Rare 3 × 3 Layering in High‐Pressure Pseudo‐Perovskite Sr4Te4O15

TL;DR

A new strontium tellurate compound, Sr4Te4O15, was synthesized under high pressure and temperature, revealing unique structural and electronic properties.

Contribution

The study reports the synthesis and characterization of a rare pseudo-perovskite with 3 × 3 corrugated oxotellurate layers.

Findings

Sr4Te4O15 crystallizes in the orthorhombic space group Pnma with specific unit cell parameters.

The compound exhibits direct and indirect bandgaps of 3.17 and 2.97 eV, respectively.

Thermal analysis shows Sr4Te4O15 transforms to SrTeO4 at elevated temperatures.

Abstract

A novel strontium tellurate, Sr4Te4O15, has been synthesized under high‐pressure/high‐temperature conditions using multianvil techniques (9 GPa, 700°C). This unique composition constitutes both Te4+ and Te6+ sites and crystallizes in the orthorhombic space group Pnma with unit cell parameters a = 1332.4(1), b =1236.4(1), and c = 748.02(9) pm. The crystal structure of Sr4Te4O15 consists of heavily corrugated oxotellurate sheets made up of corner‐sharing [TeO6]6− octahedra and [TeO4]4− bisphenoids separated by layers of Sr2+. Some of its structural features are comparable to those of 2D perovskites, and the deviation of the compound from this class is discussed in terms of both structure and composition. The stability of this compound was investigated through high‐temperature powder X‐ray diffraction and thermal analysis revealing its transformation to SrTeO4 at elevated temperatures. Comprehensive characterization, including UV‐Vis, IR, and Raman spectroscopy, as well as density functional theory calculations, was performed to investigate its electronic structure, vibrational properties, and electron localization of the Te4+ lone electron pairs. The direct and indirect bandgaps were determined to be 3.17 and 2.97 eV, respectively. The study provides insights into the structural and electronic properties of Sr4Te4O15, contributing to the understanding of rare corrugated 2D pseudo‐perovskite systems.

High‐pressure/high‐temperature synthesis, crystal structure determination, and characterization of Sr4Te4O15, a rare (110)‐oriented pseudo‐perovskite with 3 × 3 corrugated oxotellurate layers, are presented. Thermal stability was studied by temperature‐dependent X‐ray powder diffraction. UV‐Vis, IR, and Raman spectroscopy, as well as density functional theory calculations, were performed to investigate the electronic structure, vibrational properties, and electron localization of the TeIV lone electron pairs.© 2026 WILEY‐VCH GmbH

Figures

Schematic representation of flat (100)‐oriented, 2 × 2 and 3 × 3 corrugated ((110)‐oriented) perovskite sheets cut out from a 3D perovskite model.

PXRD pattern (Mo‐K α1 radiation) and Rietveld plot of Sr4Te4O15 (R exp = 0.88%, R wp = 1.36%, R p = 0.93% and GoF = 1.54). Asterisk marks indicate minor reflections belonging to SrTe3O8.

Coordination geometries and bond lengths of both Te and Sr sites of Sr4Te4O15. Distances are given in pm and all standard deviations are below 0.15 pm. Sr2 and Sr3 are displayed in an orientation that makes it easier to visualize their augmented trigonal prismatic geometry (trigonal prism composed of O2‐O5‐O6 and O3‐O2‐O1 faces respectively).

Crystal structure diagram of Sr4Te4O15 showing a 3 × 3 corrugation pattern of its OS stacked along the a‐axis.

Top: A comparison of an OS of Sr4Te4O15 (top left) with an equivalent hypothetical 3 × 3 cutout diagram of the 3D perovskite SrTiO3 (top right). The aqua‐colored octahedra of SrTiO3 indicate sites replaced by bisphenoidally coordinated Te6+ in Sr4Te4O15. The numbering of the octahedra for SrTiO3 (in yellow) aids the visualization of the 3 × 3 nature of the corrugation. Bottom: An enlarged section of the OS of Sr4Te4O15 is shown. Teal colored bonds delineate tilting with respect to both the bc and ac planes. Full and faded out teal‐colored bonds show tilting into and out of the bc plane (refer to Figure 7 for tilting of Te1‐Te2 units). The shortest octahedral edge of the structure is marked at 257.7 pm for the edge O1–O1 for [Te1O6]6−.

Diagram showing the interaction of the OS with Sr2+ (Sr2 and Sr3) positioned in the grooves created by the crests/troughs of the OS. Red colored arrows denote the tilting directions of O6 and O3 towards Sr2+ parallel to both the bc and ac plane. The tilting of crest/trough octahedra parallel to the ac plane is not clear here but is shown more clearly in Figure 7. The [Te4+O4]4− polyhedra have been omitted from this diagram to allow clear visualization of the Sr2+ coordination.

Diagram showing a side (ac plane) view of the crest/trough of the corrugated structure of Sr4Te4O15. Te6+ (dark blue) and Te4+ (aqua) consist of Te1 and Te2 respectively. Only the oxygen sites involved in coordination with Sr2+ sitting at the crest/trough are marked. The red arrows show the tilting direction of the respective oxygen atoms towards the corresponding Sr2+ sites. This tilting causes a Te1–O8–Te2 bending angle of 129.8° and a Te2–O7–Te1 bending angle of 129.3°. The total distance between the tilted oxygen atoms O7 and O8 is shown as 440.3 pm.

HT‐PXRD pattern of Sr4Te4O15 showing its thermally induced transformation to ambient pressure strontium tellurate phases.

Overlay of the experimental (black) and theoretical (red) infrared spectra of Sr4Te4O15. The absorbance scaling of the theoretical spectrum has been adjusted for clearer visualization of the correlation between the two spectra, although both spectra have been standardized relative to their strongest absorption peaks in the region displayed.

Overlay of the experimental (black) and theoretical (red) Raman spectra of Sr4Te4O15. The absorbance scaling of the theoretical spectrum has been adjusted for clearer visualization of the correlation between the two spectra, although both spectra have been standardized relative to their strongest absorption peaks in the region displayed. The Raman bands from 200 to 600 cm−1 are listed in this figure caption matched with the labeling in the figure: (i) 210 cm−1, (ii) 236 cm−1, (iii) 252 cm−1, (iv) 268 cm−1, (v) 288 cm−1, (vi) 319 cm−1, (vii) 331 cm−1, (viii) 338 cm−1, (ix) 379 cm−1, (x) 403 cm−1, (xi) 421 cm−1, (xii) 432 cm−1, (xiii) 472 cm−1, (xiv) 493 cm−1, (xv) 504 cm−1, (xvi) 597 cm−1.

UV–Visible spectrum of a sample of Sr4Te4O15 with an insertion showing the Tauc plot for the estimation of the direct (3.17 eV) and indirect (2.97 eV) bandgaps.

Electronic band structure (left) and associated density of states (right) of Sr4Te4O15. obtained at HSEsol level of theory. A direct bandgap energy of 2.97 eV can be determined at the Γ‐point.

Graphic showing a visualization of the ELF in relation to the determined crystal structure and Te4+ positions. Encircled regions denote an allocation of the ELF graphic to the respective crystal structure viewpoint to simplify interpretation. Both the crystal structure and the ELF images were obtained in the direction shown by the coordinates (i.e., structure rotated by approximately 45° around a).

Tables

TABLE 1. Crystal data, data collection, and structure refinement results for Sr4Te4O15 (standard deviation in parentheses).

Empirical formula	Sr 4 Te 4 O 15
Molar mass, g·mol−1	1100.88
Crystal system	orthorhombic
Space group	Pnma
Cell formula units, Z	4
Powder data:
Powder diffractometer	STOE Stadi P
Radiation	Mo‐K α1 (λ = 70.93 pm)
a, pm	1330.85(2)
b, pm	1234.73(2)
c, pm	747.32(1)
V, nm3	1.22803
Single‐crystal data:
Single‐crystal diffractometer	Bruker D8 Quest
Radiation	Mo‐K α (λ = 71.073 pm)
a, pm	1332.4(1)
b, pm	1236.4(1)
c, pm	748.02(9)
V, nm3	1.2322
Calculated density, g·cm−3	5.934
Crystal size, mm3	0.05 × 0.04 × 0.025
Temperature, K	299.00
Absorption coefficient, mm−1	26.595
F(000), e	1920
Detector distance, mm	38
θ range, deg	3.058–37.496
Range in hkl	±22, ±21, ±12
Total no. reflections	118 616
Data/ref. parameters	3361/119
Reflections with I > 2σ(I)	3246
R int /R σ	0.0434/0.0124
Goodness‐of‐fit on F 2	1.174
Absorption correction	multi‐scan
R 1/wR 2 for I > 2σ(I)	0.0109/0.0234
R 1/wR 2 for all data	0.0117/0.0235
Transmission max./min.	0.5022/0.3226
Largest diff. peak/hole/e·Å−3	0.719/−0.807

TABLE 2. Charge distribution calculated with BL/BS (ΣV) and Chardi (ΣQ).

