Resistance variable memory device with sputtered metal-chalcogenide region and method of fabrication

Abstract

A chalcogenide-based programmable conductor memory device and method of forming the device, wherein a chalcogenide glass region is provided with a plurality of alternating tin chalcogenide and metal layers proximate thereto. The method of forming the device comprises sputtering the alternating tin chalcogenide and metal layers.

Description

FIELD OF THE INVENTION

The invention relates to the field of random access memory (RAM) devices formed using a resistance variable material.

BACKGROUND

Resistance variable memory elements, which include chalcogenide-based programmable conductor elements, have been investigated for suitability as semi-volatile and non-volatile random access memory devices. A typical such device is disclosed, for example, in U.S. Pat. No. 6,849,868 to Campbell, which is incorporated by reference.

In a typical chalcogenide-based programmable conductor memory device, a conductive material, such as silver, is incorporated into a chalcogenide glass. The resistance of the chalcogenide glass can be programmed to stable higher resistance and lower resistance states. An unprogrammed chalcogenide-based programmable conductor memory device is normally in a higher resistance state. A write operation programs the chalcogenide-based programmable conductor memory device to a lower resistance state by applying a voltage potential across the chalcogenide glass. The chalcogenide-based programmable conductor memory device may then be read by applying a voltage pulse of a lesser magnitude than required to program it; the resistance across the memory device is then sensed as higher or lower to define the ON and OFF states.

The programmed lower resistance state of a chalcogenide-based programmable conductor memory device can remain intact for an indefinite period, typically ranging from hours to weeks, after the voltage potentials are removed. The chalcogenide-based programmable conductor memory device can be returned to its higher resistance state by applying a reverse voltage potential of about the same order of magnitude as used to write the device to the lower resistance state. Again, the higher resistance state is maintained in a semi- or non-volatile manner once the voltage potential is removed. In this way, such a device can function as a variable resistance memory having at least two resistance states, which can define two respective logic states, i.e., at least a bit of data.

One exemplary chalcogenide-based programmable conductor memory device uses a germanium selenide (i.e., Ge_xSe_100-x) chalcogenide glass as a backbone. The germanium selenide glass has, in the prior art, incorporated silver (Ag) and silver selenide (Ag₂Se).

Previous work by the inventor, Kristy A. Campbell, has been directed to chalcogenide-based programmable conductor memory devices incorporating a silver-chalcogenide material as a layer of silver selenide (e.g., Ag₂Se) or silver sulfide (e.g., Ag₂S) in combination with a silver-metal layer and a chalcogenide glass layer. The silver-chalcogenide materials are suitable for assisting in the formation of a conducting channel through the chalcogenide glass layer for silver ions to move into to form a conductive pathway.

Tin (Sn) has a reduced thermal mobility in Ge_xSe_100-x compared to silver and the tin-chalcogenides are less toxic than the silver-chalcogenides, therefore tin-chalcogenides (e.g., SnSe) have also been found to be useful in chalcogenide-based programmable conductor memory devices to replace silver selenide. However, sputtering of tin selenide to form such devices has proven difficult due to the increased density of the sputtered layers. This increased density (e.g., ˜6 g/cm³ sputtered compared to ˜3 g/cm³ evaporated) can prevent the motion of silver ions into the chalcogenide glass, thereby preventing the memory device from functioning. Therefore, evaporative deposition techniques have been used to deposit such material, which is generally a less efficient, more costly, slower, and less controlled technique for deposition. However, evaporation deposition of tin selenide and silver also incorporates some oxygen into the resulting layer, which provides for the lower density and allows for more mobility of silver ions.

SUMMARY

In an exemplary embodiment, the invention provides a chalcogenide-based programmable conductor memory device having a layered stack with a region containing tin-chalcogenide and silver proximate a chalcogenide glass layer. The device comprising a chalcogenide glass layer and the region of tin-chalcogenide and silver is formed between two conductive layers or electrodes. The tin-chalcogenide and silver region is formed by sputter deposition of tin-chalcogenide and silver.

In an exemplary embodiment of the invention, the chalcogenide-based programmable conductor memory device contains alternating layers of tin selenide (e.g., Sn_xSe, where x is between about 0 and 2) and silver.

In an exemplary embodiment of the invention, the tin-chalcogenide and silver region is formed by alternation of sputtering of tin selenide and silver layers over the chalcogenide glass layer.

