Pith. sign in

REVIEW 3 major objections 6 minor 38 references

Physics adapted generative AI for metal insulator transition materials under label scarcity

T0 review · 3 major / 6 minor · reviewed 2026-07-12 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Generative AI for metal–insulator materials should invent switchable phase pairs, not just stable crystals, under severe label scarcity.

desk verdict Clear Perspective that reframes MIT generation as phase-pair hypotheses plus a verification ladder; the modular adapter claim is coherent but untested. read the letter →

arxiv 2607.03578 v1 pith:BNIPALKU submitted 2026-07-03 cond-mat.mtrl-sci

Gourab Datta , Sarah Sharif , Zhisheng Shi , Yaser Banad This is my paper · ORCID
classification cond-mat.mtrl-sci
keywords metal–insulatortransitiongenerativeAIphysics-adaptedmaterialsdiscoveryquantumcrystaldiffusionmodelslow-rankadaptationtransition-metaloxidesandchalcogenidesphase-transitionhypotheses
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This Perspective argues that discovering metal–insulator-transition (MIT) materials is a different problem from ordinary crystal generation. An MIT is not a property of one structure; it is a relation between at least two electronically distinct phases linked by an accessible control parameter such as temperature, strain, or field. Verified examples are scarce—on the order of dozens of clear positives in curated sets—so models cannot be trained directly on the MIT class. The authors therefore propose treating generated structures as mechanism-informed transition hypotheses and steering a pretrained crystal generator with lightweight domain and physics-prior adapters, then filtering candidates through a staged verification ladder of thermodynamic access, phase competition, electronic contrast, pathway plausibility, and device-relevant control. If that program works, generative models become proposal engines for rare switchable electronic matter rather than black-box structure inventers.

What carries the argument

A modular physics-adapted generative stack: a pretrained crystal diffusion model supplies broad inorganic structure priors; domain adapters bias toward transition-metal oxides and chalcogenides; mechanism-prior adapters (implemented via parameter-efficient methods such as low-rank adaptation) steer sampling using scalable physics proxies for correlation, dimerization, charge/orbital order, excitonic or density-wave routes; and a verification ladder tests phase competition, electronic contrast, pathway plausibility, and control accessibility before any discovery claim.

What would settle it

Run the modular generator with and without mechanism-prior adapters on the same chemical space, then measure enrichment: if adapters do not raise the fraction of candidates that pass quantitative gates on phase-pair energy separation, electronic contrast (gap or N(EF) change), soft-mode or pathway connection, and accessible control relative to a generic generator baseline, the central claim fails.

Watch

Extended reading notes

Core claim

The paper’s central claim is that generative AI for MIT discovery should target mechanism-informed phase-transition hypotheses—candidate crystals plus competing phases, electronic contrast, a switching coordinate, and a realistic control route—rather than valid or stable crystals alone, and that a modular physics-adapted generator plus staged verification can do this under severe label scarcity.

Load-bearing premise

The load-bearing premise is that physics-proxy adapters trained on larger oxide and chalcogenide pools—not on the tiny set of verified MIT compounds—will actually enrich the fraction of generated candidates that survive phase-pair and switching tests.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. This Perspective argues that generative AI for metal–insulator-transition (MIT) discovery should target mechanism-informed phase-transition hypotheses rather than stable single crystals alone. Under severe label scarcity (a curated set of ~343 MIT-related entries with only ~68 verified MIT positives), the authors propose a modular pipeline: a pretrained crystal diffusion generator, a domain adapter for transition-metal oxides/chalcogenides, and lightweight mechanism-prior adapters (correlation, Peierls/dimerization, charge/orbital order, excitonic/CDW, disorder) trained from scalable physics proxies rather than from the small verified MIT set. Generated candidates are treated as hypotheses and filtered by a staged verification ladder (chemical validity, thermodynamic accessibility, phase competition, electronic contrast, pathway plausibility, experimental/device accessibility), with explicit early-triage energy and contrast gates. The manuscript synthesizes the multiphysics character of known MITs (VO2, nickelates, 1T-TaS2, Ta2NiSe5, CuIr2S4, etc.), frames inverse design as search over switchable phase landscapes, and outlines a closed-loop outlook for datasets, phase-pair benchmarks, and standardized enrichment reporting.

