1887
banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Spatiotemporal drift-diffusion simulations of analog ionic memristors
Rent:
Rent this article for
USD
10.1063/1.4815942
/content/aip/journal/jap/114/3/10.1063/1.4815942
http://aip.metastore.ingenta.com/content/aip/journal/jap/114/3/10.1063/1.4815942
View: Figures

Figures

Image of FIG. 1.
FIG. 1.

Spatiotemporal lithium concentration profile for case 1A. The homogeneous lithium concentration initially changes only at the interfaces, eventually approaching a linear profile.

Image of FIG. 2.
FIG. 2.

Spatiotemporal lithium concentration profile for case 1B. The lower initial lithium concentration compared to case 1A leads to decreased steady-state lithium concentrations.

Image of FIG. 3.
FIG. 3.

Spatiotemporal electric field profile evolution for case 1A. The complex electric field evolution arises from the creation/destruction of acceptors at the anode/cathode.

Image of FIG. 4.
FIG. 4.

Spatiotemporal case 1B electric field profile. The electric field is dominated by the creation and destruction of acceptors as lithium is drifted in the device.

Image of FIG. 5.
FIG. 5.

Resistance transient response for the case 1A and 1B simulated memristors. The increased initial lithium concentration of case 1A leads to a faster and larger resistance change.

Image of FIG. 6.
FIG. 6.

Spatiotemporal lithium concentration profile for mobility as a function of acceptor concentration, case 2. Case 1B and case 2 have similar lithium concentration profiles due to minimal hole mobility differences.

Image of FIG. 7.
FIG. 7.

Spatiotemporal electric field profile for case 2. The anodic and cathodic regions of the case 2 simulated memristor are less and more conductive than the case 1B memristor, respectively, due to slight increases in cathodic hole mobility and decreases in anodic hole mobility.

Image of FIG. 8.
FIG. 8.

Spatiotemporal hole mobility profile for case 2. As lithium is drifted in the analog memristor, the hole mobility is increased (decreased) at the cathode (anode). The average hole mobility in the analog memristor increases as time increases.

Image of FIG. 9.
FIG. 9.

Transient resistance response for the case 2 simulated memristor. The steady-state resistance change has slightly increased from case 1B due to the local anode mobility reduction in spite of the overall spatially integrated increase in hole mobility as shown in Fig. 8 .

Image of FIG. 10.
FIG. 10.

Transient resistance response for the case 3 memristors with varied constrictive flow geometries. The resistance response for case 1B is included for comparison. Greater resistance changes are achieved as the top to bottom contact width ratio is increased.

Image of FIG. 11.
FIG. 11.

Spatiotemporal lithium concentration for the 2× geometry memristor. The constrictive flow geometry facilitates higher local cathodic steady-state lithium concentrations than the Case 1B simulated memristor.

Image of FIG. 12.
FIG. 12.

2× geometry memristor spatiotemporal electric field. The inhomogeneous initial electric field profile resulting from current crowding at the cathode magnifies in time as the concentration of lithium vacancies at the cathode decreases.

Image of FIG. 13.
FIG. 13.

Lithium concentration spatiotemporal response for the 5× geometry case. As the contact width ratio is increased to 5×, higher amounts of lithium accumulate at the cathode and deplete from the anode, thus facilitating a greater change in resistance.

Image of FIG. 14.
FIG. 14.

Spatiotemporal electric field progression for the 5× geometry case. Compared to the 2× case, higher electric fields are facilitated as the contact width ratio increases due to greater current crowding and lower concentrations of lithium vacancies (acceptors) at the cathode.

Image of FIG. 15.
FIG. 15.

Spatiotemporal lithium concentration for the 10× geometry case. As the contact width ratio is increased to 10×, the maximum steady-state cathodic lithium concentration and thus resistance change begins to saturate due to volumetric constraints as stoichiometric lithium levels are approached.

Image of FIG. 16.
FIG. 16.

Spatiotemporal electric field for the 10× geometry case. As the contact width ratio is increased to 10×, the maximum electric fields in the device increase due to higher lithium concentrations and magnified current crowding facilitated by the constrictive flow geometry.

Image of FIG. 17.
FIG. 17.

Spatiotemporal lithium concentration profile for constant ion flux, case 4. A non-volatile analog memristive device is simulated by incorporating lithium extraction at the cathode.

Image of FIG. 18.
FIG. 18.

Spatiotemporal electric field profile for case 4. As lithium is extracted from nearly stoichiometric levels, the electric field rapidly changes due to the highly non-linear relationship between lithium concentration and resistivity, Eq. (11) .

Image of FIG. 19.
FIG. 19.

Transient resistance response for the case 4 biased memristor. Ion extraction facilitates over two orders of magnitude resistance change and is the preferred mechanism for analog ionic memristor operation provided the ionic extraction and replacement is repeatable without crystalline damage.

Loading

Article metrics loading...

/content/aip/journal/jap/114/3/10.1063/1.4815942
2013-07-19
2014-04-16
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Spatiotemporal drift-diffusion simulations of analog ionic memristors
http://aip.metastore.ingenta.com/content/aip/journal/jap/114/3/10.1063/1.4815942
10.1063/1.4815942
SEARCH_EXPAND_ITEM