Atom	ΣV	ΣQ	Atom	ΣV	ΣQ
Te1	+5.91	+6.03	O2	−1.92	−1.96
Te2	+3.80	+3.67	O3	−1.91	−2.03
Te3	+5.84	+6.10	O4	−1.99	−1.92
Te4	+5.80	+6.19	O5	−2.04	−2.12
Sr1	+2.02	+2.03	O6	−2.04	−1.93
Sr2	+2.21	+1.91	O7	−2.19	−2.35
Sr3	+2.51	+2.04	O8	−1.83	−1.61
O1	−2.11	−1.92	O9	−2.05	−2.29

TABLE 3. Comparison of the lattice parameters (in pm) obtained via energy minimization at HSEsol level of theory (space group Pnma) with the experimentally measured data.

	a	b	c	T/K
HSEsol/triple‐zeta	1327.82	1233.08	743.96	0
HSEsol/double‐zeta	1334.48	1237.60	747.03	0
Exper.	1332.38(14)	1236.38(14)	748.02(9)	299

TABLE 4. Hirshfeld partial charges of atoms in the asymmetric unit, in units of the elementary charge e.

Atom	Hirshfeld partial charge	Atom	Hirshfeld partial charge
Te1	4.03	O2	−1.56
Te2	2.56	O3	−1.52
Te3	4.02	O4	−1.50
Te4	3.98	O5	−1.61
Sr1	2.09	O6	−1.57
Sr2	2.08	O7	−1.42
Sr3	2.11	O8	−1.59
O1	−1.59	O9	−1.23

Full text

Introduction

1

The family of oxotellurates constitute a promising material class for the quest of finding novel materials, particularly for future multiferroic [1,2] and nonlinear optical [3, 4, 5, 6] materials. Oxotellurates also achieve remarkable diversity in their crystal structures as they readily share corners and edges of their [TeO_ n ] polyhedra, while also forming structures containing both Te^4+^ and Te^6+^ oxidation states. [Te^4+^O_4]^4−^ polyhedra are stereochemically active due to asymmetry imparted by their lone electron pair and show greater variations in their geometry and bond lengths, allowing stabilization in environments inaccessible to more rigid octahedrally coordinated Te^6+^.

High‐pressure/high‐temperature (HP/HT) synthesis has been used to access new compositions and metastable oxotellurate crystal structures that are mostly inaccessible under ambient conditions using conventional synthesis methods. Some of these phases show technologically interesting magnetoelectric and nonlinear optical properties as exemplified by HP‐Co_3_TeO_6_ [7] and HP‐Mg_3_TeO_6_ [8], respectively. The formation of unprecedented crystal structures and chemical compositions under HP can also be shown by the HP/HT‐synthesis of the first ternary tungsten tellurate (WTe_2_O_7_) [9].

In the aim of finding novel oxotellurate phases, this study used HP/HT synthesis to produce a new phase of strontium tellurate, Sr_4_Te_4_O_15_. Among the known strontium oxotellurates(IV), SrTeO_3_ [10, 11] was intensely studied particularly due to its ferroelectric properties at elevated temperatures, but its space group symmetry was heavily subjected to debate [11, 12, 13, 14]. It is now established that at ambient temperature, α‐SrTeO_3_ crystallizes in the monoclinic, noncentrosymmetric space group C2 [15]. Despite its stoichiometry, it does not crystallize in a perovskite type structure. Instead, it consists of trigonal pyramidal [TeO_3_]^2−^ units which connect via bridging oxygen atoms to the inner sides of a tunnel‐like framework made of corner and face‐sharing [SrO_ n ] polyhedra. SrTeO_3 is highly polymorphic with several high‐temperature forms discovered (β‐, γ‐, and δ‐SrTeO_3_) [16, 17, 18] as well as ε‐SrTeO_3_ isolated through dehydration of SrTeO_3_(H_2_O) [19]. A tunnel‐like framework composed of [SrO_ n ] polyhedra is also shared by another strontium tellurate(IV), Sr_3_Te_4_O_11, which was shown to crystallize in the triclinic space group P1 [20]. SrTe_5_O_11_ is a metastable phase with higher symmetry crystallizing in Fm 3¯ m in the CaF_2_‐type structure and was produced via solid‐state quenching. It exhibits a statistical distribution of Sr^2+^ and Te^4+^ on the Ca^2+^ sites where Te^4+^ was proposed to occupy an unusual hexahedral coordination [21]. The oxotellurates(VI) of strontium include SrTeO_4_, SrTe_3_O_8_, and Sr_3_TeO_6_. SrTeO_4_ is orthorhombic and isostructural to CaTeO_4_; it crystallizes in the space group Pbcn and is composed of zig–zag chains of edge‐sharing [TeO_6_]^6−^ octahedra separated by corner sharing [SrO_ n ] polyhedra [22, 23]. The strontium tellurate SrTe_3_O_8 was synthesized initially under evacuated atmosphere conditions and contains both Te^4+^ and Te^6+^ [24]. It crystallizes in a tetragonal system with the space group P4_2_/m and consists of edge‐sharing bisphenoidal units combined with corner‐sharing octahedral units [24, 25]. Sr_3_TeO_6_ is triclinic and has a double perovskite structure, crystallizing in the space group P1¯ as a superstructure [26]. Not much is known about oxotellurate perovskite materials under high‐pressure. However, Attfield et al. reported magnetic frustration in HP double perovskite Mn_2_MnTeO_6_ and about the conversion of a defective pyrochlore structure to a double perovskite upon reduction of Te^6+^ [27, 28].

The title compound Sr_4_Te_4_O_15_ bears some similarity to perovskite‐related crystal structures, in particular to 2D layered perovskites, and contains both Te^4+^ and Te^6+^. 2D perovskites are a diverse group of materials fundamentally made up of sheets of corner‐shared metal–oxide or metal‐halide octahedra. Such 2D perovskites are derivatives of the conventional 3D perovskite model since cleaving the latter along the (100), (110) and (111) directions yield 2D layers of corner‐sharing octahedra separated by charge‐balancing cations. Cleaving along (100) and both (110) and (111) yields planar and corrugated sheets of corner‐sharing octahedra, respectively. A schematic for cleaving a 3D perovskite along the (100) and (110) to produce flat and corrugated layers is shown in Figure 1.

Schematic representation of flat (100)‐oriented, 2 × 2 and 3 × 3 corrugated ((110)‐oriented) perovskite sheets cut out from a 3D perovskite model.

Further details on these perovskite‐like structures and their outstanding physical properties, particularly optical properties, can be found in several comprehensive reviews dealing with low‐dimensional perovskites which include both fully inorganic [29, 30] and hybrid organic–inorganic forms [31, 32, 33, 34, 35, 36]. Examples of (100)‐oriented perovskites are numerous, whereas the (110)‐oriented forms are more rare [31]. The most common of the currently synthesized (110)‐oriented perovskites consist of a 2 × 2 corrugation in which the crest and trough octahedra are directly connected in cis fashion (see Figure 1) [29, 30, 32, 37, 38, 39, 40, 41]. Other rare variations of the corrugations have recently been found for hybrid organic–inorganic systems, including a few 3 × 3 [42, 43, 44, 45] and 4 × 4 forms [42, 46]. In 3 × 3 systems, one trans‐connected octahedron separates cis‐connected troughs and crests, whereas in 4 × 4 forms, two trans octahedra separate crest and trough. The attention given to these classes of structures stems from their potential for optoelectronic applications, such as in white‐light photoluminescence for development of white‐light LEDs [44]. These novel optoelectronic properties arise from their reduced dimensionality as well as high octahedral distortion [31, 44, 47]. Throughout the length of this study, the authors have so far only found one fully‐inorganic compound showing a perovskite‐like 3 × 3 corrugation pattern in the literature, this being Na_5_Mn_3_F_14_ [48] with the corrugated layers composed mostly of corner‐sharing [MnF_6_]^3−^ octahedra with some Mn^3+^ sites substituted by Na^+^. The layered perovskites Sr_5_Os_3_O_13_ [49] and Sr_9_Os_5_O_23_ [50] show another kind of corrugation, this being a 3 × 2 type, also known for several recently synthesized organic–inorganic systems [51].

