Thulium sulfide (Tm2S3)

• CAS: 12166-30-2
• Molecular Formula: S3Tm2
• Molecular Weight: 434.1 g/mol
• SMILES: [S-2].[S-2].[S-2].[Tm+3].[Tm+3]
• InChI: InChI=1S/3S.2Tm/q3*-2;2*+3
• InChIKey: CVSQSQKQZKJEGY-UHFFFAOYSA-N

Rare-earth sesquisulfide procurement often ignores polymorph-specific properties-risking device failure. Tm₂S₃ (CAS 12166-30-2) is the quantitatively characterized δ-phase semiconductor with validated performance metrics. • 5-order photocurrent enhancement at δ→α phase transition (~5 GPa) with 3-order responsivity retention post-release • Lattice parameters a=10.0626Å, b=3.9675Å, c=17.396Å, β=98.655° for epitaxial lattice matching • Congruent melting point 1665°C; melting enthalpy 46.9 kJ mol⁻¹; Tm₂S₃-specific sulfur overpressure protocols for crystal growth. Supplied as phase-pure powder with full analytical documentation.

Thulium Sulfide (Tm₂S₃) Baseline Characterization

Thulium sulfide (Tm₂S₃, CAS 12166-30-2) is a heavy rare-earth sesquisulfide semiconductor belonging to the Ln₂S₃ family, crystallizing in multiple polymorphic forms depending on synthesis temperature and pressure conditions. The δ-Tm₂S₃ phase adopts the monoclinic δ-Ho₂S₃ structure type (space group P2₁/m, a = 10.0626(14) Å, b = 3.9675(5) Å, c = 17.396(2) Å, β = 98.655(11)°), characterized by Tm³⁺ coordination numbers of 6 (distorted octahedron) and 7 (mono-capped trigonal prism) [1]. Its optical band gap, measured at 2.96 eV at 1.2 GPa by UV-vis absorption spectroscopy, places it among the wide-gap semiconductors with relevance to photodetector and optoelectronic applications [2]. Unlike more commonly investigated rare-earth sulfides such as Y₂S₃ or Gd₂S₃, Tm₂S₃ exhibits a distinctive combination of a moderate direct band gap, pressure-tunable photoelectric response, and well-characterized thermal stability, making it a specialized candidate for high-pressure photonic devices and solid-state ion conductors where precise structural and electronic control is required [3].

• [1] J. Kończyk, P. Demchenko, O. Bodak, G. Demchenko, B. Marciniak, W. Prochwicz, L. Muratova. The crystal structure of the d-modification of thulium sesquisulfide, d-Tm₂S₃. Chem. Met. Alloys 1 (2008) 38-42. DOI: 10.30970/cma1.0021. View Source
• [2] Zhi-Wei Shen, Zhong-Yan Wu, Shao-Jie Wang, He-Chong Wang, Hong-Kai Li, Jing Song, Guo-Ying Gao, Lin Wang, Yong-Jun Tian. Pressure-Induced Structural Transition and Enhanced Photoelectric Properties of Tm₂S₃. Chinese Physics Letters 41 (2024) 117101. DOI: 10.1088/0256-307X/41/11/117101. View Source
• [3] T.V. Mikhailichenko, L.A. Kalinina, Yu.N. Ushakova, I.V. Murin. Ion conductivity type and thermodynamics of dissolution of binary sulfides Sm₂S₃ and Tm₂S₃ in BaSm₂S₄. Russian Journal of Electrochemistry 49 (2013) 776-782. DOI: 10.1134/S1023193513080132. View Source

Why Generic Substitution Fails for Tm₂S₃

Rare-earth sesquisulfides (Ln₂S₃) cannot be treated as interchangeable procurement commodities because their structural polymorphism, band gap, melting behavior, and ionic transport properties are exquisitely sensitive to the specific rare-earth cation radius and 4f electronic configuration. Within the δ-structure type spanning Dy to Tm, Tm₂S₃ occupies a unique position as the heaviest lanthanide to adopt this monoclinic phase before the structural transition to ε-type (corundum) occurs at Yb and Lu [1]. Its melting temperature (1665 °C) and enthalpy of melting (46.9 kJ mol⁻¹) are distinct from its nearest δ-phase neighbors, δ-Ho₂S₃ (1697 °C, 47.7 kJ mol⁻¹) and δ-Er₂S₃ (1662 °C, 42.2 kJ mol⁻¹), reflecting measurable thermodynamic differences with direct implications for high-temperature processing and phase stability [2]. Furthermore, the pressure-induced δ→α phase transition at approximately 5 GPa yields a five-order-of-magnitude photocurrent enhancement in Tm₂S₃—a quantitatively characterized behavior that cannot be assumed for other Ln₂S₃ compounds without analogous experimental validation [3]. Generic substitution therefore risks both structural mismatch in crystalline devices and unverified photoelectric performance under operational conditions.

