RRAM Materials and Devices

Abstract

Methods for the manufacture of stable strontium titanate nanocube sols are disclosed. The sols are useful in the manufacture of switchable layers suitable for RRAM applications and the switching performance is stable and reproducible. The RRAM layers comprise a mixture of strontium titanate nanocubes and surfactant.

Claims

1. A method for the synthesis of a strontium titanate nanocube dispersion, which method comprises preparing a mixture comprising a source of strontium, a source of titanium, a source of hydroxyl ion, a surfactant and organic solvent, and heating of the mixture under ambient pressure with stirring to a temperature and for a time sufficient to induce hydrolysis and formation of dispersed strontium titanate nanocubes, wherein the surfactant has a molecular weight of greater than 80,000 g.Math.mol.sup.−1.

2. A method as claimed in claim 1, wherein the source of hydroxyl ion is added to the mixture of other components dropwise with stirring before commencement of thermal hydrolysis.

3. A method as claimed in claim 1, wherein the mixture is heated up to a temperature of between 150 and 200° C. over a period of greater than two hours.

4. (canceled)

5. A method as claimed in claim 1, wherein the source of hydroxyl ion is NH.sub.4OH.

6. A method as claimed in claim 1, wherein the source of hydroxyl ion is not organic.

7. (canceled)

8. A method as claimed in claim 3, wherein the mixture is heated up to the desired temperature over a period of three hours or greater.

9. A method as claimed in claim 1 wherein the solvents are oxygenated solvents.

10. A method as claimed in claim 9, wherein the solvent is one or more solvents selected from alcohols, glycol ethers, methyl acetate, ethyl acetate, ketones, esters, and glycol ether/ester derivatives.

11. A method as claimed in claim 9, wherein the solvents are glycols of ethylene or propylene and most preferably ethylene.

12. A method as claimed in claim 11, wherein the solvent is tri ethylene glycol (TEG).

13. A method as claimed in claim 1, wherein the surfactant has a molecular weight of 100,000 g.Math.mol.sup.−l or greater, 200,000 g.Math.mol.sup.−1 or greater, or 300,000 g.Math.mol.sup.−1 or greater.

14. (canceled)

15. (canceled)

16. A method as claimed in claim 1, wherein the surfactant is a polymer or co-polymer prepared from N-vinylpyrrolidone.

17. A method as claimed in claim 16, wherein the polymer is polyvinylpyrrolidone (PVP).

18. A strontium titanate nanocube sol comprising strontium titanate nanocubes, organic solvent and surfactant of a molecular weight of greater than 80,000 g.Math.mol−1.

19. A strontium titanate nanocube sol as prepared by any of the methods of claims 1 to 17.

20. A coating dispersion or ink, which comprises a sol according to claim 18 diluted with one or more solvents.

21. (canceled)

22. A coating dispersion or ink as claimed in claim 20, comprising an alcohol preferably ethanol and/or a glycol derivative.

23. (canceled)

24. (canceled)

25. A method for the manufacture of a strontium titanate nanocube layer, which method comprises depositing a coating dispersion or printing an ink of strontium titanate nanocubes as claimed in claims 22 to 24 onto a suitable substrate and drying of the deposited or printed strontium titanate nanocube layer.

26. A metal oxide layer comprising strontium titanate nanocubes as prepared according to the method of claim 1.

27. A metal oxide layer comprising strontium titanate nanocubes and surfactant residue and/or glycol solvent residue and/or carbonaceous deposits.

28. A metal oxide layer comprising strontium titanate nanocubes and having an HRS/LRS ratio of at least 100.

29. A metal oxide layer comprising strontium titanate nanocubes and having a stable set and reset voltage of 2 volts or less.

30. An electronics device or RRAM comprising one or more metal oxide layers as claimed in claim 27.

31. (canceled)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The present invention will be further illustrated with reference to the drawings, in which:

[0060] FIG. 1 Shows the I-V curve for the device of Example 1, before annealing,

[0061] FIG. 2 Shows the I-V curve for the device of Example 1, after annealing at 450° C. for 1 hour,

[0062] FIG. 3 Shows an SEM image of the device of Example 1, after annealing

[0063] FIG. 4 Shows a high definition SEM image illustrating cracks in the device of Example 1 after annealing,

[0064] FIG. 5 Shows a TEM of the nanocubes isolated from the reaction mixture of Example 1, via water precipitation,