Hence, such heavily corrugated 2D perovskites seem to be quite rare not only for hybrid organic–inorganic systems but also for fully inorganic systems. In this study, Sr_4_Te_4_O_15_ is presented as a 2D pseudo‐perovskite rather than a true perovskite, displaying a 3 × 3 corrugation pattern. 2D pseudo‐perovskites are known in the literature for hybrid organic–inorganic systems showing replacement of octahedral groups against “highly distorted tetrahedra [52]. The rarity of such heavily corrugated configurations, especially in fully inorganic systems, and their importance for optoelectronic applications has led us to further investigate, characterize, and understand the crystal structure of Sr_4_Te_4_O_15_ and attempt to rationalize its formation under HP conditions.

Results and Discussion

2

Syntheses, Composition, and Structure Refinement

2.1

Heating the reactants for only 5 min at 8.8 GPa and 800°C gave both Sr_4_Te_4_O_15_ and the known ambient pressure tellurate SrTe_3_O_8_ in significant amounts. Obtaining a nearly X‐ray‐pure sample of Sr_4_Te_4_O_15_ was only possible through HP experiments involving prolonged heating times of up to 1 h. A lower temperature of 700°C was used in these experiments to disfavor interfering reactions of the platinum capsule with tellurium oxides that would form several platinum telluride contaminants. In this way, we were able to get an almost X‐ray‐pure sample which was confirmed by Rietveld refinement (see Figure 2). Reflections of a minor contamination of the sample with SrTe_3_O_8_ are highlighted with an asterisk. The unit cell parameters from Rietveld refinement agreed with those from single‐crystal X‐ray data collection (see Table 1). Data collection details, refined atom positions, and isotropic displacement parameters are given in Tables 1 and S1. Relevant atomic distances and anisotropic displacement parameters are given in Tables S2 and S3 (Supporting Information). Deposition Number CSD 2 494 062 (Sr_4_Te_4_O_15_) contains the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe (http://www.ccdc.cam.ac.uk/structures).

*PXRD pattern (Mo‐K

α1 radiation) and Rietveld plot of Sr4Te4O15 (R exp = 0.88%, R wp = 1.36%, R p = 0.93% and GoF = 1.54). Asterisk marks indicate minor reflections belonging to SrTe3O8.*

The atomic ratio of strontium to tellurium in Sr_4_Te_4_O_15_ was confirmed using semiquantitative EDX spectroscopy, which revealed an almost 1:1 ratio of these elements (Sr : Te = 17.7(1)at% : 18.5(1)at%). The ratio is close to the theoretical ratio of 1 for this compound (Sr : Te = 17.4 at% : 17.4 at%). These values were obtained by averaging out individual data points from different crystal surfaces (Figures S1–4, Supporting Information). The individual spectra and quantification results are available in Figures S5–S16 and Tables S4–S15 (Supporting Information). The light atom, oxygen, could not be located accurately by this semiquantitative measurement method.

Crystal Structure Description

2.2

Sr_4_Te_4_O_15_ crystallizes in an orthorhombic crystal system with the space group Pnma (62) and unit cell parameters of a = 1332.4(1), b = 1236.4(1), and c = 748.02(9) pm. Based on a thorough literature review and an ICSD [53] search against Wyckoff sequence (d ^7^ c ^7^ ba) and the unit cell parameters, no similar structures could be found. It is therefore thought that Sr_4_Te_4_O_15_ consists of a unique composition and crystal structure accessible only under HP/HT conditions. The structure has four tellurium sites (Te1‐Te4) and three strontium sites (Sr1‐Sr3). Te1, Te3, and Te4 occur as octahedrally oxygen‐coordinated Te^6+^, whereas Te2 is bisphenoidally oxygen‐coordinated Te^4+^ (see Figure 3).

Coordination geometries and bond lengths of both Te and Sr sites of Sr4Te4O15. Distances are given in pm and all standard deviations are below 0.15 pm. Sr2 and Sr3 are displayed in an orientation that makes it easier to visualize their augmented trigonal prismatic geometry (trigonal prism composed of O2‐O5‐O6 and O3‐O2‐O1 faces respectively).

Sr_4_Te_4_O_15_ is built up of infinite and corrugated [Te_4_O_15_]^8−^ oxotellurate sheets (OS) spreading parallel to the bc plane and stacked along the a‐axis. Layers of charge‐balancing Sr^2+^ separate the stacked sheets (see Figure 4). Each OS consists entirely of corner‐sharing octahedral oxygen coordinated Te^6+^ and bisphenoidal oxygen coordinated Te^4+^.

Crystal structure diagram of Sr4Te4O15 showing a 3 × 3 corrugation pattern of its OS stacked along the a‐axis.

In this study, Sr_4_Te_4_O_15_ is presented as a 2D pseudo‐perovskite. The layers in Sr_4_Te_4_O_15_ show a 3 × 3 corrugation arrangement seen in a few hybrid organic–inorganic 2D perovskites. As is shown by the chemical formula of Sr_4_Te_4_O_15_, octahedrally coordinated Te^6+^ is three times more predominant compared to bisphenoidally coordinated Te^4+^, and therefore, the OS resembles the sheets of 2D perovskites which are entirely composed of corner‐sharing octahedra (see Figure 5(top)). Due to the replacement of an octahedrally oxygen‐coordinated Te^6+^ site with a bisphenoidally coordinated Te^4+^ site in the OS, Sr_4_Te_4_O_15_ cannot be referred to as a true 2D perovskite but could be better referred to as a 2D pseudo‐perovskite. This is also reflected in the chemical formula of Sr_4_Te_4_O_15_: hypothetically replacing bisphenoidally coordinated Te^4+^ with octahedrally coordinated Te^6+^ in the chemical formula would require an additional O^2−^ ion for charge balance.

Top: A comparison of an OS of Sr4Te4O15 (top left) with an equivalent hypothetical 3 × 3 cutout diagram of the 3D perovskite SrTiO3 (top right). The aqua‐colored octahedra of SrTiO3 indicate sites replaced by bisphenoidally coordinated Te6+ in Sr4Te4O15. The numbering of the octahedra for SrTiO3 (in yellow) aids the visualization of the 3 × 3 nature of the corrugation. Bottom: An enlarged section of the OS of Sr4Te4O15 is shown. Teal colored bonds delineate tilting with respect to both the bc and ac planes. Full and faded out teal‐colored bonds show tilting into and out of the bc plane (refer to Figure 7 for tilting of Te1‐Te2 units). The shortest octahedral edge of the structure is marked at 257.7 pm for the edge O1–O1 for [Te1O6]6−.

This would lead to the chemical formula of SrTeO_4_, which represents the stoichiometry for the class of layered perovskites with the general formula ABX_4_ [29, 54, 55]. However, SrTeO_4_ is a known stoichiometry for tellurates synthesized at ambient pressure which does not adopt a perovskite‐like structure (see Introduction). Therefore, in addition to its layers of predominantly corner‐sharing octahedra, Sr_4_Te_4_O_15_ can be considered to represent a crystal structure related to ABX_4_‐type perovskites based on a chemical formula rather similar to that of ABX_4_ layered perovskites. The possible reasons for the pseudo structure and for the accessibility under HP conditions are discussed in Section 2.3.

Individual Te1–Te2 units consisting of corner‐sharing octahedrally coordinated Te^6+^ and bisphenoidally coordinated Te^4+^ constitute the crests and troughs of the corrugated structure running in the c direction of the OS. Multiple Te1–Te2 units are connected to each other via weak secondary bonds with Te2–O8 distances of 242.4 pm (see Figure 5). Therefore, Te1–Te2 units are not connected to each other in a chain‐like manner. The crests and troughs are connected to each other through one set of octahedrally coordinated Te^6+^ chains consisting of corner‐sharing Te3–O6–Te4–O6–Te3 sequences. [Te3O_6_]^6−^ and [Te4O_6_]^6−^ octahedra connect in a trans configuration through oxygen atoms lying at opposite ends of their octahedra to Te1 and Te2, respectively. This fulfills the typical cis‐trans‐cis linkage pattern of a 3 × 3 perovskite, as shown in Figure 1. Sr^2+^ is found in a similarly corrugated arrangement filling the grooves created by the crest and troughs of the corrugated structure. The [SrO_ n _] polyhedra are highly irregular owing to the coordinative flexibility of Sr^2+^. Sr1 has the highest coordination number of 10 and a geometry resembling a bicapped cube antiprism. Both Sr2 and Sr3 have a coordination number of 8 with geometries resembling a biaugmented triangular prism. Coordination geometries and distances are given in Figure 3. A table of selected distances of Te–O and Sr–O including individual standard deviations can be found in Table S2 (Supporting Information).