• [1] Crystal structure and microstructure of δ-Er₂S₃. Journal of Alloys and Compounds 323-324 (2001) 47-51. DOI: 10.1016/S0925-8388(01)00981-1 (referencing δ-Ho₂S₃ structure type spanning Dy to Tm). View Source
• [2] Temperatures and enthalpies of melting of Ln₂S₃ (Ln = Gd, Tb, Dy, Ho, Er, Tm, Yb, and Lu) compounds. Journal of Thermal Analysis and Calorimetry 131 (2018) 1057-1063. DOI: 10.1007/s10973-017-6620-X. View Source
• [3] Zhi-Wei Shen, Zhong-Yan Wu, Shao-Jie Wang, He-Chong Wang, Hong-Kai Li, Jing Song, Guo-Ying Gao, Lin Wang, Yong-Jun Tian. Pressure-Induced Structural Transition and Enhanced Photoelectric Properties of Tm₂S₃. Chinese Physics Letters 41 (2024) 117101. View Source

Tm₂S₃ Quantitative Evidence for Procurement

Pressure-Induced Photocurrent Enhancement: δ→α Transition

Under applied pressure, δ-Tm₂S₃ undergoes a first-order structural transition to the α-phase at approximately 5 GPa. This transition is accompanied by a dramatic enhancement of photoelectric performance: the photocurrent density and responsivity of α-Tm₂S₃ achieve a five orders of magnitude (10⁵×) increase compared with the initial δ-Tm₂S₃ baseline, and the α-phase retains a photocurrent responsivity three orders of magnitude above baseline after complete pressure release (unloading) [1]. The optical band gap of δ-Tm₂S₃ measured at 1.2 GPa is 2.96 eV by the Tauc plot method, confirming wide-gap semiconductor behavior appropriate for visible-blind UV detection [1]. In contrast, Y₂S₃—another δ-structure sesquisulfide—exhibits an irreversible first-order phase transition from monoclinic to orthorhombic at approximately 10 GPa, i.e., at roughly twice the transition pressure of Tm₂S₃, with a qualitatively different pressure-response profile . This lower transition threshold and the retention of enhanced responsivity post-unloading represent a quantifiable operational advantage for Tm₂S₃ in pressure-cycled or strain-engineered photodetector architectures.

Evidence Dimension	Pressure-induced photocurrent enhancement factor (α-phase vs. δ-phase baseline, same compound) and phase transition pressure vs. structural analog
Target Compound Data	Tm₂S₃: 10⁵× photocurrent enhancement at δ→α transition (~5 GPa); α-phase retains 10³× responsivity after unloading; optical band gap 2.96 eV at 1.2 GPa
Comparator Or Baseline	Y₂S₃ (δ-structure analog): irreversible monoclinic→orthorhombic transition at ~10 GPa (no reported comparable magnitude of photocurrent enhancement); baseline δ-Tm₂S₃ photocurrent prior to transition
Quantified Difference	Tm₂S₃ phase transition at ~5 GPa vs. Y₂S₃ at ~10 GPa (factor of ~2 lower threshold); 100,000× photocurrent increase in Tm₂S₃ vs. pre-transition baseline
Conditions	High-pressure diamond anvil cell; electrical impedance, UV-vis absorption, Raman spectroscopy, XRD, and photoelectric measurements under compression to >5 GPa; room temperature

Why This Matters

The five-orders-of-magnitude photocurrent enhancement at half the transition pressure of Y₂S₃ makes Tm₂S₃ the quantitatively superior candidate for strain-tunable photodetectors where lower actuation pressure and persistent post-release responsivity reduce device engineering complexity.

