Methods of forming vertical field-effect transistor with self-aligned contacts for memory devices with planar periphery/array and intermediate structures formed thereby

Abstract

Methods of forming a memory device having an array portion including a plurality of array transistors and a periphery region including peripheral circuit transistor structures of the memory device, where an upper surface of the periphery region and an upper surface of the array portion are planar (or nearly planar) after formation of the peripheral circuit transistor structures and a plurality of memory cells (formed over the array transistors). The method includes forming the peripheral circuit transistor structures in the periphery region, forming the plurality of array transistors in the array portion and forming a plurality of memory cells over respective vertical transistors. Structures formed by the method have planar upper surfaces of the periphery and array regions.

Claims

1. A method of forming a structure of a memory device having an array portion including a plurality of vertical array transistors and a periphery region including peripheral circuit transistors, the method comprising: forming at least one trench in the periphery region of a substrate and at least one trench in the array portion of the substrate concurrently, the at least one trench in the periphery region producing one or more fins in the periphery region and the at least one trench in the array portion producing one or more fins in the array portion; forming each of the peripheral circuit transistors over a fin of the one or more fins in the periphery region; forming the plurality of vertical array transistors from the one or more fins in the array portion, each of the plurality of vertical array transistors having a self-aligned contact; and forming a phase change memory cell over the self-aligned contact of each of the vertical array transistors.

2. The method of claim 1, wherein an upper surface of the peripheral circuit transistors and an upper surface of the phase change memory cells are within 500 of average elevation.

3. The method of claim 1, wherein an upper surface of the peripheral circuit transistors and an upper surface of the phase change memory cells are within 200 of average elevation.

4. The method of claim 1, wherein an upper surface of the peripheral circuit transistors and an upper surface of the phase change memory cells are coplanar.

5. The method of claim , wherein forming the phase change memory cells comprises: forming a plurality of silicide contacts by silicidation of the self-aligned contact of each of the vertical array transistors; and depositing a phase-change material over each of the plurality of silicide contacts.

6. The method of claim 1, wherein forming the phase change memory cells comprises: depositing metal on the self-aligned contact of each of the vertical array transistors; forming a recess in an upper surface of the deposited metal on each of the vertical array transistors; and depositing a phase-change material over the deposited metal on each of the vertical array transistors.

7. A method of forming a structure of a memory device having an array portion including a plurality of vertical array transistors and a periphery region including peripheral circuit transistor structures, the method comprising: forming the peripheral circuit transistor structures in the periphery region; forming the plurality of vertical array transistors in the array portion, each of the plurality of vertical array transistors having a self-aligned contact; and forming a phase change memory cell over the self-aligned contact of each of the vertical array transistors, wherein forming the plurality of vertical array transistors comprises: doping an upper surface of a substrate in the array portion; forming an oxide over the array portion and the periphery region, wherein the oxide extends above an upper surface of the peripheral circuit transistor structures; forming a first plurality of trenches in the array portion, the first plurality of trenches extending through the oxide to the doped upper surface of the substrate; growing epitaxial silicon within the first plurality of trenches; forming a patterned resist material mask on the array portion, in a direction perpendicular to a direction of formation of the first plurality of trenches, and forming the resist material mask over the entire periphery region; etching unmasked areas of the oxide and unmasked areas of the epitaxial silicon within the first plurality of trenches to form a second plurality of trenches, the second plurality of trenches extending through the oxide and epitaxial silicon to the doped upper surface of the substrate, resulting in a plurality of silicon pillars in the array portion; removing the resist material; forming a gate oxide on exposed surfaces of the silicon pillars and on exposed portions of the doped upper surface of the substrate; and forming side gates on the gate oxide.

8. The method of claim 7, wherein doping the upper surface of the substrate comprises an n-type doping.

9. The method of claim 7, wherein the oxide comprises silicon dioxide.

10. The method of claim 7, wherein forming the side gates comprises: depositing a gate material on sides of the gate oxide covered silicon pillars; and etching the gate material to be recessed below a top surface of the gate oxide covered silicon pillars.

11. The method of claim 10, wherein the gate material is TiN.

12. The method of claim 10, wherein depositing the gate material comprises atomic layer deposition and etching the gate material comprises a spacer etch process.

13. The method of claim 7, wherein forming the phase change memory cells comprises: forming a plurality of silicide contacts by silicidation of the self-aligned contact of each of the vertical array transistors; and depositing a phase-change material over each of the plurality of silicide contacts.

14. The method of claim 13, wherein the phase-change material comprises germanium, antimony, and tellurium.

15. The method of claim 7, wherein forming the phase change memory cells comprises: depositing metal on the self-aligned contact of each of the vertical array transistors; forming a recess in an upper surface of the deposited metal on each of the vertical array transistors; and depositing a phase-change material over the deposited metal on each of the vertical array transistors.

