RELATED APPLICATION DATA
The benefit of U.S. Provisional Application No. 60/630,123; filed 22 Nov. 2004, entitled SIDE WALL ACTIVE PHASE CHANGE RAM AND MANUFACTURING METHOD, is hereby claimed.
PARTIES TO A JOINT RESEARCH AGREEMENT
International Business Machines Corporation, a New York corporation; Macronix International Corporation, Ltd., a Taiwan corporation, and Infineon Technologies A.G., a German corporation, are parties to a Joint Research Agreement.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to methods for manufacturing integrated circuits and other devices, and more particularly to methods for making very small pin-shaped structures.
2. Description of Related Art
A need arises in integrated circuit manufacturing processes for making very small structures. For example, small elements comprising chalcogenide materials or other phase change materials can be caused to change phase by application of electrical current. This property has generated interest in using phase change materials to form nonvolatile memory circuits.
One direction of development has been toward using small quantities of programmable resistive material, particularly in small pores. Patents illustrating development toward small pores include: Ovshinsky, “Multibit Single Cell Memory Element Having Tapered Contact,” U.S. Pat. No. 5,687,112, issued Nov. 11, 1997; Zahorik et al., “Method of Making Chalogenide [sic] Memory Device,” U.S. Pat. No. 5,789,277, issued Aug. 4, 1998; Doan et al., “Controllable Ovonic Phase-Change Semiconductor Memory Device and Methods of Fabricating the Same,” U.S. Pat. No. 6,150,253, issued Nov. 21, 2000.
My U.S. Patent application Publication No. US-2004-0026686-A1 describes a phase change memory cell in which the phase change element comprises a side wall on an electrode/dielectric/electrode stack. Data is stored by causing transitions in the phase change material between amorphous and crystalline states using current. Current heats the material and causes transitions between the states. The change from the amorphous to the crystalline state is generally a lower current operation. The change from crystalline to amorphous, referred to as reset herein, is generally a higher current operation. It is desirable to minimize the magnitude of the reset current used to cause transition of phase change material from crystalline state to amorphous state. The magnitude of the reset current needed for reset can be reduced by reducing the size of the phase change material element in the cell and of the contact area between electrodes and the phase change material.
Other applications for small structures arise in integrated circuit manufacturing, and it is desirable to provide new manufacturing techniques and structures satisfying this need.
SUMMARY OF THE INVENTION
The present invention includes methods to form a narrow side wall spacer or pin. A method of forming a memory cell based on such a narrow side wall spacer or pin is described which comprises forming a stack comprising a first electrode, an insulating layer over the first electrode, and a second electrode over the insulating layer, with a side wall on at least the insulating layer of the stack. A side wall spacer comprising a programmable resistive material in electrical communication with the first and second electrodes is formed. The side wall spacer has a length extending from the first electrode to the second electrode along the side wall, a width generally orthogonal to the length, and a thickness determined by the thickness of a layer of programmable resistive material used to form the side wall spacer. The side wall spacer is formed by depositing a layer of programmable resistive material over the side wall of the stack, anisotropically etching the layer of programmable resistive material to remove it in areas away from the side wall, and selectively etching the programmable resistive material according to a pattern to define the width of the side wall spacer. In embodiments described herein, the width is less than 50 nanometers, and more preferably about 40 nanometers or less.
In order to selectively etch the programmable resistive material according to a pattern to define a side wall spacer with such a narrow width, one technique includes forming an etch mask having a lithographic pattern to define a lithographic width, and then trimming the etch mask to provide a trimmed mask to define the pattern used for defining the width of the side wall spacer. In one example, the etch mask comprises a photoresist, which is etched anisotropically to form the trimmed mask using an oxygen based plasma etch. In another example, the etch mask comprises a hard mask defined using a lithographic process, which is etched to reduce its width to form the trimmed mask.
The three dimensions that define the size of the active region in the phase change pin for the cell described herein are preferably less than 50 nanometers, and can all be less than the minimum feature size of the lithographic process applied to make the cell. The dimensions are defined in technology described herein, by the thin film thickness of phase change material, the inter-electrode dielectric thin film thickness, and the trimmed mask. As a result, the cell size (the volume of the phase change material) is very small (smaller than F3, where F is the minimum lithographic feature size for the process used to manufacture the memory cell). The resulting cell of phase change material comprises a narrow pin on the side wall of an electrode stack. The contact area between at least one of the top and bottom electrodes and the phase change material pin is also defined sub-lithographically by electrode layer thicknesses for the heights, and the photo-resist pattern trimming process for the width of the contacts. The small cell and small contact region allow implementation of a memory with very small reset current and low power consumption.
