pith. sign in

arxiv: 2606.10397 · v1 · pith:3LEOUBVCnew · submitted 2026-06-09 · ❄️ cond-mat.str-el · cond-mat.mtrl-sci

Improved selector behavior in ultrathin chromium-doped V₂O₃ films

Johannes Mohr , Tyler Hennen , Yudi Wang , Xiaoyu Xu , Loc Vinh , Dirk J. Wouters , Rainer Waser , Joyeeta Nag
show 1 more author
Daniel Bedau
This is my paper

Pith reviewed 2026-06-27 11:59 UTC · model grok-4.3

classification ❄️ cond-mat.str-el cond-mat.mtrl-sci
keywords selector devicesnegative differential resistanceCr-doped V2O3ultrathin filmsforming stepTiN interfaceTi diffusion
0
0 comments X

The pith

Chromium-doped V2O3 films maintain and improve selector switching down to 5 nm thickness.

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

The paper establishes that the negative differential resistance switching effect in Cr-doped V2O3, useful for selector elements in memory and neuromorphic tech, persists in ultrathin films as thin as 5 nm. At these thicknesses, the films show better electrical properties including lower leakage current and more abrupt transitions compared to thicker ones. Notably, crystalline and amorphous films behave similarly, both requiring a forming step, which is linked to the formation of a thin amorphous layer at the TiN electrode interface. Ti diffusion into the film from the electrode may account for the improvements.

Core claim

Electrical measurements demonstrate that the switching effect is maintained for very thin films down to 5 nm, with improved properties such as low leakage current and abrupt transition. For these thicknesses, crystalline and amorphous films show very similar behavior, both requiring a forming step. Transmission electron microscopy indicates this is due to a thin amorphous layer at the TiN electrode interface. Elemental mapping reveals complex chromium dopant distribution and Ti diffusion from the electrode, potentially responsible for the improved properties.

What carries the argument

The thin amorphous layer at the TiN electrode interface, which induces the forming step and allows Ti diffusion to improve electrical characteristics in ultrathin Cr-doped V2O3 films.

If this is right

  • Selector devices based on Cr-doped V2O3 can be scaled to 5 nm films while retaining functionality.
  • Amorphous films can be used interchangeably with crystalline ones at ultrathin thicknesses for selector applications.
  • Ti diffusion from electrodes can be leveraged to enhance device performance in thin film selectors.
  • Forming steps become necessary even for initially crystalline films when scaled to 5 nm due to interface effects.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Interface engineering at the electrode could be key to controlling forming voltage and leakage in these devices.
  • Similar interfacial effects might appear in other oxide-based selector materials when thinned below 10 nm.
  • Optimizing electrode materials to control Ti or other metal diffusion could further improve selector metrics.

Load-bearing premise

That the observed thin amorphous layer at the electrode interface is what causes the forming step to be required in both crystalline and amorphous films.

What would settle it

Observation via TEM of no amorphous interfacial layer in a 5 nm crystalline film that does not require a forming step, or electrical tests showing no Ti diffusion correlating with the improved leakage and abruptness.

Figures

Figures reproduced from arXiv: 2606.10397 by Daniel Bedau, Dirk J. Wouters, Johannes Mohr, Joyeeta Nag, Loc Vinh, Rainer Waser, Tyler Hennen, Xiaoyu Xu, Yudi Wang.

Figure 1
Figure 1. Figure 1: a) SEM image of the finished structure. The device is visible in the center as a small depression in the top platinum electrode. b) Sketch of the cross-section of the stack along a horizontal line in the center of a). The devices were created by blanket deposition of material on top of coupons diced from a nano-via vehicle wafer. The coupons were cleaned in an ultrasonic bath for 10 min in acetone and isop… view at source ↗
read the original abstract

Devices based on the negative differential resistance effect in chromium doped V$_2$O$_3$ are considered to be promising as selector elements for use in emerging memory technologies, as well as for neuromorphic applications. It is shown by electrical measurements, that the switching effect is maintained for very thin films down to 5 nm, and even improved properties such as a low leakage current and an abrupt transition are observed. For these thicknesses, the behavior of crystalline and amorphous films becomes very similar; most strikingly, a forming step is required in both. Transmission electron microscopy reveals this to be likely due to a thin amorphous layer that forms at the interface to the TiN electrode. Elemental mapping further shows a complex distribution of the chromium dopants, as well as a diffusion of Ti into the layer from the electrode, which might be responsible for the improved properties.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript reports that Cr-doped V₂O₃ films retain negative differential resistance selector behavior down to 5 nm thickness, with improved leakage current and transition abruptness relative to thicker films. At these thicknesses, crystalline and amorphous films exhibit similar electrical characteristics, including the requirement for a forming step in both cases. TEM imaging identifies a thin amorphous layer at the TiN electrode interface as the probable origin of the forming step, while EDX mapping shows Cr redistribution and Ti diffusion from the electrode, which the authors suggest may account for the enhanced electrical properties.