[0065] FIG. 6 Shows the XRD peaks for the materials of Examples 2, 3 and 4,

[0066] FIG. 7 Shows a TEM of the material of Example 4,

[0067] FIG. 8 FIG. 8(a) shows the I-V curve for the device pf Example 4 and FIG. 8(b) shows the retention of switching properties through switching cycles for the device of Example 4,

[0068] FIG. 9. Shows the typical I-V curves for the device of Example 5

[0069] FIG. 10 Shows the switching cycle for the device of Example 5,

[0070] FIG. 11 Shows consecutive switching cycles for the device of Example 5,

[0071] FIG. 12 Shows the endurance performance of the device of Example 5,

[0072] FIG. 13 Shows the complete set of 100 I-V cycles for the device of Example 6,

[0073] FIG. 14 FIGS. 14(a) and 14(b) show the first 10 and last 10 I-V cycles respectively for the device of Example 6,

[0074] FIG. 15 Shows the HRS and LRS values for each cycle for the device of Example 6,

[0075] FIG. 16 Shows the set and reset voltages over 100 cycles for the device of Example 7,

[0076] FIG. 17 Shows the HRS and LRS values over 100 cycles for the device of Example 7.

[0077] FIG. 18 Shows 500 I-V cycles for the device of Example 9, between −3V and 2V with the compliance current in positive sweep of 0.1 mA,

[0078] FIG. 19 Shows the retention test data of the device of Example 9, in ON and OFF status under a bias of 0.1 V (10 s interval, 10 ms holding time),

[0079] FIG. 20 Shows 126 I-V cycles for the device of Example 10, between −3V and 2V with the compliance current in positive sweep of 1 mA, also shown is the first I-V cycle,

[0080] FIG. 21 Shows the retention test data of the device of Example 10, in ON and OFF status under a bias of 0.05 V (10 s interval, 10 ms holding time),

[0081] FIG. 22 Shows 20 I-V cycles for a test point of the device of Example 11,

[0082] FIG. 23 Shows 20 individual I-V cycles for 20 as manufactured cells in the device of Example 11,

[0083] FIG. 24 Shows 220 I-V cycles for a test point of the device of Example 11,

[0084] FIG. 25 Shows the retention test data of the device of Example 11, in ON and OFF status, and

[0085] FIG. 26 Shows 15 I-V cycles for a test point of the device of Example 12.

[0086] The present invention will now be further illustrated by exemplification by means of the following procedures and examples.

Sample Preparation

Bottom Electrode Deposition

[0087] 1. The substrate material used was typically Si/PET/FTO Glass. or other materials such as elastomeric materials and normally the dimensions of substrate are 2-2.5 cm (length)×1.5 cm (width). The used substrate is 1.5 cm (length)×1.5 cm (width).
2. At the first step electrode material e.g. metal (Au) is deposited on a substrate to provide a conductive substrate. A sputter coater (Leica sputter coater) was used to deposit bottom electrode at 40 mA for 3-4 minutes. In addition, the conductive substrates such as FTO glass, ITO coated PET and doped Si can be directly used as bottom electrode, and the electrode deposition process accordingly is avoided.
3. The electrode sputtered sample was then treated under a UV light source for at least 30 minutes prior to deposition of the strontium titanate layer.

Spin Coating:

[0088] 4. A coating suspension solution is prepared using the as prepared strontium titanate sol with solvent and it is spin coated onto the substrate with bottom electrode to fabricate thin films of strontium titanate nanocubes.
5. Typically, 100 μl solution is spin coated over substrate with spin coating speed of 4000 RPM for 40 seconds unless a different staged pinning regime is used.
6. Step 5 may be repeated a number of times to provide the desired layer thickness for the nanocube film.
7. After step 6, the sample was treated in an oven to dry. After treatment at 160° C. for 1 h, no obvious liquid residue is observed in the sample.
8. After step 7 the sample was annealed at 450° C. for 1 hour. The sample was heated at a rate of 5° C./min up to 450° C. and then the furnace was allowed to cool to room temperature.

Top Electrode Deposition

[0089] 1. In order to deposit the top electrode a metal shadow mask is used to deposit circular shaped top electrode (Au/Ag) via Leica sputter coater using same conditions described in steps 2-3 for the bottom electrode. The size of electrodes varies from 50 μm to 250 μm. In addition, inkjet printing may be used to deposit top electrodes by a piezoelectric inkjet printer (Fujifilm Dimatix DMP-2831).
2. The bottom gold electrode typically has a thickness of around 50˜200 nm. The top electrode typically has a thickness of around 50˜300 nm. In relation to the bottom electrode this can be made from a range of different materials including, but not limited to platinum, iridium, silver, gold, copper, ITO, FTO or any combination thereof. Preferred top electrodes are silver and copper
After deposition of the top electrode the device is ready to conduct testing measurements.

