Chalcogenide glass constant current device, and its method of fabrication and operation

Abstract

The invention is related to methods and apparatus for providing a two-terminal constant current device, and its operation thereof. The invention provides a constant current device that maintains a constant current over an applied voltage range of at least approximately 700 mV. The invention also provides a method of changing and resetting the constant current value in a constant current device by either applying a positive potential to decrease the constant current value, or by applying a voltage more negative than the existing constant current's voltage upper limit, thereby resetting or increasing its constant current level to its original fabricated value. The invention further provides a method of forming and converting a memory device into a constant current device. The invention also provides a method for using a constant current device as an analog memory device.

Description

FIELD OF THE INVENTION

The present invention relates to a constant current device, and more specifically, to a constant current device formed of chalcogenide materials.

BACKGROUND OF THE INVENTION

Programmable resistive materials based on chalcogenide glass are being explored for use as non-volatile memory elements. By applying an external stimulus, such as different voltages of selected polarities to a chalcogenide glass in the presence of an available metal such as silver, the internal structure of the chalcogenide glass can be modified to produce high or low resistance states.

One specific example of a chalcogenide glass being investigated for memory use is germanium-selenide (Ge_xSe_100−x). Typically, the chalcogenide glass has an associated layer for supplying a metal, which becomes incorporated with the glass matrix to change resistance states. As examples, the associated layer may be a layer of silver or a layer of silver-selenide (Ag₂Se).

While the current focus is on using chalcogenide glass for memory devices, the inventors have discovered another use for chalcogenide glass, namely, as a constant current device.

BRIEF SUMMARY OF THE INVENTION

In one aspect, the invention provides a two-terminal constant current device formed of a chalcogenide glass material, and its method of formation and operation. The device comprises a metal-containing layer formed adjacent at least one chalcogenide glass layer which is biased into a constant current state. The constant current device maintains a constant current over a range of applied voltages.

In another aspect, the invention provides a constant current device and a method of forming and operating such a device in which at least one layer of silver-selenide is formed between a first germanium-selenide layer and a second germanium-selenide layer. These layers are provided between a first and a second electrode. A bias voltage is applied to the electrodes sufficient to place the device in a constant current state. The constant current device maintains a constant current over a range of applied voltages.

In another aspect, the invention provides a constant current device and a method of forming and operating such a device in which at least one layer of silver-selenide is formed between a first germanium-selenide layer, a layer of silver, and a second germanium-selenide layer. These layers are provided between a first and a second electrode. A bias voltage is applied to the electrodes sufficient to place the device in a constant current state. The constant current device maintains a constant current over a range of applied voltages.

In another aspect, the invention provides a constant current device and a method of forming and operating such a device in which at least one metal-containing layer, such as silver, is formed on a chalcogenide glass layer such as a germanium-selenide layer. A bias voltage is applied to the layers sufficient to place the device in a constant current state. The constant current device maintains a constant current over a range of applied voltages.

In another aspect, the invention provides a constant current device and a method of forming and operating such a device in which at least one metal-containing layer, such as silver-selenide, is formed with a chalcogenide glass layer such as a germanium-selenide layer, and a layer of silver. A bias voltage is applied to these layers sufficient to place the device in a constant current state. The constant current device maintains a constant current over a range of applied voltages.

In another aspect, the invention provides a method of converting a device that has been exhibiting memory behavior comprised of at least one chalcogenide glass layer and a metal-containing layer, for example, of silver or silver-selenide to a constant current device. A bias voltage is applied to the layers sufficient to place the device in a constant current state. The constant current device maintains a constant current over a range of applied voltages.

In another aspect, the invention provides a method of altering the current characteristics of a constant current device formed with at least one chalcogenide glass layer by manipulation of an applied bias voltage.

In another aspect, the invention provides a method of resetting or raising the current characteristics of a constant current device formed with at least one chalcogenide glass layer to a prior state by manipulation of an applied bias voltage.

These and other features and advantages of the invention will be better understood from the following detailed description, which is provided in connection with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates a cross-sectional view of a chalcogenide glass constant current device fabricated in accordance with a first embodiment of the invention.

FIG. 1 a illustrates a cross-sectional view of a chalcogenide glass constant current device fabricated in accordance with a first exemplary embodiment of the first embodiment of the invention.

FIG. 1 b illustrates a cross-sectional view of a chalcogenide glass constant current device fabricated in accordance with a second exemplary embodiment of the first embodiment of the invention.

FIG. 1 c illustrates a cross-sectional view of a chalcogenide glass constant current device fabricated in accordance with a third exemplary embodiment of the first embodiment of the invention.

FIG. 1 d illustrates a cross-sectional view of a chalcogenide glass constant current device fabricated in accordance with yet another exemplary embodiment of the first embodiment of the invention

FIG. 2 illustrates a cross-sectional view of a chalcogenide glass constant current device fabricated in accordance with a second embodiment of the invention.

FIG. 2 a illustrates a cross-sectional view of a chalcogenide glass constant current device fabricated in accordance with a first exemplary embodiment of the second embodiment of the invention.

FIG. 2 b illustrates a cross-sectional view of a chalcogenide glass constant current device fabricated in accordance with a second exemplary embodiment of the second embodiment of the invention.

FIG. 3 is a graph illustrating a current-voltage (I-V) curve of a chalcogenide constant current device constructed in accordance with the invention.

FIG. 4 is a graph illustrating the application of different positive voltages to decrease a constant current device's constant current amplitude.

FIG. 5 is a graph illustrating the constant current achieved as a result of applying the voltage illustrated in FIG. 4 to a chalcogenide constant current device.

FIG. 6 is a graph illustrating the breakdown voltage of a chalcogenide constant current device.

FIG. 7 is a graph illustrating the I-V characteristics of a chalcogenide constant current device after a bias voltage described with reference to FIGS. 5-6 is applied.

FIG. 8 is a graph illustrating the original positive voltages applied to a chalcogenide constant current device and the resulting constant current I-V characteristics.