16. A method of forming a structure of a memory device having an array portion including a plurality of vertical array transistors and a periphery region including peripheral circuit transistor structures, the method comprising: forming the peripheral circuit transistor structures in the periphery region; forming the plurality of vertical array transistors in the array portion, each of the plurality of vertical array transistors having a self-aligned contact; and forming a phase change memory cell over the self-aligned contact of each of the vertical array transistors, wherein forming the plurality of vertical array transistors comprises: doping an upper surface of a substrate in the array portion; forming an oxide over the array portion and the periphery region, wherein the oxide extends above an upper surface of the peripheral circuit transistor structures; forming a plurality of trenches through the oxide to the doped upper surface of the substrate; growing epitaxial silicon within the plurality of trenches to form a plurality of silicon pillars; etching the oxide to expose sides of the plurality of silicon pillars; forming a gate oxide on exposed surfaces of the silicon pillars and on exposed portions of the doped upper surface of the substrate; and forming all around gates on the silicon pillars.

17. The method of claim 16, wherein forming the all around gates comprises: depositing a gate material on the gate oxide; and etching the gate material to be recessed below a top surface of the gate oxide.

18. The method of claim 17, wherein the gate material is TiN.

19. A method of forming a structure of a memory device having an array portion including a plurality of vertical array transistors and a periphery region including peripheral circuit transistor structures, the method comprising: forming the peripheral circuit transistor structures in the periphery region; forming the plurality of vertical array transistors in the array portion, each of the plurality of vertical array transistors having a self-aligned contact; and forming a phase change memory cell over the self-aligned contact of each of the vertical array transistors, wherein forming the peripheral circuit transistor structures and forming the plurality of vertical array transistors comprises: forming a pad oxide on a silicon substrate; forming a nitride on the pad oxide; forming at least one first trench in the periphery region, for formation of shallow trench isolation regions, and a plurality of second trenches in the array portion; doping the silicon substrate in the periphery region for transistor structure formation; doping the silicon substrate in the array portion; depositing an oxide into the at least one first trench and the plurality of second trenches, and planarizing the oxide to a top surface of the nitride; forming a patterned resist material mask on the array portion, in a direction perpendicular to a direction of formation of the plurality of second trenches, and forming the resist material mask over the entire periphery region; etching unmasked areas of the array portion to form a plurality of third trenches, the plurality of third trenches extending through the nitride, and the pad oxide to the doped array portion of the silicon substrate, resulting in a plurality of silicon pillars in the array portion; removing the resist material; forming a gate oxide on exposed surfaces of the silicon pillars and on exposed portions of the doped upper surface of the silicon substrate; forming gates on the silicon pillars; removing the nitride and the pad oxide from the periphery region; and forming the self-aligned contacts by removing the nitride and the pad oxide from the array portion.

20. The method of claim 19, wherein forming the gates comprises: depositing a gate material on sides of the gate oxide formed on the silicon pillars; and etching the gate material to be recessed below a top surface of the silicon pillars, thereby forming the gates.

21. The method of claim 19, wherein the gates comprise TiN.

22. A method of forming a structure of a memory device having an array portion including a plurality of vertical array transistors and a periphery region including peripheral circuit transistor structures, the method comprising: forming the peripheral circuit transistor structures in the periphery region; forming the plurality of vertical array transistors in the array portion, each of the plurality of vertical array transistors having a self-aligned contact; and forming a phase change memory cell over the self-aligned contact of each of the vertical array transistors, wherein forming the peripheral circuit transistor structures and forming the plurality of vertical array transistors comprises: forming shallow trench isolation regions in a substrate in the periphery region; forming a gate oxide; forming a gate stack over the gate oxide in the periphery region; forming a nitride spacer on a side surface of the gate stack between the periphery region and the array portion; doping an upper surface of the substrate in the array portion; growing epitaxial silicon in the array portion, wherein the epitaxial silicon is grown to a level of an upper surface of the gate stack; forming a nitride cap over the array portion and the periphery region; forming at least one transistor structure in the periphery region; patterning the epitaxial silicon in the array portion; forming a gate oxide on exposed surfaces of the epitaxial silicon; forming gates; depositing an oxide to gap fill the array portion and planarizing the oxide to the nitride cap; and forming the self-aligned contacts for the plurality of vertical array transistors.

23. The method of claim 22, wherein forming the at least one transistor structure comprises: forming a dummy hedge between the periphery region and the array portion.

24. The method of claim 22, wherein forming the gate stack comprises: forming a polysilicon in the periphery region and array portion; forming a metal in the periphery region and array portion; forming a nitride in the periphery region and array portion; and etching the polysilicon, metal, and nitride to remove the polysilicon, metal, and nitride from above the array portion, stopping on the gate oxide.