• [1] Zhi-Wei Shen, Zhong-Yan Wu, Shao-Jie Wang, He-Chong Wang, Hong-Kai Li, Jing Song, Guo-Ying Gao, Lin Wang, Yong-Jun Tian. Pressure-Induced Structural Transition and Enhanced Photoelectric Properties of Tm₂S₃. Chinese Physics Letters 41 (2024) 117101. DOI: 10.1088/0256-307X/41/11/117101. View Source

δ-Tm₂S₃ vs. δ-Er₂S₃ Lattice Precision

The δ-Tm₂S₃ crystal structure has been determined by single-crystal X-ray diffraction with high precision: space group P2₁/m, a = 10.0626(14) Å, b = 3.9675(5) Å, c = 17.396(2) Å, β = 98.655(11)°, with refinement residuals R1 = 0.0425 and wR2 = 0.0834 from 1425 unique reflections [1]. The isostructural δ-Er₂S₃, refined from a single crystal with R = 0.037 for 1745 observed reflections, yields lattice parameters a = 10.072(1) Å, b = 3.976(2) Å, c = 17.389(2) Å, β = 98.66(1)° [2]. The difference in unit cell volume reflects the lanthanide contraction: Tm³⁺ has a smaller ionic radius than Er³⁺, resulting in a measurably contracted b-axis (3.9675 Å vs. 3.976 Å) in Tm₂S₃. This subtle but crystallographically significant contraction directly affects Tm–S bond distances, coordination polyhedron distortion, and consequently the electronic band structure near the absorption edge.

Evidence Dimension	Unit cell lattice parameters and refinement quality for isostructural δ-phase sesquisulfides
Target Compound Data	δ-Tm₂S₃: a = 10.0626(14) Å, b = 3.9675(5) Å, c = 17.396(2) Å, β = 98.655(11)°; R1 = 0.0425, wR2 = 0.0834 (1425 unique reflections)
Comparator Or Baseline	δ-Er₂S₃: a = 10.072(1) Å, b = 3.976(2) Å, c = 17.389(2) Å, β = 98.66(1)°; R = 0.037 (1745 observed reflections)
Quantified Difference	b-axis contraction: 3.9675 Å (Tm) vs. 3.976 Å (Er), difference 0.0085 Å (~0.2%); systematic volume reduction consistent with lanthanide contraction from Er to Tm
Conditions	Single-crystal X-ray diffraction; automatic diffractometer; room temperature; δ-Ho₂S₃ structure type; monoclinic P2₁/m

Why This Matters

The precisely characterized lattice contraction in Tm₂S₃ relative to Er₂S₃ enables predictable epitaxial strain engineering in heterostructure devices, where even sub-picometer lattice mismatch critically determines interface quality and electronic transport.

• [1] J. Kończyk, P. Demchenko, O. Bodak, G. Demchenko, B. Marciniak, W. Prochwicz, L. Muratova. The crystal structure of the d-modification of thulium sesquisulfide, d-Tm₂S₃. Chem. Met. Alloys 1 (2008) 38-42. DOI: 10.30970/cma1.0021. View Source
• [2] Crystal structure and microstructure of δ-Er₂S₃. Journal of Alloys and Compounds 323-324 (2001) 47-51. DOI: 10.1016/S0925-8388(01)00981-1. View Source

Melting Behavior vs. Heavy Lanthanide Sesquisulfides

Differential scanning calorimetry (DSC) of heavy rare-earth sesquisulfide polymorphs provides the first systematic melting data for the Ln₂S₃ series. δ-Tm₂S₃ exhibits congruent melting at tm = 1665 °C with a melting enthalpy of ΔH = 46.9 kJ mol⁻¹ [1]. Its nearest δ-phase neighbors display clearly distinguishable thermal signatures: δ-Ho₂S₃ melts higher (tm = 1697 °C, ΔH = 47.7 kJ mol⁻¹), while δ-Er₂S₃ melts at a comparable temperature but with a significantly lower enthalpy (tm = 1662 °C, ΔH = 42.2 kJ mol⁻¹). The next heaviest Ln₂S₃, ε-Lu₂S₃, crystallizes in a different structure type (corundum-type ε-phase) and melts substantially higher (tm = 1755 °C, ΔH = 53.6 kJ mol⁻¹). Notably, δ-Tm₂S₃ experiences a greater mass loss of 0.77 mass% at 1500–1670 °C (acquiring the concentration Tm₂S₂.₉₁), compared to 0.1–0.24 mass% for the other Ln₂S₃ compounds [1], indicating a quantitatively distinct thermal dissociation behavior near the melting point.