The average Te^6+^–O bond lengths of Te1 (193.2 pm), Te3 (193.9 pm), and Te4 (194.4 pm) are close to the typical average Te^6+^–O bond length found in literature for the alkali‐earth metal strontium tellurates(VI) SrTeO_4_ (194.3 pm) [22], BaTeO_4_ (194.3 pm) [56], and CaTeO_4_ (195.2 pm) [22]. The Te^4+^–O distances in the bisphenoidal unit range from 182.8 to 218.7 pm and are comparable to the distances found in literature for oxotellurates(IV) containing the bisphenoidally coordinated Te^4+^ unit such as Sr_3_Te_4_O_11_, (177.8–292.2 pm) [20], Ba_3_Te_4_O_11_ (183.0–225.0 pm) [57], or Ca_4_Te_5_O_11_ (185.5–234.9 pm) [58]. The longer Te^4+^–O distance of 242.4 pm (Te2–O8) is considered as a secondary bond of Te^4+^ according to investigations by Christy et al. [59]. The shortest Te–O bond in this structure is Te2–O9 involving the Te^4+^ cation with a bond length of 182.8 pm. In contrast to almost all the oxygen sites which experience fourfold coordination sphere that includes two tellurium cations, O9 is only coordinated by three cations, two of which are with Sr^2+^. The Te2–O9 bond could therefore be compensating for the reduced coordination of O9 by shortening to lower its valence. The validity of the structure obtained through refinement was checked via bond‐length/bond‐strength (BL/BS) [60, 61, 62] and charge distribution (Chardi) [63] calculations which indicate that valences of the different ions are relatively well satisfied according to their expected charges (see Table 2). As can be clearly seen, the atom O9 (discussed above), as well as O7, show larger deviations from the ideal valence when calculated according to Chardi. This is where the simple concept reaches its limits. However, the calculation according to the BL/BS concept yields good valences for oxide anions. MAPLE (Madelung Part of Lattice Energy) [64, 65, 66] calculations were also carried out for verification of the validity of the structure of Sr_4_Te_4_O_15_. Only a 0.65% difference between the MAPLE value of 104,269.25 kJmol^−1^ obtained for the structure model presented compared to a stoichiometric summation of individual MAPLE values of starting materials used exists. A breakdown of this result is shown in the Supporting Information. These BL/BS, Chardi, and MAPLE results suggest that the structure of Sr_4_Te_4_O_15_ is indeed a valid structure (Table 2).

A striking feature of the structure is its complicated topology of its OS which constitutes buckled parts through the tilting of its constituent polyhedra. [Te3O_6_]^6−^ and [Te4O_6_]^6−^ octahedra are tilted in directions within and perpendicular to the equatorial bc plane of the OS. This is shown by Te3–O6–Te4–O6 tilt axes with a Te–O–Te bending angle of 139.0° (see Figure 6). The bending perpendicular to the bc plane is shown by alternating full and faded Te3–O6 and Te4–O6 bonds in Figure 5(bottom). The tilt axes of these chains are made up of some of the longer bonds of the structure, namely Te4–O6 which is the longest Te^6+^–O bond (201.5 pm). The [Te1O_6_]^6−^ octahedra constituting the crests and troughs of the structure also experience tilting of their O7 and O8 apices only perpendicular to the bc plane. Here it is tilted in−phase with the bc plane tilting of Te3–O6–Te4–O6 axes. The [Te2O_4_]^4−^ bisphenoid forms a 129.3° bending angle with Te1 by tilting its O8–Te2–O7 axis in the opposite direction through the bc plane (tilt is shown more clearly in Figure 7). Apart from the tilting, the octahedra also vary in their relative degree of rotation around their tilt axes. This can be seen especially in comparison to the regular 3 × 3 corrugated layer of the perovskite SrTiO_3_ in Figure 5. For [Te1O_6_]^6−^, the octahedral edges O5‐O5 and O1‐O1 are aligned parallel to the bc plane, and the edges O5‐O1 are parallel to the ac plane. In contrast, the octahedral edges of [Te3O_6_]^6−^ and [Te4O_6_]^6−^ are rotated round their Te3–O6–Te4–O6 tilt axes such that the edges appear to go into and out of the bc plane. Furthermore, [Te3O_6_]^6−^ and [Te4O_6_]^6−^ octahedra are rotated in a staggered manner relative to each other along the c‐axis of Te3‐O6‐Te4 chains in Figure 5.

Diagram showing the interaction of the OS with Sr2+ (Sr2 and Sr3) positioned in the grooves created by the crests/troughs of the OS. Red colored arrows denote the tilting directions of O6 and O3 towards Sr2+ parallel to both the bc and ac plane. The tilting of crest/trough octahedra parallel to the ac plane is not clear here but is shown more clearly in Figure 7. The [Te4+O4]4− polyhedra have been omitted from this diagram to allow clear visualization of the Sr2+ coordination.

Crystal Structure Discussion and Thermal Analysis

2.3

The currently known 3 × 3 hybrid organic–inorganic perovskites [42, 43, 44, 45] are stabilized through hydrogen bonding ability of the organic portion separating the corrugated layers. The importance of this stabilizing effect was demonstrated through the synthesis of a series of 2D perovskites by Mao et al. which included α‐(DMEN)PbBr_4_ displaying a 3 × 3 configuration [44]. The hydrogen bonding ability of this compound is due to two ammonium groups of DMEN which efficiently stabilized the oxygen anions of both the cis and trans octahedra. These forces acting on the sheets of corner‐sharing octahedra were thought to promote the ~90° folding to produce the 3 × 3 corrugated topology, as the organic cation “pulls” the sheet towards itself from multiple directions. As the separation between the two ammonium groups increased through an increase in alkyl chain length, the system favored the formation of (100)‐oriented perovskites. The increased separation prevented the additional ammonium group from contributing to further stabilization of trans‐connected octahedra thus leading to flat 2D perovskite sheets. The same concepts could be applied to rationalize the formation of Sr_4_Te_4_O_15_: as with hybrid organic–inorganic perovskites, the negative charge of both the cis and trans octahedra of the OS require efficient stabilization. The large ionic size of Sr^2+^ in Sr_4_Te_4_O_15_ and hence its long Sr–O contacts as well as the higher coordination numbers preferred under pressure favor the folding of the OS into a corrugated form. The long Sr–O contacts together with the flexible coordination geometry and diffuse nature of Sr^2+^ allows Sr2 and Sr3 to stabilize the OS from every direction, i.e., the trans Te3‐O6‐Te4 chains and cis Te1‐Te2 units (see Figure 6).

Considering the rarity of structures such as Sr_4_Te_4_O_15_ in inorganic systems and the useful optical applications of such configurations, it is interesting to attempt to rationalize its formation under pressure. This could be done by considering its structural motifs. A principal feature of the corrugated OS sheets of Sr_4_Te_4_O_15_ is their buckling. Figures 5 and 6 show the interaction of the buckled OS with neighboring Sr^2+^ cations with tilting of oxygen atoms occurring toward Sr^2+^ parallel to both the ac and bc planes. Tilting is known to occur in perovskites for the 3D congeners when the A cation is too small for its space, hence requiring a shortening of A–O bonds for better stabilization of A atoms [67]. The same effect may be present in Sr_4_Te_4_O_15_ where the OS adopts a heavily buckled configuration to better stabilize Sr^2+^ in both the ac and bc planes. As Sr_4_Te_4_O_15_ seems to be accessible only under HP conditions, it could be postulated that HP conditions induce this octahedral tilting and hence buckling of the OS. This is based on observations related to 3D perovskites for which pressure, in some cases, is known to increase the octahedral tilt [68]. This postulation could be explored further by considering similar layered perovskites that show buckling and octahedral tilting in their perovskite sheets such as La_2_NiO_4_. Buckling is a measure through which highly strained crystals achieve structural relaxation [69]. This phenomenon was studied extensively for the layered perovskite La_2_NiO_4_ which crystallizes as (100)‐oriented sheets of Ni^2+^ octahedra. Its behavior is summarized below from studies carried out by Brown I.D, 2016 [69, 70], but greater detail is provided in the respective references. At high temperature, La_2_NiO_4_ has higher symmetry with overly compressed Ni–O sheets and stretched La–O layers. During cooling, buckling lengthens overly compressed Ni–O equatorial bonds, thus relaxing the system while lowering the symmetry. Concomitant with this buckling is the relaxation of the La–O layer which occurs through shortening of La–O bonds induced by the Ni–O layer buckling thus increasing the valence on La^3+^. Owing to the predominance of octahedral tilt and OS buckling in the structure of Sr_4_Te_4_O_15_ it could be speculated that, similar to La_2_NiO_4_, Sr_4_Te_4_O_15_ is derived from a more symmetric but highly strained intermediate where the compressive impact of pressure on bonds (rather than temperature as in La_2_NiO_4_) stabilize shorter Te–O bonds that relax by tilting when the temperature is lowered under high‐pressure conditions, thus buckling the OS. In this case, pressure plus additional temperature would favor a higher symmetry precursor that relaxes to form Sr_4_Te_4_O_15_. However, it is generally known that pressure and temperature are opposing parameters. Relaxation into the low symmetry tilted structure, as occurs with a decrease in temperature in La_2_NiO_4_, can therefore also be achieved with an increase in pressure for Sr_4_Te_4_O_15_. Furthermore, pressure does not always favor the formation of more symmetric perovskite phases but instead favors tilted structures with lower symmetry. Performing in situ measurements may allow detection of any intermediary species, possibly of higher symmetry preceding the formation of Sr_4_Te_4_O_15_. This could help determine the precise origin of chain/unit buckling and the role of pressure in the formation of this 3 × 3 structure.

