Exact solution for stationary states of a closed memristor with mobile charged vacancies

Irina V. Boylo; Konstantin L. Metlov

REVIEW 2 major objections 6 minor 27 references

Exact solution for stationary states of a closed memristor with mobile charged vacancies

T0 review · 2 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read For a closed memristor with mobile charged vacancies, the stationary end concentrations and the limiting on/off resistances are independent of the strength of electrostatic interaction among the vacancies, even though the interior vacancy…

desk verdict The alpha-independence claim is correct and survives the typo in Eq. (9), but the printed quadrature is broken and must be fixed before this is citable. read the letter →

arxiv 2608.12264 v1 pith:6XGL243L submitted 2026-08-12 cond-mat.mes-hall math-phmath.MP

Irina V. Boylo , Konstantin L. Metlov This is my paper · ORCID

classification cond-mat.mes-hallmath-phmath.MP

keywords memristorchargedvacancieselectrostaticinteractionstationarystatesPoissonequationBurgerslogittransformresistanceswitching

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 paper extends a nonlinear model of a closed memristor—one that exchanges only electrons, not atoms—to include electrostatic repulsion among mobile charged vacancies, and asks what that interaction does to the stationary vacancy profiles that set the device's on and off resistances. The authors obtain exact analytical expressions for these stationary states, showing that the interaction reshapes the interior profile and creates an intermediate electrically neutral region between the vacancy-rich and vacancy-poor zones. Despite this strong reshaping, the vacancy concentrations at the two ends, $c(0)$ and $c(1)$, come out independent of the interaction strength $\alpha$, and so do the limiting on/off resistances computed in the linear approximation. The reason is that the logit boundary constant $\Lambda_0$ is fixed by the potential boundary conditions and vacancy conservation alone, with no $\alpha$ dependence. If this is right, the static resistance contrast of such a memristor is a boundary and contact property, while the Coulomb interaction only rearranges the internal charge distribution.

What carries the argument

The load-bearing object is the logit transform $\Lambda=\log(c/(1-c))$, which linearizes the vacancy flux equation, reduces the stationary problem to the autonomous ODE $\Lambda_{xx} = \frac{\alpha}{2}\big(e^\Lambda/(1+e^\Lambda)-r\big)$, and yields the solution as a quadrature with two integration constants $\Lambda_0$ and $\upsilon$. The ideal-contact potential conditions translate into $\Lambda(1)-\Lambda(0)=p$, and the explicit formula for $\Lambda_0$ shows why $\alpha$ drops out of the endpoint concentrations. The derivative-scaling factor $\upsilon$, defined as the slope of the profile at $c=r$ normalized to the $\alpha=0$ Burgers solution, carries the electrostatic reshaping of the interior profile.

What would settle it

Compute stationary solutions of Eqs. (4) with the potential boundary conditions changed to $\varphi(0)=V_0$ and $\varphi(1)=V_1$, or to a Robin condition representing contact capacitance, holding $p$, $r$, and $\alpha$ fixed, and check whether $\Lambda_0$ and the endpoint concentrations stay independent of $\alpha$; any $\alpha$ variation would falsify the claim for that boundary class. Experimentally, one could vary $\alpha$ through temperature or through the dielectric constant of the host while measuring $R_{\rm on}$ and $R_{\rm off}$ in a closed memristor with high-quality contacts: the claim predicts no change in the limiting resistances.

Watch

Extended reading notes

Core claim

The paper's central claim is that in the stationary state of a closed memristor with mobile charged vacancies, the endpoint concentrations $c(0)$ and $c(1)$ are independent of the electrostatic interaction strength $\alpha$, and therefore the limiting on and off state resistances are also $\alpha$-independent to leading linear order. The argument runs through the logit transformation $\Lambda=\log(c/(1-c))$, which turns the zero-flux condition into $\Lambda = p x - \alpha\varphi/2 + \Lambda_0$ and the Poisson equation into an autonomous second-order ODE solvable by quadrature. The ideal-contact conditions $\varphi(0)=\varphi(1)=0$ force $\Lambda(1)-\Lambda(0)=p$, and conservation of the total vacancy number then fixes $\Lambda_0 = \log\big((e^{pr}-1)/(e^p-e^{pr})\big)$, which contains no $\alpha$. Consequently the end concentrations, and hence the resistance combination $\sigma = (c(0)/C_{\max})\cos^2\theta + (c(1)/C_{\max})\sin^2\theta$, do not change when the Coulomb interaction is turned on, even though the full profile develops an intermediate neutral region and the interior slope at the filling point is reduced by a factor $\upsilon$ that lies strictly between 0 and 1.