The above 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

FIGS. 1 and 2 show exemplary embodiments of memory devices in accordance with the invention.

FIGS. 3-6 show exemplary sequential stages of processing during the fabrication of a memory device as in FIG. 2, in accordance with the invention.

FIG. 7 shows an exemplary processor-based system incorporating a memory device in accordance with the invention.

DETAILED DESCRIPTION

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 semiconductor substrate that has an exposed substrate surface. A semiconductor substrate should be understood to include silicon, epitaxial 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. When reference is made to a semiconductor substrate or wafer in the following description, previous process steps may have been utilized to form regions or junctions in or over the base semiconductor or foundation. The substrate need not be semiconductor-based, but may be any support structure suitable for supporting an integrated circuit, including, but not limited to, metals, alloys, glasses, polymers, ceramics, and any other supportive materials as is known in the art.

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

The term “tin-chalcogenide” is intended to include various alloys, compounds, and mixtures of tin and chalcogens (e.g., sulfur (S), selenium (Se) tellurium (Te), polonium (Po), and oxygen (O)), including some species which have an excess or deficit of tin. For example, tin selenide, a species of tin-chalcogenide, is a preferred material for use in the invention and may be represented by the general formula Sn_+/−Se. Though not being limited by a particular stoichiometric ratio between Sn and Se, devices of the present invention typically comprise an Sn_xSe species where x ranges between about 0 and about 2, e.g., SnSe.

The term “chalcogenide glass” is intended to include glasses that comprise at least one element from group VIA (also know as group 16) of the periodic table. Group VIA elements (e.g., O, S, Se, Te, and Po) are also referred to as chalcogens.

The invention is now explained with reference to the figures, which illustrate exemplary embodiments and throughout which like reference numbers indicate like features. FIG. 1 shows an exemplary embodiment of a memory device 100 constructed in accordance with the invention. The device 100 shown in FIG. 1 is supported by a substrate 10. Over the substrate 10, though not necessarily directly so, is a conductive address line 12, which serves as an interconnect for the device 100 shown and a plurality of other similar devices of a portion of a memory array of which the shown device 100 is a part. It is possible to incorporate an optional insulating layer (not shown) between the substrate 10 and address line 12, and this may be preferred if the substrate 10 is semiconductor-based.

The conductive address line 12 can be any material known in the art as being useful for providing an interconnect line, such as doped polysilicon, silver (Ag), gold (Au), copper (Cu), tungsten (W), nickel (Ni), aluminum (Al), platinum (Pt), titanium (Ti), and other materials. Over the address line 12 is a first electrode 16, which can be defined within an insulating layer 14, if desired, and which is also over the address line 12. This electrode 16 can be any conductive material that will not migrate into chalcogenide glass, but is preferably tungsten (W). The insulating layer 14 should not allow the migration of silver (or other metal, e.g., copper) ions and can be an insulating nitride, such as silicon nitride (Si₃N₄), a low dielectric constant material, an insulating glass, or an insulating polymer, but is not limited to such materials.

A memory element, i.e., the portion of the memory device 100 which stores information, is formed over the first electrode 16. In the embodiment shown in FIG. 1, a layer of chalcogenide glass 18, preferably a germanium chalcogenide such as germanium selenide (Ge_xSe_100-x), can be provided over the first electrode 16. The germanium selenide can be within a stoichiometric range of about Ge₂₀Se₈₀ to about Ge₄₃Se₅₇, preferably about Ge₄₀Se₆₀, i.e., Ge₂Se₃. The layer of chalcogenide glass 18 can be between about 100 Å and about 1000 Å thick, preferably about 300 Åthick. Layer 18 need not be a single layer of glass, but may also be comprised of multiple sub-layers of chalcogenide glass having the same or different stoichiometries. This layer of chalcogenide glass 18 is in electrical contact with the underlying electrode 16.