A possible thought experiment involves the substitution of the Te^4+^ by Te^6+^ in this structure to make it completely made of Te^6+^ (which would be equivalent to the chemical formula SrTeO_4_ after accounting for charge balance by O^2−^) and reflect on why this would be unfavorable for the crystal. The incorporation of Te^4+^ in place of Te^6+^ by the crystal could be interpreted as a form of electronic relaxation where cations lower their valence to increase their bond length and better adapt to their coordination environment. The tilting of the O7 and O8 apices of Te1–Te2 units causes a large separation of 440.3 pm between O7 and O8 (see Figure 7) which could favor the incorporation of Te^4+^ (Te2) rather than Te^6+^, as Te^4+^ could flexibly coordinate in this position with both short or long bonds including secondary bonding. Incorporation of Te^6+^ at this site in octahedral geometry would require Te^6+^ to form bonds that are at least 220.15 pm in length (Figure 7), which is far from the Te^6+^–O length of 192.1 pm of unit valence [60] and would therefore lead to highly strained bonds. Therefore, the crystal structure avoids such an unfavorable scenario by electronic relaxation which compensates for the high bond strain required to sustain a high degree of octahedral tilting and OS buckling. The Global Instability Index (GII) measures the level of lattice strain of a structure, and hence its extent of violation of the valence sum rule [71]. A number below 0.05 indicates no significant lattice strain whereas a number above 0.2 indicates a structure that cannot possibly form due to the extreme strain. The GII is given by Equation (1), in which the extent of violation of the valence over all atoms of the asymmetric unit is averaged. N stands for the number of atoms in the asymmetric unit (equal to 16 in the case of Sr_4_Te_4_O_15_), s _ ij _ stands for the calculated bond valence between atoms i and j (hence ∑jsij represents the actual valence on each atom based on experimental data), and v _ i _ represents the expected, ideal valence of each atom.

Diagram showing a side (ac plane) view of the crest/trough of the corrugated structure of Sr4Te4O15. Te6+ (dark blue) and Te4+ (aqua) consist of Te1 and Te2 respectively. Only the oxygen sites involved in coordination with Sr2+ sitting at the crest/trough are marked. The red arrows show the tilting direction of the respective oxygen atoms towards the corresponding Sr2+ sites. This tilting causes a Te1–O8–Te2 bending angle of 129.8° and a Te2–O7–Te1 bending angle of 129.3°. The total distance between the tilted oxygen atoms O7 and O8 is shown as 440.3 pm.

$$[eqn]$$

Sr_4_Te_4_O_15_ has a GII value of 0.18 which indicates that there is not enough strain to render this structure unfeasible. However, this value is much higher than 0.05 hence the structure is significantly strained which may be expected from a metastable phase obtainable under special HP conditions. To investigate the stability, we have carried out thermal analysis of a sample of Sr_4_Te_4_O_15_, the results of which are available further below in this text. Strained crystals are also expected to possess distorted polyhedra, where the observed bond lengths deviate significantly from their ideal values. To further quantify the strain in the structure of Sr_4_Te_4_O_15_, the octahedral bond‐length (Δd) [72] and angular distortion (σ ^2^using the following Equations (2) and (3):

$$[eqn]$$ $$[eqn]$$

For Te1, Te3, and Te4, Δd is at 9.2 × 10^−4^, 9.5 × 10^−4^, and 12.8 × 10^−4^ and σ ^2^ oct at 19.0, 14.4, and 14.7, respectively. The average values for Δd and σ ^2^ oct of these octahedra lies at 10.5 × 10^−4^ and 16.0, displaying a moderate level of distortion [72, 74]. Another parameter relevant to strain involves anion–anion repulsion which may also be an important factor for Sr_4_Te_4_O_15_ due to its shortest octahedral edge lying at 257.7 pm between adjacent O1 atom sites bonded to Te1. This short edge creates the cis connection of the structure and is significantly smaller than the ideal edge length of 270 pm for Te^6+^–O octahedra suggested by Brown [69]. Hence, it is expected that higher anion–anion repulsion will be at this edge, although the extent of the repulsion is uncertain.

To further understand the nature of the crystal structure of Sr_4_Te_4_O_15_, high‐temperature powder X‐ray diffraction (HT‐PXRD) was performed. Metastable HP phases typically transform back to ambient pressure phases when heated to sufficiently high temperatures, releasing energy in this process. The HT‐PXRD results are shown in Figure 8 which shows the transformative changes of a sample of Sr_4_Te_4_O_15_ up till 925°C, beyond which reactions with the SiO_2_ mark capillary predominate. Sr_4_Te_4_O_15_ starts to transform at a temperature just below 675°C. In the same temperature range, SrTeO_4_ [22] starts to form. The disappearance of Sr_4_Te_4_O_15_ concomitant with the appearance of SrTeO_4_ is a gradual oxidative step experienced by Sr_4_Te_4_O_15_ shown by the concurrent reduction and increase of reflection intensities of Sr_4_Te_4_O_15_ and SrTeO_4_, respectively. Sr_4_Te_4_O_15_ reflections disappear completely between the temperatures just below 725°C at which SrTeO_4_ predominates. The reflection patterns of Sr_4_Te_4_O_15_ and SrTeO_4_ are clearly different, and no reflections show a continuous transition between the phases, which indicates a reconstructive phase transition of first order. SrTeO_4_ remains stable up till 825°C which leads to a sharper transition to SrTeO_3_ due to reduction by decomposition. The reflections from this phase of SrTeO_3_ match best with those of the δ‐phase of SrTeO_3_ [18], studied at 507°C. The powder XRD pattern of both SrTeO_4_ and SrTeO_3_ obtained from the single HT‐PXRD ranges in comparison with the referenced literature data are given in Figures S18 and S19 (Supporting Information).

HT‐PXRD pattern of Sr4Te4O15 showing its thermally induced transformation to ambient pressure strontium tellurate phases.

Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) of a sample of Sr_4_Te_4_O_15_ were also carried out to further understand these phase transitions observed by HT‐PXRD. The DSC and TGA plots are available in Figure S20 (Supporting Information). Following from the reconstructive nature of the phase transition from Sr_4_Te_4_O_15_ to SrTeO_4_, a DSC signal is expected at the lower temperature range close to 675°C. However, the DSC curve is largely featureless up until the first endothermic maximum of 838°C. The proximity of this temperature to the phase transition temperature of SrTeO_4_ to SrTeO_3_ obtained from HT‐PXRD leads to the conclusion that a direct transformation from Sr_4_Te_4_O_15_ to SrTeO_3_ takes place and the phase SrTeO_4_ is omitted under these conditions. Possible causes for the absence of a feature for this transition could include the different environments HT‐PXRD and DSC/TGA are carried out in, with the former done in air and the latter done in an inert argon atmosphere. Consequently, the oxidation step to SrTeO_4_ would have been suppressed under Ar atmosphere. Another reason could also be due to the DSC/TGA being carried out in corundum crucibles which may have been too insensitive for very weak phase transition signals that may characterize the phase transition of Sr_4_Te_4_O_15_ to SrTeO_4_. Carrying out the thermal analysis in platinum crucibles could be an alternative for observation of weaker signals due to the enhanced heat conduction of such crucibles. However, this was avoided here due to the propensity of tellurium to react with platinum to form platinum tellurides at even moderate temperatures. The measurements carried out here do not support or refute the hypothesis that Sr_4_Te_4_O_15_ is a metastable high‐pressure phase. An exothermic DSC signal is expected for the reverse conversion of a metastable phase into a related thermodynamically stable phase of similar composition. For Sr_4_Te_4_O_15_, only an endothermic signal was observed at 838°C, which corresponds to its reductive decomposition to SrTeO_3_. However, since Sr_4_Te_4_O_15_ could not be produced under ambient pressure in our experiments, the parameter of pressure during synthesis must be considered essential.