Load-bearing premise

The entire $\alpha$-independence result rests on ideal contacts that pin the electrostatic potential to zero at both ends, together with the global-neutrality background; if real contacts hold a finite surface charge, the cancellation no longer follows.

Editorial extensions

If this is right

The static on/off resistance contrast of such a closed memristor is fixed by the contact asymmetry and the total vacancy filling, not by the strength of vacancy-vacancy repulsion.
Strong Coulomb interaction can be diagnosed through the interior profile: a neutral plateau forms between the depleted and enriched zones, and its width grows with $\alpha$.
The interior slope at the filling point is always reduced by electrostatic interaction ($0<\upsilon<1$), and it approaches the Burgers value in the large-current limit $p\to\infty$.
The quadrature solution plus the rational approximation for $\upsilon$ give analytic estimates for stationary profiles over the whole parameter range without full numerics.

Reading between the lines

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

My inference: if the $\alpha$-independence is understood as a boundary effect, then engineering the contacts—changing $\theta$, $R_1$, or $R_2$—is the only static lever on the on/off resistance ratio, while tuning the dielectric environment should affect switching dynamics but not the limiting resistance values.
My expectation: repeating the stationary calculation with non-ideal contacts, for example finite surface capacitance or a Schottky barrier replacing $\varphi(0)=\varphi(1)=0$, will make $c(0)$ and $c(1)$ depend on $\alpha$; the reported cancellation is specific to ideal contacts.
A testable extension: the same quadrature method could be adapted to slow time-dependent switching around the stationary state, connecting the $\alpha$-independent endpoints to the switching kinetics studied in the earlier linear-resistance model.

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.

Desk Editor's Note

read the letter

Bottom line: the headline physical claim is right, and the centerpiece formula as printed is wrong. The stress-test note is correct about Eq. (9). With V(t)=log(1+e^t)-rt, the printed form x = integral of dt/(A+alpha V) gives Lambda_x = A+alpha V, so Lambda_xx = alpha(c-r)(A+alpha V), which cannot match Eq. (8)'s (alpha/2)(c-r) unless A+alpha V = 1/2 identically. The alpha=0 limit makes the same point: Eq. (9) would give Lambda = Lambda0 + A x, and the boundary conditions force A=p, while Eq. (10) at upsilon=1 gives A=p^2. The missing square root is not cosmetic; it undermines the profiles in Fig. 1, the numerical determination of upsilon, and the asymptotics in Appendix A as printed.

What is actually new survives this. The model extends the authors' earlier Burgers-equation memristor by adding the Poisson coupling, and the stationary profiles plus the intermediate neutral region are new. More importantly, the endpoint-independence result is real. It does not use the quadrature at all: integrating Eq. (8) gives Lambda_x(1)=Lambda_x(0), the first integral gives Lambda_x^2 - alpha V = const, and with Lambda(1)-Lambda(0)=p this yields Lambda0 independent of alpha. That argument is exact and internally sound, and it is the paper's most valuable contribution. The paper is also honest about the ideal-contact boundary condition as a modeling assumption; the scope limitation is stated, not hidden.

Soft spots, in proportion: Eq. (9) is a load-bearing typo that must be corrected, but it is repairable and the central claim does not depend on it. The bound 0 < upsilon < 1 is asserted rather than proved, and the derivation of Eq. (12) is omitted; both should be supplied. The "exact solution" still needs a numerical root-find for upsilon, so it is exact in the quadrature sense, not fully closed-form; that is a fair characterization but worth stating precisely. The flux model in Eq. (1) is carried over from prior work without microscopic derivation; I do not see that as a flaw here since this paper is about the stationary consequences of that model.

Who is this for? People working on analytical memristor device theory, especially on vacancy drift-diffusion models. It will not reshape the field, but the endpoint-independence result is a clean, surprising simplification that deserves to be in the literature. I would send it to peer review with a directive to fix Eq. (9) and supply the missing derivations; after that, I would accept it. I would also cite it if I worked on these models.

