy, S. Banerjee, A. Sundaresan, M. Liang, P. S. Halasyamani and S. Natarajan, Chem. – Asian J., 2024, 19(6), e202301113.
37. E. Kroumova, M. I. Aroyo, J. M. Perez-Mato, A. Kirov, C. Capillas, S. Ivantchev and H. Wondratschek, Phase Transitions, 2003, 76(1–2), 155–170.
38. T. A. Sidorov, Russ. J. Inorg. Chem., 2007, 52, 1586–1594.
39. D. L. Rousseau, R. P. Bauman and S. P. S. Porto, J. Raman Spectrosc., 1981, 10(1), 253–290.
40. M. Bagheri and H. Komsa, J. Raman Spectrosc., 2024, 55(10), 1113–1123.
41. G. Ulian, D. Moro and G. Valdrè, Appl. Clay Sci., 2020, 197, 105779.
42. S. Dey, R. A. Ricciardo, H. L. Cuthbert and P. M. Woodward, Inorg. Chem., 2014, 53(9), 4394–4399.
43. G. Pfaff, ChemTexts, 2022, 8(3), 15.
44. W. Zhou, J. Ye, Z. Liu, L. Wang, L. Chen, S. Zhuo, Y. Liu and W. Chen, Inorg. Chem., 2021, 61(1), 693–699.
45. V. Naresh and S. Buddhudu, Ferroelectrics, 2012, 437(1), 110–125.
46. A. Akhrouf, F. E. Lahnine, Y. Boudad, M. Salmi, A. Amarray, N. Sajai, R. Fakhreddine, S. El Ghachtouli and S. Krimi, Mater. Res. Bull., 2024, 180, 113020.
47. A. Bellmann, C. Rautenberg, U. Bentrup and A. Brückner, Catalysts, 2020, 10(1), 1.
48. A. Ibrahim and M. S. Sadeq, Ceram. Int., 2021, 47(20), 28536–28542.
49. M. Dondi, M. Ardit, G. Cruciani and C. Zanelli, Am. Mineral., 2014, 99(8–9), 1736–1745.
50. Z. Wang, Q. Wang, Q. Liu, S. Suthirakun, W. Kaewraung, P. Jiang, H. Zhao, X. Xin and M. A. Subramanian, ACS Appl. Opt. Mater., 2024, 2(2), 313–322.
51. S. Tamilarasan, D. Sarma, S. Bhattacharjee, U. V. Waghmare, S. Natarajan and J. Gopalakrishnan, Inorg. Chem., 2013, 52(10), 5757–5763.
52. D. N. Petrov, N. T. Dang, N. D. Co, B. D. Tu, N. D. Lam, T. V. Quang, V. Q. Nguyen, J. H. Lee, B. T. Huy, D. S. Yang, D. T. Khan and T. L. Phan, J. Mater. Sci., 2024, 59(21), 9217–9236.
53. P. K. Thejus, B. Koley and K. G. Nishanth, Dyes Pigm., 2018, 158, 267–276.
54. D. Malik and S. Natarajan, Eur. J. Inorg. Chem., 2023, 2023(31), e202300305.
55. S. Zhao, P. Gong, S. Luo, L. Bai, Z. Lin, C. Ji, T. Chen, M. Hong and J. Luo, J. Am. Chem. Soc., 2014, 136(24), 8560–8563.