FIG. 9 illustrates a computer system incorporating a chalcogenide constant current device formed in accordance with the invention.

DETAILED DESCRIPTION OF THE INVENTION

In the following detailed description, reference is made to various specific embodiments of the invention. These embodiments are described with sufficient detail to enable those skilled in the art to practice the invention. It is to be understood that other embodiments may be employed, and that various structural, logical, and electrical changes may be made without departing from the spirit or scope of the invention.

The term “substrate” used in the following description may include any supporting structure including but not limited to a glass, plastic, or semiconductor substrate that has an exposed substrate surface. A semiconductor substrate should be understood to include silicon, silicon-on-insulator (SOI), silicon-on-sapphire (SOS), doped and undoped semiconductors, epitaxial layers of silicon supported by a base semiconductor foundation, and other semiconductor structures which may not be silicon-based. When reference is made to a semiconductor substrate in the following description, previous process steps may have been utilized to form regions or junctions in and/or over the base semiconductor or foundation.

The term “silver” is intended to include not only elemental silver, but silver with other trace metals or in various alloyed combinations with other metals as is known in the semiconductor industry, as long as such silver alloy is conductive, and as long as the physical and electrical properties of the silver remain unchanged.

The term “silver-selenide” is intended to include various species of silver-selenide, including some species which may have a slight excess or deficit of silver, for instance, Ag₂Se, Ag_2+xSe, and Ag_2−xSe.

The term “chalcogenide glass” is intended to include glasses that comprise an element from group VIA (or group 16) of the periodic table. Group VIA elements, also referred to as chalcogens, include sulfur (S), selenium (Se), tellurium (Te), polonium (Po), and oxygen (O).

The present invention is directed toward chalcogenide glass constant current devices and their methods of formation and operation. The chalcogenide constant current structures disclosed herein can be utilized for any number of applications where a constant current is needed over a range of applied voltages.

The invention will now be explained with reference to FIGS. 1-9, which illustrate exemplary embodiments of a chalcogenide constant current device 100 and 101, and its method of formation and operation, in accordance with the invention.

FIG. 1 depicts a first embodiment of a chalcogenide constant current device constructed in accordance with the invention. A first electrode 2 is formed over a substrate 1. The first electrode 2 may comprise any conductive material so long as the constant current device's electrical properties are not altered. For example, various metals, such as, one or more of tungsten, nickel, tantalum, aluminum, platinum, or titanium nitride among many others. In addition, the first electrode 2 can comprise a conductively-doped semiconductive material. However, care must be used since some metals, such as Ag, Au, or Cu, may migrate into a subsequently deposited glass layer and alter the electrical behavior of the constant current device.

For purposes of a simplified description, the first electrode 2 is described as comprising tungsten (W). Although FIG. 1 illustrates a first electrode 2 provided on substrate 1, it should be appreciated that additional layers may be provided beneath electrode 2, between it and the substrate 1. For instance, a semiconductor substrate containing circuit layers covered with an insulating layer can be provided below first electrode 2 if desired. The presence of additional underlying layers between electrode 2 and substrate 1 does not affect the utility of the invention.

Next, a first chalcogenide glass layer 4 is formed over the first conductive electrode 2. The chalcogenide glass layer 4 is electrically coupled to electrode 2.

The first chalcogenide glass layer 4 is preferably a-germanium-selenide glass having a Ge_xSe_100−x stoichiometry. The stoichiometric range for the first chalcogenide glass layer 4 is from about Ge₁₈Se₈₂ to about Ge₄₃Se₅₇, but is preferably from about Ge₂₅Se₇₅ to about Ge₄₀Se₆₀, and is more preferably about Ge₄₀Se₆₀.

The first chalcogenide glass layer 4 has a thickness that ranges from about 150 Angstroms (Å) to about 400 Å. Preferably, the first chalcogenide glass layer 4 has a thickness of about 250 Å to about 300 Å. For purposes of a simplified description, the first chalcogenide glass layer 4 is described further below as a Ge₄₀Se₆₀ layer. The first chalcogenide glass layer 4 acts as a glass backbone and allows a metal-containing layer, such as a silver-selenide or a chalcogenide-doped with silver (Ag) layer, to be directly formed thereon.

The formation of the first chalcogenide glass layer 4, having a stoichiometric composition, such as Ge₄₀Se₆₀ in accordance with one exemplary embodiment of the invention, can be accomplished by any suitable method. For instance, evaporation, co-sputtering germanium and selenium in the appropriate ratios, sputtering using a germanium-selenide target having the desired stoichiometry, or chemical vapor deposition with stoichiometric amounts of GeH₄ and SeH₂ gases (or various compositions of these gases), which result in a germanium-selenide film of the desired stoichiometry, are non-limiting examples of methods which can be used to form the first chalcogenide glass layer 4.

Still referring to FIG. 1, a metal-containing layer 6, preferably a silver-selenide layer, is formed over the first chalcogenide glass layer 4. However, any suitable metal-containing layer 6 may be used so long as it contains an adequate source of metal ions, which can move in and out of chalcogenide glass layers during device operation. For instance, besides silver-selenide, the metal-containing layer 6 may be silver. Other suitable metal-containing layers 6 include all chalcogenide layers doped Ag such as chalcogenides O, S, Se, Te, and Po. Silver sulfide, silver oxide, and silver telluride, for example, among others, are all suitable silver-chalcogenides that may be used as metal-containing layer 6.

A variety of processes can be used to form the metal-containing layer 6. Some non-limiting examples are physical vapor deposition techniques such as evaporative deposition, sputtering, co-sputtering, starting with an Ag containing chalcogenide source. Other non-limiting processes such as chemical vapor deposition, co-evaporation, depositing a layer of selenium above a layer of silver to form silver-selenide, or chemical bath deposition of a silver-selenide layer can also be used. FIG. 1 illustrates that the metal-containing layer 6 is in contact with the upper surface of the first chalcogenide glass layer 4; however, intervening layers may also be provided between layers 4 and 6, as long as they permit the resulting device to operate in a constant current mode.