Significance. The paper addresses a genuine and under-articulated gap: modern crystal generators and stability models optimize single-structure properties, whereas MIT function is a relation between states under a control parameter and device window. The scarcity argument is well grounded in the Georgescu/Rondinelli dataset, the multiphysics examples match the literature, and the verification-first stance (including honest failure modes such as proxy bias and collapse to familiar chemistries) is a useful corrective to overclaiming in generative materials AI. Strengths include a clear modular design philosophy, a concrete physics-prior table (Table 2), operational reporting gates, and a substantial supplementary compilation of the 68 MIT-positive compounds with temperatures and mechanism notes. If adopted, the framework could reorient evaluation of generative models from validity/novelty/stability toward enrichment of testable phase-pair hypotheses. As a Perspective without a trained model or pilot enrichment study, the contribution is conceptual and agenda-setting rather than empirical; its lasting value depends on whether the community can operationalize the proposed proxies and ladder.

major comments (3)
  1. The load-bearing premise of §§3–4 and Table 2—that mechanism-prior adapters trained or conditioned on scalable proxies (U/W, metal–metal distances, bond disproportionation, soft modes, nesting, etc.) over broad oxide/chalcogenide pools will meaningfully enrich accessible, electronically contrasting phase-pair hypotheses—is asserted but not demonstrated. The manuscript correctly reserves the ~68 verified positives for calibration and enrichment evaluation and flags proxy bias and training-set collapse as failure modes, yet it provides no enrichment metric definition, baseline comparison (generic vs domain-adapted vs mechanism-adapted generation), or even a small retrospective pilot against the Georgescu set / Supplementary Table S1. For a Perspective this need not be a full model, but the central claim remains open without a concrete, falsifiable success criterion (e.g., fraction of candi
  2. Section 4 proposes operational triage gates (~50 meV atom−1 of the hull as strong metastability; ~25–50 meV per formula unit or active metal site for phase-pair separation; order-of-magnitude N(EF) suppression or gap opening for contrast). These are useful reporting conventions, but their calibration against known MIT energy scales is thin. Figure 3 and Table 1 emphasize that many canonical transitions sit at tens of meV or less (Fe3O4 ~10 meV, BaVS3 ~6 meV, V2O3 ~13 meV), while high-T systems such as NbO2 are much larger; free-energy, magnetic, and entropic contributions are acknowledged only qualitatively. Please either (i) justify the numerical windows more carefully against computed or experimental energy scales for a subset of the 68 positives, or (ii) reframe them more clearly as provisional reporting gates rather than physics-derived thresholds, and discuss how magnetic order, U c
  3. The manuscript repeatedly states that the generated object should be a phase pair plus control parameter and operating constraint (Abstract; Fig. 4a; §3), yet the proposed implementation builds on single-structure crystal diffusion models with LoRA-style adapters. The architectural path from single-structure generation to joint sampling or sequential proposal of related metallic/insulating (or high-T/low-T) phases is underspecified. Without a clearer sketch—e.g., paired generation under a shared composition, distortion-coordinate conditioning, or post-generation competing-phase search—the claim that the generator itself proposes transition hypotheses (as opposed to single crystals later checked for competition) risks overstating what current modular adapters deliver. Please clarify the intended generation object and the minimal algorithmic steps that turn adapter-biased samples into phas
minor comments (6)
  1. Figure 2’s caption and body text describe a data-scarcity vs target-richness mismatch, but the figure itself (as rendered in the manuscript) is dominated by a schematic of what an MIT label encodes; consider aligning the visual more tightly with the scarcity hierarchy stated in the text (general structures → MIT-related labels → verified positives → mechanism-specific cases).
  2. Table 1 is valuable but mixes control abbreviations, approximate kBT scales, and mechanism status unevenly across rows; a consistent column for approximate energy scale (meV) and a short note on whether the mechanism is contested would improve usability.
  3. Supplementary Figure S1 and Table S1 are substantial assets; the main text should point more explicitly to them when discussing the 68 positives and near-room-temperature coverage (e.g., in §2 and Fig. 3).
  4. A few repeated or near-duplicate references appear (e.g., Shao et al. on VO2 appears more than once; Verwey 1939 and Morin 1959 are cited in both main and supplementary lists). Consolidate where possible.
  5. Terminology alternates between “mechanism-prior adapters,” “physics-prior adapters,” and “mechanism-aware” generation; pick one primary term and use it consistently after first definition.
  6. The device-facing layer (operating window, hysteresis, thin-film compatibility, cycling) is invoked throughout but is thinner than the mechanism layer in Table 2 and Fig. 4b. A short paragraph or table row on how device constraints would enter as adapters or filters would balance the framework.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: Perspective proposes an untested modular framework without fitting MIT labels or reducing predictions to inputs by construction.

