Nowadays flash memory is one of the most frequently used nonvolatile memories in electronic devices. However, since flash memory is based on Si transistors with floating gates which can store electronic charges, it has basic limitations in its speed and density. It takes longer than 1 lsec for electronic charges to be stored in a floating gate in one cell of flash memory. In addition, we'll reach density limitation in flash memory in the near future by conventional scaling methods, such as decrease in gate length or increase in dielectric constant of the gate oxide, which are commonly applied to Sibased 2-dimensional devices. Thus, in order to overcome the limitations of flash memory, we require a new nonvolatile memory which is not based on Si devices with electronic charge storing phenomena. Here we introduce a next generation nonvolatile memory consisting of two oxide resistors, NiO and VO 2 , where the former is a memory element storing data by utilizing so called bi-stable resistance switching and the latter is a switch element controlling access using the related threshold switching. Since the memory only utilizes resistance switching behaviors of the two oxide resistors, writing and reading times are around several 10s of ns. In addition, it overcomes density limitations by its compatibility with 3-dimensional stack structures due to its low processing temperature lower than 300°C. High performance tests show the feasibility of a universal memory which has advantages of both flash and static random access memories.Si-based flash memory has become the standard for nonvolatile memory which does not lose information in the absence of an external bias. Nonetheless it faces several barriers as cell size is reduced beyond the sub-micrometer region (currently having realized a 40 nm pattern for 32 gigabit NAND flash memory) [1] due to charge leakage across the tunnel oxide. In addition, it needs a little longer time (> 1 ls) to write information by storing charges in a floating gate of flash memory. The efforts of the semiconductor industries have been focused not only on developing scaling methods or modifying device structures for Si-based flash memories [1] but on finding a next generation memory using materials which can circumvent the fundamental limits of Si. The goal of a next generation memory is both to surpass flash memory for nonvolatile memory applications and to realize a universal memory which combines the advantages of nonvolatile slow memory such as flash memory and volatile fast memory such as static random access memory. In order to accomplish this, a class of materials and structures which have easy scalability and rapid programming speed in addition to nonvolatility and low power consumption must be developed. In general, nonvolatile memory consists of a memory element with bi-stable states under zero bias and a switch element with resistance controlled by external bias. The memory element stores the information and the switch element controls access to a specific memory element. Several gro...

For high density of resistive random access memory applications using NiOx films, understanding of the filament formation mechanism that occurred during the application of electric fields is required. We show the structural changes of polycrystalline NiOx (x=1–1.5) film in the set (low resistance), reset (high resistance), and switching failed (irreversible low resistance) states investigated by simultaneous high-resolution transmission electron microscopy and electron energy-loss spectroscopy. We have found that the irreversible low resistance state facilitates further increases of Ni filament channels and Ni filament density that resulted from the grain structure changes in the NiOx film.

A one-bit cell of a general nonvolatile memory consists of a memory element and a switch element. Several memory elements have been tried given that any bistable states, that is, two charging states, two spin states, or two resistance states, can be used for a memory element. On the other hand, silicon-based transistors have been the most popularly used switch element. However, silicon-based transistors do not conform to high-density, nonvolatile memories with three-dimensional (3D) stack structures due to their high processing temperatures and the difficulty of growing high-quality epitaxial silicon over metals. Here, we show a low-temperaturegrown oxide diode, Pt/p-NiO x /n-TiO x /Pt, applied as a switch element for high-density, nonvolatile memories. The diode exhibits good rectifying characteristics at room temperature: a rectifying ratio of 10 5 at ± 3 V, a forward current density of up to ∼ 5×10 3 A cm -2 , an ideality factor of 4.3, and a turn-on voltage of 2 V. Furthermore, we verify its ability to allow and deny access to the Pt/NiO/Pt memory element with two stable resistance states. Under the forward-bias condition, we could access the memory element and change the resistance state, although access was denied under the reverse bias condition. This one-diode/one-resistor (1D/1R) structure could be a promising building block for high-density, nonvolatile random-access memories with 3D stack structures.

Universal memory, which combines the high-speed performance of present-day static random access memory (SRAM) [1] with the non-volatility of Flash [1] must realize several goals such as low operating current, size scalability, and compatibility with mass production to become a feasible memory alternative. The best approach towards the goal of high density is to utilize stackable structures with a crossbar geometry, [2] and to achieve low-temperature fabrication [3] while still retaining a selective switch (transistor or diode) as the data storage element. Thus for high-density applications, crossbar structures are ideal, whereas for non-volatility, resistance-change materials show the best promise. In order to realize the fabrication of universal memory elements, it is imperative to develop a class of materials and structures that combine robust processibility, strong scalability, and rapid programming speed with non-volatility and low power consumption. In our work, we have focused on defining just the storage node portion of the devices, which utilize the resistance change within the film to store information via two different stable resistance states. Here, we have attempted to determine the properties of such structures and to study the mechanisms behind resistance RAM (RRAM) storage. Our Ti-doped (0.1 wt %) NiO samples deposited at room temperature show favorable node characteristics such as the lowest write current reported thus far for a unipolar switching resistance-change-based device (ca. 10 lA). In addition, the programming speed is comparable to the write time of SRAM (10 ns). By combining this node element with an appropriate select switch, such as a high-performance diode, a threshold device, or a two-terminal non-ohmic device, it becomes possible to fabricate high-density universal memory. Indeed, the fabrication of universal memory as the next generation of non-volatile memory is the logical goal for research in this field. In comparison to Flash and dynamic RAM (DRAM), which are the current industry standards, next generation memories must combine the non-volatility of Flash with the high-speed performance of SRAM.[1] Several emerging non-volatile memory architectures have been investigated in order to fabricate materials that fit these specifications. [1,[4][5][6] For example, phase-change RAM (PRAM) [7] utilizes resistance switching accompanied by full or partial phase changes in chalcogenide materials induced by electrical pulses as a method for storing information. Recently, much effort has been devoted to investigations of magnetic race-track memory, a new concept in magnetic non-volatile memory involving the storage of information in the domain walls of materials. [8,9] Also, RRAM [3,[10][11][12][13][14][15][16][17][18] has been studied as a possible candidate for new memory storage devices. RRAM is based on either transition metal oxides that exhibit unipolar switching properties [10][11][12] or perovskite materials displaying bipolar switching properties; [13][14][15] essentially, this is...

We investigated the resistance switching (RS) phenomenon in epitaxial NiO (epi-NiO) films by employing different types of top electrodes (TEs). Epi-NiO showed successive bipolar RS when Pt and CaRuO3 (CRO) were used as the TEs, but not when Al and Ti were used. We studied the temperature dependence of the current–voltage (I–V) characteristics for various TEs and resistance states to understand the conduction properties of TE/epi-NiO. Pristine CRO/epi-NiO showed metallic behavior, while pristine Pt/epi-NiO and Al/epi-NiO showed insulating behavior. Pt/epi-NiO and Al/epi-NiO, however, switched to a metallic or non-insulating state after electroforming. Transmission electron microscopy (TEM) images revealed the presence of a distinct stable interfacial AlO x layer in pristine Al/epi-NiO. On the other hand, the interfacial metal oxide layer was indistinguishable in the case of pristine Pt/epi-NiO and CRO/epi-NiO. Our experimental results suggested that epi-NiO has an oxygen defect on its surface and therefore the various TE/epi-NiO interfaces characterized in this study adopt distinctive electrical states. Further, the bipolar RS phenomenon can be explained by the voltage-polarity-dependent movement of oxygen ions near the interface.