Metal-containing layer 6 is formed to a thickness that ranges from about 200 Å to about 2000 Å. Preferably, the metal-containing layer 6 is about 600 Å thick.

Still referring to FIG. 1, a second chalcogenide glass layer 8 is formed over the metal-containing layer 6. FIG. 1 illustrates that the second chalcogenide glass layer 8 is in contact with the upper surface of the metal-containing layer 6; however, intervening layers may also be provided between layers 6 and 8, so long as they permit the resulting device to operate in a constant current mode.

The second chalcogenide glass layer 8 is preferably a germanium-selenide glass having a Ge_xSe_100−x stoichiometry. The stoichiometric range for the second chalcogenide glass layer 8 is from about Ge₁₈Se₈₂ to about Ge₄₃Se₅₇, preferably from about Ge₂₅Se₇₅ to about Ge₄₀Se₆₀, and is more preferably about Ge₄₀Se₆₀. The second chalcogenide glass layer 8 preferably has a thickness that ranges from about 50 Å to about 500 Å. More preferably, the second chalcogenide glass layer 8 is about 150 Å thick.

Although the first chalcogenide glass layer 4 and second chalcogenide glass layer 8 are described-above as having a stoichiometry similar to each other, e.g., about Ge₄₀Se₆₀, it should be appreciated that the first chalcogenide glass layer 4 and the second chalcogenide glass layer 8 can possess different stoichiometries from each other, and they can even be different glasses. For instance, the first chalcogenide glass layer 4 can possess a stoichiometry of Ge₄₀Se₆₀ while the second chalcogenide glass layer 8 can possess a stoichiometry of Ge₂₅Se₇₅. For purposes of a simplified description, the second chalcogenide glass layer 8 is described further below as having a stoichiometry of Ge₄₀Se₆₀.

The formation of the second chalcogenide glass layer 8, can be accomplished by any suitable method as described above with reference to formation of the first chalcogenide glass layer 4.

As an alternative embodiment, depicted in FIG. 1a, an additional layer 9, such as Ag, can be provided above the second chalcogenide glass layer 8. Alternatively, layer 9 can be provided above (FIG. 1b) or below (FIG. 1c) metal-containing layer 6, e.g., Ag₂Se, rather than above the second chalcogenide glass layer 8 (FIG. 1a). In FIGS. 1a-1c, layer 9 is preferably an Ag layer that is less than or equal to about 500 Å thick, and more preferably is an Ag layer that is approximately 200 Å thick. Layer 9 can be deposited using any techniques well-known in the art. It is not ideal to have an Ag source provided strictly by electrode 10. Accordingly, care must be taken to limit the Ag source for long-term temperature stability and endurance. The presence of layer 9 addresses this concern. However, the presence of layer 9 is not required.

Further, in yet another alternative embodiment of FIG. 1, depicted in FIG. 1d, a constant current device can be fabricated where a third glass layer 7, analogous to glass layer 8, can be provided between layer 9 and the top electrode 10. Similar to the embodiments depicted in FIGS. 1a-1c, layer 9 is preferably an Ag layer that is less than or equal to about 500 Å thick, and more preferably is an Ag layer that is approximately 200 Å thick.

Referring back to FIG. 1, a second electrode 10 is formed over the second chalcogenide glass layer 8, to complete the formation of the chalcogenide constant current device 100. The second electrode 10 may comprise any conductive material so long as the constant current device's electrical properties are not altered. For example, various metals, such as, one or more of tungsten, nickel, tantalum, aluminum, platinum, silver, or titanium nitride among many others. However, care must be used since some metals, such as Ag, Au, or Cu, may migrate into a subsequently deposited glass layer and alter the electrical behavior of the constant current device.

In addition, the second electrode 10 can comprise a conductively-doped semiconductive material. FIG. 1 illustrates that the second electrode 10 is in contact with an upper surface of the second chalcogenide glass layer 8; however, intervening layers may also be provided between layers 8 and 10, as long as they permit the resulting device to operate in a constant current mode.

It should be appreciated that the first electrode 2 and the second electrode 10 can comprise the same material or different materials. As examples, the first electrode 2 and second electrode 10 can each be made of tungsten, or preferably, the first electrode 2 can comprise tungsten and the second electrode 10 can comprise silver. As described previously, it is not ideal to have an Ag source provided strictly by electrode 10. Accordingly, care must be taken to limit the Ag source for long-term temperature stability and endurance.

After forming the second electrode 10, a negative electrical pulse of absolute amplitude greater than the erase potential of the device is applied to structure 100 in order to have the device exhibit constant current behavior. Conventionally, a structure 100 fabricated as shown in FIG. 1, can be operated as a normal PCRAM memory device. For example, in DC operation, the device 100 can be read when a potential of approximately 100 mV is applied across conductors 10 and 2, a voltage of approximately 250 mV or greater is applied across conductors 10 and 2 to write it, and a negative potential of absolute amplitude greater than approximately negative 80 mV is applied to conductors 10 and 2 to erase it.

However, Applicants have discovered that when a negative electrical pulse that is significantly more negative than the erase potential is applied across conductors 10 and 2, the structure's electrical behavior is altered and it exhibits constant current source behavior rather than memory behavior. When in the constant current mode, a wide range of negative voltages can be applied across such a device without changing the device's current flow. The amplitude of the negative pulse signal that switches a device into constant current mode, e.g., significantly more negative than the PCRAM device's erase potential, will vary depending upon the thickness of the respective chalcogenide glass layers and the amount of metal, e.g., silver (Ag) present in the chalcogenide layer 4.

A first exemplary structure 100 was formed in accordance with the FIG. 1 embodiment of the invention in which the first electrode 2 comprised tungsten, the first chalcogenide glass layer 4 comprised a Ge₄₀Se₆₀ layer about 300 Å thick, the metal-containing layer 6 comprised an Ag₂Se layer about 600 Å thick, the second chalcogenide glass layer 8 comprised a Ge₄₀Se₆₀ layer about 150 Å thick, and the second electrode 10 comprised an Ag electrode.