Testing Procedures

IV Curves

[0090] 1. For basic memory testing a voltage sweep mode is first run to extract current voltage measurements. Once a discrete transition from a conductive state to another conductive state (could be low or high), depending upon current variation as compared to previous state, is observed this measurement is recorded as a conventional voltage sweep IV curve. (Test-1, T-1)

Endurance Testing

[0091] 1. For the endurance test, the voltages in T-1 are considered and where jumps or declines hi current level of the sample are observed one more sweep is carried out to achieve the first transition point. After achieving this transition point constant voltage pulses are imposed (relatively small voltage to set potential) (0.1-0.5V) and width (0.01 s to 0.00015), for a number of cycles (e.g. 1000 or so). In conjunction to those pulses, the current level of the sample (device) is recorded. This current value Vs. cycles (number of pulses) will provide an endurance plot of a single state [Test-2, T-2].
2. After T-2 is completed one more sweep measurement is run to determine another current transition state which should be different from the state observed in T-2. The endurance measurements are carried out at this second current transition as described in 1. This provides a further current level Vs. cycles for a said number of cycles. [Test-3, T-3].
3. Steps 1 and 2 are repeated if more than two current transitions were observed in voltage sweep measurements (Test-1).
4. After measuring at all current levels for a given number of cycles a plot is made of all the states (current/resistance levels by calculating resistance using ohm's law) Vs. number of cycles and this plot will provide a complete detailed endurance test of a given device for a said number of cycles.

Retention Testing

[0092] 1. To conduct retention test, the same sequence, as described hi points 1-4 above under Endurance Testing is run but the only difference is that the experiment is run in a way such that only one constant read pulse (with defined width) is used and the sample's response is recorded for a period of time (e.g. 4000 seconds), [Test-4, T-4].
2. After having all the current/resistance levels for a given period of time the date is re-plotted in a single graph to show that the device can sustain all states for a given period of time.
Note: The purpose of endurance and retention tests are to observe sample's response in stress atmosphere. By imposing different number of pulses its fatigue test (endurance) is determined and by imposing a constant pulse for a longer period its retention (ability to sustain its data or information for longer time) is determined.

Example 1—Prior Art

[0093] The general procedure as described by Yanan Hao, et. al. (Highly dispersed SrTiO3 nanocubes from rapid sol-precipitation method, Nanoscale, 2014, 6, 7940) was used to prepare a dispersion of strontium titanate nanocubes

[0094] A strontium titanate sol was synthesized as follows:

[0095] To 40 mL of TEG solvent (Triethylene-glycol, 99 wt %, Sigma) was added 16 mmol Sr(OH).sub.28H.sub.2O (95% Sigma), 16 mmol Ti(OBu).sub.4 (97 wt %, Sigma) 8 mL NH.sub.4OH (30%, AR, Chem-supply) and 0.74 g PVP (Mw ˜55,000 g.Math.mol.sup.−1 Aldrich Code 56568) to form a mixture. The obtained mixture was heated to 160° C., over 30 mins and maintained at that temperature for 2 hours with stirring. After 2 hours the reaction mixture was allowed to cool under ambient conditions before being used further.

[0096] A coating dispersion was prepared by taking the reaction mixture and diluting it with ethanol in a weight ratio of 1:1 solvent:reaction mixture.

[0097] A device was manufactured by spin coating a 100 ul coating dispersion solution onto an FTO substrate at 4000 rpm for 40 seconds. After annealing at 160° C. for 1 h or 450° C. for 1 h (5° C./min to 450° C. and furnace cooled to room temperature), silver electrodes were printed onto the sample by inkjet printing. The device was further dried at 150° C. for 3 h.

[0098] The device was analyzed and also tested for its switching performance and these results are as indicated in FIGS. 1, 2, 3, 4 and 5.

[0099] FIG. 1 shows the I-V curve for the device before annealing. This demonstrates a very weak switching performance. The device can hardly be switched during the positive sweep. FIG. 2 shows the I-V curve after annealing at 450° C. for 1 hour. The switching profile has completely changed and shows a weak switching performance. The on state is difficult to be maintained.

[0100] FIG. 3 shows the SEM image of the film after annealing. There are obvious significant variations in the film surface, which is not smooth and is cracked. One of these cracks is shown in FIG. 4.