56. S. Volkov, R. Bubnova, A. Povolotskiy, V. Ugolkov and M. Arsent'ev, J. Solid State Chem., 2020, 281, 121023.
57. W. Ma, J. Zhang, F. Yu and B. Dai, Inorg. Chem., 2022, 61(42), 16533–16538.
58. G. Blasse, Prog. Solid State Chem., 1988, 18(2), 79–171.
59. X. Qin, X. Liu, W. Huang, M. Bettinelli and X. Liu, Chem. Rev., 2017, 117(5), 4488–4527.
60. S. SeethaLekshmi, A. R. Ramya, M. L. P. Reddy and S. Varughese, J. Photochem. Photobiol., C, 2017, 33, 109–131.
61. Y. Zhang, W. Gong, J. Yu, Y. Lin and G. Ning, RSC Adv., 2015, 5(117), 96272–96280.
62. K. Pavani, J. S. Kumar and L. R. Moorthy, J. Alloys Compd., 2014, 586, 722–729.
63. I. Ayoub, U. Mushtaq, G. B. Nair, G. Sundaram, H. C. Swart and V. Kumar, Small, 2025, 21(13), e2500285.
64. J. Zheng, Q. Cheng, S. Wu, C. Rong, L. Cai and C. Chao, J. Rare Earths, 2015, 33(9), 933–938.
65. X. Ju, X. Li, W. Li, C. Tao and J. Yang, Mater. Lett., 2012, 77, 35–37.
66. A. S. Altowyan, M. B. Çoban, U. H. Kaynar, J. Hakami, E. A. Çin, S. C. Kaynar, M. Ayvacikli and N. Can, Ceram. Int., 2024, 50(13), 24036–24044.
67. W. Ma and X. Wang, Luminescence, 2022, 37(5), 722–728.
68. N. Kucuk, Ü. H. Kaynar, S. Akca, Y. Alajlani, L. Yin, Y. Wang, J. G. Guinea, K. Bulcar, T. Dogan and Y. Karabulut, J. Alloys Compd., 2020, 838, 155587.
69. G. Blasse and B. C. Grabmaier, A general introduction to luminescent materials, Springer, 1994.
70. L. Zhang, S. Yi, X. Hu, B. Liang, W. Zhao and Y. Wang, Mod. Phys. Lett. B, 2017, 31(08), 1750051.
71. N. Guo, Q. Liang, S. Li, R. Ouyang and W. Lü, Opt. Mater., 2017, 73, 570–576.
72. I. Carrasco, F. Piccinelli and M. Bettinelli, J. Lumin., 2017, 189, 71–77.
73. B. Wang, Q. Ren, O. Hai and X. Wu, RSC Adv., 2017, 7(25), 15222–15227.
74. V. Naresh, K. Gupta, C. P. Reddy and B. S. Ham, Spectrochim. Acta, Part A, 2017, 175, 43–50.
75. X. Huo, Z. Wang, C. Tao, N. Zhang, D. Wang, J. Zhao, Z. Yang and P. Li, J. Alloys Compd., 2022, 902, 163823.
76. Q. Ren, Y. Zhao, X. Wu and O. Hai, Opt. Laser Technol., 2021, 136, 106774.
77. G. Blasse, Phys. Lett. A, 1968, 28(6), 444–445.
78. Y. Feng, J. Huang, C. Li, G. Hu, J. Liu and X. Yu, J. Alloys Compd., 2017, 706, 478–484.
79. Y. Liu, J. Wang, X. Liu, H. Zhang and H. Liu, Opt. Mater., 2019, 95, 109194.
80. M. Mungra, F. Steudel, B. Ahrens and S. Schweizer, J. Lumin., 2017, 192, 71–76.
81. M. S. Mendhe, S. P. Puppalwar and S. J. Dhoble, Opt. Mater., 2018, 82, 47–55.
82. M. Xia, X. Wu, Y. Zhong, H. T. B. Hintzen, Z. Zhou and J. Wang, J. Mater. Chem. C, 2019, 7(10), 2927–2935.
83. B. Li, X. Huang, H. Guo and Y. Zeng, Dyes Pigm., 2018, 150, 67–72.
84. (a) S. Mugiraneza and A. M. Hallas, Commun. Phys., 2022, 5(1), 95; (b) H. J. Koo, R. K. Kremer and M.-H. Whangbo, Inorg. Chem., 2020, 59(24), 18319–18324.
85. Y. Zheng, R. Gao, Y. Qiu, L. Zheng, Z. Hu and X. Liu, Inorg. Chem., 2021, 60(7), 5252–5263.
86. B. Qiu, C. Wang, N. Zhang, L. Cai, Y. Xiong and Y. Chai, ACS Catal., 2019, 9(7), 6484–6490.
87. Z. Liu, G. Wang, X. Zhu, Y. Wang, Y. Zou, S. Zang and S. Wang, Angew. Chem., 2020, 132(12), 4766–4772.
88. J. Wang, R. Gao, D. Zhou, Z. Chen, Z. Wu, G. Schumacher, Z. Hu and X. Liu, ACS Catal., 2017, 7(10), 6533–6541.
89. X. Han, G. He, Y. He, J. Zhang, X. Zheng, L. Li, C. Zhong, W. Hu, Y. Deng and T. Ma, Adv. Energy Mater., 2018, 8(10), 1702222.