Vibrational Spectroscopy

2.4

The infrared spectrum of Sr_4_Te_4_O_15_ consists of several broad bands in the region spanning 400 to 800 cm^−1^ (see Figure 9). The spectrum is featureless beyond 1000 cm^−1^, and therefore only this region is shown below for easier visualization of the bands. Due to the complexity of the structure, including four tellurium sites with differing oxidation states, it is in general challenging to individually assign each band at middle wavenumbers precisely to each vibration.

Overlay of the experimental (black) and theoretical (red) infrared spectra of Sr4Te4O15. The absorbance scaling of the theoretical spectrum has been adjusted for clearer visualization of the correlation between the two spectra, although both spectra have been standardized relative to their strongest absorption peaks in the region displayed.

The strongest band in the spectrum is observed at its lowest end at 400 cm^−1^, although the maximum of this absorption is not clearly visible due to the limited measurement range of the spectrometer used. This maximum should therefore lie at slightly lower wavenumbers as the curve starts to level off at 400 cm^−1^. Intense absorption around this region is also documented for other tellurate crystal structures and was attributed to the δ Te–O bending mode of [TeO_6_]^6−^ octahedra [75, 76]. Other Te–O octahedral vibrations are visible at lower wavenumbers, one of these being found at the band split at 464 and 472 cm^−1^ which can be similarly ascribed to the Te–O bending mode based on observations of similar bands by Siebert [77], the splitting of which could reflect the presence of multiple Te sites. The absorptions at 530, 567, and 596 cm^−1^ could be ascribed to Te–O stretching vibrations corresponding to similar vibrations in this region for other tellurates [77]. Strong vibrations associated with axial symmetric and asymmetric Te–O bond stretches of [TeO_4_]^4−^ range from 620 to 720 cm^−1^ for tellurates consisting of bisphenoidal [TeO_4_]^4−^ units, these being α‐TeO_2_, TiTe_3_O8, ZrTe_3_O_8_, and SnTe_3_O_8_ studied by Arnaudov et al. [78]. For these same cations in the literature, equatorial Te–O bond stretches of [TeO_4_]^4−^ polyhedra range between 700 and 800 cm^−1^. However, several further vibrations relating to Te^6+^–O stretches are known ranging from 600 until 800 cm^−1^ [77]. Therefore, it is impossible to assign these vibrations for [TeO_4_]^4−^ precisely to the bands in the IR spectrum of Sr_4_Te_4_O_15_. Comparison of the theoretical and experimental infrared spectra show a good correlation, but with a shift of the theoretical spectrum to higher wavenumbers compared to the experimental spectrum. Such a shift can be attributed to the methodological differences: in contrast to experimental measurements, theoretical calculations assume a perfect crystal with infinite periodicity.

The most prominent Raman bands (see Figure 10) lie at higher wavenumbers ranging from 688 to 793 cm^−1^, with an overlapping shoulder band occurring at 764 cm^−1^ next to the absorption at 754 cm^−1^.

Overlay of the experimental (black) and theoretical (red) Raman spectra of Sr4Te4O15. The absorbance scaling of the theoretical spectrum has been adjusted for clearer visualization of the correlation between the two spectra, although both spectra have been standardized relative to their strongest absorption peaks in the region displayed. The Raman bands from 200 to 600 cm−1 are listed in this figure caption matched with the labeling in the figure: (i) 210 cm−1, (ii) 236 cm−1, (iii) 252 cm−1, (iv) 268 cm−1, (v) 288 cm−1, (vi) 319 cm−1, (vii) 331 cm−1, (viii) 338 cm−1, (ix) 379 cm−1, (x) 403 cm−1, (xi) 421 cm−1, (xii) 432 cm−1, (xiii) 472 cm−1, (xiv) 493 cm−1, (xv) 504 cm−1, (xvi) 597 cm−1.

In this region, Raman absorptions from both [TeO_4_]^4−^ and [TeO_6_]^6−^ groups are again expected and hence a precise assignment of the bands is not possible. However, as observed by Frost and Keefe for the mineral tlapallite [75, 76] with mixed tellurate anions, four primary bands are observed corresponding to the symmetric and asymmetric stretch vibrations of [TeO_6_]^6−^ and [TeO_4_]^4−^ groups ([TeO_3_]^2−^ for tlapallite). The same logical assignment as in Frost and Keefe can be followed with bands of higher wavenumbers belonging to [TeO_6_]^6−^ groups and bands of lower wavenumbers belonging to [TeO_4_]^4−^ groups. Therefore, the bands at 754 and 793 cm^−1^ are assigned to the symmetric and asymmetric stretch of [TeO_6_]^6−^ and the bands at 764 and 688 cm^−1^ assigned to the symmetric and asymmetric stretch of [TeO_4_]^4−^ groups. Te–O bending modes for both [TeO_4_]^4−^ and [TeO_6_]^6−^ also occur in similar regions to each other, with the bands ranging from 421 to 504 cm^−1^ being assigned to bending vibrations from both groups in Sr_4_Te_4_O_15_. As with the infrared spectrum, the theoretical Raman spectrum of Sr_4_Te_4_O_15_ shows good correlation to the experimental spectrum but is shifted to higher wavenumbers due to ideal conditions assumed in the theoretical calculations.

UV–Visible Spectroscopy

2.5

Figure 11 shows the diffuse UV–Visible reflectance spectrum of Sr_4_Te_4_O_15_ across the wavelength range of 2500–250 nm. Approximately 80% of the radiation across almost the entire measured wavelength range is reflected. The absorption of the sample starts to increase drastically at just below the region of 440 nm and achieves a 90% absorption at 250 nm. This decay also agrees well with the calculated bandgap values. The bandgap of Sr_4_Te_4_O_15_ was determined using the Kubelka‐Munk (K‐M) [79] function and Tauc plots [80]. The results of this calculation are shown in the insert in Figure 11: the values determined for the direct and indirect bandgap are 3.17 and 2.97 eV, respectively. These values are relatively close to each other which is in accordance with the finding from the theoretical bandgap modeling presented in the next section. As seen in the DFT‐based band structure calculation discussed below, a direct bandgap at the Γ‐point of 2.97 eV was determined in very good agreement with the estimations obtained from the Tauc method.

UV–Visible spectrum of a sample of Sr4Te4O15 with an insertion showing the Tauc plot for the estimation of the direct (3.17 eV) and indirect (2.97 eV) bandgaps.

Quantum Chemical Calculations

2.6

The lattice parameters a, b, and c obtained from the energy minimization applying both the double‐ and triple‐zeta basis sets agree well with the experimentally measured unit cell parameters (Table 3) with deviations being less than 0.5%. Since theoretically optimized geometries correspond to structures at 0 K conditions, slightly smaller unit cells can be expected and the observed deviations between the theoretical and experimental values are well within the expected range of the thermal expansion of crystalline compounds. In order to further assess the structural prediction by the HSEsol calculation, the associated powder X‐ray diffraction pattern was compared to the experimental pattern (see Figure S17, Supporting Information). The patterns show a high correlation.

The calculated Hirshfeld partial charges are listed in Table 4. Overall, tellurium and strontium atoms show positive partial charges ranging from 2.08 to 4.03 e, while oxygen atoms are negative with charges ranging from −1.61 to −1.23 e, respectively. Closer inspection of the charges obtained for tellurium shows three irreducible atoms carrying a partial of approximately 4.0 e, while in the case of the fourth atom, only a charge of 2.56 e was found, pointing towards the presence of Te atoms in different oxidation states.

To assess the electronic structure of the novel Sr_4_Te_4_O_15_ compound band structure calculations and density of states were performed at HSEsol level of theory (see Figure 12). The associated path along the high‐symmetry points of the Brillouin zone is based on the convention proposed by Setyawan and Curtarolo [81]. A direct bandgap has been identified at the Γ‐point with the estimated bandgap energy being 2.97 eV. This value is just below the commonly applied threshold of 3 eV above which solid‐state systems are isolators. This is consistent with the findings of the Hirshfeld population analysis highlighting a dominant ionic character. The calculated band structure shown in Figure 12 features flat bands with low dispersion, including bands at the Fermi level. This is typically associated to confinement of the electrons associated to geometric strain as well as the presence of lone electron pairs in the crystal structure [44].

Electronic band structure (left) and associated density of states (right) of Sr4Te4O15. obtained at HSEsol level of theory. A direct bandgap energy of 2.97 eV can be determined at the Γ‐point.