When a negative voltage typically within the range from about negative 1.0 V (−1.0 V) to about negative 2.0 V (−2.0 V), with a pulse of about 8 ns to at least about 30 ns, was applied across electrodes 10 and 2, the exemplary structure 100 operated as a constant current device. A device fabricated in this manner provided a constant current of about negative 8 μA across an applied voltage range of about negative 100 mV (−100 mV) to at least about negative 800 mV (−800 mV), as graphically shown in FIG. 3. When the voltage becomes more negative than approximately −800 mV, the constant current behavior of the exemplary structure 100 begins to deteriorate. It was also found that devices fabricated in this manner permanently converted from a memory device into a constant current operation device.

Although FIG. 3 illustrates a constant current of about negative 8 μA across a voltage range of about −100 mV to about −800 mV, it should be appreciated that the exact constant current value and voltage range will depend upon the structural properties of the device, such as the precise composition and thickness of layers 4, 6, and 8, and the total amount of metal, e.g., silver, present in the chalcogenide layer 4.

Further, as noted above regarding the FIG. 1 embodiment, the first chalcogenide layer 4 and second chalcogenide layer 8 do not have to possess the same stoichiometries, nor do the first chalcogenide layer 4 and second chalcogenide layer 8 have to possess the same thickness. A constant current device 100 fabricated as described with reference to FIG. 1 can be of any geometric shape as desired or needed for a particular application.

As a second and more specific exemplary example, a constant current device was fabricated in accordance with a FIG. 1 embodiment of the invention in which the first electrode 2 comprised tungsten, the first chalcogenide layer 4 comprised a Ge₄₀Se₆₀ glass about 300 Å thick, the metal-containing layer 6 comprised an Ag₂Se layer about 600 Å thick, the second glass layer 8 comprised a Ge₂₅Se₇₅ glass about 150 Å thick, and the second electrode 10 comprised tungsten.

When a negative potential of 1.5 V (−1.5 V) with a 30 ns pulse was applied to the fabricated device, the device exhibited constant current behavior rather than memory behavior. Applicants further learned that applying a negative potential anywhere in the voltage ranges from about −1.0 V to about −2.0 V with a 30 ns pulse to the fabricated device, also generated constant current behavior. However, it was found that applying a negative 900 mV, 30 ns pulse did not cause the fabricated device exhibit constant current behavior.

Similar to the first exemplary device, the second exemplary structure 100, maintained a constant current over an applied negative voltage range of about −100 mV to about −800 mV. Accordingly, the constant current observable voltage range was at least approximately 700 mV once a negative electrical 30 ns pulse in the range of about −1.0 V to about −2.0 V was applied across conductors 10 and 2.

In a third exemplary example of the FIG. 1 embodiment, a structure 100 was fabricated including a first electrode 2 comprising tungsten, a layer 4 comprising Ge₄₀Se₆₀ glass about 150 Å thick, a metal-containing layer 6 comprising Ag₂Se about 600 Å thick, a glass layer 8 comprising Ge₄₀Se₆₀ glass about 150 Å thick, and a second electrode 10 comprising silver. A negative 8 ns electrical pulse of about negative 800 mV (−800 mV) was then applied to the fabricated device.

As a result, this device also exhibited constant current behavior rather than memory behavior. The fabricated device maintained a constant current over an applied negative voltage range of at least approximately 700 mV, specifically from about −100 mV to about −800 mV, after a negative 8 ns electrical pulse of about −800 mV was applied across electrodes 10 and 2 to convert the structure to a constant current mode. As is evident from the third exemplary example, the total amount of silver present in the device seems to affect the magnitude of the negative potential that needs to be applied to switch a device 100 into constant current mode.

Reference is now made to FIG. 2 which shows another exemplary embodiment of the invention and its method of formation.

The FIG. 2 embodiment illustrates a first electrode 2 provided over a substrate 1. The first electrode 2 may comprise any of the conductive materials listed above for the same electrode as in the FIG. 1 embodiment. For purposes of a simplified description, the first electrode 2 is described as tungsten (W). Similar to the FIG. 1 embodiment, additional layers may be provided between electrode 2 and substrate 1.

Next, a chalcogenide glass layer 4 is formed over the first electrode 2. Although FIG. 2 is described without intervening layers between the first electrode 2 and chalcogenide glass layer 4, intervening layers may be present so long as they do not prevent operation of the FIG. 2 structure in a constant current mode. The chalcogenide glass layer 4 is preferably a germanium-selenide glass having a Ge_xSe_100−x stoichiometry. The stoichiometric range for the chalcogenide glass layer 4 is from about Ge₁₈Se₈₂ to about Ge₄₃Se₅₇, is preferably from about Ge₂₅Se₇₅ to about Ge₄₀Se₆₀, and is more preferably about Ge₄₀Se₆₀. The chalcogenide glass layer 4 has a thickness that ranges from about 150 Å to about 400 Å. Preferably, the chalcogenide glass layer 4 in the FIG. 2 embodiment has a thickness of about 250 Å to about 300 Å.

The formation of the chalcogenide glass layer 4, having a stoichiometric composition, such as Ge₄₀Se₆₀ in accordance with one exemplary embodiment of the invention, can be accomplished by any of the methods described above for forming glass layers 4 or 8 of FIG. 1.

Still referring to FIG. 2, a metal-containing layer 6, preferably silver-selenide, is deposited over the chalcogenide glass layer 4. However, any suitable metal-containing layer 6 may be used so long as it contains an adequate source of metal ions, e.g., silver, which can move in and out of chalcogenide glass layer 4 during device operation. For instance, besides silver-selenide, the metal-containing layer 6 may be silver. Other suitable metal-containing layers include all chalcogenide layers containing Ag, e.g., chalcogenides O, S, Se, Te, and Po. Silver sulfide, silver oxide, and silver telluride, for example, among others, are all suitable silver-chalcogenides that may be used as metal-containing layer 6, e.g., a chalcogenide layer containing Ag.