A memory device is also described that includes a stack including a first electrode, an inter-electrode insulating member over the first electrode, and a second electrode over the inter-electrode insulating member. The stack has a side wall over at least the insulating member. A spacer comprising programmable resistive material on the side wall is in electrical communication with the first and second electrodes. The spacer has a length extending from the first electrode to the second electrode along the side wall on the insulating layer, which is generally orthogonal to the length and a thickness. The width and thickness of the spacer are less than 40 nanometers in embodiments of the technology described herein. The programmable resistive material comprises a phase change material, which is reversibly programmable.
The method described herein for formation of the phase change material pin can be used to make a very small pin for other nano-technology uses on an integrated circuit or other device, using materials other than phase change materials, like metals, dielectrics, organic materials, semiconductors, and so on. The small dimension side wall pin can be formed on structures other than that used for the phase change memory cell described herein, such as structures comprising other types of stacks of thin films, such as stacks of thin film dielectrics, with and without an electrode layer for contact to the pin.
Other aspects and advantages of the technology described herein can be understood with reference to the figures and the detailed description which follow.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 illustrates an embodiment of a side wall active pin phase change memory element.
FIG. 2 is a schematic diagram for a memory array comprising phase change memory elements.
FIG. 3 is a block diagram of an integrated circuit device including a thin film fuse phase change memory array and other circuitry.
FIG. 4 is a cross-section of the final array structure for an embodiment of the invention.
FIG. 5 is a cross-section of the structure after front-end-of-line processing and formation of electrode stack thin film layers.
FIG. 6A and FIG. 6B show top and cross-sectional views respectively after electrode stack etching the structure from FIG. 5.
FIG. 7 shows phase change material thin film deposition on the structure of FIG. 6B.
FIG. 8A and FIG. 8B show top and cross-sectional views respectively after GST thin film spacer etching.
FIG. 9 shows a cross-sectional view after dielectric fill layer formation.
FIG. 10 shows a cross-sectional view after chemical mechanical polishing for planarization and exposure of the phase change material side wall.
FIG. 11 shows a top view after formation of a photoresist pattern, and trimming for definition of phase change side wall pin width.
FIG. 12A and FIG. 12B show top and cross-sectional views respectively after selective etching of the phase change material side wall to define a phase change side wall pin width dimension.
FIG. 13 shows a top view after removal of the photoresist, with the resulting phase change material side wall pin.
FIG. 14 shows a cross-sectional view after filling in the small seam left by removal of the phase change material side wall, and subsequent oxide deposition.
FIG. 15 shows top and cross-sectional views after via formation and metallization for definition of the bit lines.
FIG. 16 illustrates an embodiment in which the thin film phase change material side wall is partially etched.
FIG. 17 illustrates a first stage in a representative process for manufacturing a small pin on an integrated circuit.
FIG. 18 illustrates a second stage in a representative process for manufacturing a small pin on an integrated circuit.
FIG. 19 illustrates a third stage in a representative process for manufacturing a small pin on an integrated circuit.
FIG. 20 illustrates a fourth stage in a representative process for manufacturing a small pin on an integrated circuit.
FIGS. 21 and 22 illustrate cross-section and perspective views, respectively, of a fifth stage in a representative process for manufacturing a small pin on an integrated circuit.
FIG. 23 illustrates a sixth stage in a representative process for manufacturing a small pin on an integrated circuit.
FIG. 24 illustrates a seventh stage in a representative process for manufacturing a small pin on an integrated circuit.
FIG. 25 illustrates an eighth stage in a representative process for manufacturing a small pin on an integrated circuit.
FIG. 26 illustrates a small pin on made as described herein.
The following detailed description is made with reference to the figures. Preferred embodiments are described to illustrate the present invention, not to limit its scope, which is defined by the claims. Those of ordinary skill in the art will recognize a variety of equivalent variations on the description that follows.
FIG. 1 is a perspective view of a side wall active pin memory cell 10. The cell includes a narrow side wall spacer, referred to as a pin 5 on a side wall of an electrode stack that includes a thin film electrode 6, and a thin film electrode 7 separated by an inter-electrode dielectric layer 8. A dielectric 9 overlies the electrode stack in the illustrated embodiment. The pin 5 consists of a programmable resistive material, such as a phase change material. The pin 5 has an active region, within which the phase change is confined, with a length L between a first electrode 6 and a second electrode 7 which is determined by the thickness of the inter-electrode dielectric layer 8. The active region of the pin 5 has a thickness T determined by the thickness of a thin film formed on the side wall of the electrode stack. The electrode stack is made using a photolithographic process or other type of lithographic process so that its width is about equal to the minimum feature size specified for the lithographic process. For advanced lithographic processes the width W of the electrode stack may be on the order of 90 nanometers. The active region of the pin 5 has a width which is less than the minimum feature size for the lithographic process used to define the electrode stack. In embodiments described herein, the width of the active region of the pin 5 is about 40 nanometers or less.