25. The method of claim 22, wherein patterning the epitaxial silicon in the array portion comprises: forming a first plurality of trenches through the nitride and the epitaxial silicon; depositing an oxide material for gap filling the first plurality of trenches and covering the peripheral circuit transistor structures; forming a second plurality of trenches through the deposited oxide material, the nitride and the epitaxial silicon, the second plurality of trenches arranged perpendicularly to the first plurality of trenches; and removing the oxide material from above the nitride.

26. The method of claim 25, further comprising: removing the oxide material from between silicon pillars formed by the first plurality of trenches and the second plurality of trenches.

27. The method of claim 26, wherein forming the gates comprises: depositing a gate material in spaces between the silicon pillars; and spacer etching the gate material to be recessed below a surface of the silicon pillars.

28. The method of claim 26, wherein forming the plurality of self-aligned contacts comprises: forming an oxide cap over the peripheral circuit transistor structures; removing the nitride cap from above the silicon pillars; and doping an upper surface of the silicon pillars.

29. The method of claim 25, wherein forming the gates comprises: depositing a gate material in the second plurality of trenches; and spacer etching the gate material to be recessed below a surface of the epitaxial silicon.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a simplified diagram of a typical memory device.

(2) FIG. 2 illustrates a vertically oriented semiconductor transistor device.

(3) FIG. 3A illustrates an example memory cell and a memory access device (VFET).

(4) FIG. 3B is a schematic diagram of the example memory cell and a memory access device of FIG. 3A.

(5) FIG. 4Ai is a cross-sectional view of a step in the fabrication of a memory device in accordance with a first disclosed embodiment.

(6) FIG. 4Aii is a top-down view of FIG. 4Ai.

(7) FIG. 4Bi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 4Ai.

(8) FIG. 4Bii is a top-down view of FIG. 4Bi.

(9) FIG. 4Ci is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 4Bi.

(10) FIG. 4Cii is a top-down view of FIG. 4Ci.

(11) FIG. 4Di is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 4Ci.

(12) FIG. 4Dii is a top-down view of FIG. 4Di.

(13) FIG. 4Ei is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 4Di.

(14) FIG. 4Eii is a top-down view of FIG. 4Ei.

(15) FIG. 4Fi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 4Ei.

(16) FIG. 4Fii is a top-down view of FIG. 4Fi.

(17) FIG. 4Gi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 4Fi.

(18) FIG. 4Gii is a perpendicular cross-sectional view of the step shown in FIG. 4Gi, in the array portion of the memory device.

(19) FIG. 4Giii is a top-down view of FIGS. 4Gi and 4Gii.

(20) FIG. 4Hi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 4Gi.

(21) FIG. 4Hii is a perpendicular cross-sectional view of the step shown in FIG. 4Hi, in the array portion of the memory device.

(22) FIG. 4Hiii is a top-down view of FIGS. 4Hi and 4Hii.

(23) FIG. 4I is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 4Hi.

(24) FIG. 5Ai is a cross-sectional view of a step in the fabrication of a memory device in accordance with a second disclosed embodiment.

(25) FIG. 5Aii is a top-down view of FIG. 5Ai.

(26) FIG. 5Bi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 5Ai.

(27) FIG. 5Bii is a top-down view of FIG. 5Bi.

(28) FIG. 5Ci is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 5Bi.

(29) FIG. 5Cii is a top-down view of FIG. 5Ci.

(30) FIG. 5Di is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 5Ci.

(31) FIG. 5Dii is a top-down view of FIG. 5Di.

(32) FIG. 5Ei is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 5Di.

(33) FIG. 5Eii is a top-down view of FIG. 5Ei.

(34) FIG. 5Fi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 5Ei.

(35) FIG. 5Fii is a top-down view of FIG. 5Fi.

(36) FIG. 6A is a cross-sectional view of a step in the fabrication of a memory device in accordance with a third disclosed embodiment.

(37) FIG. 6Bi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 6A.

(38) FIG. 6Bii is a top-down view of FIG. 6Bi.

(39) FIG. 6Ci is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 6Bi.

(40) FIG. 6Cii is a top-down view of FIG. 6Ci.

(41) FIG. 6Di is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 6Ci.

(42) FIG. 6Dii is a top-down view of FIG. 6Di.

(43) FIG. 6Ei is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 6Di.

(44) FIG. 6Eii is a top-down view of FIG. 6Ei.

(45) FIG. 6F is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 6Ei.

(46) FIG. 6Gi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 6Fi.

(47) FIG. 6Gii is a perpendicular cross-sectional view of the step shown in FIG. 6Gi, in the array portion of the memory device.