In Figure 13 the electron localization function (ELF) is visualized using an isosurface of 0.70, respectively. While in case of Sr and O spherical distributions centered at the atoms were observed, hemispherical residuals are visible in the ELF near the Te^4+^ ions indicating the presence of stereochemically active lone electron pairs highlighted by arrows in Figure 13. The calculated ELF provides direct evidence for the proposed bisphenoidal coordination motif of the Te^4+^ atoms. The hemispherical electron density attributed to the lone electron pair can be observed to lie away from the coordination sphere of Te^4+^ and therefore roughly perpendicular to the crest/groove direction along b. This is as expected for the behavior of lone electron pairs for the minimization of electron repulsion.

Graphic showing a visualization of the ELF in relation to the determined crystal structure and Te4+ positions. Encircled regions denote an allocation of the ELF graphic to the respective crystal structure viewpoint to simplify interpretation. Both the crystal structure and the ELF images were obtained in the direction shown by the coordinates (i.e., structure rotated by approximately 45° around a).

Conclusion

3

In this study, the novel compound Sr_4_Te_4_O_15_ was successfully synthesized under extreme HP/HT conditions, revealing a unique 3 × 3 corrugated 2D pseudo‐perovskite structure. The crystal structure, characterized by corrugated oxotellurate sheets and charge‐balancing Sr^2+^ layers, represents a rare example of a fully inorganic 2D pseudo‐perovskite. Theoretical and experimental investigations confirmed the stability of the structure under specific conditions and provided detailed insights into its electronic and vibrational properties. The thermal behavior of Sr_4_Te_4_O_15_ was investigated using high‐temperature powder X‐ray diffraction. It begins to transform into SrTeO_4_ at temperatures just below 675°C, with a complete phase transition occurring by 725°C. The transformation of Sr_4_Te_4_O_15_ to SrTeO_4_ at elevated temperatures is consistent with the typical behavior of HP phases when subjected to heating at ambient pressure. Although similar behavior is expected from metastable phases, the DSC results do not indicate any compelling evidence towards classifying Sr_4_Te_4_O_15_ as a metastable high‐pressure phase. This is also despite signs of a large GII value indicating unfavorable crystallographic stability. The direct and indirect bandgaps of 3.17 and 2.97 eV, respectively, suggest potential applications as wide bandgap materials. The formation of Sr_4_Te_4_O_15_ under HP/HT conditions highlights the role of extreme synthesis environments in stabilizing novel and complex crystal structures, which are otherwise currently inaccessible under ambient conditions. This work not only expands the library of pseudo‐perovskite materials but also provides a foundation for further exploration of their unique properties and potential applications. Encouraged by the low‐dimensional structure with the additional presence of lone electron pair cations from Te^4+^, photoluminescence investigations performed under cryogenic conditions are planned to further understand the contribution of the structural features to potential optical properties of the structure.

Experimental Section

4

High‐Pressure/High‐Temperature Synthesis

4.1

To synthesize Sr_4_Te_4_O_15_, a stoichiometric ratio of SrO (purity 99.5%, Apollo Scientific, United Kingdom), TeO_3_ and TeO_2_ (purity >98%, TCI Deutschland GmbH, Germany) as shown by Equation 1 were weighed and ground together using an agatate mortar until a well homogenized powder was obtained. SrO was stored and weighed in a glovebox. TeO_3_ was synthesized through decomposition of telluric acid (purity >99%, TCI Deutschland GmbH, Germany) in an alumina crucible at a temperature of 350°C for a duration of 24 h using a muffle furnace.

$$[eqn]$$

The ground mixture was then filled into a platinum capsule (99.95%, Ögussa, Vienna, Austria), which was then placed inside a crucible and covered with a lid both made of hexagonal boron nitride (Henze Boron Nitride Products AG, Kempten, Germany). This assembly was then fitted with a short and long graphite sleeve as resistive heaters, a ZrO_2_ cylinder as an insulator preventing heat loss, and two MgO spacers placed at both ends of the cylinder each topped by a molybdenum plate for electrical conduction. The assembly was then inserted inside a 5% Cr_2_O_3_‐doped MgO octahedron (18 mm MgO octahedra doped with 5% Cr_2_O_3_, Ceramic Substrates & Components Ltd, Newport Isle of Wight, United Kingdom) which functioned as the pressure medium. The pressure was delivered through two stages: an 8‐anvil setup of truncated edge tungsten carbide cubes (HM‐type ha‐7% Co, Hawedia, Marklkofen, Germany) hosting the octahedron directly at the center and a 6‐steel‐anvil setup as part of a Walker‐type module of a 1000 t uniaxial press (Max Voggenreiter GmbH, Mainleus, Germany). More details concerning the experimental setup can be found in several literature sources [82, 83]. The pressure was ramped up to 8.8 GPa (280 bar oil pressure) within 230 min. At peak pressure, the temperature was ramped up to 700°C in 10 min and kept at maximum temperature. After 60 min, the temperature was lowered to 320°C within 30 min to allow better crystallization of the material, followed by quenching to room temperature. The pressure was then brought down to ambient pressure in 690 min. The platinum crucible was extracted from the octahedron and disassembled by unwrapping the platinum foil, thus releasing the sample. The sample appeared colorless and crystalline and was stable under air.

X‐Ray Powder Diffraction and Temperature‐Dependent X‐Ray Powder Diffraction

4.2

An X‐ray powder pattern of the sample was collected at both ambient and high‐temperature of up to 950°C using a transmission geometry setup on a STOE Stadi P diffractometer (STOE & Cie GmbH, Darmstadt, Germany) with (111) curved Ge‐monochromatized Mo‐K _ α1_ radiation (*λ *= 70.93 pm). A 1280‐strip DECTRIS Mythen 1K microstrip detector (Dectris AG, Baden‐Daettwil, Switzerland) was used to collect the diffraction intensities. The sample was well ground and mounted centrally in between two thin acetate foils with the help of grease for random orientation of the crystallites and for the foils to stick together. Data was collected in a 2.0° –70.0° 2Theta range with a step‐size of 0.015° and an exposure time of 25 s for each step. The software package associated with the diffractometer was WinXPOW 3.07 [84]. For high‐temperature X‐ray diffraction (HT‐PXRD), the STOE Stadi P powder diffractometer was fitted with a STOE resistance graphite furnace. A 0.3 mm Mark capillary was filled with the sample while making sure that the sample was well compacted without any gaps left. The capillary placed in the graphite furnace and heated until 950°C. The heating rate of the sample was performed at a rate of 50°C min^−1^ when going from 100°C–300°C and a rate of 25°C min^−1^ when going from 300°C–950°C. After each temperature step, data was collected from 9° to 32° 2Theta. Rietveld refinement was carried out with the software Diffrac^plus^‐Topas 4.2 (Bruker AXS, Karlsruhe, Germany) whereby the initial atomic coordinates were derived from the single‐crystal structure model presented here. Modified Thompson–Cox–Hastings pseudo‐Voigt profiles [85, 86] were used for peak shape modeling. Instrumental contributions on reflection profiles were corrected from the refinement of a LaB_6_ standard [87]. Background fitting was done using Chebychev polynomials up to the 8^th^ order.

Single‐Crystal X‐Ray Diffraction

4.3

Crystals were first selected under a polarizing microscope (Leica M125) by their dispersion in a drop of perfluoropolyalkylether (viscosity 1800) on a glass slide and mounted onto the tip of MicroMounts (MiTeGen, LLC, Ithaca, NY, USA). Single‐crystal data for Sr_4_Te_4_O_15_ was collected at room temperature using a Bruker D8 Quest (Bruker, Billerica, USA) with a PhotonIII C14 area detector. Data collection, unit‐cell refinement, data reduction, and multiscan absorption correction were performed using the programs SAINT [88], APEX4 [89], and SADABS [90]. The structural data were solved using intrinsic phasing with the program ShelXT and refined with the program ShelXL 2018/3 using the least‐squares refinement method, both embedded in the Olex2 software package [91]. The space group Pnma was chosen and Platon [92] was used to verify that no additional symmetry was missed.

Energy‐Dispersive X‐Ray Spectroscopy (EDX)

4.4

Several crystals of Sr_4_Te_4_O_15_ were embedded in epoxy resin, polished with diamond material (1/4 µm), and coated with a conductive carbon layer of 25 nm thickness. Semiquantitative EDX‐analysis was performed on a JEOL JXA‐iSP100 electron microprobe equipped with a fully integrated dry SD30 EDX‐detector unit (30 mm^2^ active area). Operating conditions were 15 kV acceleration voltage and 10 nA beam current. Deducting the dead time of approximately 25%, the live measurement time was 20 sec with a count rate of 31 000–34 000 cps. The atomic mass ratio of the cations was determined by using the standardless integrated JEOL software, based on the PRZ quantification method. Carbon, as the coating material, was considered for the matrix correction, but excluded from the semiquantitative results.