As illustrated, the active region of the pin 5 has a length L defined by a thin film thickness of the inter-electrode dielectric 8, which in embodiments of the invention can range between about 20 and 50 nanometers. Likewise, the active region of the pin 5 has a thickness T which is defined by the thin film thickness of the material used to form the side wall pin, which in embodiments of the invention can range between about 10 and 50 nanometers. Accordingly, all three dimensions of the pin 5 are less than 50 nanometers in embodiments of the present invention, and more preferably about 40 or less nanometers.
In embodiments of the invention, the programmable resistive material comprises a phase change material, such as Ge2Sb2Te5 or other materials described below. The volume of material within the pin 5, in which the phase change is induced in the structure illustrated in FIG. 1, is therefore very small. For embodiments in which the length L, the width W and the thickness T of the active region of the pin 5 are less than 40 nanometers, the volume of the active region is less than 64×10−24 m3. Accordingly, the magnitude of the reset current required for changing the phase is very small.
Embodiments of the memory cell include phase change based memory materials, including chalcogenide based materials and other materials, for the side wall pin 5. Chalcogens include any of the four elements oxygen (O), sulfur (S), selenium (Se), and tellurium (Te), forming part of group VI of the periodic table. Chalcogenides comprise compounds of a chalcogen with a more electropositive element or radical. Chalcogenide alloys comprise combinations of chalcogenides with other materials such as transition metals. A chalcogenide alloy usually contains one or more elements from column six of the periodic table of elements, such as germanium (Ge) and tin (Sn). Often, chalcogenide alloys include combinations including one or more of antimony (Sb), gallium (Ga), indium (In), and silver (Ag). Many phase change based memory materials have been described in technical literature, including alloys of: Ga/Sb, In/Sb, In/Se, Sb/Te, Ge/Te, Ge/Sb/Te, In/Sb/Te, Ga/Se/Te, Sn/Sb/Te, In/Sb/Ge, Ag/In/Sb/Te, Ge/Sn/Sb/Te, Ge/Sb/Se/Te and Te/Ge/Sb/S. In the family of Ge/Sb/Te alloys, a wide range of alloy compositions may be workable. The compositions can be characterized as TeaGebSb100−(a+b), where a and b represent atomic percentages that total 100% of the atoms of the constituent elements. One researcher has described the most useful alloys as having an average concentration of Te in the deposited materials well below 70%, typically below about 60% and ranged in general from as low as about 23% up to about 58% Te and most preferably about 48% to 58% Te. Concentrations of Ge were above about 5% and ranged from a low of about 8% to about 30% average in the material, remaining generally below 50%. Most preferably, concentrations of Ge ranged from about 8% to about 40%. The remainder of the principal constituent elements in this composition was Sb (Ovshinsky '112 patent, cols 10-11). Particular alloys evaluated by another researcher include Ge2Sb2Te5, GeSb2Te4 and GeSb4Te7 (Noboru Yamada, “Potential of Ge—Sb—Te Phase-Change Optical Disks for High-Data-Rate Recording”, SPIE v.3109, pp. 28-37 (1997).) More generally, a transition metal such as chromium (Cr), iron (Fe), nickel (Ni), niobium (Nb), palladium (Pd), platinum (Pt) and mixtures or alloys thereof may be combined with Ge/Sb/Te to form a phase change alloy that has programmable resistive properties. Specific examples of memory materials that may be useful are given in Ovshinsky '112 at columns 11-13, which examples are hereby incorporated by reference.
Phase change materials are capable of being switched between a first structural state in which the material is in a generally amorphous solid phase, and a second structural state in which the material is in a generally crystalline solid phase in its local order in the active channel region of the cell. These phase change materials are at least bistable. The term amorphous is used to refer to a relatively less ordered structure, more disordered than a single crystal, which has the detectable characteristics such as higher electrical resistivity than the crystalline phase. The term crystalline is used to refer to a relatively more ordered structure, more ordered than in an amorphous structure, which has detectable characteristics such as lower electrical resistivity than the amorphous phase. Typically, phase change materials may be electrically switched between different detectable states of local order across the spectrum between completely amorphous and completely crystalline states. Other material characteristics affected by the change between amorphous and crystalline phases include atomic order, free electron density and activation energy. The material may be switched either into different solid phases or into mixtures of two or more solid phases, providing a gray scale between completely amorphous and completely crystalline states. The electrical properties in the material may vary accordingly.