Evidence Dimension	Congruent melting temperature and enthalpy of melting for heavy Ln₂S₃ polymorphs
Target Compound Data	δ-Tm₂S₃: tm = 1665 °C; ΔH = 46.9 kJ mol⁻¹; mass loss 0.77 mass% (1500–1670 °C), yielding Tm₂S₂.₉₁
Comparator Or Baseline	δ-Ho₂S₃: tm = 1697 °C, ΔH = 47.7 kJ mol⁻¹; δ-Er₂S₃: tm = 1662 °C, ΔH = 42.2 kJ mol⁻¹; ε-Lu₂S₃: tm = 1755 °C, ΔH = 53.6 kJ mol⁻¹; all others mass loss 0.1–0.24 mass%
Quantified Difference	Tm vs. Er: Δtm = +3 °C, Δ(ΔH) = +4.7 kJ mol⁻¹ (11.1% higher enthalpy); Tm vs. Ho: Δtm = −32 °C; Tm mass loss ~3-8× greater than other Ln₂S₃
Conditions	DSC melting traces; samples prepared from Ln₂O₃ in CS₂ + H₂S flow at 1050–1150 °C; polycrystalline samples melted and recrystallized in DSC crucibles; argon atmosphere

Why This Matters

The 11.1% higher melting enthalpy of Tm₂S₃ compared to its closest δ-phase neighbor Er₂S₃, combined with its distinct mass loss profile, has direct implications for crystal growth process design—requiring tailored thermal budgets that cannot be extrapolated from Er₂S₃ or Ho₂S₃ parameters.

• [1] Temperatures and enthalpies of melting of Ln₂S₃ (Ln = Gd, Tb, Dy, Ho, Er, Tm, Yb, and Lu) compounds. Journal of Thermal Analysis and Calorimetry 131 (2018) 1057-1063. DOI: 10.1007/s10973-017-6620-X. View Source

Ion Conductivity: Tm₂S₃ vs. Sm₂S₃ in BaSm₂S₄

The thermodynamics of dissolution and ion conductivity type have been directly compared for Tm₂S₃ and Sm₂S₃ as binary sulfide components dissolved in the BaSm₂S₄ matrix (CaFe₂O₄ structure type). Solid solutions of both Sm₂S₃ and Tm₂S₃ based on BaSm₂S₄ were synthesized and characterized by XRD and microprobe analysis, with the length of the solid solution region determined for each [1]. Partial molar enthalpies of dissolution of the binary sulfides in the basic thiosamarate host were measured using a modified EMF technique, enabling quantitative comparison of the thermodynamic driving force for incorporation [1]. A possible mechanism of defect formation was proposed, linking the observed ion conductivity type to the specific rare-earth cation identity [1]. The comparison between Sm₂S₃ (a lighter lanthanide with larger ionic radius) and Tm₂S₃ (a heavier lanthanide with smaller ionic radius) directly tests how cation size modulates ionic transport in this ternary sulfide electrolyte system, establishing Tm₂S₃ as the heavy-lanthanide-end member reference point for systematic composition-property mapping.

Evidence Dimension	Thermodynamics of dissolution and ion conductivity type of binary sulfides in BaSm₂S₄ solid electrolyte matrix
Target Compound Data	Tm₂S₃ in BaSm₂S₄: ion conductivity type characterized; partial molar enthalpy of dissolution measured by modified EMF technique; solid solution region length determined by XRD
Comparator Or Baseline	Sm₂S₃ in BaSm₂S₄: same measurements performed in parallel study; provides light-lanthanide vs. heavy-lanthanide comparison within identical host matrix
Quantified Difference	Differential thermodynamic behavior attributed to cation size difference (Sm³⁺ vs. Tm³⁺ ionic radius); specific numerical values for partial molar enthalpies available in full article
Conditions	Solid-state synthesis; BaSm₂S₄ host (CaFe₂O₄ structure type); microprobe analysis and XRD phase identification; modified EMF technique for thermodynamic measurements

Why This Matters

The availability of direct head-to-head thermodynamic and conductivity data for Tm₂S₃ vs. Sm₂S₃ in the identical BaSm₂S₄ matrix provides procurement justification for Tm₂S₃ as the heavy-lanthanide reference material in systematic studies of cation-radius-dependent ion transport in ternary rare-earth sulfide solid electrolytes.

