Tracking oxygen vacancy migration in memristor devices using operando hard X-ray photoelectron spectroscopy
Fabrication of ultrathin (sub-2 nm) oxide semiconductor memristors poses the fundamental challenge of achieving oxide growth with atomic precision in terms of electronic structure and defect formation. Recently, ultrathin memristors consisting of bilayers of mixed [Formula: see text] [Formula: see text] and MgO atomic layers were fabricated using an in vacuo atomic layer deposition process. This approach offers a unique platform for precise atomic control of oxygen vacancies in the device in which the vacancies are introduced by placing MgO atomic layers between pristine [Formula: see text] [Formula: see text] layers. In this work, we present a systematic operando Hard X-ray PhotoElectron Spectroscopy (HAXPES) study of the switching of such memristors, combined with complementary current-voltage and capacitance-voltage (C-V) measurements. We used a memristor stack of [Formula: see text] [Formula: see text]/MgO atomic layers, with the MgO-containing oxide deposited on the bottom Al metal electrode and a pure [Formula: see text] [Formula: see text] layer below the top Pd electrode. HAXPES analysis shows a substantial change in the chemical shift of the Aluminum oxide when switching between the "OFF" and "ON" states indicative of a redistribution of oxygen vacancies in the device active layer. Interestingly, subsequent switching to the OFF state shows hysteretic behavior indicating the retention of some oxygen vacancies in the top [Formula: see text] [Formula: see text] layer. This vacancy retention can be correlated with the stochastic behavior of the switching voltage observed in these devices. C-V measurements show a clear frequency-dependent response in the OFF state, consistent with enhanced polarization and vacancy trapping at low frequencies.
Authors
- Sandeep Miryala (ORCID: https://orcid.org/0000-0003-1277-1745)
- J. C. Woicik (ORCID: https://orcid.org/0000-0002-2167-2186)
- Ian Harding (ORCID: https://orcid.org/0000-0003-4917-2834)
- Abdul K. Rumaiz (ORCID: https://orcid.org/0000-0002-9907-299X)
- Conan Weiland (ORCID: https://orcid.org/0000-0001-6808-1941)
- Mohamed Boukhicha (ORCID: https://orcid.org/0000-0002-8866-5714)
- G. Carini
- Ryan Goul (ORCID: https://orcid.org/0000-0003-0386-9986)
- Jiacong Wu (ORCID: https://orcid.org/0009-0000-6837-5608)
- D. P. Siddons (ORCID: https://orcid.org/0000-0003-1482-9867)
- Supriya Karmakar (ORCID: https://orcid.org/0000-0002-5797-7141)
- Arif Ozbay (ORCID: https://orcid.org/0000-0003-3734-1952)
- A. Kuczewski (ORCID: https://orcid.org/0009-0003-0566-5298)
- A. Marshall
- Francesca Capocasa (ORCID: https://orcid.org/0009-0004-4403-1797)
Institutions
- National Institute of Standards and Technology (US)
- University of Kansas (US)
- Farmingdale State College (US)
- Brookhaven National Laboratory (US)
- Tübitak National Metrology Institute (TR)
Publication Details
- Journal
- Scientific Reports
- Published
- 2025-12-02
- DOI
- https://doi.org/10.1038/s41598-025-30246-4
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00