full rationale

This is a Perspective, not a derivation paper. It argues that generative AI for MIT discovery should propose mechanism-informed phase-transition hypotheses (phase pair + control + contrast) rather than stable crystals alone, using a pretrained crystal generator, domain adapters, and mechanism-prior adapters trained on scalable physics proxies (U/W, metal–metal distances, bond disproportionation, soft modes; Table 2), with verified MIT compounds reserved for calibration and enrichment evaluation rather than sole supervised training (§3). Success is defined as enrichment of candidates that survive a staged verification ladder (thermodynamic accessibility, phase competition, electronic contrast, pathway, control; §4–5), not as regenerating known MIT labels. There is no fitted parameter renamed as a prediction, no equation that equals its input by construction, no uniqueness theorem imported from the present authors, and no load-bearing self-citation chain. Citations (Georgescu MIT dataset, crystal diffusion models, LoRA, DFT/phonon methods) are external literature framing. Residual open risk is empirical (whether proxy-trained adapters actually enrich true phase pairs), which is correctness/evidence, not circularity. Score 0; steps empty.

Assumptions & free parameters 0 free parameters · 5 assumptions · 2 invented entities

The load-bearing content is a design argument, not a fitted theory. It rests on standard domain facts about MIT multiphysics and generative modeling, plus the paper-specific premise that proxy-trained mechanism adapters plus a verification ladder are the right way to handle label scarcity. No free parameters are fitted; no new physical entities are postulated.

assumptions (5)
  • domain assumption Verified thermally driven MIT labels are scarce (order 10^2 positives) and composite (phase pair + control + measurable contrast), so direct supervised generation on the MIT class is statistically fragile.
    Invoked throughout Sections 1–2 and tied to the Georgescu et al. 343/68 dataset.
  • domain assumption MIT mechanisms are multiphysics and often intertwined (correlation, lattice, charge/orbital order, CDW/excitonic, disorder), so rigid supervised mechanism classes on the small positive set are misleading.
    Sections 1–2 and Table 1; examples VO2, nickelates, 1T-TaS2, Ta2NiSe5.
  • domain assumption A pretrained crystal diffusion generator plus parameter-efficient adapters (e.g., LoRA) can be steered by scalable physics proxies without full retraining for each route.
    Section 3; cites crystal diffusion models and LoRA literature.
  • ad hoc to paper Staged computational gates (hull energy, phase-pair ΔE, electronic contrast, soft modes/pathways, control accessibility) are sufficient early filters before experiment.
    Section 4 verification ladder and quantitative triage thresholds (~50 meV/atom, ~25–50 meV pair separation).
  • ad hoc to paper Mechanism-prior adapters should be trained from proxies over broad oxide/chalcogenide pools rather than by partitioning the verified MIT set.
    Core design choice in Section 3; load-bearing for the claim that the framework works under label scarcity.
invented entities (2)
  • mechanism-prior adapters (correlation, dimerization, charge/orbital, excitonic/CDW, mixed)
    purpose: Controllable generative biases that steer a pretrained crystal generator toward testable MIT phase-transition hypotheses without hard mechanism labels.
    Named modular components of the proposed framework (Fig. 4, Table 2); engineering modules rather than new physical particles, but they are paper-specific postulated objects whose effectiveness is not demonstrated here.
  • MIT verification ladder (chemical validity → thermodynamics → phase competition → electronic contrast → pathway → experimental/device accessibility)
    purpose: Staged filter that converts generated structures into ranked transition hypotheses and defines enrichment metrics for generative models.
    Section 4 and Fig. 5; a workflow construct, not a physical entity, with no empirical enrichment study in this manuscript.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Physics adapted generative AI for metal insulator transition materials under label scarcity." pith.science (2026). https://pith.science/paper/BNIPALKU

