Pith. sign in

REVIEW 3 major objections 6 minor 89 references

Elemental Germanium Phase-Change Memory

T0 review · 3 major / 6 minor · reviewed 2026-07-30 · grok-4.5

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

Pith's one-line read Elemental germanium is a working CMOS-native phase-change memory that crystallizes in 240 ps, keeps data above 110 °C for a projected decade, and drifts less than GST.

desk verdict Working monatomic Ge PCM with solid STEM/4D-STEM mechanism proof and real CMOS-compatibility upside; the >110 °C/10-year number is the softest claim and should be read as a lower-bound projection, not a measured figure. read the letter →

arxiv 2607.23709 v1 pith:BCZTAXXF submitted 2026-07-26 cond-mat.mtrl-sci physics.app-ph

Till Zellweger , Marko Mladenović , Kevin Portner , Christoph Weilenmann , Michael Stiefel , Hanglin He , Klemens Bauer , Luiz Felipe Aguinsky
show 2 more authors
Mathieu Luisier Alexandros Emboras
This is my paper · ORCID
classification cond-mat.mtrl-sciphysics.app-ph
keywords phase-changememoryelementalgermaniumnon-volatilesingle-elementPCMCMOScompatibilityresistancedriftmelt-quenchcrystallization
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

Standard phase-change memory relies on chalcogenide alloys such as GST. Those alloys segregate their atoms under cycling, drift in resistance, and bring Sb and Te that contaminate ordinary CMOS fabs. This paper shows that pure germanium, already used in mainstream semiconductor lines, can itself serve as the active phase-change layer in a simple vertical tungsten–germanium–tungsten cell. The cells crystallize in an instrument-limited 240 picoseconds, remain non-volatile with thermal stability projecting past 110 °C for ten years, and show a high-resistance drift coefficient about 60 percent lower than typical GST—while containing neither antimony nor tellurium. If the result holds under industrial scaling, phase-change memory can be built inside ordinary semiconductor facilities instead of dedicated chalcogenide fabs.

What carries the argument

A vertical self-heating W/Ge/W cell: a thin sputtered Ge film between tungsten electrodes is Joule-heated by unipolar pulses so that melt-quench amorphization and recrystallization switch resistance by orders of magnitude; STEM, 4D-STEM, quantum-transport heating maps, and melt-quench molecular dynamics confirm the phase-change mechanism.

What would settle it

Measure the crystallization activation energy on the same W/Ge/W devices or films used for retention, then bake programmed high-resistance cells long enough to observe actual failure times at several temperatures; if the device Ea is substantially below ~3 eV, the projected >110 °C decade retention fails.

Watch

Extended reading notes

Core claim

The authors establish that elemental germanium functions as a non-volatile phase-change material: short electrical pulses melt-quench a nanometric Ge volume into a stable amorphous high-resistance state and recrystallize it into a low-resistance state, with direct STEM and diffraction evidence that the two resistance states are crystalline and amorphous Ge. In their vertical self-heating W/Ge/W cells this yields sub-nanosecond crystallization (240 ps), projected 10-year retention above 110 °C, and a resistance-drift coefficient ν≈0.045, all without Sb or Te.

Load-bearing premise

The claim of more than 110 °C ten-year retention rests on importing a literature crystallization activation energy of about 3 eV for germanium and back-solving an Arrhenius model from a single 150 °C bake in which no failure was seen.

Editorial extensions

If this is right

  • Phase-change memory arrays could be fabricated inside standard CMOS lines without Sb/Te contamination controls or dedicated fabs.
  • Single-element Ge cells remove stoichiometry constraints and elemental segregation that limit cycling of multi-component alloys.
  • The combination of ≤240 ps write speed and high thermal stability opens high-temperature embedded and automotive non-volatile memory uses where GST is marginal.
  • A lower resistance-drift coefficient without doping or confinement improves analog multi-level and in-memory computing accuracy.
  • Adding a dedicated nanoheater and tighter thermal confinement is expected to raise endurance toward the 10^6–10^9 cycle range of optimized GST cells.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If explosive crystallization plus temperature-dependent Ea truly decouples high-T kinetics from low-T stability, other covalently bonded elemental semiconductors may be worth screening as PCM candidates.
  • CMOS-native Ge PCM could be co-integrated with Ge photodetectors or strained-Ge channels on the same wafer, collapsing optical, logic, and non-volatile memory process flows.
  • Because the active volume is monatomic, aggressive lateral scaling below stoichiometry-limited alloy cells may be limited mainly by thermal confinement rather than composition control.
  • Independent multi-temperature failure statistics on sub-50 nm Ge cells would be the decisive industrial gate for the retention and endurance claims.
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. The manuscript introduces elemental germanium as a single-element, CMOS-compatible phase-change memory material, demonstrated in a vertical W/Ge/W cell (30/20/30 nm). The authors show threshold switching with snapback at ~2.2 V, a pristine HRS/LRS contrast exceeding four orders of magnitude, reversible unipolar switching over 1,500 cycles with a well-defined amorphization window, and instrument-limited crystallization in 240 ps. Phase change as the switching mechanism is supported by convergent structural evidence: cross-sectional STEM lattice fringes in the LRS versus a disordered HRS, 4D-STEM sharp spots versus diffuse rings, EDX showing no W–Ge interdiffusion (ruling out metallization-based switching), thin-film R–T crystallization near 490 °C corroborated by Raman and XRD, plus quantum-transport self-heating simulations showing Ge melting under RESET bias and melt-quench MD simulations showing the amorphous phase can be stabilized at device-relevant quench rates. The headline quantitative claims are 240 ps SET speed, a resistance drift coefficient ν≈0.045 (~60% below GST), and a projected 10-year retention temperature above 110 °C.