Metal-containing layer 6 can be formed by any of the methods described above in connection with forming metal-containing layer 6 of the FIG. 1 embodiment. FIG. 2 illustrates that the metal-containing layer 6 is in contact with the upper surface of the chalcogenide glass layer 4; however, intervening layers may be present below metal-containing layer 6 so long as they do not prevent operation of the FIG. 2 structure in a constant current mode. Metal-containing layer 6 is formed to a thickness that ranges from about 200 Å to about 2000 Å. Preferably, the metal-containing layer 6 is about 600 Å thick.

As an alternative embodiment depicted in FIG. 2a, an additional layer 9, such as Ag, can be provided above the metal-containing layer 6, e.g., Ag₂Se, between the second electrode 10 and the metal-containing layer 6. Alternatively, an additional layer 9, similar to the FIG. 2a embodiment, can be provided below metal-containing layer 6, between the first chalcogenide layer 4 and metal-containing layer 6. In FIG. 2a, layer 9 is preferably an Ag layer that is less than or equal to 500 Å thick, and more preferably an Ag layer that is approximately 200 Å thick. Similar to the concerns for the exemplary FIG. 1a-1d embodiments, care must be taken to limit the Ag source for long-term temperature stability and endurance. The presence of layer 9 addresses this concern. Again, the presence of layer 9 is not required.

Referring back to FIG. 2, a second electrode 10 is formed over the metal-containing layer 6 to complete the formation of the chalcogenide constant current device 101, in accordance with the second embodiment. Further, although FIG. 2 is described without intervening layers between the metal-containing layer 6 and the second electrode 10, intervening layers may be present so long as they do not prevent operation of the FIG. 2 structure in a constant current mode. The second electrode 10 may comprise any of the materials described above for electrode 10 in the FIG. 1 embodiment. However, when the second electrode 10 of FIG. 2 is in direct contact with metal-containing layer 6, the second electrode 10 should preferably not comprise silver unless it is sputtered onto an Ag₂Se layer acting as metal-containing layer 6.

It should be appreciated that the first electrode 2 and the second electrode 10 of FIG. 2 can comprise the same material or different materials. As examples, the first electrode 2 and second electrode 10 can each be made of tungsten, or alternatively, the first electrode 2 can comprise tungsten and the second electrode 10 can comprise silver.

After the FIG. 2 structure 101 is formed, a negative electrical potential of sufficient magnitude can be applied across electrodes 10 and 2 to cause the device to exhibit constant current behavior. Conventionally, a device fabricated as shown in FIG. 2, can be operated as a normal PCRAM memory device. For example, in DC operation, the device can be read when a potential of approximately 100 mV is applied across conductors 10 and 2, a voltage of approximately 250 mV or greater is applied across conductors 10 and 2 to write it, and a negative potential of absolute amplitude greater than approximately negative 80 mV is applied to conductors 10 and 2 to erase it.

However, when a negative electrical pulse that is significantly more negative than the erase potential of structure 101 is applied across conductors 10 and 2, the structure's electrical behavior is altered and it exhibits constant current source behavior rather than memory behavior. When in the constant current mode, a wide range of negative voltages can be applied across such a device without changing the device's current flow. The amplitude of the negative pulse signal that switches a device into constant current mode, e.g., significantly more negative than the PCRAM device's erase potential, will vary depending upon the thickness of the respective chalcogenide glass layer and the amount of metal, e.g., silver (Ag) present in the chalcogenide layer 4.

In a first exemplary example of the FIG. 2 embodiment, a constant current device was fabricated in which the first electrode 2 comprised tungsten, the first layer 4 comprised a Ge₄₀Se₆₀ glass about 300 Å thick, the metal-containing layer 6 comprised an Ag₂Se layer about 600 Å thick, and the second electrode 10 comprised tungsten.

A negative electrical 30 ns potential of 1.5 V (−1.5 V) was applied to the fabricated exemplary device. As a result, this device exhibited constant current behavior rather than memory behavior. It was further found that applying an electrical negative 30 ns potential that ranged from about −1.0 V to about −2.0 V to the fabricated device also generated constant current behavior. However, when a more positive electrical pulse than −1.0 V was applied, for example, a negative 900 mV, 30 ns pulse, it was observed that this did not convert the exemplary device 101 into a constant current mode.

The exemplary device 101 maintained a constant current over an applied negative voltage range of about −100 mV to about −800 mV. Accordingly, the constant current observable voltage range is at least approximately 700 mV once a negative voltage in the range of about −1.0 V to about −2.0 V is applied across conductors 10 and 2.

In a second exemplary example of the FIG. 2 embodiment, a structure 101 was fabricated including a first electrode 2 comprising tungsten, a layer 4 comprising Ge₄₀Se₆₀ glass about 150 Å thick, a metal-containing layer 6 comprising Ag₂Se about 600 Å thick, and a second electrode 10 comprising silver. A negative potential of 800 mV (−800 mV) with a 8 ns pulse was applied to the fabricated device. As a result, this device also exhibited constant current behavior rather than memory behavior. As is evident from the second exemplary example, the total amount of silver present in the device 101 seems to affect the magnitude of the negative potential that needs to be applied to switch a device into constant current mode.

The fabricated device maintained a constant current over an applied negative voltage range of at least approximately 700 mV, specifically from about −100 mV to about −800 mV, when a negative electrical potential of about −800 mV was applied across electrodes 10 and 2. As with the FIG. 1 embodiment, the FIG. 2 embodiment can be of any geometric shape depending upon the desired or needed application.

Applicants have discovered that devices 100 and 101 fabricated in accordance with the embodiments of the present invention, which normally exhibit PCRAM memory behavior, can be permanently converted to a constant current device by applying a negative voltage across the device in excess of the negative voltage required to switch the device from a memory device to a constant current device. Stated another way, applying a significantly more negative electrical pulse than the device's erase potential converts the memory device into a constant current device. It was also observed that the greater the amount of Ag present in the constant current devices 100 and 101, a larger absolute amplitude, negative potential 8 ns pulse or a pulse of larger absolute amplitude at various widths could also be applied to devices 100 and 101 to induce constant current behavior.