Over the chalcogenide glass layer 18 is a region 20 of tin-chalcogenide; preferably tin selenide (Sn_xSe, where x is between about 0 and 2), and silver, which are layered as shown in FIG. 14. Alternating layers of tin selenide and silver are utilized to provide a region 20 incorporating both materials, wherein the silver is dispersed throughout the tin selenide. It is also possible that other chalcogenide materials may be substituted for selenium here, such as sulfur, oxygen, or tellurium; however, selenium is preferred and the remainder of the description will describe the invention utilizing tin selenide. The tin selenide and silver region 20 is preferably about 1,000 Å to about 2,000 Å thick; however, its thickness depends, in part, on the thickness of the underlying chalcogenide glass layer 18. The ratio of the thickness of the tin selenide and silver region 20 to that of the underlying chalcogenide glass layer 18 can be at least about 1:1, preferably about 3.33:1 to about 6.67:1.

Still referring to FIG. 1, over the tin selenide and silver region 20 is a second electrode 24. The second electrode 24 can be made of the same material as the first electrode 16, but is not required to be so. In the exemplary embodiment shown in FIG. 1, the second electrode 24 is preferably tungsten (W). The device(s) may be isolated by an insulating layer 26. The memory device 100 shown in FIG. 1 is a simplified exemplary embodiment of the invention. Other alternative embodiments may have more glass layers, as shown, for example, in FIG. 2, or may be provided within a via or may be made of blanket layers over an electrode such as electrode 16. Also, alternative embodiments may provide a common electrode in place of the dedicated electrode 16, shown in FIG. 1.

In accordance with the embodiment shown at FIG. 1, in a completed memory device 100, the tin selenide and silver region 20 provides a source of tin selenide and silver, which is incorporated into chalcogenide glass layer 18 during a conditioning step after formation of the memory device 100. The tin selenide and silver region 20 may also provide silver selenide (Ag₂Se) and silver tin selenide (Ag_xSn_ySe_z) to condition the chalcogenide glass layer 18. Specifically, the conditioning step comprises applying a potential across the memory element structure of the device 100 such that material from the region 20 is incorporated into the chalcogenide glass layer 18, thereby forming a conducting channel in the chalcogenide glass layer 18. Movement of silver ions into or out of the conducting channel during subsequent programming respectively forms or dissolves a conductive pathway, which causes a detectable resistance change across the memory device 100.

FIG. 2 shows another exemplary embodiment of a memory device 101 constructed in accordance with the invention. Memory device 101 has many similarities to memory device 100 of FIG. 1 and layers designated with like reference numbers are preferably the same materials and have the same thicknesses as those described in relation to the embodiment shown in FIG. 1. The primary difference between device 100 and device 101 is the addition to device 101 of an optional second chalcogenide glass layer 18a, a metal layer 22, and an optional third chalcogenide glass layer 18b.

The optional second chalcogenide glass layer 18a is formed over the tin selenide and silver region 20, is preferably Ge₂Se₃, and is preferably about 150 Å thick. Over this optional second chalcogenide glass layer 18a is a metal layer 22, which is preferably silver (Ag) and is preferably about 500 Å thick. Over the metal layer 22 is an optional third chalcogenide glass layer 18b, which is preferably Ge₂Se₃ and is preferably about 100 Å thick. The optional third chalcogenide glass layer 18b provides an adhesion layer for subsequent electrode formation. As with layer 18 of FIG. 1, layers 18a and 18b are not necessarily a single layer, but may be comprised of multiple sub-layers. Additionally, the optional second and third chalcogenide layers 18a and 18b may be a different chalcogenide glass from the first chalcogenide glass layer 18 or from each other.

Over the optional third chalcogenide glass layer 18b is a second electrode 24, which may be any conductive material, but is preferably not one that will migrate into the memory element stack and alter memory operation (e.g., not Cu or Ag), as discussed above for the preceding embodiments. Preferably, the second electrode 24 is tungsten (W).

FIGS. 3-6 illustrate a cross-sectional view of a wafer during the fabrication of a memory device 101 as shown by FIG. 2. Although the processing steps shown in FIGS. 3-6 most specifically refer to memory device 101 of FIG. 2, the methods and techniques discussed may also be used to fabricate other memory device structures, such as shown in FIG. 1, as would be understood by a person of ordinary skill in the art based on a reading of this specification.

As shown by FIG. 3, a substrate 10 is provided. As indicated above, the substrate 10 can be semiconductor-based or another material useful as a supporting structure for an integrated circuit, as is known in the art. If desired, an optional insulating layer (not shown) may be formed over the substrate 10; the optional insulating layer may be silicon nitride or other insulating materials used in the art. Over the substrate 10 (or optional insulating layer, if desired), a conductive address line 12 is formed by depositing a conductive material, such as doped polysilicon, aluminum, platinum, silver, gold, nickel, but preferably tungsten, patterning one or more conductive lines, for example, with photolithographic techniques, and etching to define the address line 12. The conductive material maybe deposited by any technique known in the art, such as sputtering, chemical vapor deposition, plasma enhanced chemical vapor deposition, evaporation, or plating.