Significance. If the results hold, this is a genuinely significant contribution: the first demonstration of a CMOS-native elemental phase-change material, eliminating Sb/Te contamination concerns and elemental redistribution under cycling, and apparently breaking the speed–stability tradeoff that governs the Ge–Sb–Te system. The paper ships several notable strengths: (i) the phase-change mechanism is established by multiple independent structural probes rather than inferred from electrical data alone; (ii) the 240 ps figure is honestly reported as instrument-limited, with the true speed likely lower; (iii) the benchmarking in Fig. 4b/d/e is unusually careful about restricting comparisons to matched device sizes and measurement techniques; (iv) the MD melt-quench result is validated against two independent interatomic potentials (SW and GAP); and (v) limitations (endurance of ~5×10³ cycles versus 10⁶–10⁹ for optimized GST) are disclosed rather than buried. The principal risk attaches to one of the three headline numbers: the >110 °C/10-year retention projection.

major comments (3)
  1. The >110 °C/10-year retention figure rests on a single bake temperature (150 °C, the setup maximum), a single HRS device, no observed failure within 10^5 s, and a literature activation energy Ea≈3 eV imported from 1970s resistance measurements on uncapped evaporated/sputtered Ge thin films (refs 42, 43). τ0 is then back-solved by setting t_fail = 10^5 s in Eq. (1). One temperature cannot constrain an Arrhenius temperature dependence, so the ~3-decade extrapolation in time is entirely model assumption. The τ0 construction is conservative only conditional on Ea being correct: with the identical back-solving procedure, Ea = 2 eV instead of 3 eV lowers the 10-year temperature to roughly 95 °C, erasing the claimed margin over GST (~87 °C) and the 85 °C benchmark. The paper itself supplies two reasons to doubt that the thin-film nucleation-limited Ea transfers to this device: (a) the authors'
  2. Fig. 3e shows a residual crystalline spot persisting in the HRS diffraction pattern, and the MD protocol of §4.8/Fig. 3g explicitly models recrystallization proceeding by growth from a crystalline seed. A seeded, growth-dominated device crystallization process generically carries a lower effective activation energy than the nucleation-limited kinetics of the uncapped films from which Ea≈3 eV was taken — i.e., the error direction is the one that degrades retention. Further, the text (§2.4) notes Ea falls to ~1.4 eV above ~360 °C, demonstrating that Ge's crystallization energetics are regime- and geometry-dependent and must be measured, not imported. This is fixable within the manuscript's scope: either (i) measure the device-relevant Ea directly (accelerated bakes at two or more temperatures, e.g., 150/175/200 °C, on multiple HRS devices, or a Kissinger/Arrhenius analysis on the capped W/
  3. Given that the abstract and conclusion both carry '>110 °C for 10 years' as a headline result, the retention claim should either be supported by a device-measured Ea with a stated uncertainty and sensitivity analysis, or the claim should be softened to what the data directly show: HRS stability at 150 °C over 10^5 s (itself a strong result that already exceeds GST, which fully crystallizes under these conditions) with the Arrhenius projection presented as conditional on the literature Ea. At minimum, report the implied 10-year temperature for a plausible Ea range (e.g., 2–3 eV) and state n (number of devices baked) in §4.3 — currently only one HRS and one LRS device are described.
minor comments (6)
  1. [§2.2] Fig. 2d: the cycled on/off ratio is ~100, versus the >4×10⁴ pristine contrast of Fig. 2b. A brief discussion of the origin of this contrast loss after the first SET (partial crystallization, filamentary volume, series resistance of the W/Ge contacts) would help readers judge multi-level prospects.
  2. [§2.4] Fig. 4c: drift is measured only at room temperature over 10³ s. Since GST's ν is known to increase strongly with temperature (Fig. 4d, refs 59, 60), at least one elevated-temperature drift measurement would strengthen the '~60% lower than GST' claim; otherwise note that the comparison is strictly valid at RT only.
  3. [§2.4, Fig. 4b] Fig. 4b mixes bottom-electrode diameter (mushroom cells) with the lateral Ge width (this work) as 'device size'. The justification given (both define the thermally active volume) is reasonable, but the equivalence is approximate; a caveat sentence would be appropriate, particularly since the record 242 ps Sb point is at 60 nm.
  4. [§2.3, Fig. 3e] The residual crystalline fraction in the HRS (Fig. 3e) should be quantified if possible (e.g., azimuthally integrated spot-to-ring intensity or crystallinity mapping from the 4D-STEM dataset), since it bears on both the retention mechanism and the MD seed model.
  5. [various] Typographical: 'A WG' appears with a space throughout (§2.4, Methods §4.2). Fig. 2c reports n=5 devices per configuration; state this in the caption. The endurance data in Supplementary Fig. S4 (failure at up to 5,000 cycles, 10 devices) are important enough that the failure mode (stuck-LRS vs stuck-HRS) deserves one sentence in the main text.
  6. [§4.7, Fig. 3f] Methods §4.7: the QT simulation uses a crystalline 20 nm Ge region at 3 V to show melting during RESET. Since the HRS that must be RESET is amorphous with much higher resistance, clarify why crystalline-Ge self-heating is the relevant configuration (presumably the post-threshold-switching state); a sentence on the assumed conduction state during the pulse would close this gap.