Referee Report

2 major / 6 minor

Summary. The paper studies a one-dimensional closed memristor with mobile charged vacancies, adding electrostatic interaction to the authors' earlier nonlinear model. It derives stationary-state equations for the vacancy concentration and the electrostatic potential, introduces the logit variable Λ, and reduces the problem to an autonomous second-order ODE. The authors claim an exact quadrature solution, give an explicit expression for the endpoint logit Λ0, and conclude that the endpoint concentrations c(0) and c(1), and hence the limiting resistances in the on and off states, are independent of the electrostatic interaction strength α in the linear approximation (3). Figures show vacancy profiles for α=0 and α=1000, and an appendix provides asymptotic and rational approximations for a derivative scaling factor υ.

Significance. If correct, the alpha-independence of the limiting resistances is a nontrivial exact result that is robust to the strength of the electrostatic interaction, and the analytical stationary profiles are a useful contribution to memristor modeling. A strength of the derivation is that the independence of the endpoint concentrations follows directly from the first integral of Eq. (8) and the boundary conditions, without relying on the printed quadrature. However, the printed central quadrature contains an error that affects the quantitative results, and some supporting derivations are omitted, so the manuscript requires revision before the claims can be fully accepted.

major comments (2)

[Section III, Eq. (9)] The printed quadrature is dimensionally inconsistent with Eq. (8) and is missing a square root in the denominator. Differentiating the printed Eq. (9) gives Λ_x = A + αV(Λ) with V(t) = log(1+e^t)−rt, and differentiating again yields Λ_xx = α(c−r)(A+αV(Λ)), which can agree with Eq. (8) only if A+αV(Λ) = 1/2 identically, which is impossible. The α=0 limit of Eq. (9) gives A = p from the boundary condition Λ(1)−Λ(0)=p, whereas Eq. (10) with υ=1 gives A = p², a direct contradiction. The correct first integral is Λ_x² − αV(Λ) = A, so the quadrature should be ∫_{Λ0}^{Λ} dt / sqrt(A + αV(t)) = x. This error undermines the numerical determination of υ from x=1, the profiles in Fig. 1, and the asymptotics in Appendix A, all of which must be recomputed with the corrected formula.
[Section III, Eq. (12)] The central claim that Λ0, and hence c(0) and c(1), are independent of α is based entirely on Eq. (12), but the derivation of Eq. (12) from the boundary conditions and particle-number conservation is not shown in the manuscript. Since this is the load-bearing step for the main result, a short derivation should be provided so that the reader can verify the claimed independence.

minor comments (6)

[Section III, Eq. (11) and Fig. 1 caption] The parameter υ is defined analytically in Eq. (11) as the ratio of derivatives at c=r, but the caption of Fig. 1 defines it geometrically as υ = tanφ/tanφ_B without introducing this geometric definition in the text; these definitions should be reconciled.
[Section III, after Eq. (11)] The statement 'It can be shown that always 0 < υ < 1' is made without proof or reference; a proof or a citation should be provided.
[Appendix A, Eq. (A5)] Equation (A5) contains a misplaced parenthesis in 'h(576 +ap(1 + (r−1)r)))' and the notation 'a/b=' is ambiguous; write the expressions for a and b separately.
[General] The full Russian translation of the manuscript is appended after the English references, which is unconventional for a journal submission and should be removed or moved to supplementary material.
[Throughout] Please specify that all logarithms are natural logarithms, and clarify the notation 'cx|c=r' and 'bx|b=r' in Eq. (11) as derivatives with respect to x evaluated where c=r and b=r, respectively.
[Section III, paragraph on transition region] The sentence 'It seems (although the authors did not emphasize this) ...' refers to the authors in the third person; rephrase to 'we did not emphasize' or use another suitable first-person construction.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the alpha-independence of endpoint concentrations is derived from the stated flux model, Poisson equation, boundary conditions, and vacancy conservation, not fitted or assumed.