Phase change materials can be changed from one phase state to another by application of electrical pulses. It has been observed that a shorter, higher amplitude pulse tends to change the phase change material to a generally amorphous state. A longer, lower amplitude pulse tends to change the phase change material to a generally crystalline state. The energy in a shorter, higher amplitude pulse is high enough to allow for bonds of the crystalline structure to be broken and short enough to prevent the atoms from realigning into a crystalline state. Appropriate profiles for pulses can be determined empirically, without undue experimentation, specifically adapted to a particular phase change alloy.
In following sections of the disclosure, the phase change material is referred to as GST, and it will be understood that other types of phase change materials can be used. A material useful for implementation of a memory cell as described herein is Ge2Sb2Te5.
Useful characteristics of the programmable resistive material, like a phase change material include the material having a resistance which is programmable, and preferably in a reversible manner, such as by having at least two solid phases that can be reversibly induced by electrical current. These at least two phases include an amorphous phase and a crystalline phase. However, in operation, the programmable resistive material may not be fully converted to either an amorphous or crystalline phase. Intermediate phases or mixtures of phases may have a detectable difference in material characteristics. The two solid phases should generally be bistable and have different electrical properties. The programmable resistive material may be a chalcogenide material. A chalcogenide material may include GST. Alternatively, it may be one of the other phase change materials identified above.
FIG. 2 is a schematic illustration of a memory array, which can be implemented as described herein. In the schematic illustration of FIG. 2, a common source line 28, a word line 23 and a word line 24 are arranged generally parallel in the Y-direction. Bit lines 41 and 42 are arranged generally parallel in the X-direction. Thus, a Y-decoder and a word line driver in block 45 are coupled to the word lines 23, 24. An X-decoder and a set of sense amplifiers in block 46 are coupled to the bit lines 41 and 42. The common source line 28 is coupled to the source terminals of access transistors 50, 51, 52 and 53. The gate of access transistor 50 is coupled to the word line 23. The gate of access transistor 51 is coupled to the word line 24. The gate of access transistor 52 is coupled to the word line 23. The gate of access transistor 53 is coupled to the word line 24. The drain of access transistor 50 is coupled to the bottom electrode member 32 for side wall pin memory cell 35, which has top electrode member 34. The top electrode member 34 is coupled to the bit line 41. Likewise, the drain of access transistor 51 is coupled to the bottom electrode member 33 for side wall pin memory cell 36, which has top electrode member 37. The top electrode member 37 is coupled to the bit line 41. Access transistors 52 and 53 are coupled to corresponding side wall pin memory cells as well on bit line 42. It can be seen that the common source line 28 is shared by two rows of memory cells, where a row is arranged in the Y-direction in the illustrated schematic. In other embodiments, the access transistors can be replaced by diodes, or other structures for controlling current flow to selected devices in the array for reading and writing data.
FIG. 3 is a simplified block diagram of an integrated circuit according to an embodiment of the present invention. The integrated circuit 74 includes a memory array 55 implemented using side wall active pin phase change memory cells, on a semiconductor substrate. A row decoder 56 is coupled to a plurality of word lines 62, and arranged along rows in the memory array 55. A column decoder 63 is coupled to a plurality of bit lines 64 arranged along columns in the memory array 55 for reading and programming data from the side wall pin memory cells in the array 55. Addresses are supplied on bus 58 to column decoder 63 and row decoder 56. Sense amplifiers and data-in structures in block 59 are coupled to the column decoder 63 via data bus 67. Data is supplied via the data-in line 71 from input/output ports on the integrated circuit 75 or from other data sources internal or external to the integrated circuit 75, to the data-in structures in block 59. In the illustrated embodiment, other circuitry is included on the integrated circuit, such as a general purpose processor or special purpose application circuitry, or a combination of modules providing system-on-a-chip functionality supported by the thin film fuse phase change memory cell array. Data is supplied via the data-out line 72 from the sense amplifiers in block 59 to input/output ports on the integrated circuit 75, or to other data destinations internal or external to the integrated circuit 75.
A controller implemented in this example using bias arrangement state machine 69 controls the application of bias arrangement supply voltages 68, such as read, program, erase, erase verify and program verify voltages. The controller can be implemented using special-purpose logic circuitry as known in the art. In alternative embodiments, the controller comprises a general-purpose processor, which may be implemented on the same integrated circuit, which executes a computer program to control the operations of the device. In yet other embodiments, a combination of special-purpose logic circuitry and a general-purpose processor may be utilized for implementation of the controller.