Circularity Check

1 steps flagged · score 1.0 of 10

Primarily self-contained experimental device work; only mild model-extrapolation circularity in the 10-year retention temperature projection.

  1. fitted input called prediction [Methods §4.3 (Device Retention Measurements), Eq. 1; Fig. 2f; Abstract]
    "Since no crystallization was observed over the full 10^5 s measurement window at 150 °C, the pre-exponential factor τ0 was obtained by setting t_fail = 10^5 s and T = 150 °C in Eq. 1. The 10-year retention temperature was then obtained as a lower bound by solving Eq. 1 for T with t_fail = 10 years. ... Extrapolation using the Ge crystallization activation energy of 3 eV projects stable retention at >110 °C for 10 years."

    τ0 is fixed by construction from the single non-failure datum (t_fail:=10^5 s at 150 °C) together with an externally imported Ea≈3 eV. With those inputs locked, solving Eq. 1 for T at t_fail=10 years yields “>110 °C” as a deterministic rewrite of the assumptions, not an independently measured or cross-validated quantity. The projection is therefore a fitted-input extrapolation labeled as a retention prediction. Mild and confined to one metric; Ea was not fitted to the authors’ own device data.

full rationale

The paper’s central claims—that elemental Ge in a W/Ge/W cell is a functional non-volatile PCM with instrument-limited 240 ps crystallization, directly measured HRS drift ν≈0.045, reversible unipolar switching, and STEM/4D-STEM-confirmed amorphous↔crystalline contrast—are grounded in independent electrical measurements, structural probes, and external thin-film literature, not in a self-referential fit or author-only uniqueness theorem. Quantum-transport and MD simulations use established external potentials (Stillinger–Weber, GAP) and are used as supporting feasibility checks, not as fitted predictors of the measured device metrics. The sole mild circularity is the Arrhenius 10-year retention projection: τ0 is back-solved by construction from a single non-failure observation (10^5 s at 150 °C) plus an imported literature Ea≈3 eV, so the numerical “>110 °C for 10 years” is an arithmetic consequence of those inputs under Eq. 1 rather than an independently constrained prediction. That step affects only one headline metric and does not underwrite the phase-change mechanism, speed, or drift results. No self-citation load-bearing chain, uniqueness import, or ansatz smuggling is present. Score 1 reflects one minor fitted-input extrapolation that is not load-bearing for the paper’s core demonstration.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central experimental claim (Ge switches by phase change) rests on standard PCM device physics plus materials assumptions about bonding, melting, and crystallization kinetics. The headline retention number additionally imports a literature activation energy and an Arrhenius failure model. Simulations add potential- and Hamiltonian-level assumptions that support feasibility but are not required once TEM evidence is accepted. No new physical entities are postulated.