full rationale

The paper's central result—that c(0) and c(1), and hence the on/off resistances in Eq. (3), do not depend on the interaction strength alpha—is obtained by integrating the stated stationary equations, not by fitting or by definitional identity. From J=0 one gets Eq. (4a), and the logit transform gives Eq. (8); the boundary condition Phi(0)=Phi(1)=0 yields Lambda(1)-Lambda(0)=p, and conservation (5) forces the integrated charge to vanish. The first integral of Eq. (8) then gives Lambda_x(1)=Lambda_x(0) and V(Lambda(1))=V(Lambda(0)) with V(t)=log(1+e^t)-r t; combined with Lambda(1)-Lambda(0)=p this yields Eq. (12) with no alpha dependence. This is a genuine derivation from the model assumptions. The flux law (1) and resistance formula (3) are taken explicitly and transparently from the authors' prior work [18,19] as model input, not presented as derived in this paper, so the self-citations are not load-bearing circularity. The printed quadrature Eq. (9) appears to omit a square root and therefore looks incorrect as written, but that is a mathematical typo/correctness issue rather than a circular reduction; the alpha-independence claim does not rely on Eq. (9). No fitted parameter is renamed as a prediction, and no uniqueness result is imported from the authors' prior work to force the conclusion.

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

The central claim rests on the memristor model from the authors' prior work: the specific nonlinear flux, the mean-field Poisson equation with global-neutrality background r, and ideal-contact boundary conditions. These are domain assumptions, not derived in this paper. There are no free parameters fitted to data; the integration constant upsilon is fixed by boundary conditions and conservation, and the approximate expressions in Appendix A are matched to asymptotics, not to external data.

assumptions (5)

domain assumption Vacancy flux has the nonlinear form J = -D grad C + C(1-C/Cmax) 2DqE/(kBT).
Taken from the authors' prior model [18]; the entire stationary problem is built on this constitutive law, and no microscopic derivation is given in this paper.
domain assumption Poisson equation with background concentration C0 equal to the global average r.
Section II, Eq. (2); assumes global charge neutrality and a fixed uniform neutralizing background.
domain assumption Ideal contacts set Phi=0 at both ends.
Section II; this boundary condition is essential for the alpha-independence of endpoint concentrations.
domain assumption Stationary states correspond to zero vacancy current J=0.
Section III; ignores possible leakage or nonuniform steady currents.
domain assumption Resistance depends linearly on endpoint concentrations via Eq. (3).
Taken from [19]; the conclusion about limiting resistances is only as good as this linearized phenomenological relation.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Exact solution for stationary states of a closed memristor with mobile charged vacancies." pith.science (2026). https://pith.science/paper/6XGL243L

@misc{pith2026260812264,
  author       = {Pith},
  title        = {Pith review of: Exact solution for stationary states of a closed memristor with mobile charged vacancies},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6XGL243L}},
  note         = {Machine review of arXiv:2608.12264}
}

read the original abstract

A nonlinear model of a memristor based on charged mobile vacancies is considered taking into account the electrostatic interaction between them. This interaction significantly affects the stationary (limiting) vacancy distributions formed under the action of the electric current flowing through the memristor, for which analytical expressions are obtained in this work. Between the regions with reduced and increased vacancy concentrations, an intermediate electrically neutral region is formed due to the electrostatic interaction. Interestingly, the limiting resistances in the ``on'' and ``off'' states of such a memristor do not depend on the strength of the electrostatic interaction, at least in the leading first order in the vacancy concentration.

Figures

Figures reproduced from arXiv: 2608.12264 by the authors.

**Figure 2.** Derivative scaling factor s as a function of the reduced current strength p in memristors with different vacancy fillings at α = 1000. The curves for fillings r and 1−r coincide. The inset shows the dependence of the derivative of the vacancy distribution profile dc/ dx at the point c = r as a function of the memristor filling factor r at p = 50. The dependence of the derivative scaling factor υ on the current stren… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 26 canonical work pages

[10]

M. J. Rozenberg, M. J. Sánchez, R. Weht, C. Acha, F. Gomez-Marlasca, and P. Levy, Mechanism for bipolar resistive switching in transition-metal oxides, Phys. Rev. B81, 115101 (2010)

work page 2010
[11]

Ghenzi, M

N. Ghenzi, M. J. Sánchez, F. Gomez-Marlasca, P. Levy, and M. J. Rozenberg, Hysteresis switching loops in Ag- manganite memristive interfaces, J. Appl. Phys.107, 093719 (2010)

work page 2010
[12]

Larentis, F

S. Larentis, F. Nardi, S. Balatti, D. C. Gilmer, and D. Ielmini, Resistive switching by voltage-driven ion mi- gration in bipolar rram—part ii: Modeling, IEEE Trans. Electron Devices59, 2468–2475 (2012)

work page 2012
[13]