[0101] FIG. 5 is a TEM of the nanocubes isolated from the reaction mixture via water precipitation. The particles exhibit a highly variable and nonuniform morphology in which a large amount of organic material is observed.

[0102] These result show that the reaction mixture sols and materials prepared according to this Nanoscale reference do not provide suitable materials for RRAM use.

Example 2

[0103] All starting materials were purchased from Sigma and used without further purification. A strontium titanate nanocube dispersion was prepared as follows:

[0104] A 50 ml polytetrafluoroethylene reaction vessel was prepared with a polytetrafluoroethylene stirrer (500˜1300 rpm). To this reaction vessel was introduced 20 mL of triethylene glycol (TEG). Then, 0.37 g of polyvinylpyrrolidone (PVP Mw 360,000 g.Math.mol.sup.−1) was added to the TEG solution and was dissolved at 85° C. under stirring after about 30 min. Then, 8 mmol strontium hydroxide hexahydrate was added and dissolved at 85° C. under stirring after about 15 min. Then, 8 mmol Ti(OBu).sub.4 was added at 85° C. under stirring after about 10 min. Then, 5 ml of NH.sub.4OH was injected drop by drop with a 10 ml syringe. After the addition of the NH.sub.4OH was completed the reaction temperature was increased from 85° C. to 160° C. After reaction at 160° C. for 4 h, the vessel was taken out from the heater and cooled to room temperature.

[0105] The resulting reaction mixture (sol) was light yellow in colour and transparent.

[0106] Strontium titanate nanocubes were isolated from the sol by precipitation with water at a ratio of water:reaction mixture of 30:1 and separated from the supernatant by high speed centrifuge. The precipitated nanocubes were washed with water, and then washed with water and ethanol for 5 times.

[0107] The key XRD peaks for the material are shown in FIG. 6. The XRD spectrum shows good crystallinity of STO phase in these nanocubes. TEM shows that the product is composed of uniform nanocubes.

Example 3—Lanthanum Doped

[0108] Example 2 was repeated for the preparation of a lanthanum doped strontium titanate sol and isolated nanocubes.

[0109] All materials and conditions were identical apart from the following: 25 mL of triethylene glycol (TEG) was used as solvent. 1.701 g (6.4 mmol) of strontium hydroxide hexahydrate (Sr(OH).sub.2.8H.sub.2O) was added. 0.693 g (1.6 mmol) of lanthanum(III) nitrate hexahydrate (La(NO.sub.3).sub.3.6H.sub.2O was added before the Ti(OBu).sub.4.

[0110] The XRD data for this sample is shown in FIG. 6. The XRD data indicates that no second phase is found in doped products.

Example 4—Iron Doped

[0111] Example 3 was repeated for the preparation of an iron doped strontium titanate sol and isolated nanocubes.

[0112] All materials and conditions were identical apart from the following: The lanthanum dopant was replaced with 0.647 g (1.6 mmol) of iron(III) nitrate nonahydrate (Fe(NO.sub.3).sub.3.9H.sub.2O).

[0113] The XRD for this material is shown in FIG. 6. In addition, the TEM for this material is shown in FIG. 7.

[0114] This material was tested for its switching behavior by manufacture of a device.

[0115] Washed iron doped strontium titanate nanocubes (0.25 g) were placed into a centrifuge tube and a total of 3 mL solvents and agents (2% Oleic Acid, as agent, which can avoid the agglomeration, and toluene or ethanol as solvents).

[0116] The lips of centrifuge tubes were tightened, and these were then placed in a beaker half filed with water and sonicated with ultrasonic for 10 mins.

[0117] The samples were then centrifuged for 4 mins at 16000 rpm the supernatant layer was collected into labelled bottles. All solutions were sonicated by ultrasonic for 10 mins before coating.

[0118] For coating 20 μL of each solution was dropped onto the centre of a substrate surface. This layer was allowed to dry completely and then a further layer was deposited. A total of three layers were deposited.

[0119] Top Au electrodes were sputtered in the gold-sputtering device, covered by a 10 mm×10 mm round patterning (diameter=50 μm) shadow mask.

[0120] FIG. 8(a) shows the I-V curve for this device. FIG. 8(b) shows the retention of switching properties through switching cycles. The device shows stable ON/OFF ratios.

Example 5

[0121] A strontium titanate sol was prepared as Example 2.