Applicants further discovered that devices 100 and 101 fabricated in accordance with the invention, can be operated as a constant current device without first operating as a memory device, upon application of a negative electrical pulse sufficient to place the device in a constant current mode of operation. Conversely, devices 100 and 101 fabricated in accordance with the invention may be operated as a memory device without first operating then as a constant current device by application of voltages below the threshold required to place the device in a constant current mode. As a result, upon application of a negative pulse sufficient to place the memory device in a constant current mode of operation, the device converts to a constant current mode and remains in that state even when voltage is removed.

In another aspect, the invention provides methods for changing and resetting the current level which passes through constant current devices 100 and 101 such as described in FIGS. 1 and 2, when in a constant current operation mode.

For purposes of a simplified description, the methods of changing and resetting are described with respect to a fabricated 100 structure of FIG. 1. However, the methods are equally effective when applied to the FIG. 2 structure 101.

In particular, an exemplary structure 100 formed in accordance with FIG. 1 is described below in reference to FIG. 3, comprises a bottom tungsten electrode 2, a Ge₄₀Se₆₀ first glass layer 4 about 300 Å thick, a Ag₂Se layer 6 about 600 Å thick, a Ge₄₀Se₆₀ second glass layer 8 about 150 Å thick, and a top silver electrode 10. Thus, when the structure 100 of FIG. 1 is in a constant current mode and a DC voltage in the range of approximately 0 to approximately −1.0 V is applied across electrodes 10 and 2, the device has the current/voltage curve (I/V curve) as illustrated in FIG. 3.

FIG. 3 illustrates an observed region A beyond about −800 mV where the constant current value of −8 μA begins to decline for the constant current devices of the invention. Stated in another way, region A is where the constant current value is no longer maintained at −8 μA. It has been observed that Region A corresponds to a reset region or breakdown area of a chalcogenide constant current device 100. Specifically, when structure 100 is in constant current operation mode, as depicted in FIG. 3, the constant current value of the device can be reset to its initial constant current value, for example −8 μA, by applying a more negative voltage greater than the negative voltage threshold of the device. The initial constant current value is the initial current value of the chalcogenide structure when it is first switched to exhibit constant current behavior.

A constant current device can prematurely degrade if the applied potential is increased too much above the point where the constant current device starts to reduce the constant current value as illustrated in FIG. 3. The degradation of the constant current device can eventually lead to DNR behavior. However, the increase in the constant current's value can occur with a repeated pulse of similar or less amplitude as initially used to cause the device to exhibit constant current behavior. Further, after a certain negative electrical potential corresponding to region A, a constant current device will not reset to its initial constant current value. The constant current device's function is killed. A more preferable method of resetting a constant current device's constant current value is applying a more positive electrical potential.

Still referring to FIG. 3, it should be appreciated that although the constant current level is depicted as −8 μA, the constant current value will vary depending upon the thickness of the glass stack fabricated in accordance with the embodiments of FIGS. 1 and 2, and also depending upon the total amount of Ag in devices 100 and 101. Further, although region A is observed beyond about −800 mV in FIG. 3, region A's location can also vary, e.g., be greater or less than the value depicted in FIG. 3, depending upon the thickness of the glass stack, total amount of silver, and applied negative voltage. Still further, although FIG. 3 illustrates that the device maintains a constant current of at least 700 mV, the thickness of the glass stack, total amount of silver, and magnitude of the applied negative voltage, will also affect the device's range for maintaining a constant current value, e.g., a constant current device can possess a greater constant current range than depicted.

Thus, in one method aspect provided by the invention, region A of FIG. 3 corresponds to a negative breakdown voltage which can be applied to a constant current structure 100 or 101 to reset the structure's constant current value to the original value observed when structures 100 or 101 was initially switched to constant current mode. In another aspect provided by the invention, FIG. 3 also illustrates a method for changing the constant current value of device 100.

Applicants have discovered that applying a negative potential, up to a certain value increases the constant current's amplitude. Repeated current pulses of similar or less amplitude can further increase the constant current's amplitude. The negative potential necessary to increase the constant current's amplitude will vary depending upon the structural characteristics of the constant current device Conversely, applying a positive potential decreases the constant current's amplitude. Similarly, repeated current pulses of similar or less amplitude can further decrease the constant current's amplitude. Again, the positive potential necessary to decrease the constant current's amplitude will vary depending on the structural characteristics of the constant current device. Further details are provided below.

Typically, when structure 100, operating in constant current mode, is written with a positive voltage 60 which approximately is of larger absolute amplitude than a voltage threshold (Vt) value shown in FIG. 4; this signal decreases the value of the constant current limit, e.g., makes smaller in magnitude, as shown in FIG. 5. For purposes of a simplified description, this voltage is described as V₁. FIG. 5 illustrates a constant current limit 61, after a positive voltage 60, e.g., V₁ is applied to structure 100 that is larger than the Vt shown in FIG. 4.

It should be appreciated that, as FIG. 4 illustrates, the magnitude of V₁'s which can be applied to constant current structure 100 can be of a larger absolute amplitude than positive Vt voltage 60. Referring now to FIG. 5, which illustrates various V₁'s, graph 61 corresponds to the applied positive Vt voltage 60 of FIG. 4, graph 71 corresponds to the applied positive voltage 70 of FIG. 4, and graph 81 corresponds to the applied positive voltage 80 of FIG. 4. FIG. 4 illustrates only a small number of V₁'s that can be applied to induce a larger constant current value, e.g., decreasing the constant current's amplitude. As a result, FIG. 5 illustrates only a small number of constant current values that can result from an applied positive voltage to structure 100 operating in constant current mode. Further, FIG. 5 illustrates that repeated current pulses of similar or less amplitude can further decrease the constant current's amplitude as graphs 61, 71, and 81 illustrate.