FIG. 4 depicts a cross-section of a plurality of side wall active pin phase change random access memory cells 100-103. The cells 100-103 are formed on a semiconductor substrate 110. Isolation structures such as shallow trench isolation STI dielectric trenches 111 and 112 isolate pairs of rows of memory cell access transistors. The access transistors are formed by common source region 116 in the substrate 110, and drain regions 115 and 117 in the substrate 110. Polysilicon word lines 113 and 114 forms the gates of the access transistors. The dielectric fill layer 118 is formed over the polysilicon word lines 113, 114. Contact plug structures 141 and 120 contact individual access transistor drains, and common source line 119 contacts source regions along a row in the array. The common source line 119 contacts the common source region 116. The plug structure 120 contacts a bottom electrode 121 of cell 101. The cell 101, like cells 100, 102 and 103, includes a thin film bottom electrode 121, a thin film inter-electrode dielectric layer 122, a thin film top electrode 123, and a side wall pin 124 comprising GST or another phase change material. A dielectric fill layer 127 overlies the cells 100-103. Tungsten plug 126 contacts the top electrode 123. A patterned metal layer providing contacts 129, 130, 131, 132 overlies the dielectric fill layer 127. Typically the contacts 129-132 are part of a single bit line extending to decoding circuits as can be seen with reference to FIG. 2. A thin oxide layer 125 is shown overlying the top electrode 123. The layer 125 is used for process margin as described below.
In representative embodiments, the patterned metal layer (contacts 129-132) comprises copper metallization. Other types of metallization, including aluminum and aluminum alloys, could be utilized as well. The top and bottom electrodes (e.g. 121, 123) comprise TiN or TaN with a thickness of 10 to 30 nm. Alternatively, the electrodes may be TiAlN or TaAlN, or may comprise one or more elements selected from the group consisting of Ti, W, Mo, Al, Ta, Cu, Pt, Ir, La, Ni, Ru and O. The inter-electrode insulating layer may be silicon oxide, silicon oxynitride, silicon nitride, Al2O3, other low K dielectrics, or an ONO or SONO multi-layer structure. Alternatively, the inter-electrode insulating layer may comprise one or more elements selected from the group consisting of Si, Ti, Al, Ta, N, O, and C. The inter-electrode thickness may be 10 to 200 nm, and more preferably 50 nanometers or less. The second electrode may be TiN or TaN.
FIG. 5 illustrates a structure 99 after front-end-of-line processing, forming the standard CMOS components in the illustrated embodiment corresponding to the word lines, the source line, and the access transistors in the array shown in FIG. 2. In FIG. 5, source line 119 overlies doped region 116 in the semiconductor substrate, where the doped region 116 corresponds with the source terminal of a first access transistor on the left in the figure, and of a second access transistor on the right in the figure. In this embodiment, the source line 119 extends to the top surface of the structure 99. In other embodiments the source line does not extend all the way to the surface. Doped region 115 corresponds with the drain terminal of the first access transistor. A word line including polysilicon 113, and silicide cap (not shown), acts as the gate of the first access transistor. Dielectric layer 118 overlies the polysilicon word line 113. Plug 120 contacts doped region 115, and provides a conductive path to the surface of the structure 99 for contact to a memory cell electrode as described below. The drain terminal of the second access transistor is provided by doped region 117. A word line including polysilicon line 114, and the silicide cap (not shown) acts as the gate for the second access transistor. Plug 141 contacts doped region 117 and provides a conductive path to the top surface of the structure 99 for contact to a memory cell electrode as described below. Isolation trenches 111 and 112 separate the two-transistor structure coupled to the plugs 120 and 141, from adjacent two-transistor structures. The structure 99 illustrated in FIG. 5 provides a substrate for formation of memory cell components, as described in more detail below.
After formation of the plugs 120, 141 and source line 119 for the structure 99, a multilayer, thin film structure is formed including bottom electrode thin film 150, top electrode thin film 152, inter-electrode dielectric 151, and protective top dielectric 153. The bottom electrode film 150 has a thickness less than 50 nanometers, and preferably in the range of 10 to 30 nanometers. The top electrode film 152 has a thickness less than 50 nanometers, and preferably in the range of 10 to 30 nanometers, and can be different than that of the bottom electrode film. For example, the thickness of the top electrode film 152 can be slightly greater than that of the bottom electrode, in order to improve process margin for reliable contacts using tungsten plug technology and the like. The top dielectric 153 provides process margin for use of chemical mechanical polishing for planarization, variations in side wall spacer etching, and the like. Alternative embodiments without the top dielectric 153 might be implemented.