free parameters (3)
  • Crystallization activation energy Ea used for 10-year retention extrapolation = ≈3 eV (literature; not fitted on these devices)
    Methods §4.3 sets Ea≈3 eV from literature thin-film Ge studies, fixes t_fail=10^5 s at 150 °C to obtain τ0, then solves for T at t_fail=10 years. The numerical >110 °C claim moves directly with this chosen Ea.
  • Resistance drift exponent ν in R(t)=R0(t/t0)^ν = ν≈0.045 (HRS); lower states down to ~3e-3
    Standard empirical drift law; ν is fitted per programmed state in Fig. 4c and compared to literature GST values. The ‘~60% lower than GST’ headline uses the HRS fit ν≈0.045 versus a typical GST ν≈0.11.
  • RESET/SET pulse amplitudes and widths for cycling and amorphization window = geometry-specific pulse recipes
    Operating points (e.g. 2.5 V/25 ns RESET, 0.8 V/200 ns SET; amorphization map in Fig. 2e) are empirically chosen for the 200×200 nm², 20 nm geometry and define reported endurance and energy.
assumptions (6)
  • domain assumption Reversible resistance contrast in chalcogenide and related materials can be produced by melt-quench amorphization and recrystallization under Joule heating (standard PCM operating model).
    Used throughout §2 to interpret I–V snapback, SET/RESET polarity, and the amorphization window as phase change rather than electrochemical metallization.
  • domain assumption Arrhenius form t_fail(T)=τ0 exp(Ea/kT) with constant Ea adequately projects long-term retention from a single-temperature bake when no failure is observed.
    Methods §4.3; underpins the >110 °C / 10-year claim. Paper itself notes Ge’s Ea is temperature-dependent (~1.4 eV at high T vs ~3 eV at low T), which weakens constant-Ea extrapolation.
  • ad hoc to paper Literature crystallization Ea≈3 eV for elemental Ge thin films applies to the confined W/Ge/W cell with SiNx encapsulation.
    Invoked without device-level Kissinger/Arrhenius extraction on the fabricated stack; interfaces and residual crystalline seeds (seen in HRS diffraction) could alter kinetics.
  • domain assumption Stillinger–Weber and GAP interatomic potentials, and the sp3d5s* TB + VFF NEGF electro-thermal model, are sufficiently accurate to decide whether Ge reaches Tm≈1210 K and can be melt-quenched amorphous on 0.1–1 ns scales.
    §2.3 and Methods §4.7–4.8; used as supporting feasibility evidence alongside experiment.
  • domain assumption W does not form interfering intermetallics with Ge under the thermal budget of device operation, so electrodes remain chemically inert boundaries.
    Cited from Gaudet et al.; motivates electrode choice and is consistent with EDX maps showing separated W and Ge.
  • standard math Standard mathematical/statistical treatment of means, standard deviations, and power-law drift fits.
    Fig. 2c error bars; Fig. 4c drift fits.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Elemental Germanium Phase-Change Memory." pith.science (2026). https://pith.science/paper/BCZTAXXF

@misc{pith2026260723709,
  author       = {Pith},
  title        = {Pith review of: Elemental Germanium Phase-Change Memory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BCZTAXXF}},
  note         = {Machine review of arXiv:2607.23709}
}
abstract

Phase-change memory (PCM) is a mature technology for fast, scalable, non-volatile data storage, with applications spanning embedded memory, as well as in-memory and neuromorphic computing. PCM predominantly relies on chalcogenide alloys, with $\mathrm{Ge_2Sb_2Te_5}$ (GST) as the industry standard. Yet in these alloys, the individual Ge, Sb, and Te atoms redistribute upon cycling, causing stochastic operation and ultimately device failure. To address this issue, elemental antimony was proposed as a PCM material, but it exhibits a metastable amorphous state that prevents reliable data retention. Moreover, tellurium and antimony can contaminate complementary metal-oxide-semiconductor (CMOS) production lines or act as unintended dopants, restricting manufacturing of PCM to dedicated fabs. Here we introduce elemental germanium (Ge) as a CMOS-native phase-change material that overcomes these fundamental limitations. In a vertical PCM cell architecture, Ge enables sub-nanosecond crystallization (240 ps, 40 times faster than GST), non-volatile data storage with excellent thermal stability ($>$110 {\deg}C for 10 years vs. $\sim$87 {\deg}C for GST), and a resistance drift coefficient approximately 60% lower than in GST. These results establish pure Ge, a standard semiconductor, as an alternative to chalcogenide phase-change materials, achieving superior performance in key metrics and enabling phase-change memory to be fabricated in standard semiconductor facilities.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

89 extracted references

  1. [1]

    Memory technology enabling future computing systems.APL Machine Learning3, 026101 (2025)

    Fantini, P. Memory technology enabling future computing systems.APL Machine Learning3, 026101 (2025)

    2025

  2. [2]

    Zhang, W., Mazzarello, R., Wuttig, M. & Ma, E. Designing crystallization in phase-change materials for universal memory and neuro-inspired computing.Nature Reviews Materials4, 150–168 (2019)

    2019

  3. [3]

    S., Le Gallo, M

    Syed, G. S., Le Gallo, M. & Sebastian, A. Phase-change memory for in-memory computing. Chemical reviews125, 5163–5194 (2025)

    2025

  4. [4]

    Nature Machine Intelligence1939–1953 (2025)

    Portner, K.et al.Actor–critic networks with analogue memristors mimicking reward-based learning. Nature Machine Intelligence1939–1953 (2025)