@misc{pith2026260703578,
  author       = {Pith},
  title        = {Pith review of: Physics adapted generative AI for metal insulator transition materials under label scarcity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BNIPALKU}},
  note         = {Machine review of arXiv:2607.03578}
}
read the original abstract

Metal-insulator-transition (MIT) materials are promising candidates for switchable electronics, neuromorphic hardware, and reconfigurable photonics, yet experimentally verified examples remain limited and the underlying mechanisms are often complex. We argue that generative AI for MIT discovery should move beyond the search for stable crystal structures alone and instead prioritize mechanism-informed phase-transition hypotheses. A modular physics-adapted framework, combined with staged verification of phase competition, electronic contrast, transition-pathway plausibility, and control accessibility, can guide credible discovery under severe label scarcity.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references

  1. [1]

    Chemistry of Materials 33(14), 5591–5605 (2021)

    Georgescu, A.B., Ren, P., Toland, A.R., Zhang, S., Miller, K.D., Apley, D.W., Olivetti, E.A., Wagner, N., Rondinelli, J.M.: Database, features, and machine learning model to identify thermally driven metal –insulator transition com - pounds. Chemistry of Materials 33(14), 5591–5605 (2021)

    2021

  2. [2]

    Australian journal of chemistry 49(2), 205–217 (1996)

    Rossell, H.J., Goodman, P., Bulcock, S., March, R.H., Kennedy, S.J., White, T.J., Lincoln, F.J., Murray, K.S.: Structural and solid-state examination of ca4mn3o10 and sr4mn3o10. Australian journal of chemistry 49(2), 205–217 (1996)

    1996

  3. [3]

    Journal of Magnetism and Magnetic Materials 511, 166949 (2020)

    Periyasamy, M., Fjell v˚ag, Ø.S., Fjell v˚ag, H., Sj˚astad, A.O.: Coupling of magne- toresistance switching and glassy magnetic state at the metal–insulator transition in ruddlesden-popper manganite ca4mn3o10. Journal of Magnetism and Magnetic Materials 511, 166949 (2020)

    2020

  4. [4]

    Physics Letters A 33(6), 370 –371 (1970)

    Okinaka, H., Kosuge, K., Kachi, S., Nagasawa, K., Bando, Y., Takada, T.: Electrical properties of v8o15 single crystal. Physics Letters A 33(6), 370 –371 (1970)

    1970

  5. [5]

    Journal of Solid State Chemistry 6(2), 258–270 (1973)

    Kachi, S., Kosuge, K., Okinaka, H.: Metal -insulator transition in vno2n - 1. Journal of Solid State Chemistry 6(2), 258–270 (1973)

    1973

  6. [6]

    Structural Science 25(4), 781–787 (1969)

    Gardner, R., Vlasse, M., Wold, A.: Preparation, properties and crystal structure of barium vanadium sulfide, bavs3. Structural Science 25(4), 781–787 (1969)