Still referring to FIG. 3, over the address line 12 is formed an insulating layer 14. This layer 14 can be silicon nitride, a low dielectric constant material, or many other insulators known in the art that do not allow silver ion migration, and may be deposited by any method known in the art. An opening 14a in the insulating layer is made, for example, by photolithographic and etching techniques, thereby exposing a portion of the underlying address line 12. Over the insulating layer 14, within the opening 14a, and over the address line 12 is formed a conductive material, preferably tungsten (W). A chemical mechanical polishing (CMP) step may then be utilized, using the insulating layer 14 as a stop, to remove the conductive material from over the insulating layer 14, to leave it as a first electrode 16 over the address line 12, and planarize the wafer.

FIG. 4 shows the cross-section of the wafer of FIG. 3 at a subsequent stage of processing. A series of layers making up the memory device 101 (FIG. 2) are blanket-deposited over the wafer. A chalcogenide glass layer 18 is formed to a preferred thickness of about 300 Å over the first electrode 16 and insulating layer 14. The chalcogenide glass layer 18 is preferably Ge₂Se₃. Deposition of this chalcogenide glass layer 18 may be accomplished by any suitable method, such as evaporative techniques or chemical vapor deposition using germanium tetrahydride (GeH₄) and selenium dihydride (SeH₂) gases; however, the preferred technique utilizes either sputtering from a germanium selenide target having the desired stoichiometry or co-sputtering germanium and selenium in the appropriate ratios.

Still referring to FIG. 4, the tin selenide and silver region 20 is formed over the chalcogenide glass layer 18. To form region 20, alternating layers of tin selenide 20a and silver 20b are deposited by sputtering. Each tin selenide layer 20a is preferably between about 200 Å and about 400 Å. Each silver layer 20b is preferably between about 50 Å and about 100 Å. Sputtering these layers is preferred to evaporation deposition because of the increased efficiency, cost effectiveness, speed of fabrication, control of deposition rate, control of layer thickness, and control of layer properties provided by sputtering. Although the sputtered tin selenide and silver region 20 is denser than a like evaporated region, the proximity of the silver and the tin selenide layers 20b and 20a, respectively, allows for mixing and migration of these materials, heretofore not available to such sputtered regions.

Again, the thickness of region 20 is selected based, in part, on the thickness of layer 18; therefore, where the chalcogenide glass layer 18 is preferably about 300 Å thick, the alternating tin selenide layers 20a and silver layers 20b should make for a region 20 that is about 1,000 Å to about 2,000 Å thick. It should be noted that, as the processing steps outlined in relation to FIGS. 3-6 may be adapted for the formation of other devices in accordance the invention.

Still referring to FIG. 4, a second chalcogenide glass layer 18a is formed over the tin selenide and silver region 20, which can be sputtered similarly to the formation of layer 18. The second chalcogenide glass layer 18a is preferably a germanium selenide layer with a stoichiometry of Ge₂Se₃ and is preferably about 150 Å thick. Over the second chalcogenide glass layer 18a, a metal layer 22 is formed. The metal layer 22 is preferably silver (Ag), or at least contains silver, and is formed to a preferred thickness of about 500 Å. The metal layer 22 may be deposited by any technique known in the art. A third chalcogenide glass layer 18b is formed over the metal layer 22. This third chalcogenide glass layer 18b is also preferably germanium selenide with a stoichiometry of Ge₂Se₃, can be about 100 Å thick, and is preferably deposited by sputtering.

Still referring to FIG. 4, over the third chalcogenide glass layer 18b, a conductive material is deposited to form a second electrode 24 layer. Again, this conductive material may be any material suitable for a conductive electrode, but is preferably tungsten; however other materials may be used such as titanium nitride or tantalum, for example.