• [1] T.V. Mikhailichenko, L.A. Kalinina, Yu.N. Ushakova, I.V. Murin. Ion conductivity type and thermodynamics of dissolution of binary sulfides Sm₂S₃ and Tm₂S₃ in BaSm₂S₄. Russian Journal of Electrochemistry 49 (2013) 776-782. DOI: 10.1134/S1023193513080132. View Source

Optical Band Gap Systematics in Ln₂S₃

Optical band gaps of rare-earth sesquisulfides have been systematically measured by diffuse reflectance spectroscopy across a complete suite of Ln₂S₃ compounds of varying structure type. Band gaps range from 2.2 to 3.5 eV, with the primary determinant being the crystal structure type (α, δ, ε, τ) rather than the specific rare-earth ion identity [1]. Within a given structure type, compounds exhibit similar band gaps independent of the particular lanthanide. Tm₂S₃, as a δ-structure compound, falls within the band gap range characteristic of this monoclinic structure type. The ordered α-structure consistently exhibits a substantially lower band gap than the defect γ-structure for all compounds examined except La₂S₃ [2]. This structure-dominated band gap systematics means that selecting Tm₂S₃ (δ-type) versus, for example, ε-Yb₂S₃ or ε-Lu₂S₃ (corundum-type) is a deliberate choice of optical absorption edge range—not merely a cation substitution—since the band gap is primarily a function of the crystallographic polymorph rather than the rare-earth identity.

Evidence Dimension	Optical band gap as a function of crystal structure type across the Ln₂S₃ series
Target Compound Data	Tm₂S₃ (δ-structure type): band gap within the δ-phase range (~2.2–3.5 eV across Ln₂S₃ series; experimentally 2.96 eV at 1.2 GPa)
Comparator Or Baseline	α-structure Ln₂S₃: substantially lower band gap than γ-structure for all compounds except La₂S₃; ε-structure (Yb₂S₃, Lu₂S₃): distinct band gap range associated with corundum-type polymorph
Quantified Difference	Band gap varies by structure type: α < γ for all Ln (except La); δ-type Tm₂S₃ occupies an intermediate band gap regime within the 2.2–3.5 eV Ln₂S₃ range, distinct from both α and ε polymorphs
Conditions	Diffuse reflectance spectroscopy on powder samples; band gaps extracted from absorption edge spectra; room temperature; compounds prepared by oxide sulfidization

Why This Matters

The structure-dominated band gap systematics means that Tm₂S₃ cannot be substituted by an ε-phase sulfide (Yb₂S₃, Lu₂S₃) without changing the optical absorption edge, nor by an α-phase sulfide without accepting a significantly lower band gap—making polymorph identity a critical procurement specification for optical applications.

• [1] Oleg Schevciw, William B. White. The optical absorption edge of rare earth sesquisulfides and alkaline earth - rare earth sulfides. Materials Research Bulletin 18 (1983) 1059-1068. DOI: 10.1016/0025-5408(83)90147-2. View Source
• [2] Optical absorption edge in rare earth sesquisulfides. Materials Research Bulletin 41 (2006) 448-454. DOI: 10.1016/j.materresbull.2005.07.035. View Source

Si-Doped Tm₂S₃ Thin-Film Transport Properties

The electrical transport properties of Si-doped θ-Tm₂S₃ thin films have been systematically characterized as a function of temperature. Temperature-dependent specific resistivity and thermo-EMF (Seebeck coefficient) were measured in the range 90–500 K for Si-doped θ-Tm₂S₃ films [1]. Spectral dependences of photoconductivity and photo-EMF were investigated in the photon energy range 0.2–3.3 eV [1]. The results suggest that acceptor levels formed by vacancies in the cationic sublattice play a substantial role in the photoconductivity process, with silicon atoms acting as compensating dopants [1]. This doping study provides the only available quantitative transport data for intentionally doped Tm₂S₃ thin films, establishing a baseline for thermoelectric and photoconductive device evaluation that is specific to the θ-polymorph of Tm₂S₃ and not transferable to other Ln₂S₃ compositions without independent verification.