Accordingly, applying a positive voltage, e.g., V₁, greater than Vt to structure 100 when operating in the constant current mode results in a decreased constant current amplitude than initially observed when structure 100 was first switched to operate in constant current mode. For example, if the constant current device has an initial constant current value of −8 μA as illustrated in FIG. 3, a positive voltage 70 can be applied to achieve a new constant current value illustrated in FIG. 5, graph 71. Stated another way, V₁ is applied to the structure reducing the constant current strength. The constant current value can be continuously decreased if desired. Therefore, applying a positive voltage, e.g., V₁, to constant current devices 100 and 101 allows one to change the constant current value to a lower constant current value than the device initially exhibited.

In another method aspect, the invention provides a method of increasing or resetting the high constant current value achieved with the methods described previously in reference to FIGS. 4 and 5. To increase the high constant current values illustrated in FIG. 5 from graphs 61, 71, and 81, a voltage more negative than a negative threshold voltage (Vt) needs to be applied to structure 100. This voltage is also known as the breakdown voltage of the device. For purposes of a simplified description, this voltage is described as V₂.

FIG. 6 illustrates in region A (analogous to region A of FIG. 3), the breakdown voltage of a constant current device 100 utilizing the constant current limit graph 60 of FIG. 5. A negative threshold voltage (V₂) is applied at a voltage slightly more negative than the upper voltage limit range for the existing constant current level, defined as the breakdown voltage of the device.

For instance, FIG. 6 illustrates that negative 800 mV (−800 mV) is about the upper voltage limit range for the existing constant current value of structure 100. Therefore, a V₂ must be applied that is slightly more negative than −800 mV as FIG. 6 illustrates. Specifically, the breakdown voltage of constant current device 100 is provided by Region A, which is slightly more negative than −800 mV.

Accordingly, FIG. 6 illustrates applying V₂ which is more negative than about −800 mV to constant current structure 100. Preferably, a more negative V₂ of equal to or greater than −1.0 V is applied to reset the constant current value of structure 100. However, other negative V₂ values can be used so long as they are greater than the constant current value's upper voltage limit range of structure 100, e.g., a V₂ greater than at least −800 mV. For instance, referring back to FIG. 3, FIG. 3 illustrates that the constant current's upper voltage limit range is approximately −800 mV for constant current structure 100. Thus, a slightly more negative V₂ is applied. This is illustrated by Region A. As discussed previously, the exact breakdown voltage and the amplitude of this negative signal, e.g., V₂, will vary depending upon the thickness of the glass layers and the amount of Ag within the constant current device. Further, applying a more negative electrical signal than −800 mV serves to increase the constant current's negative value.

Thus, applying a V₂ which corresponds to the breakdown voltage of the constant current device raises the existing constant current value, e.g., the existing constant current value becomes more negative or increases the constant current's amplitude. In other words, applying a more negative voltage that corresponds to region A of FIGS. 3 and 6, can either reset the constant current limit of structure 100 to the original constant current limit or at least raise the constant current value, e.g., make more negative, of the constant current device as FIG. 7 illustrates. Further, repeated current pulses of similar or less amplitude can further increase the constant current's amplitude as desired.

Specifically, FIG. 7 illustrates that the constant current value of structure 100 is raised, e.g., from about −8 μA as FIG. 6 illustrates, to about negative 800 nA (−800 nA). As a result, a new higher constant current value is achieved, e.g., more negative constant current value. Particularly, the higher constant current value achieved corresponds to the original negative constant current value of the constant current device. As FIG. 7 illustrates, the constant current value is significantly higher, e.g., more negative.

Typically, applying a negative V₂ that is more negative than from about −800 mV to about −2.0 V results in a breakdown voltage allowing the constant current limit to be reset or raised in a constant current device. However, in general, applying a more negative V₂ greater than −2.0 V results in the constant current device being destroyed. Therefore, it is preferable that a breakdown voltage within region A (which will vary as described above for each constant current device) that ranges from about −800 mV to about −2.0 V be applied to a constant current device such as devices 100 and 101 to reset or raise their constant current values, e.g., make more negative.

Although applying a more negative V₂ greater than −2.0 V typically results in the constant current device being destroyed, this is not always the case. This value e.g., −2.0 V, may vary depending upon the thickness of the layers within the constant current device. For instance, the thicker the device, a more negative V₂ may be applied e.g., more negative than −2.0 V without destroying the device. Further, applying a more negative V₂ greater than −2.0 V may result in the device resetting to a very low constant current value at which point the device can be pulsed and programmed as a constant current source again rather than destroying the device. Again, this is dependent on the thickness of the layers present in the constant current device. Thus, in one aspect, applying a more negative V₂ greater than −2.0 V can result in the device resetting to a low constant current value.

In effect, the methods disclosed above allow a constant current device's constant current level to be changed and reset. Changing or resetting the constant current value is non-destructive to the chalcogenide constant current device 100 or 101. Therefore, the constant current limit's value can be decreased, e.g., made more positive than the original constant current level of the constant current device by applying a V₁ voltage, and conversely, the constant current level can be reset or raised, e.g., made more negative than the original constant current value. It should be appreciated that the methods of the present invention are applicable to any constant current device and not limited to devices 100 and 101 constructed in accordance with FIGS. 1 and 2.

In another aspect, the invention also provides a method of utilizing a constant current device such that it functions as an analog memory device. This specific aspect provides a method of examining the constant current values which were applied to a constant current device. Once these constant current values are determined, the positive voltages, e.g., V₁, that were applied to achieve the constant current values can also be determined. In turn, these values can be read such that a memory state can correspond to each constant current value. Thus, when device 100 operates with different constant current values, the constant current device can be read such that a plurality of memory states can be stored. Each memory state corresponds to a specific constant current value. For instance, repeated current pulses of similar or less amplitude can further increase or decrease the constant current's amplitude. Each of these values can be read and stored for an analog assortment of varying states.