Now referring to FIG. 5, a layer of photoresist 30 is deposited over the top electrode 24 layer, masked and patterned to define the stacks for the memory device 101, which is one of a plurality of like memory devices of a memory array. An etching step is used to remove portions of layers 18, 20a, 20b, 18a, 22, 18b, and 24, with the insulating layer 14 used as an etch stop, leaving stacks as shown in FIG. 5. The photoresist 30 is removed, leaving a substantially complete memory device 101, as shown by FIG. 6. An insulating layer 26 may be formed over the device 101 to achieve a structure as shown by FIG. 2. This isolation step can be followed by the forming of connections to other circuitry of the integrated circuit (e.g., logic circuitry, sense amplifiers, etc.) of which the memory device 101 is a part, as is known in the art.

A conditioning step is performed by applying a voltage pulse of a given duration and magnitude to incorporate material from the tin selenide and silver region 20 into the chalcogenide glass layer 18 to form a conducting channel in the chalcogenide glass layer 18. The conducting channel will support a conductive pathway during operation of the memory device 101, the presence or lack of which provides at least two detectable resistance states for the memory device 101.

The embodiments described above refer to the formation of only a few possible chalcogenide-based programmable conductor memory device in accordance with the invention, which may be part of a memory array. It must be understood, however, that the invention contemplates the formation of other memory structures within the spirit of the invention, which can be fabricated as a memory array and operated with memory element access circuits.

FIG. 7 illustrates a processor system 400 which includes a memory circuit 448 employing chalcogenide-based programmable conductor memory devices (e.g., device 100 and 101) fabricated in accordance with the invention. A processor system, such as a computer system, generally comprises a central processing unit (CPU) 444, such as a microprocessor, a digital signal processor, or other programmable digital logic devices, which communicates with an input/output (I/O) device 446 over a bus 452. The memory circuit 448 communicates with the CPU 444 over bus 452, typically through a memory controller.

In the case of a computer system, the processor system may include peripheral devices, such as a floppy disk drive 454 and a compact disc (CD) ROM drive 456, which also communicate with CPU 444 over the bus 452. Memory circuit 448 is preferably constructed as an integrated circuit, which includes one or more resistance variable memory devices, e.g., device 101. If desired, the memory circuit 448 may be combined with the processor, for example CPU 444, in a single integrated circuit.

The above description and drawings should only be considered illustrative of exemplary embodiments that achieve the features and advantages of the invention. Modification and substitutions to specific 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 forming a memory device, comprising: providing a first electrode;forming a chalcogenide glass layer over said first electrode;sputtering a plurality of alternating tin chalcogenide layers and silver layers over said chalcogenide glass layer; andproviding a second electrode over said plurality of alternating tin chalcogenide layers and silver layers.

2. The method of claim 1, wherein said tin chalcogenide is tin selenide.

3. The method of claim 1, wherein each sputtered tin chalcogenide layer is about 200 Å to about 400 Å thick.

4. The method of claim 1, wherein each sputtered silver layer is about 50Å to about 100 Å thick.

5. The method of claim 1, wherein said plurality of alternating tin chalcogenide and silver layers forms a region that is about 1,000 Å to about 2,000 Å thick.

6. The method of claim 1, wherein said chalcogenide glass comprises germanium selenide.

7. The method of claim 6, wherein said germanium selenide has a stoichiometry of about Ge2Se3.

8. The method of claim 1, further comprising: forming a second chalcogenide glass layer over said plurality of alternating tin chalcogenide and silver layers;forming a metal layer over said second chalcogenide glass layer, said metal layer comprising silver; andforming a third chalcogenide glass layer over said metal layer.

9. A method of forming a memory device, comprising: providing a first electrode;providing a first germanium selenide layer over said first electrode;sputtering a plurality of alternating tin selenide layers and silver layers over said first germanium selenide layer;providing a second germanium selenide layer over said plurality of alternating tin selenide layers and silver layers;providing a metal layer over said second germanium selenide layer, said metal layer comprising silver;providing a third germanium selenide layer over said metal layer; andproviding a second electrode over said third germanium selenide layer.

10. The method of claim 9, wherein said plurality of alternating tin seleinde layers and silver layers form a region about 1,000 Å to about 2,000 Å thick.

11. The method of claim 9, wherein said first germanium selenide layer is about 300 Å thick.

12. The method of claim 9, wherein said tin selenide layers are each about 200 Å to about 400 Å thick.

13. The method of claim 9, wherein said silver layers are each about 50 Å to about 100 Å thick.