S. Kim, S. Choi, and W. Lu, Comprehensive physical model of dynamic resistive switching in an oxide memris- tor, ACS Nano8, 2369–2376 (2014)

work page 2014
[14]

Marchewka, R

A. Marchewka, R. Waser, and S. Menzel, A 2d axisymmet- ric dynamic drift-diffusion model for numerical simulation of resistive switching phenomena in metal oxides, in2016 International Conference on Simulation of Semiconductor Processes and Devices (SISPAD)(2016) p. 145–148

work page 2016
[16]

N. V. Agudov, A. V. Safonov, A. V. Krichigin, A. A. Kharcheva, A. A. Dubkov, D. Valenti, D. V. Guseinov, A. I. Belov, A. N. Mikhaylov, A. Carollo, and B. Spag- nolo, Nonstationary distributions and relaxation times in a stochastic model of memristor, J. Stat. Mech.2020, 024003 (2020)

work page 2020
[19]

Boilo and K

I. Boilo and K. Metlov, Switching of a closed mobile vacancy based memristor whose specific resistance lin- early depends on local vacancy concentration, JETP,168, 713–719 (2025)

work page 2025
[21]

Chua, Memristor-the missing circuit element, IEEE Trans

L. Chua, Memristor-the missing circuit element, IEEE Trans. Circuit Theory18, 507–519 (1971)

work page 1971

Show all 27 references

[22]

Chua and S

L. Chua and S. M. Kang, Memristive devices and systems, Proceedings of the IEEE64, 209–223 (1976)

1976
[23]

D. B. Strukov, G. S. Snider, D. R. Stewart, and R. S. Williams, The missing memristor found, Nature453, 80–83 (2008)

2008
[24]

J. C. Bruyere and B. K. Chakraverty, Switching and negative resistance in thin films of nickel oxide, Applied Physics Letters16, 40–43 (1970)

1970
[25]

Sawa, Resistive switching in transition metal oxides, Mater

A. Sawa, Resistive switching in transition metal oxides, Mater. Today11, 28–36 (2008)

2008
[26]

Fujii, M

T. Fujii, M. Kawasaki, A. Sawa, H. Akoh, Y. Kawazoe, and Y. Tokura, Hysteretic current-voltage characteristics and resistance switching at an epitaxial oxide schottky junction SrRuO 3/SrTi0.99Nb0.01O3, Applied Physics Letters86, 10.1063/1.1845598 (2004)

2004 doi
[27]

Emmerich, Y

T. Emmerich, Y. Teng, N. Ronceray, E. Lopriore, R. Chiesa, A. Chernev, V. Artemov, M. Di Ventra, A. Kis, and A. Radenovic, Nanofluidic logic with mechano–ionic memristive switches, Nature Electronics7, 271–278 (2024)

2024
[28]

Shooshtari, T

M. Shooshtari, T. Serrano-Gotarredona, and B. Linares- Barranco, Review of memristors for in-memory computing and spiking neural networks, Advanced Intelligent Systems8, 10.1002/aisy.202500806 (2025)

2025 doi
[29]

D. B. Strukov and R. S. Williams, Exponential ionic drift: fast switching and low volatility of thin-film memristors, Appl. Phys. A94, 515–519 (2009)

2009
[30]

M. J. Rozenberg, M. J. S´ anchez, R. Weht, C. Acha, F. Gomez-Marlasca, and P. Levy, Mechanism for bipolar resistive switching in transition-metal oxides, Phys. Rev. B81, 115101 (2010)

2010
[31]

Ghenzi, M

N. Ghenzi, M. J. S´ anchez, F. Gomez-Marlasca, P. Levy, and M. J. Rozenberg, Hysteresis switching loops in Ag- manganite memristive interfaces, J. Appl. Phys.107, 5 093719 (2010)

2010
[32]

Larentis, F

S. Larentis, F. Nardi, S. Balatti, D. C. Gilmer, and D. Ielmini, Resistive switching by voltage-driven ion migration in bipolar rram—part ii: Modeling, IEEE Trans. Electron Devices59, 2468–2475 (2012)

2012
[33]

S. Kim, S. Choi, and W. Lu, Comprehensive physical model of dynamic resistive switching in an oxide memristor, ACS Nano8, 2369–2376 (2014)

2014
[34]