Evidence Dimension	Temperature dependence of specific resistivity and thermo-EMF (Seebeck coefficient) for Si-doped thin films
Target Compound Data	Si-doped θ-Tm₂S₃ thin films: resistivity and thermo-EMF measured 90–500 K; photoconductivity and photo-EMF spectral response 0.2–3.3 eV
Comparator Or Baseline	No direct comparator data available for identically doped Ln₂S₃ thin films under identical measurement conditions; this evidence establishes the Tm₂S₃-specific baseline for future comparative studies
Quantified Difference	Quantitative values for resistivity, thermo-EMF, photoconductivity, and photo-EMF available in the full article; no directly comparable dataset exists for other doped Ln₂S₃ thin films in the 90–500 K range
Conditions	Thin films prepared by vacuum-thermal evaporation; Si doping by co-deposition; temperature range 90–500 K; spectral range 0.2–3.3 eV photon energy

Why This Matters

This is the only available systematic transport dataset for intentionally doped Tm₂S₃ thin films, providing essential baseline parameters for evaluating Tm₂S₃ as a thermoelectric or photoconductive thin-film material—parameters that cannot be inferred from bulk crystal or other Ln₂S₃ data.

• [1] Z.U. Dzhabua, M.G. Teteloshvili, A.V. Gigineishvili. Silicon doping and photoelectric properties of thulium sesquisulfide thin films. Physical Surface Engineering 12(2) (2014) 232-236. View Source

Tm₂S₃ Application Scenarios

Strain-Engineered Photodetectors

Tm₂S₃ is the quantitatively demonstrated candidate for photodetectors employing pressure-induced band gap engineering. The five-orders-of-magnitude photocurrent enhancement at the δ→α phase transition (~5 GPa), combined with the retention of three-orders-of-magnitude responsivity after pressure release [1], enables device architectures where a single pressure-cycling step permanently enhances detector sensitivity. The lower transition pressure (~5 GPa) compared to Y₂S₃ (~10 GPa) reduces the mechanical requirements for device packaging and actuation. This scenario is directly supported by the evidence in Section 3 (Evidence Item 1).

Epitaxial Heterostructures with δ-Phase Sesquisulfides

The precisely determined lattice parameters of δ-Tm₂S₃ (a = 10.0626 Å, b = 3.9675 Å, c = 17.396 Å, β = 98.655°) [2] enable calculated lattice mismatch engineering when paired with isostructural δ-Er₂S₃ or δ-Ho₂S₃. The ~0.2% b-axis contraction relative to δ-Er₂S₃ provides a quantifiable strain budget for coherent epitaxial growth of Tm₂S₃/Er₂S₃ superlattices. This scenario is directly supported by the evidence in Section 3 (Evidence Item 2).

Heavy-Lanthanide Reference for Solid Electrolyte Development

Tm₂S₃ serves as the heavy-lanthanide-end-member reference in systematic studies of cation-radius-dependent ion conductivity in BaLn₂S₄-type solid electrolytes. The parallel thermodynamic and conductivity characterization of Tm₂S₃ and Sm₂S₃ in identical BaSm₂S₄ matrices [3] provides the experimental framework for composition-property mapping across the lanthanide series, where Tm³⁺ represents the small-cation extreme of the δ-phase stability field. This scenario is directly supported by the evidence in Section 3 (Evidence Item 4).

Crystal Growth Process Design Using Thermal Data

Crystal growers targeting δ-Tm₂S₃ single crystals for research or device prototyping can use the validated congruent melting point (1665 °C) and melting enthalpy (46.9 kJ mol⁻¹) [4] to design thermal profiles. The quantitatively distinct mass loss behavior (0.77 mass% vs. 0.1–0.24 mass% for other Ln₂S₃) necessitates Tm₂S₃-specific sulfur overpressure compensation during melt growth—a requirement that cannot be met by extrapolating from Er₂S₃ or Ho₂S₃ growth protocols. This scenario is directly supported by the evidence in Section 3 (Evidence Item 3).

• [1] Zhi-Wei Shen et al. Pressure-Induced Structural Transition and Enhanced Photoelectric Properties of Tm₂S₃. Chinese Physics Letters 41 (2024) 117101. View Source
• [2] J. Kończyk et al. The crystal structure of the d-modification of thulium sesquisulfide, d-Tm₂S₃. Chem. Met. Alloys 1 (2008) 38-42. View Source
• [3] T.V. Mikhailichenko et al. Ion conductivity type and thermodynamics of dissolution of binary sulfides Sm₂S₃ and Tm₂S₃ in BaSm₂S₄. Russian Journal of Electrochemistry 49 (2013) 776-782. View Source
• [4] Temperatures and enthalpies of melting of Ln₂S₃ (Ln = Gd, Tb, Dy, Ho, Er, Tm, Yb, and Lu) compounds. Journal of Thermal Analysis and Calorimetry 131 (2018) 1057-1063. View Source