    1969

  7. [7]

    Journal of Physics and Chemistry of Solids 40(8), 573–577 (1979)

    Massenet, O., Since, J., Mercier, J., Avignon, M., Buder, R., Nguyen, V., Kelber, J.: Magnetic and electrical properties of bavs3 and bavxti1- xs3. Journal of Physics and Chemistry of Solids 40(8), 573–577 (1979)

    1979

  8. [8]

    Journal of Solid State Chemistry 140(2), 182–193 (1998)

    Takeda, T., Nagata, M., Kobayashi, H., Kanno, R., Kawamoto, Y., Takano, M., Kamiyama, T., Izumi, F., Sleight, A.: High-pressure synthesis, crystal structure, and metal –semiconductor transitions in the tl2ru2o7 - δpyrochlore. Journal of Solid State Chemistry 140(2), 182–193 (1998)

    1998

Show all 38 references
  1. [9]

    journal of the physical society of japan 76(4), 043703– 043703 (2007)

    Yamamoto, A., Sharma, P.A., Okamoto, Y., Nakao, A., Katori, H.A., Niitaka, S., Hashizume, D., Takagi, H.: Metal –insulator transition in a pyrochlore -type ruthenium oxide, hg2ru2o7. journal of the physical society of japan 76(4), 043703– 043703 (2007)

    2007

  2. [10]

    Nature 144(3642), 327–328 (1939)

    Verwey, E.: Electronic conduction of magnetite (fe3o4) and its transition point at low temperatures. Nature 144(3642), 327–328 (1939)

    1939

  3. [11]

    Gossard, A., Di Salvo, F., Erich, L., Remeika, J., Yasuoka, H., Kosuge, K., Kachi, S.: Microscopic magnetic properties of vanadium oxides. ii. v 3 o 5, v 5 o 9, v 6 o 11, and v 6 o 13. Physical Review B 10(10), 4178 (1974) 10

    1974

  4. [12]

    Physical Review B 45(14), 8209 (1992)

    Torrance, J., Lacorre, P., Nazzal, A., Ansaldo, E., Niedermayer, C.: Systematic study of insulator-metal transitions in perovskites r nio 3 (r= pr, nd, sm, eu) due to closing of charge-transfer gap. Physical Review B 45(14), 8209 (1992)

    1992

  5. [13]

    Physical review B 46(8), 4414 (1992)

    Garc´ıa-Mu n˜oz, J., Rodr´ıguez -Carvajal, J., Lacorre, P., Torrance, J.: Neutron - diffraction study of r nio 3 (r= la, pr, nd, sm): Electronically induced structural changes across the metal -insulator transition. Physical review B 46(8), 4414 (1992)

    1992

  6. [14]

    Physical review letters 3(1), 34 (1959)

    Morin, F.: Oxides which show a metal -to-insulator transition at the neel temperature. Physical review letters 3(1), 34 (1959)

    1959

  7. [15]

    Physical Review B 14(4), 1429 (1976)

    Lakkis, S., Schlenker, C., Chakraverty, B., Buder, R., Marezio, M.: Metal - insulator transitions in ti 4 o 7 single crystals: Crystal characterization, specific heat, and electron paramagnetic resonance. Physical Review B 14(4), 1429 (1976)

    1976

  8. [16]

    -I., Yamauchi, T., Ueda, H., Gotou, H., Yagi, T., Ueda, Y.: Observation of metal –insulator transition in hollandite vanadate, k2v8o16

    Isobe, M., Koishi, S., Kouno, N., Yamaura, J. -I., Yamauchi, T., Ueda, H., Gotou, H., Yagi, T., Ueda, Y.: Observation of metal –insulator transition in hollandite vanadate, k2v8o16. Journal of the Physical Society of Japan 75(7), 073801 (2006)

    2006

  9. [17]

    Physical Review B 6(4), 1402 (1972)