FIG. 6A shows a mask pattern in top view including a first rectangle 155 and a second rectangle 156 for etching the multilayer thin film structure of FIG. 5, to form the electrode stacks 60, 65, as shown in cross-section in FIG. 6B. The electrode stack 60 includes the bottom electrode 121, inter-electrode dielectric 122, and the top electrode 123. The electrode stack 60 has side wall 61. Likewise, the electrode stack 65 has side wall 66. Reactive ion etching REI is utilized in order to establish side walls 61, 66 as vertical as possible. Although not shown in the diagram, the reactive ion etching RIE may over-cut into the dielectric fill layer 118. In a representative process the over-cut is about 20 nanometers. BCl3 and/or Cl2 based recipes for the processes can be used.
FIG. 7 illustrates a structure after depositing, by sputtering for example, a conformal layer 170 of GST, or other programmable resistive material, over the stacks 60, 65. GST can be deposited using sputtering without collimation at about 250 degrees C. This results in a thin film having a thickness on the top of the electrode stacks of about 60 to 80 nanometers, a thickness on the side walls of about 20 to 30 nanometers, and a thickness between the stacks of about 60 to 80 nanometers, when using Ge2Sb2Te5 as the phase change material. Various embodiments of the process can sputter the entire wafer to thickness of 40 to 100 nanometers on the flat surfaces.
FIG. 8A illustrates the results of side wall etching in plan view, by an etching processes which remove the GST layer from the flat surfaces, and leave side walls 171 on stack 60 and side walls 172 on stack 65, completely surrounding the stacks 60, 65. Anisotropic Cl2 and/or BCl3 recipe RIE processes can be used. FIG. 8B shows the side walls 171 and 172 in cross-section. The side walls have tops slightly below the surface of the top dielectric layer 160, due to slight over-etching to ensure total removal from the surface 173 of the structure 99.
FIG. 9 illustrates a dielectric fill-in process. The process involves formation of a low-temperature liner oxide, a silicon nitride layer or silicon oxide layer (not shown), using a process temperature less than about 200 degrees C., over the phase change material side walls. One suitable process is to apply silicon dioxide using plasma enhanced chemical vapor deposition PECVD. After formation of the liner, the dielectric fill 180 is implemented using a higher temperature process such as high-density plasma HDP CVD of silicon dioxide or other similar material.
As illustrated in FIG. 10, an oxide chemical mechanical polishing CMP process is applied to planarize the structure, and to expose the tops 181, 182 of the GST side walls 171, 172. The top dielectric layer on the electrode stack ensures that the CMP does not touch the top electrode material, such as TiN, and protects it from following RIE processes or other etching steps.
FIG. 11 illustrates photoresist pattern trimming for the purpose of forming a sub-lithographic mask to trim the side walls 171, 172. A photo resist pattern is formed using lithographic techniques involving transferring a pattern from a mask or set of masks to the photoresist layer, including a rectangular extension 185 over the stack 60 and a rectangular extension 186 over the stack 65 as shown in dotted line outline. The width W1 of the extensions 185, 186 after development of the photoresist, is close to the minimum feature size for the lithographic process utilized to form the pattern extension 185, 186. Next, the width W1 of the extensions 185, 186 is reduced to sub-lithographic width W2 by photoresist trimming to leave a narrow trimmed mask 187, 188. For example, the photoresist is etched anisotropically using an oxide plasma to trim the width and thickness of the patterned photoresist, down to a width W2 less than 50 nanometers in exemplary embodiments, and to width W2 of for example about 40 nanometers in a 0.2 micron (200 nanometer) minimum feature size lithographic process environment.
In an alternative embodiment, a hard mask layer (not shown), such as a low temperature deposited layer of SiN or SiO2, can be put between the photoresist pattern and the surface of the stacks 60, 65, to prevent etching damage of the cell, if the photoresist is not thick enough after the trimming process, or selective etching of the GST and the hard mask is improved by the hard mask.
FIG. 12A illustrates side wall cell width etching in plan view according to the trimmed mask 187, 188, using for example a chlorine based reactive ion etch so that the dielectric fill 180 is not etched. The etch removes the exposed GST, leaving a narrow side wall pin 124 shown in cross-section in FIG. 12B, on the electrode stack. A seam 190 around the stack 60 and the stack 65 is left in the dielectric layer 180, which preferably extends to the top surface 173 of the structure 99 with complete removal of the GST. In embodiments of the process, all the GST in the seam 190 need not be removed. Rather it is sufficient that a significant portion of the GST in the seam 190 is removed, so that current between the bottom and top electrodes is concentrated in a narrow pin on the inter-electrode dielectric layer of the stack.