Marchewka, R

A. Marchewka, R. Waser, and S. Menzel, A 2d axisymmetric dynamic drift-diffusion model for numerical simulation of resistive switching phenomena in metal oxides, in2016 International Conference on Simulation of Semiconductor Processes and Devices (SISPAD)(2016) p. 145–148

2016
[35]

I. V. Boylo, Resistive switching in manganese oxide with nonlinear dependence of the local resistivity on the oxygen- vacancy concentration, Phys. Status Solidi B Basic Res. 254, 1600698 (2017)

2017
[36]

N. V. Agudov, A. V. Safonov, A. V. Krichigin, A. A. Kharcheva, A. A. Dubkov, D. Valenti, D. V. Guseinov, A. I. Belov, A. N. Mikhaylov, A. Carollo, and B. Spagnolo, Nonstationary distributions and relaxation times in a stochastic model of memristor, J. Stat. Mech.2020, 024003 (2020)

2020
[37]

Agudov, A

N. Agudov, A. Dubkov, A. Safonov, A. Krichigin, A. Kharcheva, D. Guseinov, M. Koryazhkina, A. Novikov, V. Shishmakova, I. Antonov, A. Carollo, and B. Spagnolo, Stochastic model of memristor based on the length of conductive region, Chaos, Solitons Fractals150, 111131 (2021)

2021
[38]

I. V. Boylo and K. L. Metlov, Nonlinear effects in memristors with mobile vacancies, R. Soc. Open Sci.8, 210677 (2021), arXiv:2004.04465 [cond-mat.mes-hall]

2021 arXiv
[39]

Бойло and К

И. Бойло and К. Метлов, Переключение закрытого мемристора на подвижных вакансиях в материале с линейной зависимостью удельного сопротивления от их концентрации, ЖЭТФ168, 569–575 (2025)

2025
[40]

Sukhanov, V

A. Sukhanov, V. Valeev, V. Nuzhdin, and R. Khaibullin, Magnetic resonance studies of ar-ion irradiated rutile, Magnetic resonance in solids25, 10.26907/mrsej-23102 (2023)

2023 doi

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [10]

M. J. Rozenberg, M. J. Sánchez, R. Weht, C. Acha, F. Gomez-Marlasca, and P. Levy, Mechanism for bipolar resistive switching in transition-metal oxides, Phys. Rev. B81, 115101 (2010)

work page 2010

[2] [11]

Ghenzi, M

N. Ghenzi, M. J. Sánchez, F. Gomez-Marlasca, P. Levy, and M. J. Rozenberg, Hysteresis switching loops in Ag- manganite memristive interfaces, J. Appl. Phys.107, 093719 (2010)

work page 2010

[3] [12]

Larentis, F

S. Larentis, F. Nardi, S. Balatti, D. C. Gilmer, and D. Ielmini, Resistive switching by voltage-driven ion mi- gration in bipolar rram—part ii: Modeling, IEEE Trans. Electron Devices59, 2468–2475 (2012)

work page 2012

[4] [13]

S. Kim, S. Choi, and W. Lu, Comprehensive physical model of dynamic resistive switching in an oxide memris- tor, ACS Nano8, 2369–2376 (2014)

work page 2014

[5] [14]

Marchewka, R

A. Marchewka, R. Waser, and S. Menzel, A 2d axisymmet- ric dynamic drift-diffusion model for numerical simulation of resistive switching phenomena in metal oxides, in2016 International Conference on Simulation of Semiconductor Processes and Devices (SISPAD)(2016) p. 145–148

work page 2016

[6] [16]

N. V. Agudov, A. V. Safonov, A. V. Krichigin, A. A. Kharcheva, A. A. Dubkov, D. Valenti, D. V. Guseinov, A. I. Belov, A. N. Mikhaylov, A. Carollo, and B. Spag- nolo, Nonstationary distributions and relaxation times in a stochastic model of memristor, J. Stat. Mech.2020, 024003 (2020)

work page 2020

[7] [19]

Boilo and K

I. Boilo and K. Metlov, Switching of a closed mobile vacancy based memristor whose specific resistance lin- early depends on local vacancy concentration, JETP,168, 713–719 (2025)

work page 2025

[8] [21]

Chua, Memristor-the missing circuit element, IEEE Trans

L. Chua, Memristor-the missing circuit element, IEEE Trans. Circuit Theory18, 507–519 (1971)

work page 1971