FIG. 8 illustrates the method for reading the constant current values of a constant current device such that it functions as an analog memory device. For example, when there are three available constant current values such as 0.1 μA, 0.5 μA, and 1 μA, one of the three different signal levels can be stored and read by measuring the current value. As described above in FIGS. 4 and 5, the same constant current device can possess different constant current values. Applying a positive voltage (V₁), as FIG. 4 illustrates, to the constant current device decreases the value of the constant current limit, e.g., makes the constant current amplitude decrease, as FIG. 5 illustrates. Applying a positive voltage such as lines 60, 70, and 80 of FIG. 4 can result in lines 61, 71, and 81 of FIG. 5. At least one of these constant current values can be stored and read individually. As noted previously, a plurality of constant current values can be achieved in a single constant current device.

FIG. 8 graphically illustrates V_a, V_b, and V_c. Depending upon the values of V_a, V_b, and V_c, the positive voltage which was applied to achieve V_a, V_b, and V_c can be determined. For instance, the constant current value V_a allows the calculation of the positive V₁ that was applied to cause V_a. Similarly, utilizing the constant current values of V_b and V_c allows the calculation of the positive V₁ that was applied to cause V_b and V_c. Each corresponding constant current value, V_a, V_b, and V_c for instance, can then be stored as a separate memory state for the device.

It should be appreciated that although the methods of the present invention described above refer to a chalcogenide constant current structure 100, the methods of the invention are equally applicable to any constant current structure, such as constant current structure 101, and are not just limited to the constant current structures disclosed within.

Further, although the embodiments described above refer to the formation of only one chalcogenide constant current structure 100 or 101, it must be understood that the invention contemplates the formation of any number of such chalcogenide constant current structures. A plurality of chalcogenide constant current structures can be fabricated and provided and operated with a plurality of elements as a non-limiting example. Thus, the chalcogenide glass constant current structure 100 or 101 can be utilized in many electronic devices. Specifically, the methods and operation of the structures disclosed above, can be used in any device whenever it is desired to have a constant current maintaining device that maintains a constant current over a wide range of applied voltages.

A typical processor based system which includes a constant current device according to the present invention is illustrated generally at 500 in FIG. 9. A processor based system is exemplary of a system having digital circuits which could include constant current devices. Without being limiting, such a system could include a computer system, camera system, scanner, machine vision, vehicle navigation, video phone, surveillance system, auto focus system, star tracker system, motion detection system, image stabilization system and data compression system for high-definition television, all of which can utilize the present invention.

A processor based system, such as a computer system, for example generally comprises a central processing unit (CPU) 544, for example, a microprocessor, that communicates with an input/output (I/O) device 546 over a bus 552. The constant current device 542 also communicates with the system over bus 552. The computer system 500 also includes random access memory (RAM) 548, and, in the case of a computer system may include peripheral devices such as a floppy disk drive 554 and a compact disk (CD) ROM drive 556 which also communicate with CPU 544 over the bus 552. Constant current device 542 is preferably constructed as an integrated circuit containing at least one chalcogenide glass layer and a metal-containing layer, as previously described with respect to FIGS. 1-8. It may also be desirable to integrate the processor 554, constant current device 542 and memory 548 on a single IC chip.

The invention is not limited to the details of the illustrated embodiments. Accordingly, the above description and drawings are only to be considered illustrative of exemplary embodiments which achieve the features and advantages of the invention. Modifications and substitutions to specific methods, process conditions, and structures can be made without departing from the spirit and scope of the invention. Accordingly, the invention is not to be considered as being limited by the foregoing description and drawings, but is only limited by the scope of the appended claims.

Claims

1. A method of changing the constant current value in a constant current device, said method comprising: providing a constant current device between first and second electrodes, wherein the first electrode has a lower potential than the second electrode; andapplying to said first electrode having a lower potential a positive potential thereby changing a constant current value of said constant current device, wherein said positive potential is more positive than a negative threshold voltage of said device.

2. The method as in claim 1, wherein said step of applying a positive potential decreases the constant current value of said device.

3. A method of changing the constant current value in a constant current device, said method comprising: providing a constant current device between at least a first and second electrode, wherein at least one electrode has a lower potential than the other electrode; andapplying to said electrode having the lower potential a positive voltage of larger absolute amplitude than said constant current device's voltage threshold to said constant current device, wherein said positive voltage decreases a constant current amplitude.

4. A method for utilizing a constant current device as an analog memory device, said method comprising: providing a constant current device between a first and second electrode, wherein at least one electrode is connected to a voltage source;applying at least one positive voltage across said first and second electrodes to said constant current device, wherein said at least one positive voltage creates at least one constant current value;reading said at least one constant current value; and saving said at least one read constant current value.

5. The method as in claim 4, wherein said saving at least one constant current value corresponds to a memory state.

6. The method as in claim 4, further comprising storing said at least one constant current value as a memory state for said device.

7. A method for using a constant current device as an analog memory device, said method comprising: reading at least one constant current value of said constant current device; andstoring said at least one constant current value of said device, wherein said at least one constant current value corresponds to a memory state of said analog memory device.

8. The method as in claim 7, further comprising applying a plurality of repeated pulses of similar or less amplitudes that corresponds to different constant current values of said device, wherein said different constant current values of said device are read and stored as different memory states for said device.

9. The method as in claim 8, wherein said step of applying a plurality of repeated pulses results in an analog assortment of states for said device.

10. A method for decreasing the amplitude of a constant current device's constant current value, said method comprising: providing a constant current device between two electrodes, wherein at least one of the electrode has a lower potential than the other electrode; andapplying to the electrode having a lower potential a positive potential to said constant current device, wherein said applied positive voltage is more positive than a negative threshold voltage of said device.

11. The method as in claim 10, wherein repeated pulses of similar or less amplitude of said applied positive potential is applied to said device thereby further decreasing the amplitude of said device's constant current value.

12. The method as in claim 10, wherein said applied positive voltage induces a more positive constant current value in said device.