[0122] This sol was diluted at a mass ratio of 1:1 with ethanol to produce a dispersion for spin coating. This coating dispersion (100 μl) was deposited by spin coating at 4000 rpm for 30 seconds onto FTO substrate. The substrate with the spin coated strontium titanate nanocubes was transferred to an oven and heated at 160° C. for 1 h, when no obvious liquid residue is observed in the sample. The dried strontium titanate film was then annealed by heating at 5° C./min to 450° C. and furnace cooled to room temperature.

[0123] Top electrodes of silver were deposited by inkjet printing (Fujifilm Dimatix DMP-2831) as a uniform array pattern of 400 electrode points each with a diameter of about 50 μm. The resultant device was further dried at 150° C. for 3 hours.

[0124] The sample was tested for its switching performance. The results are shown in FIGS. 9 to 12.

[0125] With reference to FIG. 9 the typical I-V curves for this device are shown and there is 100% performance; this means all 40 of the tested points were switchable. Small loops can be observed in the negative voltage and the device is relatively difficult to switch back to a high resistance state.

[0126] With reference to FIG. 10 the device can be switched on and off by applying a voltage following the sequence 0 V.fwdarw.2 V.fwdarw.0 V.fwdarw.-8 V.fwdarw.0 V. A higher negative current is normally required to switch the device back to high resistance state using negative voltage.

[0127] To test the repeatability of the device, the same point on the device was successfully measured for more than 120 consecutive cycles, which can be seen in FIG. 11. The set voltage is about 1.5 to 2 V and the reset voltage is−2 to−5 V.

[0128] The endurance of the device is shown in FIG. 12 and is based in the resistances calculated by Ohm's law at 0.1 V for both high and low resistance states. As shown in the Figure, the Ron are almost kept constant with only a small variation of Roff is observed. The device exhibits and on/off ratio of >100.

[0129] The device of this example is highly stable, and the majority of the switch points can be switched on by applying a voltage of between 1-2 V. The device can be switched off by using higher negative voltages.

Example 6

[0130] Sample Reference—01(FTO)E3

[0131] A sample of strontium titanate nanocube dispersion was prepared and tested as follows:

[0132] Device Fabrication Processes

[0133] A. Substrate Preparation

1. In this example, a conductive substrate, FTO coated glass (25.4 mm (length)×25.4 mm (width)) was used and because of its properties FTO can be directly used as he bottom electrode, and further bottom electrode deposition process is accordingly avoided.
2. The substrate was ultrasonically cleaned with a standard regiment of de-ionised water with 5% concentration of deconex FPD-211 (Borer Chemie AG) and rinsed twice with de-ionised water, isopropyl alcohol (Aldrich), de-ionised water and isopropyl alcohol followed by drying with nitrogen gas (Entergris Wafergard GN Gas Filter Gun) until surface is visually dry. A UV-ozone treatment (Novascan) for 15 minutes was conducted prior to the spin coating.

[0134] B. Spin Coating

1. A coating suspension solution was prepared using the as prepared strontium titanate sol of Example 2. A strontium titanate solution (1:1 w:w in ethanol) was used to fabricate a RRAM device.
2. The strontium titanate solution was sonicated for 40 minutes prior to use.
3. After sonication, the strontium titanate solution was placed into a syringe. The solution (appox 0.5 mL) was dispensed from the syringe, without filtering, in an ‘S’ sequence onto a stationary substrate (FTO). The dispensed solution was left standing to wet the substrate for 60 s before initiating a spin-coat sequence using a Laurell/Model: WAS-650.
4. The spin coating parameters were:

i) Ramp=300 rpm for 5 s;

ii) Spin=4000 rpm for 30 s;

[0135] iii) Slow=1000 rpm for 10 s.
5. Prior to drying of the strontium titanate coated substrate on a hotplate, the edges of the substrate were carefully wiped with a cleanroom tissue (BEMCOT S-2) to expose the bottom transparent conducting oxide (TCO) electrode.
6. The strontium titanate coated substrate was then transferred to a hotplate for drying at 160° C. for approximately 10 min duration (Hotplate model: IKA HCT).
7. The dried strontium titanate coated substrate is then annealed. It was transferred to a programmable enclosed high-temperature hotplate (Detlef Gestigkeit Electritechnik/Model: PR 5-3T) for additional annealing at 450° C. for 60 min. The temperature ramp rate is 5° C. per min (equates to approximately 90 min. to raise the temperature from room temperature to the target temperature of 450° C. Total annealing time sequence exceeds 4 hours). The process in full requires approximately 90 minutes to reach the target temperature of 450° C., the sample is held at this temperature for a further 60 minutes and then is allowed to cool from this annealing temperature to ambient temperature or handling temperature within the furnace and this takes a minimum of 90 minutes.
8. The thickness for the strontium titanate coating on the FTO substrate was analysed by surface profilometry. The strontium titanate film thickness is consistent with an approximate thickness of between 80-90 nm.