Thermal Analysis

4.5

Simultaneous thermal analysis (STA) consisting of both differential scanning calorimetry (DSC) and thermogravimetric (TGA) measurements of Sr_4_Te_4_O_15_ was carried out using a Netsch STA 449F3 (Netzsch GmbH, Selb, Germany). A sample of approximately 15 mg was weighed inside an alumina crucible (Netzsch GmbH, Selb, Germany) and measured using a flowing argon atmosphere of 50 mL min^−1^ preceding a background correction scan using a clean and empty alumina crucible. The program of both the background scan and the sample measurement scan were set at a 10 °C/min ramp to 1000 °C from room temperature. The data from the background scan was then subtracted from the data for the sample measurement using the associated Proteus software [93].

Infrared Spectroscopy

4.6

The infrared spectrum of powdered Sr_4_Te_4_O_15_ was recorded in the 400–4000 cm^−1^ range using a Bruker Alpha Platinum FTIR‐ATR spectrometer (Bruker, Billerica, USA) equipped with a 2 × 2 mm diamond ATR crystal. A DTGS detector was used to acquire the signal over 24 scans. Atmospheric contributions were corrected using a background reference measurement processed with Opus software [94].

Raman Spectroscopy

4.7

Raman spectroscopic analyses were conducted at ambient conditions using a HORIBA JOBIN‐YVON LabRam‐HR800 spectrometer with an 800 mm focal length. Each Raman spectrum represents the average of three measurements, each acquired with a counting time of 1 s. The analyzed areas were excited using a green Nd‐YAG laser with an excitation wavelength of 532.18 nm. The scattered light was dispersed by an optical grating with 1800 grooves/mm, providing a spectral resolution of 1.83 cm^−1^, and detected using a cooled Andor CCD detector. An Olympus 50× objective, a 1000 μm confocal pinhole, and a 100 μm slit were used to optimize signal intensity, yielding a lateral resolution of approximately 5 μm^2^. Spectra were collected in multi‐window acquisition mode in the frequency range of 50–900 cm^−1^. Prior to analysis, the Raman mode of a Si standard wafer was measured relative to the Rayleigh line, confirming the expected Raman shift at 520.7 cm^−1^.

UV‐Vis Spectroscopy

4.8

The diffuse reflectance spectrum of powdered Sr_4_Te_4_O_15_ was measured over the 200–2500 nm range using an Agilent Cary 5000 UV‐Vis spectrometer equipped with an integrating sphere (DRA‐2500). Measurements used a D65 standard illuminant and a 10° complementary observer. A scan rate of 600 nm·min^−1^ and a data interval of 1 nm were used, with BaSO_4_ serving as the white reference. Reflectance data were converted to optical absorbance using the Kubelka–Munk (K–M) function, and the bandgap was estimated via Tauc plots [79, 80].

Quantum Chemical Calculations

4.9

Theoretical calculations of the novel Sr_4_Te_4_O_15_ were carried out using the solid‐state program Crystal23 [95] at DFT level using the hybrid HSEsol functional optimized for the treatment of solid‐state systems [96]. The basis sets developed by Heyd et al. have been applied to Sr and Te [97], for O the basis set reported by Mahmoud et al. has been used [98]. A shrinking factor of 8 was applied to set up Brillouin‐zone (BZ) sampling along with convergence criteria for energy and forces of 10^−8^ Hartree and 4.5 · 10^−4^ Hartree Bohr^−1^, respectively. Electronic band structure and density of states (DOS) calculations were executed using the resulting minimum configuration and identical calculation settings. The electron localization function (ELF) was subsequently calculated using Multiwfn [99] and visualized in VMD [100]. To investigate the ionic character of Sr_4_Te_4_O_15_, a Hirshfeld population analysis as implemented in Crystal23 was carried out [101]. The X‐ray powder‐diffraction patterns of the optimized structure was generated using RIETAN‐FP [102] available in the VESTA program [103] and compared to the experimental reference. Due to the larger number of irreducible atoms and the reduced symmetry when calculating the numerical Hessian, the application of a smaller one‐electron basis proved necessary in the computation of vibrational spectra. The double‐zeta basis sets provided by Laun et al. [104], Heyd et al. [97], and Oliveira et al*.* [105] have been applied to Sr, Te, and O atoms, respectively. The infrared and Raman spectra were calculated at the minimum configuration based on the harmonic approximation [106]. The resulting theoretical line spectra were then subject to a weighted kernel density estimation based on a Gaussian kernel to achieve a better resemblance of and comparison with experimental spectral data [107].

Supporting Information

Additional supporting information can be found online in the Supporting Information section. Supporting Fig. S1: Back scattered electron (BSE) image of Sr_4_Te_4_O_15_ showing spectral data points 2–5. Supporting Fig. S2: BSE image of Sr_4_Te_4_O_15_ showing spectral data points 6 and 7. Supporting Fig. S3: BSE image of Sr_4_Te_4_O_15_ showing spectral data points 8–11. Supporting Fig. S4: BSE image of Sr_4_Te_4_O_15_ showing spectral data points 12 and 13. Supporting Fig. S5: Semiquantitative EDX spectrum 2. Supporting Fig. S6: Semiquantitative EDX spectrum 3. Supporting Fig. S7: Semiquantitative EDX spectrum 4. Supporting Fig. S8: Semiquantitative EDX spectrum 5. Supporting Fig. S9: Semiquantitative EDX spectrum 6. Supporting Fig. S10: Semiquantitative EDX spectrum 7. Supporting Fig. S11: Semiquantitative EDX spectrum 8. Supporting Fig. S12: Semiquantitative EDX spectrum 9. Supporting Fig. S13: Semiquantitative EDX spectrum 10. Supporting Fig. S14: Semiquantitative EDX spectrum 11. Supporting Fig. S15: Semiquantitative EDX spectrum 12. Supporting Fig. S16: Semiquantitative EDX spectrum 13. Supporting Fig. S17: Overlay of the experimental single‐crystal XRD pattern with that obtained from the DFT‐optimized structure of Sr_4_Te_4_O_15_. Compared to the calculated pattern, the experimental pattern shows a clear shift towards lower angles, which can be attributed to the difference in unit cell size between the experimental measurement temperature of 299 K and the temperature of 0 K assumed by the DFT optimization. Supporting Fig. S18: HT‐PXRD and corresponding matching reflections for SrTeO_4_ at a temperature of 755°C. The reflections of the HT‐PXRD pattern are shifted slightly to the left due to unit cell expansion induced by the high temperature. Supporting Fig. S19: HT‐PXRD and corresponding matching reflections for the hight temperature polymorph δ‐SrTeO_3_ (507°C) at a temperature of 850°C. The reflections of the HT‐PXRD pattern are shifted slightly to the left due to unit cell expansion induced by the higher temperature. Supporting Fig. S20: DSC/TGA plot of a sample of Sr_4_Te_4_O_15_. Supporting Fig. S21: Powder diffractogram of sample remaining after DSC‐TGA of Sr_4_Te_4_O_15_. Supporting Table S1: Atomic coordinates, Wyckoff positions, and equivalent isotropic displacement parameters U eq [Å2] for Sr_4_Te_4_O_15_ (space group: Pnma). U eq is defined as one third of the trace of the orthogonalized U ij tensor. Site occupancy factors (SOF) are equal to 1 for each site. (standard deviations given in parentheses). Supporting Table S2: Te–O and Sr–O bond distances with standard deviations in parentheses. Supporting Table S3: Anisotropic displacement parameters U ij (Å2) for Sr_4_Te_4_O_15_ (space group Pnma). Standard deviations are given in parentheses. The anisotropic displacement factor exponent takes the form: −2π^2^[(ha*)2U 11+···+2hkabU 12]. Supporting Table S4: Semiquantitative EDX quantification results of spectrum 2. Supporting Table S5: Semiquantitative EDX quantification results of spectrum 3. Supporting Table S6: Semiquantitative EDX quantification results of spectrum 4. Supporting Table S7: Semiquantitative EDX quantification results of spectrum 5. Supporting Table S8: Semiquantitative EDX quantification results of spectrum 6. Supporting Table S9: Semiquantitative EDX quantification results of spectrum 7. Supporting Table S10: Semiquantitative EDX quantification results of spectrum 8. Supporting Table S11: Semiquantitative EDX quantification results of spectrum 9. Supporting Table S12: Semiquantitative EDX quantification results of spectrum 10. Supporting Table S13: Semiquantitative EDX quantification results of spectrum 11. Supporting Table S14: Semiquantitative EDX quantification results of spectrum 12. Supporting Table S15: Semiquantitative EDX quantification results of spectrum 13.

Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Funding

This study was supported by the Austrian Science Fund (10.55776/P35602).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material