    1953

  5. [5]

    & Yamada, N

    Wuttig, M. & Yamada, N. Phase-change materials for rewriteable data storage.Nature Materials 6, 824–832 (2007)

    2007

  6. [6]

    L., Gonze, X., Bichara, C

    Wuttig, M., Deringer, V. L., Gonze, X., Bichara, C. & Raty, J.-Y. Incipient metals: functional materials with a unique bonding mechanism.Advanced Materials30, 1803777 (2018)

    2018

  7. [7]

    Shportko, K.et al.Resonant bonding in crystalline phase-change materials.Nature Materials7, 653–658 (2008)

    2008

  8. [8]

    W.et al.Recent progress in phase-change memory technology.IEEE Journal on Emerging and Selected Topics in Circuits and Systems6, 146–162 (2016)

    Burr, G. W.et al.Recent progress in phase-change memory technology.IEEE Journal on Emerging and Selected Topics in Circuits and Systems6, 146–162 (2016)

    2016

Show all 89 references
  1. [9]

    Phase change memory applications: the history, the present and the future.Journal of Physics D: Applied Physics53, 283002 (2020)

    Fantini, P. Phase change memory applications: the history, the present and the future.Journal of Physics D: Applied Physics53, 283002 (2020)

    2020

  2. [10]

    & MacManus-Driscoll, J

    Hellenbrand, M., Teck, I. & MacManus-Driscoll, J. L. Progress of emerging non-volatile memory technologies in industry.MRS Communications14, 1099–1112 (2024)

    2024

  3. [11]

    Polignano, M.et al.Tellurium contamination in silicon.ECS Journal of Solid State Science and Technology2, N28–N34 (2013)

    2013

  4. [12]

    & Mimkes, J

    St¨ umpel, H., Vorderw¨ ulbecke, M. & Mimkes, J. Diffusion of selenium and tellurium in silicon. Applied Physics A46, 159–163 (1988)

    1988

  5. [13]

    & Current, M

    Larson, L. & Current, M. Metallic impurities and dopant cross-contamination effects in ion implanted surfaces.MRS Online Proceedings Library45, 381–388 (1985)

    1985

  6. [14]

    & Ryssel, H

    Steen, C., Pichler, P. & Ryssel, H. Segregation of antimony to Si/SiO2 interfaces.Materials Science and Engineering: B154, 264–267 (2008)

    2008

  7. [15]

    Materials Advances4, 747–768 (2023)

    Bala, N.et al.Recent advances in doped Ge 2Sb2Te5 thin film based phase change memories. Materials Advances4, 747–768 (2023)

    2023

  8. [16]

    E.et al.Interfacial phase-change memory.Nature Nanotechnology6, 501–505 (2011)

    Simpson, R. E.et al.Interfacial phase-change memory.Nature Nanotechnology6, 501–505 (2011)

    2011

  9. [17]

    Rao, F.et al.Reducing the stochasticity of crystal nucleation to enable subnanosecond memory writing.Science358, 1423–1427 (2017)

    2017

  10. [18]

    I.et al.Unveiling the effect of superlattice interfaces and intermixing on phase change memory performance.Nano Letters22, 6285–6291 (2022)

    Khan, A. I.et al.Unveiling the effect of superlattice interfaces and intermixing on phase change memory performance.Nano Letters22, 6285–6291 (2022)

    2022

  11. [19]

    Salinga, M.et al.Monatomic phase change memory.Nature Materials17, 681–685 (2018)

    2018

  12. [20]

    Shen, J.et al.Elemental electrical switch enabling phase segregation–free operation.Science374, 1390–1394 (2021). 13

    2021

  13. [21]

    Shen, J.et al.Toward the speed limit of phase-change memory.Advanced Materials35, 2208065 (2023)

    2023

  14. [22]

    Chen, B.et al.Suppressing structural relaxation in nanoscale antimony to enable ultralow-drift phase-change memory applications.Advanced Science10, 2301043 (2023)

    2023

  15. [23]

    Leng, W.et al.Crystallization kinetics of monatomic antimony.Applied Physics Letters119, 171908 (2021)

    2021

  16. [24]

    Persch, C.et al.The potential of chemical bonding to design crystallization and vitrification kinetics.Nature Communications12, 4978 (2021)

    2021

  17. [25]

    & Annunziata, R

    Redaelli, A., Petroni, E. & Annunziata, R. Material and process engineering challenges in Ge-rich GST for embedded PCM.Materials Science in Semiconductor Processing137, 106184 (2022)

    2022

  18. [26]

    & Trumbore, F

    Hassion, F., Thurmond, D. & Trumbore, F. On the melting point of germanium.The Journal of Physical Chemistry59, 1076–1078 (1955)

    1955

  19. [27]