90. C. Wei, Z. Feng, G. G. Scherer, J. Barber, Y. Shao-Horn and Z. J. Xu, Adv. Mater., 2017, 29(23), 1606800.
91. P. W. Menezes, A. Indra, A. Bergmann, P. Chernev, C. Walter, H. Dau, P. Strasser and M. Driess, J. Mater. Chem. A, 2016, 4(25), 10014–10022.
92. K. Zhang and R. Zou, Small, 2021, 17(37), 2100129.
93. B. Y. Chang and S. M. Park, Annu. Rev. Anal. Chem., 2010, 3(1), 207–229.
94. S. Anantharaj and S. Noda, ChemElectroChem, 2020, 7(10), 2297–2308.
95. M. Van Haeverbeke, M. Stock and B. De Baets, IEEE Access, 2022, 10, 51363–51379.
96. J. G. Kim, D. L. Pugmire, D. Battaglia and M. A. Langell, Appl. Surf. Sci., 2000, 165(1), 70–84.
97. P. N. Anantharamaiah and P. A. Joy, Physica B: Condens. Matter, 2019, 554, 107–113.
98. T. Baidya, T. Murayama, S. Nellaiappan, N. K. Katiyar, P. Bera, O. Safonova, M. Lin, K. R. Priolkar, S. Kundu, B. S. Rao and P. Steiger, J. Phys. Chem. C, 2019, 123(32), 19557–19571.
99. X. Duan, M. Pan, F. Yu and D. Yuan, J. Alloys Compd., 2011, 509(3), 1079–1083.
100. T. J. Chuang, C. R. Brundle and D. W. Rice, Surf. Sci., 1976, 59(2), 413–429.
101. S. A. T. Shanmugapriya, A. P. Singh, R. G. Chaudhary, S. Mondal, A. Fatehmulla, P. Hadole and A. Mondal, Ionics, 2024, 30(7), 4099–4109.
102. G. Feng, S. Li, J. Zheng, X. Zhong, S. Wang and J. Wang, J. Catal., 2025, 443, 115968.
103. S. Anantharaj, P. E. Karthik and S. Noda, Angew. Chem., Int. Ed., 2021, 60(43), 23051–23067.
104. C. C. L. McCrory, S. Jung, J. C. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2013, 135, 16977.
105. J. J. Li, J. Zeng, L. Jia and W. Fang, Int. J. Hydrogen Energy, 2010, 35(23), 12733–12740.
106. J. A. Torres-Ochoa, D. Cabrera-German, O. Cortazar-Martinez, M. Bravo-Sanchez, G. Gomez-Sosa and A. Herrera-Gomez, Appl. Surf. Sci., 2023, 622, 156960.
107. D. P. Dutta and P. Raval, J. Photochem. Photobiol., A, 2018, 357(C), 193–200.
108. E. Nicolas-Tolentino, Z. R. Tian, H. Zhou, G. Xia and S. L. Suib, Chem. Mater., 1999, 11(7), 1733–1741.
109. P. Bera, K. R. Priolkar, P. R. Sarode, M. S. Hegde, S. Emura, R. Kumashiro and N. P. Lalla, Chem. Mater., 2002, 14(8), 3591–3601.
110. M. J. Thompson, K. J. Blakeney, S. D. Cady, M. D. Reichert, J. D. Pilar-Albaladejo, S. T. White and J. Vela, Chem. Mater., 2016, 28(6), 1668–1677.
111. J. Robertson, Eur. Phys. J.: Appl. Phys., 2004, 28(3), 265–291.
112. S. Singh, P. Dey, J. N. Roy and S. K. Mandal, J. Alloys Compd., 2015, 642, 15–21.
113. M. T. Sebastian and H. Jantunen, Int. Mater. Rev., 2008, 53(2), 57–90.
114. I. S. Cho, Mater. Sci. Semicond. Process., 2022, 148, 106793.

---