    Fogle, W., Perlstein, J.H.: Semiconductor -to-metal transition in the blue potas - sium molybdenum bronze, k 0.30 mo o 3; example of a possible excitonic insulator. Physical Review B 6(4), 1402 (1972)

    1972

  10. [18]

    Journal of the Ceramic Society of Japan 122(1430), 846–851 (2014)

    Yamada, I.: High -pressure synthesis, electronic states, and structure –property relationships of perovskite oxides, acu3fe4o12 (a: divalent alkaline earth or triva- lent rare-earth ion). Journal of the Ceramic Society of Japan 122(1430), 846–851 (2014)

    2014

  11. [19]

    Chemistry–A European Journal 20(31), 9510–9513 (2014)

    Zhang, S., Saito, T., Mizumaki, M., Shimakawa, Y.: Temperature-induced inter- site charge transfer involving cr ions in a-site-ordered perovskites acu3cr4o12 (a= la and y). Chemistry–A European Journal 20(31), 9510–9513 (2014)

    2014

  12. [20]

    Physical Review B 63(19), 195104 (2001)

    Mandrus, D., Thompson, J., Gaal, R., Forro, L., Bryan, J., Chakoumakos, B., Woods, L., Sales, B., Fishman, R., Keppens, V.: Continuous metal -insulator transition in the pyrochlore cd 2 os 2 o 7. Physical Review B 63(19), 195104 (2001)

    2001

  13. [21]

    Physica B: Condensed Matter 194, 1077–1078 (1994)

    Nagata, S., Hagino, T., Seki, Y., Bitoh, T.: Metal-insulator transition in thiospinel cuir2s4. Physica B: Condensed Matter 194, 1077–1078 (1994)

    1994

  14. [22]

    Nature communications 1(1), 106 (2010)

    Wang, Z., Okude, M., Saito, M., Tsukimoto, S., Ohtomo, A., Tsukada, M., Kawasaki, M., Ikuhara, Y.: Dimensionality -driven insulator–metal transition in a-site excess non -stoichiometric perovskites. Nature communications 1(1), 106 (2010)

    2010

  15. [23]

    Journal of the Physical Society of Japan 71(8), 1848–1851 (2002)

    Isobe, M., Ueda, Y.: Observation of phase transition from metal to spin-singlet 11 insulator in mgti2o4 with s= 1/2 pyrochlore lattice. Journal of the Physical Society of Japan 71(8), 1848–1851 (2002)

    2002

  16. [24]

    Journal of the American Chemical Society 135(16), 6100–6106 (2013)

    Etani, H., Yamada, I., Ohgushi, K., Hayashi, N., Kusano, Y., Mizumaki, M., Kim, J., Tsuji, N., Takahashi, R., Nishiyama, N., et al.: Suppression of intersite charge transfer in charge-disproportionated perovskite ycu3fe4o12. Journal of the American Chemical Society 135(16), 61...

    2013

  17. [25]

    Nature 458(7234), 60–63 (2009)

    Long, Y., Hayashi, N., Saito, T., Azuma, M., Muranaka, S., Shimakawa, Y.: Temperature -induced a –b intersite charge transfer in an a -site-ordered lacu3fe4o12 perovskite. Nature 458(7234), 60–63 (2009)

    2009

  18. [26]

    Physical Review B 62(2), 844 (2000)

    Woodward, P., Cox, D., Moshopoulou, E., Sleight, A., Morimoto, S.: Structural studies of charge disproportionation and magnetic order in cafeo 3. Physical Review B 62(2), 844 (2000)

    2000

  19. [27]

    Solid state sciences 2(7), 673–687 (2000)

    Takeda, T., Kanno, R., Kawamoto, Y., Takano, M., Kawasaki, S., Kamiyama, T., Izumi, F.: Metal–semiconductor transition, charge disproportionation, and low- temperature structure of ca1–xsrxfeo3 synthesized under high-oxygen pressure. Solid state sciences 2(7), 673–687 (2000)