    & Takao, M

    Yamada, N., Ohno, E., Nishiuchi, K., Akahira, N. & Takao, M. Rapid-phase transitions of GeTe– Sb2Te3 pseudobinary amorphous thin films for an optical disk memory.Journal of Applied Physics 69, 2849–2856 (1991)

    1991

  20. [28]

    Marris-Morini, D.et al.Germanium-based integrated photonics from near-to mid-infrared applications.Nanophotonics7, 1781–1793 (2018)

    2018

  21. [29]

    & Yin, H

    Zhang, Q., Zhang, Y., Luo, Y. & Yin, H. New structure transistors for advanced technology node CMOS ICs.National Science Review11(2024)

    2024

  22. [30]

    & Lacaita, A

    Ielmini, D., Lavizzari, S., Sharma, D. & Lacaita, A. L. Physical interpretation, modeling and impact on phase change memory (PCM) reliability of resistance drift due to chalcogenide structural relaxation.2007 IEEE International Electron Devices Meeting939–942 (2007)

    2007

  23. [31]

    & Lavoie, C

    Gaudet, S., Detavernier, C., Kellock, A., Desjardins, P. & Lavoie, C. Thin film reaction of transition metals with germanium.Journal of Vacuum Science & Technology A24, 474–485 (2006)

    2006

  24. [32]

    T.et al.Phase separation in Ge-rich GeSbTe at different length scales: Melt-quenched bulk versus annealed thin films.Nanomaterials12, 1717 (2022)

    Yimam, D. T.et al.Phase separation in Ge-rich GeSbTe at different length scales: Melt-quenched bulk versus annealed thin films.Nanomaterials12, 1717 (2022)

    2022

  25. [33]

    & Jonscher, A

    Walley, P. & Jonscher, A. Electrical conduction in amorphous germanium.Thin solid films1, 367–377 (1968)

    1968

  26. [34]

    & Wuttig, M

    Friedrich, I., Weidenhof, V., Njoroge, W., Franz, P. & Wuttig, M. Structural transformations of Ge2Sb2Te5 films studied by electrical resistance measurements.Journal of Applied Physics87, 4130–4134 (2000)

    2000

  27. [35]

    & Tanaka, K

    Kato, T. & Tanaka, K. Electronic properties of amorphous and crystalline Ge 2Sb2Te5 films. Japanese Journal of Applied Physics44, 7340 (2005)

    2005

  28. [36]

    Oh, J., Banerjee, S. K. & Campbell, J. C. Metal-germanium-metal photodetectors on heteroepitax- ial Ge-on-Si with amorphous Ge Schottky barrier enhancement layers.IEEE Photonics Technology Letters16, 581–583 (2004)

    2004

  29. [37]

    Hull, E. L. & Pehl, R. H. Amorphous germanium contacts on germanium detectors.Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment538, 651–656 (2005)

    2005

  30. [38]

    & Zhang, Y

    Ielmini, D. & Zhang, Y. Analytical model for subthreshold conduction and threshold switching in chalcogenide-based memory devices.Journal of Applied Physics102(2007)

    2007

  31. [39]

    Zheng, J.et al.Extended switching endurance of phase change memory through nano-confined cell structure.Nature Communications16, 5788 (2025). 14

    2025

  32. [40]

    Stern, K.et al.Uncovering phase change memory energy limits by sub-nanosecond probing of power dissipation dynamics.Advanced Electronic Materials7, 2100217 (2021)

    2021

  33. [41]

    & Lee, Y.-K

    Choi, Y., Jung, M. & Lee, Y.-K. Effect of heating rate on the activation energy for crystallization of amorphous Ge 2Sb2Te5 thin film.Electrochemical and Solid-State Letters12, F17–F19 (2009)

    2009

  34. [42]

    & Lim, P.-K

    Chik, K. & Lim, P.-K. Annealing and crystallization of amorphous germanium thin films.Thin Solid Films35, 45–56 (1976)

    1976

  35. [43]

    & Gheorghiu, A

    Germain, P., Zellama, K., Squelard, S., Bourgoin, J. & Gheorghiu, A. Crystallization in amorphous germanium.Journal of Applied Physics50, 6986–6994 (1979)

    1979

  36. [44]

    & Waser, R.Resistive switching: from fundamentals of nanoionic redox processes to memristive device applications(John Wiley & Sons, 2015)

    Ielmini, D. & Waser, R.Resistive switching: from fundamentals of nanoionic redox processes to memristive device applications(John Wiley & Sons, 2015)

    2015

  37. [45]

    & Xide, X

    Jian, Z., Kaiming, Z. & Xide, X. Modification of Stillinger-Weber potentials for Si and Ge.Physical Review B41, 12915 (1990)

    1990

  38. [46]

    O., Zanotto, E

    Tipeev, A. O., Zanotto, E. D. & Rino, J. P. Crystal nucleation kinetics in supercooled germanium: MD simulations versus experimental data.The Journal of Physical Chemistry B124, 7979–7988 (2020)