Significance. If substantiated with quantitative statistics, the result that functional Cr:V₂O₃ selectors can be realized at 5 nm is relevant for back-end-of-line integration in crossbar memory arrays. The observation that interface amorphization equalizes the behavior of crystalline and amorphous films supplies a concrete microstructural explanation for forming-step requirements. The combination of electrical I–V data with cross-sectional TEM/EDX constitutes a standard and appropriate characterization approach for such devices.

major comments (2)
  1. [Abstract] Abstract: the assertion that the thin amorphous interface layer 'is likely' responsible for the forming step required in both crystalline and amorphous 5 nm films rests on post-fabrication correlation; no control experiments (different bottom electrode, inserted diffusion barrier, or interface-free reference stack) are described that would isolate the layer’s causal contribution versus Cr redistribution or simple thickness scaling.
  2. [Electrical measurements] Electrical measurements: the claims of 'low leakage current' and 'abrupt transition' are presented without reported error bars, number of devices measured, yield statistics, or raw I–V traces; this absence prevents quantitative assessment of the improvement relative to thicker films and weakens the central scaling claim.
minor comments (1)
  1. [Abstract] The abstract uses tentative language ('likely', 'might be responsible') that accurately reflects the inferential nature of the interface interpretation; this phrasing should be retained or strengthened with explicit caveats in the discussion section.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive report. The two major comments identify genuine limitations in the strength of evidence presented. We address each below and indicate where revisions will be made to the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the assertion that the thin amorphous interface layer 'is likely' responsible for the forming step required in both crystalline and amorphous 5 nm films rests on post-fabrication correlation; no control experiments (different bottom electrode, inserted diffusion barrier, or interface-free reference stack) are described that would isolate the layer’s causal contribution versus Cr redistribution or simple thickness scaling.

    Authors: We agree that the claim rests on correlative post-fabrication TEM/EDX observations rather than direct causal isolation via control stacks. The manuscript does not contain the suggested control experiments, and performing them would require new sample fabrication. In revision we will change the wording from 'likely' to 'consistent with' and add an explicit discussion paragraph noting that Cr redistribution and thickness scaling remain alternative or contributing factors. This is a textual clarification only. revision: partial

  2. Referee: [Electrical measurements] Electrical measurements: the claims of 'low leakage current' and 'abrupt transition' are presented without reported error bars, number of devices measured, yield statistics, or raw I–V traces; this absence prevents quantitative assessment of the improvement relative to thicker films and weakens the central scaling claim.

    Authors: The original manuscript presented only representative curves without aggregate statistics. We have access to measurements from multiple devices (approximately 20–30 per thickness and crystallinity type across several samples) and can compute standard deviations. In the revised manuscript we will add error bars to the key metrics, state the number of devices and samples measured, report device yield, and move a set of raw I–V traces to the supplementary information. This addresses the quantitative-assessment concern directly. revision: yes

Circularity Check

0 steps flagged

No circularity: purely experimental report with no derivations or self-referential predictions

full rationale

The manuscript consists exclusively of electrical measurements, TEM imaging, and EDX elemental mapping on fabricated devices. No equations, fitted models, parameter predictions, or mathematical derivations appear. Claims about forming steps and interface layers are presented as direct observations and post-fabrication inferences, not as outputs of any self-contained derivation chain. No self-citations are invoked as load-bearing uniqueness theorems or ansatzes. The paper is therefore self-contained against external benchmarks with no reduction of results to their own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper is an experimental characterization study; it introduces no free parameters, new physical entities, or non-standard axioms beyond routine assumptions of thin-film metrology.

axioms (1)
  • domain assumption TEM imaging and elemental mapping accurately capture the presence and location of the thin amorphous interface layer and Ti diffusion.
    The central interpretation of why crystalline and amorphous films behave identically rests on these microscopy results.

pith-pipeline@v0.9.1-grok · 5706 in / 1337 out tokens · 39271 ms · 2026-06-27T11:59:58.496987+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

25 extracted references

  1. [1]

    Volker Rzehak, Low -Power FRAM Microcontrollers and Their Applications , Texas Instruments

  2. [2]

    Heidecker, MRAM Technology Status, Pasadena, California, 2013

    J. Heidecker, MRAM Technology Status, Pasadena, California, 2013

    2013

  3. [3]

    Efficient and high- performing vehicle architecture: Cooperation of Continental and Infineon, 2023

    2023

  4. [4]

    Qualcomm QCC730 Dual band micro-power Wi-Fi Product Brief,

  5. [5]

    Qualcomm QCC711 Tri -core Ultra -Low Power Bluetooth Low Energy SoC Product Brief,

  6. [6]