[0136] C. Top Electrode Deposition

1. In order to deposit the square shaped top electrodes, a silver (Ag) target was accumulated atop the annealed SrTiOx layer by physical vapour deposition (PVD) in a Kitano Seiki KVD OLED Evo II system (base pressure of 10{circumflex over ( )}−6 Pa) using a stainless steel shadow mask to define the device area (0.1 mm×0.1 mm. A total of 36 sites (in a 6 by 6 array) were produced on the substrate.
2. After deposition of the top electrodes, the device is ready to conduct testing measurements.

[0137] Testing Procedures

IV Curves

[0138] 1. For basic memory testing a voltage sweep mode is first run to extract current voltage measurements. A Semiconductor Parameter Analyser (Agilent B1500A) was used to test the electrical properties of the RRAM device. Once, a discrete transition from a conductive state to another conductive state (could be low or high), depending upon current variation as compared to previous state, is observed this measurement is recorded as a conventional voltage sweep IV curve. (Test-1, T-1)

Endurance Testing

[0139] 9. For the endurance test, the voltages in T-1 are considered and where jumps or declines in current level of the sample are observed one more sweep is carried out to achieve the first transition point. After achieving this transition point constant voltage pulses are imposed (relatively small voltage to set potential) (0.1-0.5V) and width (0.01 s to 0.0001 s), for a number of cycles (e.g. 100 or so). In conjunction to those pulses, the current level of the sample (device) is recorded. This current value Vs. cycles (number of pulses) will provide an endurance plot of a single state [Test-2, T-2].
10. After T-2 is completed one more sweep measurement is run to determine another current transition state which should be different from the state observed in T-2. The endurance measurements are carried out at this second current transition as described in 1. This provides a further current level Vs. cycles for a said number of cycles. [Test-3, T-3].
11. Steps 1 and 2 are repeated if more than two current transitions were observed in voltage sweep measurements (Test-1).
12. After measuring at all current levels for a given number of cycles a plot is made of all the states (current/resistance levels by calculating resistance using ohm's law) Vs. number of cycles and this plot will provide a complete detailed endurance test of a given device for a said number of cycles.
13. The device was tested through 100 I-V cycles of a device. The results are shown in FIGS. 13 to 15. FIG. 13 shows the complete set of 100 cycles, whereas FIGS. 14(a) and 14(b) show the first 10 and last 10 I-V cycles respectively. This demonstrates a degree of stability and reproducibility for this device. FIG. 15 shows the extracted HRS and LRS for each cycle. The data for these tests is summarized in Table 1 below:

TABLE-US-00001 TABLE 1 Statistic Counts 3.sup.rd Q. Median 1.sup.st Q. Cycles 100 V.sub.SET (V) 100 1.10 1.05 1.00 V.sub.RESET (V) 98 −2.04 −2.13 −2.20 I.sub.HRS (A) @ 0.5 V 100 2.407E−04 2.305E−04 2.048E0.4 I.sub.LRS (A) @ 0.5 V 98 6.431E−04 5.590E−04 4.153E−04 r.sub.OFF/ON @ 0.5 V 98 3.2 2.5 1.8

Example 7

[0140] Sample References—B001 to B004

[0141] A sample of strontium titanate nanocube dispersion was prepared as per Example 2 and tested using the same general procedures as described for Example 6.

[0142] The only differences were that the substrate was a bottom electrode of 140 nm ITO on glass and the middle spin stage for the spin coating was varied to show the impact of coating thickness on performance.