    2020

  39. [47]

    & Sand, A

    Jin, R., Hamedani, A. & Sand, A. E. Data-efficient machine-learning interatomic potential for studying radiation effects in germanium.Machine Learning: Science and Technology7, 015012 (2026)

    2026

  40. [48]

    Wang, W.et al.Fast phase transitions induced by picosecond electrical pulses on phase change memory cells.Applied Physics Letters93, 043121 (2008)

    2008

  41. [49]

    Wang, W.et al.Enabling universal memory by overcoming the contradictory speed and stability nature of phase-change materials.Scientific Reports2, 360 (2012)

    2012

  42. [50]

    Cheng, H.-Y.et al.A high performance phase change memory with fast switching speed and high temperature retention by engineering the Ge xSbyTez phase change material 3–4 (2011)

    2011

  43. [51]

    Bruns, G.et al.Nanosecond switching in GeTe phase change memory cells.Applied physics letters 95, 043108 (2009)

    2009

  44. [52]

    Perniola, L.et al.Electrical behavior of phase-change memory cells based on GeTe.IEEE Electron Device Letters31, 488–490 (2010)

    2010

  45. [53]

    Meng, Y.et al.GeTe ultrathin film based phase-change memory with extreme thermal stability, fast set speed, and low reset power energy.AIP Advances13, 035205 (2023)

    2023

  46. [54]

    Fantini, A.et al.N-doped GeTe as performance booster for embedded phase-change memories 29–1 (2010)

    2010

  47. [55]

    B.et al.Carbon-doped GeTe: a promising material for phase-change memories

    Beneventi, G. B.et al.Carbon-doped GeTe: a promising material for phase-change memories. Solid-State Electronics65, 197–204 (2011)

    2011

  48. [56]

    & Song, Z

    Xue, Y., Yan, S., Lv, S., Song, S. & Song, Z. Ta-doped Sb2Te allows ultrafast phase-change memory with excellent high-temperature operation characteristics.Nano-micro letters13, 33 (2021)

    2021

  49. [57]

    Zeng, Y.et al.A fast and high endurance phase change memory based on In-doped Sb 2Te3.ACS Applied Nano Materials7, 13983–13990 (2024)

    2024

  50. [58]

    Science366, 210–215 (2019)

    Ding, K.et al.Phase-change heterostructure enables ultralow noise and drift for memory operation. Science366, 210–215 (2019)

    2019

  51. [59]

    & Salinga, M

    Wimmer, M., Kaes, M., Dellen, C. & Salinga, M. Role of activation energy in resistance drift of amorphous phase change materials.Frontiers in Physics2, 75 (2014). 15

    2014

  52. [60]

    S., Dirisaglik, F., Gokirmak, A

    Khan, R. S., Dirisaglik, F., Gokirmak, A. & Silva, H. Resistance drift in Ge 2Sb2Te5 phase change memory line cells at low temperatures and its response to photoexcitation.Applied Physics Letters 116, 253501 (2020)

    2020

  53. [61]

    Li, C.et al.Understanding phase-change materials with unexpectedly low resistance drift for phase-change memories.Journal of Materials Chemistry C6, 3387–3394 (2018)

    2018

  54. [62]

    Wang, Q.et al.Reliable Ge 2Sb2Te5 based phase-change electronic synapses using carbon doping and programmed pulses.Journal of Materiomics8, 382–391 (2022)

    2022

  55. [63]

    Meli, V.et al.Multi level cell reliability in Ge-rich GeSbTe-based phase change memory arrays 1–5 (2023)

    2023

  56. [64]

    Kiouseloglou, A.et al.A novel programming technique to boost low-resistance state performance in Ge-rich GST phase change memory.IEEE Transactions on Electron Devices61, 1246–1254 (2014)

    2014

  57. [65]

    Liu, B.et al.Multi-level phase-change memory with ultralow power consumption and resistance drift.Science Bulletin66, 2217–2224 (2021)

    2021

  58. [66]

    Huang, J.et al.Nanoscale chemical heterogeneity ensures unprecedently low resistance drift in cache-type phase-change memory materials.Nano Letters23, 2362–2369 (2023)

    2023

  59. [67]

    Zhou, L.et al.Resistance drift suppression utilizing GeTe/Sb 2Te3 superlattice-like phase-change materials.Advanced Electronic Materials6, 1900781 (2020)

    2020

  60. [68]

    I.et al.Ultralow–switching current density multilevel phase-change memory on a flexible substrate.Science373, 1243–1247 (2021)

    Khan, A. I.et al.Ultralow–switching current density multilevel phase-change memory on a flexible substrate.Science373, 1243–1247 (2021)

    2021

  61. [69]