    H. Lv, X. Xu, P. Yuan, D. Dong, T. Gong, J. Liu, Z. Yu, P. Huang, K. Zhang, C. Huo, C. Chen, Y. Xie, Q. Luo, S. Long, Q. Liu, J. Kang, D. Yang, S. Yin, S. Chiu, M. Liu, In 2017 IEEE International Electron Devices Meeting (IEDM) , IEEE, 2017, pp. 2.4.1-2.4.4

    2017

  7. [7]

    Sebastian, M

    A. Sebastian, M. Le Gallo, R. Khaddam -Aljameh, E. Eleftheriou, Nat Nanotechnol 2020, 15, 529

    2020

  8. [8]

    Y. Li, Z. Wang, R. Midya, Q. Xia, J. J. Yang, J Phys D Appl Phys 2018, 51, 503002

    2018

  9. [9]

    Y. Cho, J. Heo, S. Kim, S. Kim, Surfaces and Interfaces 2023, 41, 103273

    2023

  10. [10]

    E. Cha, J. Woo, D. Lee, S. Lee, J. Song, Y. Koo, J. Lee, C. G. Park, M. Y. Yang, K. Kamiya, K. Shiraishi, B. Magyari -Kope, Y. Nishi, H. Hwang, In 2013 IEEE International Electron Devices Meeting, IEEE, 2013, pp. 10.5.1-10.5.4

    2013

  11. [11]

    S. A. Chekol, J. Song, J. Park, J. Yoo, S. Lim, H. Hwang, In Memristive Devices for Brain-Inspired Computing, Elsevier, 2020, pp. 135–164

    2020

  12. [12]

    X. Peng, R. Madler, P.-Y. Chen, S. Yu, J Comput Electron 2017, 16, 1167

    2017

  13. [13]

    D. B. McWhan, J. P. Remeika, Phys Rev B 1970, 2, 3734

    1970

  14. [14]

    Stoliar, L

    P. Stoliar, L. Cario, E. Janod, B. Corraze, C. Guillot- Deudon, S. Salmon- Bourmand, V. Guiot, J. Tranchant, M. Rozenberg, Advanced Materials 2013, 25, 3222

    2013

  15. [15]

    Hennen, D

    T. Hennen, D. Bedau, J. A. J. Rupp, C. Funck, S. Menzel, M. Grobis, R. Waser, D. J. Wouters, In 2018 IEEE International Electron Devices Meeting (IEDM), 1- 5 December 2018, San Francisco, CA, USA, 2018, pp. 37.5.1-37.5.4

    2018

  16. [16]

    Hennen, D

    T. Hennen, D. Bedau, J. A. J. Rupp, C. Funck, S. Menzel, M. Grobis, R. Waser, D. J. Wouters, In 2019 IEEE 11th International Memory Workshop (IMW) , Monterey, CA, USA, 2019, pp. 1–4. 17

    2019

  17. [17]

    J. Mohr, T. Hennen, D. Bedau, R. Waser, D. J. Wouters, Advanced Physics Research 2024, 3, 2400040

    2024

  18. [18]

    Hennen, Harnessing stochasticity and negative differential resistance for unconventional computation

    T. Hennen, Harnessing stochasticity and negative differential resistance for unconventional computation. Dissertation, RWTH Aachen University, 2023

    2023

  19. [19]

    J. A. J. Rupp, Synthesis and Resistive Switching Mechanisms of Mott Insulators based on Undoped and Cr -doped Vanadium Oxide Thin Films. Dissertation, RWTH Aachen University, 2020

    2020

  20. [20]

    Trastoy, Y

    J. Trastoy, Y. Kalcheim, J. del Valle, I. Valmianski, I. K. Schuller, J Mater Sci 2018, 53, 9131

    2018

  21. [21]

    Tyler Hennen, thennen/py-ivtools, Zenodo, 2024

    2024

  22. [22]

    J. Mohr, C. Bengel, T. Hennen, D. Bedau, S. Menzel, R. Waser, D. J. Wouters, physica status solidi (a) 2024, 221, 2300405

    2024

  23. [23]

    J. A. J. Rupp, M. Querré, A. Kindsmüller, M.-P. Besland, E. Janod, R. Dittmann, R. Waser, D. J. Wouters, J Appl Phys 2018, 123, 44502

    2018

  24. [24]

    D. B. McWhan, A. Menth, J. P. Remeika, W. F. Brinkman, T. M. Rice, Phys Rev B 1973, 7, 1920

    1973

  25. [25]

    KUWAMOTO, H

    H. KUWAMOTO, H. V. KEER, J. E. KEEM, S. A. SHIVASHANKAR, L. L. VAN ZANDT, J. M. HONIG, Le Journal de Physique Colloques 1976, 37, C4. 18 Supporting Information Figure S1: Lower magnification view of the films shown in Figure 4

    1976