[0143] The key differences for these samples are indicated in Table 2:

TABLE-US-00002 TABLE 2 Bottom Drying Annealing Top Device Electrode Solution RPM Temp Temp Electrode B001 140 nm ITO SrTiO.sub.3 4000 160° C. 450° C. 200 nm Ag B002 140 nm ITO SrTiO.sub.3 3000 160° C. 450° C. 200 nm Ag B003 140 nm ITO SrTiO.sub.3 2000 160° C. 450° C. 200 nm Ag B004 140 nm ITO SrTiO.sub.3 1000 160° C. 450° C 200 nm Ag

[0144] These samples were evaluated for their I-V performance and the key results are provided in Table 3:

TABLE-US-00003 TABLE 3 Film Thickness Tested Set Reset HRS LRS HRS/ Device RPM nm Sites [V] [V] [Ω] [Ω] LRS B001 4000 80 8 0.70 −0.85   8.7 K 590 ≈10 B002 3000 100 8 0.90 −0.88   9.5 K 680 ≈10 B003 2000 115 12 0.95 −0.63  92.5 K 580 ≈160 B004 1000 150 12 1.00 −0.65 111.9 K 550 ≈200

[0145] The reproducibility of the switching characteristics were evaluated over 100 cycles and the performance is shown in Table 4 and FIGS. 16 and 17 for B003 the preferred device.

TABLE-US-00004 TABLE 4 Statistic Counts Max. 3.sup.rd Quartile Mean Median 1.sup.st Quartile Min. Cycles 100 V.sub.SET (V) 100 1.00 0.95 0.89 0.90 0.85 0.70 V.sub.RESET (V) 100 −0.20 −0.40 −0.78 −0.55 −0.76 −2.50 HRS (Ω) @ 0.2 V 100 1.143E+07 3.768E+05 4.106E+05 1.176E+05 5.576E+04 9.950E+03 LRS (Ω) @ 0.2 V 100 2.376E+03 3.825E+02 3.824E+02 2.642E+02 2.349E+02 2.068E+02 r.sub.OFF/ON @ 0.2 V 100 47144.9 1501.2 1608.7 468.3 161.6 4.5

[0146] FIG. 16 shows the set and reset voltages over 100 cycles; these are relatively stable and reproducible. FIG. 17 shows the HRS and LRS stability over 100 cycles and these are both relatively stable and reproducible.

[0147] These results show that devices manufactured from this particular sol and this method of manufacture, when deposited at an RPM of 2000 or less, have very good switching performances and stability. For this particular type of sol, the thickness under these deposition conditions is greater than 100 nm. These devices with thicker layers of strontium titanate produce good quality switchable layers that have a high HRS/LRS ratio, which is desirable and is an order of magnitude greater than layers of thickness less than 100 nm. In addition, this is maintained through 100 switching cycles. The preferred switchable layers of the present invention have a thickness of at least 100 nm and an HRS/LRS ratio of greater than 100 and have low set and reset voltages, which is highly desirable.

Example 8—Nanocube Sol Preparation

[0148] A 50 ml polytetrafluoroethylene reaction vessel was prepared with a polytetrafluoroethylene stirrer (500˜1300 rpm). To this reaction vessel was introduced 20 mL of triethylene glycol (TEG).

[0149] Then, 0.37 g of polyvinylpyrrolidone (PVP Mw ˜360,000 g.sup.−1 Sigma-Aldrich) was added to the TEG solution at room temperature and then this was dissolved at 85° C. under stirring after about 30 min.

[0150] Then, 8 mmol strontium hydroxide hexahydrate was added and dissolved at 85° C. under stirring after about 15 min.

[0151] Then, 8 mmol Ti(OBu).sub.4 was added at 85° C. under stirring after about 10 min.

[0152] Then, 5 ml of NH.sub.4OH was injected drop by drop with a 10 ml syringe.

[0153] After the addition of the NH.sub.4OH was completed the reaction temperature was increased from 85° C. to 160° C. After reaction at 160° C. for 6 h, the vessel was taken out from the heater and cooled to room temperature.

[0154] The resulting reaction mixture (sol) was light yellow in colour and transparent.

Example 9— Strontium Titanate Nanocube RRAM on Silicon

[0155] A RRAM device was prepared using the sol of Example 8.

[0156] The sol of Example 8 was diluted with ethanol at a mass ratio of 1:1.

[0157] Before the spin-coat process, a Si substrate (1.5 cm×1.5 cm) coated with SiO.sub.2 (thickness 200±10 nm) (Zhejiang Lijing Optoelectronic Technology Co., Ltd) was cleaned with deionized water and ethanol, and then treated with ultraviolet radiation for 20 min.

[0158] The cleaned substrate was deposited wit ˜100 nm Au lines (Sputter coater Leica EM ACE600) to be used as the bottom electrode.

[0159] Then 100 μl of the diluted solution was dropped onto the substrate with a 1-2 min delay before the spin coating process.

[0160] Then the solution was spin coated at the speed of 4000 rmp for 30 s.