FIG. 13 illustrates the next step in the process in plan view, which involves removal of the trimmed photoresist mask (187, 188) and hard mask layer (if any). The side wall pin 124 on stack 60 and side wall pin 124A on stack 65 have a sub-lithographic width W on the order of 40 nanometers or less in embodiments of the process.
FIG. 14 illustrates the small seam fill-in and oxide deposition step. The small seams 190 (FIG. 13B) left by the removal of the side walls can be filled with electrical and/or thermally insulating fills 193, 194, using atomic layer deposition. In representative embodiments, atomic layer deposition is used to deposit dielectric material such as AlO2, HfO2 and the like. In other embodiments, the seams can be filled by spin coating silicon oxide using inorganic spin on glass or “low K” material. In an alterative, the seams are sealed to form a void that is substantially evacuated, to provide good thermal isolation for the cells. Next, a top oxide deposition covers the electrode stacks with a layer 195 of dielectric, which is planarized in preparation for subsequent metallization. The top oxide layer is preferably formed by PECVD, or other lower temperature process.
FIG. 15 shows via formation and metallization for bit lines and contacts to the memory cells. Vias are etched in the layer 195 and filled with tungsten or other conductor material to form plugs 196 and 197 making contact to the top electrode layer 123 in stack 60 and top electrode layer 123A in stack 65. A patterned metal layer 198 provides bit lines extending in the plane of the drawing to decoding circuits. As described above, the plugs 120 and 141 provide contacts between the respective bottom electrodes of the stacks 60 and 65, to the drains 115 and 117 of the access transistors. The word lines 113, 114 are formed by the polysilicon gates on the access transistors. The common source doped region 116, and source line 119, provides for sensing current flow from the bit lines through the memory cells, to the access transistor and down the common source line.
FIG. 16 shows a cross-section of an electrode stack, such as stack 60, after deposition of the GST layer side wall etching, in an alternative embodiment, in which the GST layer is only partially etched around the periphery of the electrode stack, leaving a residual layer 203 in the bottom of the seam (190, see FIG. 12B) around the stack. In the embodiment of FIG. 16, the pin 201 has a sub-lithographic width where it contacts the top electrode layer 202, extending into the inter-electrode dielectric layer, so that current flow is concentrated in the narrow region 210 of the phase change material pin.
The phase change material pin described above, and the process for making it, are representative of technologies using nano-scale structures as described herein. FIGS. 17-25 illustrate a sequence of stages in another representative process for making a small pin. FIG. 17 illustrates a substrate that comprises a silicon wafer 300 having a first layer 301 of material and a second layer 302 of material formed thereon. In the illustrated example, the first layer 301 of material comprises “material B” and the second layer 302 of material comprises “material A,” wherein the two materials are selected so that they can be selectively etched. Representative materials include silicon nitride and silicon dioxide in integrated circuit, flat panel display and related manufacturing processes. Representative thicknesses for the first layer 301 of material range from about 50 to about 500 nanometers, and more specifically can be around 200 nanometers of the silicon nitride for the example shown. Representative thicknesses for the second layer 302 of material likewise range from about 50 to about 500 nanometers in some practical examples, and more specifically can be about 220 nanometers of silicon dioxide for the example shown.
FIG. 18 illustrates a second stage in the representative process. In this stage, the second layer 302 of material is etched according to a pattern, down to the surface 307 of the first layer 301 of material, leaving a structure 303 with a side wall 305 in the second layer 302 of material and a structure 304 with a side wall 306 in the second layer 302 of material.
FIG. 19 illustrates a third stage in the representative process. In this stage, a layer 308 of sidewall material is formed over the structures 303, 304 and the surface 307 of the first layer 301 of material, and conformal with the side walls 305, 306 in the second layer 302 of material. The sidewall material comprises a phase change material as described above in one embodiment. In other embodiments, the sidewall material comprises a metal, such as aluminum, tungsten, copper, titanium, titanium nitride, tantalum, tantalum nitride, gold, and platinum and other metals, metal compounds and metal alloys. In other embodiments, the sidewall material comprises a semiconductor, such as silicon, germanium, gallium nitride, and other compounds. In other embodiments, the sidewall material comprises a non-metal, such as aluminum oxide, titanium oxide, hafnium oxide, or other high-K, and thermally and electrically insulating materials. Materials which can be used for the sidewall material include conductors, semiconductors, and insulators. Materials used for the side wall material can be crystalline, polycrystalline and amorphous materials. Materials used for the sidewall may be active materials such as used for memory elements, transistor gates, laser diodes, quantum well devices and the like. The thickness of the layer 308 of sidewall material depends on the particular application. In representative structures, the sidewall material on the side walls 305, 306 ranges from about 10 nanometers to about 50 nanometers. In other structures, greater or lesser thicknesses of the sidewall material may be applied.