    2000

  20. [28]

    Sudo, S., Nishioka, T., Miyako, Y., Miyadai, T.: Specific heat study of metal - insulator transition in antiferromagnetic phase of nis1. 48se0. 52. Journal of the Physical Society of Japan 55(6), 1806–1809 (1986)

    1986

  21. [29]

    Physical Review B 60(12), 8422 (1999)

    Alexander, C., Cao, G., Dobrosavljevic, V., McCall, S., Crow, J., Lochner, E., Guertin, R.: Destruction of the mott insulating ground state of ca 2 ruo 4 by a structural transition. Physical Review B 60(12), 8422 (1999)

    1999

  22. [30]

    Journal of the American Chemical Society 121(20), 4754–4762 (1999)

    Alonso, J.A., Martinez-Lope, M.J., Casais, M.T., Aranda, M.A., Fernandez-Diaz, M.T.: Metal - insulator transitions, structural and microstructural evolution of rnio3 (r= sm, eu, gd, dy, ho, y) perovskites: evidence for room-temperature charge disproportionation in monoclinic h...

    1999

  23. [31]

    Physical Review B—Condensed Matter and Materials Physics 80(16), 161104 (2009)

    Shi, Y., Guo, Y., Yu, S., Arai, M., Belik, A., Sato, A., Yamaura, K., Takayama - Muromachi, E., Tian, H., Yang, H., et al.: Continuous metal-insulator transition of the antiferromagnetic perovskite naoso 3. Physical Review B—Condensed Matter and Materials Physics 80(16), 161104 (2009)

    2009

  24. [32]

    Journal of the Physical Society of Japan 16(2), 187–192 (1961)

    Murakami, M.: Anisotropy of electrical conduction in iron sulfide single crystal. Journal of the Physical Society of Japan 16(2), 187–192 (1961)

    1961

  25. [33]

    Nature Physics 16(6), 669–675 (2020) 12

    Bansal, D., Niedziela, J.L., Calder, S., Lanigan -Atkins, T., Rawl, R., Said, A.H., Abernathy, D.L., Kolesnikov, A.I., Zhou, H., Delaire, O.: Magnetically driven phonon instability enables the metal–insulator transition in h-fes. Nature Physics 16(6), 669–675 (2020) 12

    2020

  26. [34]

    Journal of Solid State Chemistry 136(1), 67–73 (1998)

    Onoda, M.: Phase transitions of ti3o5. Journal of Solid State Chemistry 136(1), 67–73 (1998)

    1998

  27. [35]

    Physical Review B 8(4), 1364 (1973)

    Shin, S., Chandrashekhar, G., Loehman, R., Honig, J.: Thermoelectric effects in pure and v-doped ti 2 o 3 single crystals. Physical Review B 8(4), 1364 (1973)

    1973

  28. [36]

    Physical Review Materials 7(2), 024415 (2023)

    Dewey, J.E., Ghosh, S., Chaturvedi, V., Postiglione, W.M., Figari, L., Jacobson, A., Korostynski, C., Charlton, T.R., Mkhoyan, K.A., Leighton, C.: Anomalous strain relaxation and its impact on the valence-driven spin-state/metal-insulator transition in epitaxial (pr 1 -y y y) ...

    2023

  29. [37]

    Physical review letters 82(19), 3871 (1999)

    Alonso, J., Garc´ıa-Mu n˜oz, J., Fern ´andez -D´ıaz, M., Aranda, M., Mart´ınez-Lope, M., Casais, M.: Charge disproportionation in r nio 3 perovskites: Simultaneous metal-insulator and structural transition in ynio 3. Physical review letters 82(19), 3871 (1999)

    1999

  30. [38]

    Physical Review B (2008) 13

    Redfern, S.: β phase and γ-β metal-insulator transition in multiferroic bifeo3. Physical Review B (2008) 13

    2008

Pith tools

Reviewed July 12, 2026 · model on record in the stance chip above.