[0161] Once spun down the thin film coated substrate was immediately transferred to an oven and heated at 100° C. for 30 min. The think film coated substrate was not subjected to a high temperature annealing stage.

[0162] Then a top surface silver line electrode with 3 layers was deposited by inkjet printing (Fujifilm Dimatix DMP-2800 inkjet printer) and further dried at 80° C. for 12 hours. Every silver layer was overlapped.

[0163] The RRAM performance of this device was evaluated, and the results are shown in FIGS. 18 and 19. In the 500 measured I-V cycles, the device showed a narrow set (0˜1V) and reset voltage (0˜2V). In addition, a good retention performance with a stable on and off resistance for 10000 s is demonstrated.

Example 10—Strontium Titanate Nanocube RRAM on ITO

[0164] A RRAM device was prepared using the sol of Example 8.

[0165] Before the spin-coat process, an ITO coated glass substrate (2.5 cm×2.5 cm) was cleaned with deionized water and ethanol, and then treated with ultraviolet radiation for 30 min.

[0166] Then 100 μl of the diluted solution was dropped onto the substrate with a 1-2 min delay before the spin coating process.

[0167] Then the solution was spin coated at the speed of 4000 rmp for 30 s.#

[0168] Once spun down the thin film coated substrate was immediately transferred to an oven and heated at 100° C. for 30 min. The think film coated substrate was not subjected to a high temperature annealing stage.

[0169] Ethanol was used to carefully wipe the STO film at one corner of the sample in order to expose the bottom electrode.

[0170] Then a top surface silver line electrode with 3 layers was deposited by inkjet printing (Fujifilm Dimatix DMP-2800 inkjet printer) and further dried at 80° C. for 12 hours. Every silver layer was overlapped.

[0171] The RRAM performance of this device was evaluated and the results are shown in FIGS. 20 and 21. In the measured 126 I-V cycles, the device can be set within 1 V and reset within−3V. Additionally, the device showed nearly constant resistance for 7000 s.

Example 11—Strontium Titanate Nanocubes—Slot Die FTO Glass

[0172] The sol of Example 8 was diluted with ethanol at a mass ratio of 1:1

[0173] An FTO glass was cleaned with deionized water and ethanol, and then treated with ultraviolet radiation for 30 min.

[0174] Then the diluted solution was printed onto the clean FTO glass substrate (5 cm×5 cm) by slot-die printing. The coating speed was 15 mm/s, and the dispense rate was 10 ul/s.

[0175] After the coating was deposited it the thin film coated substrate was immediately transferred the to an oven at 100° C. for 30 min.

[0176] Ethanol was used to carefully wipe the STO film at one corner of the sample in order to expose the bottom electrode.

[0177] Then top silver electrodes with 2 and 3 layers were deposited by inkjet printing and further dried at 80° C. for 4 hours. Every silver layer was overlapped.

[0178] The RRAM performance of this device was evaluated and is shown in FIGS. 22, 23, 24 and 25. The device could be switched with voltages greater than 2 V. The device was manufactured with 20 cells and it was found that all 20 cells were operational and could be switched, which indicated a yield of 100% on manufacture.

Example 12— Strontium Titanate Nanocubes—Slot Die ITO PET Flexible Substrate

[0179] The sol of Example 8 was diluted with ethanol at a mass ratio of 1:1.

[0180] An ITO coated flexible PET substrate was cleaned with deionized water and ethanol, and then treated with ultraviolet radiation for 30 min.

[0181] Then the diluted solution was printed onto the clean ITO coated flexible PET substrate (5 cm×5 cm) by slot-die printing. The coating speed was 15 mm/s, and the dispense rate was 10 ul/s.

[0182] After the coating was deposited it the thin film coated substrate was immediately transferred the to an oven at 100° C. for 30 min.

[0183] Ethanol was used to carefully wipe the STO film at one corner of the sample in order to expose the bottom electrode.

[0184] Then top silver electrodes with 2 and 3 layers were deposited by inkjet printing and further dried at 80° C. for 4 hours. Every silver layer was overlapped.

[0185] The RRAM performance of this device was evaluated and is shown in FIG. 26.

[0186] All of the features disclosed in this specification for each and every embodiment (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive.

[0187] Throughout the description and claims of this specification, the words “comprise” and “contain” and variations of the words, for example “comprising” and “comprises”, means “including but not limited to”, and is not intended to (and does not) exclude other components, integers or steps.

[0188] Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise. Features, integers, characteristics, compounds described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.