FIG. 20 illustrates a fourth stage in the representative process. In this stage, a fill layer 309 is formed over the layer 308 of the sidewall material. The material used for the fill layer 309 may comprise “material A,” or silicon dioxide in this example. In alternative systems, the material used for the fill layer 309 is different than “material A,” including for example a material that is the same as “material B.” The material used for the fill layer 309 is preferably suitable for an etch back and planarization process used in subsequent processing as described below.
FIGS. 21 and 22 illustrate a fifth stage in the representative process. In the fifth stage, the structure of FIG. 20 is etched back and planarized, using a procedure such as chemical mechanical polishing. The resulting structure has a planar surface including a surface 310 of structure 303, the surface 312 of the sidewall material, the surface 314 of the fill material, the surface 313 of the sidewall material, and the surface 311 of the structure 304. As can be seen in FIG. 22, side walls on the ends of the trench filled by layer 309 have surfaces 315, 316 which are flush with the surface 314 of the fill material, and with the surfaces 312, 313 of the sidewall material, after the chemical mechanical polishing or other etch back and planarization steps.
FIG. 23 shows a next stage in the representative procedure, in which a layer of photoresist is deposited, patterned and developed to form a lithographic mask 320 over at least one of the side wall spacers, such as the side wall spacer having exposed surface 313. The mask 320 extends from over the surface 314 of the fill layer 309, across the surface 313 of the sidewall spacer, over the structure 304. The width of the lithographic mask 320 is defined using a lithographic process, such as a photolithographic process and preferably has the minimum feature size specified for the lithographic process utilized. For example, modern lithographic processes may have a minimum feature size ranging from about 90 to about 200 nanometers. Advanced lithographic processes may be applied to achieve small minimum feature sizes.
FIG. 24 shows a seventh stage in the representative procedure. In the seventh stage, the lithographic mask 320 is trimmed, using an anisotropic etching procedure such as oxygen based plasma etching applied to photoresist materials. As a result of the etching of the lithographic mask 320, a sub-lithographic, trimmed mask 321 is formed having a width which is less than the minimum feature size specified for the lithographic process. The width of the trimmed mask 321 for a representative embodiment is about 40 nanometers or less, where the minimum feature size is about 200 nanometers or less. As illustrated in FIG. 24, both the width and thickness of the lithographic mask 320 are trimmed, so that the trimmed mask 321 is both more narrow, and thinner than the lithographic mask. In alternative systems, a hard mask material may be utilized along with or in place of the photoresist. For example, a layer of silicon nitride may be formed and patterned using the lithographic process to provide a lithographic mask 320 comprising silicon nitride. The silicon nitride lithographic mask 320 may then be etched to form the trimmed mask 321. After forming a trimmed mask 321, the resulting structure is selectively etched to remove the sidewall material in regions not covered by the trimmed mask 321. As a result of the selective etching, seams 322 are left surrounding the fill layer 309 having a surface 314. In a preferred embodiment, the selective etching removes all of the fill material down to the surface (not shown) of the first layer 301 of material, and results in a patch 323 of sidewall material below the fill layer 309 and continuous with the pin 325 of sidewall material left on the sidewall between the structure 304 and the fill layer 309.
FIG. 25 shows an eighth stage in the procedure. In the eighth stage, the trimmed mask 321 is removed leaving small pin 325 and the seam between the fill layer 309 and the structure 304. The length of the pin is determined by the thin film thickness after etch back of the second layer 302 of material. The thickness of the pin is determined by the thin film thickness of the sidewall material on the sidewall of the structure 304. The width of the pin is determined by the sub-lithographic, trimmed mask 321 and the etching process utilized to selectively etch according to the pattern defined by the trimmed mask 321.
FIG. 26 illustrates a pin 325 manufactured as described herein. The pin 325 consists of a narrow sidewall spacer on a structure 330 that includes a patch 323 of sidewall material, and a layer of fill material 309. The top 326 of the pin 325 is flush with the surface 314 of the fill layer 309. In the illustrated structure, the pin 325 is shown on the side of the structure 330 that comprises the fill layer 309. In alternative systems, the fill layer 309 may be removed, leaving the pin 325 on the side of the structure 304 of the second layer 302 of material.
While the present invention is disclosed by reference to the preferred embodiments and examples detailed above, it is to be understood that these examples are intended in an illustrative rather than in a limiting sense. It is contemplated that modifications and combinations will readily occur to those skilled in the art, which modifications and combinations will be within the spirit of the invention and the scope of the following claims.