    Wang, X.et al.Amorphous phase-change memory alloy with no resistance drift.Nature Materials 1–7 (2025)

    2025

  62. [70]

    & Kambe, K

    Hashimoto, M., Niizeki, T. & Kambe, K. Effect of substrate temperature on crystallization of amorphous antimony film.Japanese Journal of Applied Physics19, 21 (1980)

    1980

  63. [71]

    Zuliani, P.et al.Overcoming temperature limitations in phase change memories with optimized GexSbyTez.IEEE Transactions on Electron Devices60, 4020–4026 (2013)

    2013

  64. [72]

    Whab, A., Parveen, S., Bhatt, N.et al.Elucidating the phase change memory properties of (GeTe)1−x(GeSe2)x (0≤x≤0.21) films.Vacuum240, 114491 (2025)

    2025

  65. [73]

    Japanese Journal of Applied Physics47, 843 (2008)

    Wang, F.et al.Temperature influence on electrical properties of Sb–Te phase-change material. Japanese Journal of Applied Physics47, 843 (2008)

    2008

  66. [74]

    K., Dieker, H

    Njoroge, W. K., Dieker, H. & Wuttig, M. Influence of dielectric capping layers on the crystallization kinetics of Ag 5In6Sb59Te30 films.Journal of Applied Physics96, 2624–2627 (2004)

    2004

  67. [75]

    Chen, B.et al.Anomalous crystallization kinetics of ultrafast ScSbTe phase-change memory materials induced by nitrogen doping.Acta Materialia238, 118211 (2022)

    2022

  68. [76]

    & Jacobs, B

    Coombs, J., Jongenelis, A., van Es-Spiekman, W. & Jacobs, B. Laser-induced crystallization phe- nomena in GeTe-based alloys. I. Characterization of nucleation and growth.Journal of Applied Physics78, 4906–4917 (1995)

    1995

  69. [77]

    C., Heo, T

    Egan, G. C., Heo, T. W., Samanta, A. & Campbell, G. H. A novel liquid-mediated nucleation mechanism for explosive crystallization in amorphous germanium.Acta Materialia179, 190–200 (2019)

    2019

  70. [78]

    & Schlotzhauer, G

    Bostanjoglo, O. & Schlotzhauer, G. Impulse stimulated crystallization of Sb films investigated by time resolved TEM.physica status solidi (a)68, 555–560 (1981). 16

    1981

  71. [79]

    Pirovano, A.et al.Low-field amorphous state resistance and threshold voltage drift in chalcogenide materials.IEEE Transactions on Electron Devices51, 714–719 (2004)

    2004

  72. [80]

    Chemistry of Materials31, 8794–8800 (2019)

    Chen, B.et al.Kinetics features conducive to cache-type nonvolatile phase-change memory. Chemistry of Materials31, 8794–8800 (2019)

    2019

  73. [81]

    Orava, J., Hewak, D. W. & Greer, A. L. Fragile-to-strong crossover in supercooled liquid Ag-In- Sb-Te studied by ultrafast calorimetry.Advanced Functional Materials25, 4851–4858 (2015)

    2015

  74. [82]

    K¨ uhne, T. D.et al.CP2K: An electronic structure and molecular dynamics software package - quickstep: Efficient and accurate electronic structure calculations.The Journal of Chemical Physics 152, 194103 (2020)

    2020

  75. [83]

    P., Burke, K

    Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple.Physical Review Letters78, 1396–1396 (1997)

    1997

  76. [84]

    & Hutter, J

    VandeVondele, J. & Hutter, J. Gaussian basis sets for accurate calculations on molecular systems in gas and condensed phases.The Journal of Chemical Physics127, 114105 (2007)

    2007

  77. [85]

    https://next-gen.materialsproject.org (2026)

    Materials Project. https://next-gen.materialsproject.org (2026). Accessed: 2026-06-19

    2026

  78. [86]

    & Luisier, M

    Rhyner, R. & Luisier, M. Atomistic modeling of coupled electron-phonon transport in nanowire transistors.Physical Review B89, 235311 (2014)

    2014

  79. [87]

    B., Klimeck, G

    Boykin, T. B., Klimeck, G. & Oyafuso, F. Valence band effective-mass expressions in thesp 3d5s∗ empirical tight-binding model applied to a Si and Ge parameterization.Physical Review B69, 115201 (2004)

    2004

  80. [88]

    & Herman, I

    Sui, Z. & Herman, I. P. Effect of strain on phonons in Si, Ge, and Si/Ge heterostructures.Physical Review B48, 17938–17953 (1993)

    1993

  81. [89]

    & Luisier, M

    Rhyner, R. & Luisier, M. Minimizing self-heating and heat dissipation in ultrascaled nanowire transistors.Nano Letters16, 1022–1026 (2016). 17

    2016

Pith tools

Reviewed July 30, 2026 · model on record in the stance chip above.