(48) FIG. 6Giii is a top-down view of FIGS. 6Gi and 6Gii.

(49) FIG. 6H is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 6Gi.

(50) FIG. 7A is a cross-sectional view of a step in the fabrication of a memory device in accordance with a fourth disclosed embodiment.

(51) FIG. 7B is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7A.

(52) FIG. 7C is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7B.

(53) FIG. 7D is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7C.

(54) FIG. 7E is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7D.

(55) FIG. 7F is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7E.

(56) FIG. 7Gi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7F.

(57) FIG. 7Gii is a perpendicular cross-sectional view of the step shown in FIG. 7Gi.

(58) FIG. 7Hi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7Gi.

(59) FIG. 7Hii is a perpendicular cross-sectional view of the step shown in FIG. 7Hi.

(60) FIG. 7Ii is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7Hi.

(61) FIG. 7Iii is a perpendicular cross-sectional view of the step shown in FIG. 7Ii.

(62) FIG. 7Ji is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7Ii.

(63) FIG. 7Jii is a perpendicular cross-sectional view of the step shown in FIG. 7Ji.

(64) FIG. 7Ki is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7Ji.

(65) FIG. 7Kii is a perpendicular cross-sectional view of the step shown in FIG. 7Ki.

(66) FIG. 7Li is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7Ki.

(67) FIG. 7Lii is a perpendicular cross-sectional view of the step shown in FIG. 7Li.

(68) FIG. 7Mi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7Li.

(69) FIG. 7Mii is a perpendicular cross-sectional view of the step shown in FIG. 7Li.

(70) FIG. 8Ai is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 7F, in accordance with an alternative array processing method of the fourth embodiment

(71) FIG. 8Aii is a perpendicular cross-sectional view of the step shown in FIG. 8Ai.

(72) FIG. 8Bi is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 8Ai.

(73) FIG. 8Bii is a perpendicular cross-sectional view of the step shown in FIG. 8Bi.

(74) FIG. 8Ci is a cross-sectional view of a step in the fabrication of a memory device, following that shown in FIG. 8Bi.

(75) FIG. 8Cii is a perpendicular cross-sectional view of the step shown in FIG. 8Ci.

(76) FIG. 9 illustrates a processing system including a memory device, formed in accordance with one or more of the disclosed embodiments.

DETAILED DESCRIPTION OF THE DISCLOSED EMBODIMENTS

(77) In the following detailed description, reference is made to the accompanying drawings which form a part hereof, and in which are shown by way of illustration specific embodiments that may be practiced. It should be understood that like reference numbers represent like elements throughout the drawings. These example embodiments are described in sufficient detail to enable those skilled in the art to practice them. It is to be understood that other embodiments may be utilized, and that structural, material, and electrical changes may be made, without departing from the scope of the invention, only some of which are discussed in detail below.

(78) The term substrate used in the following description may include any supporting structure including, but not limited to, a semiconductor substrate. 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, including those made of semiconductors other than silicon. 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 also 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.

(79) Embodiments of the invention relate to methods of forming memory devices having self-aligned contacts of the VFETs and that results in the periphery and array portions of the memory cell being nearly planar with each other after formation of these portions of the memory device. Several examples of processes are discussed herein; however one of skill in the art would recognize that changes may be made in the processes as long as the nearly planar surface of the resulting memory array and peripheral circuit transistor structures is maintained. In one embodiment, the average elevations of the surfaces of the periphery and array portions of the memory cell are within 500 of each other. More preferably, the average elevations of the surfaces of the periphery and array portions of the memory cell are within 200 of each other. The VFETs of the memory array may be arranged to have a common N+ source. In other words, the VFETs may be tied together on the bottom and have a common voltage (e.g., ground). The memory elements may be connected together by a series of bit lines arranged perpendicularly to the wordlines, giving unique memory addresses of a wordline and bitline in a cross-point array.

(80) In a first embodiment, a planar periphery and array can be accomplished by first forming the periphery transistors, followed by formation of the array memory elements and access devices. Particulars of the method are now discussed with respect to FIGS. 4Ai to 4Hiii.

(81) In FIGS. 4Ai and 4Aii, pad oxide 401 and nitride 402 are deposited on a substrate 400. FIG. 4Ai shows a cross-sectional view corresponding to both a periphery and array portion of the memory device. FIG. 4Aii shows a top-down view of the section of FIG. 4Ai corresponding to the array portion of the memory device. Subsequent figures are arranged in this same manner, unless otherwise noted. The pad oxide 401 may be formed, for example, of silicon dioxide. The nitride 402 may be, for example, silicon nitride. These materials may be formed by any known methods of deposition, such as chemical vapor deposition (CVD) or atomic layer deposition (ALD).

(82) In FIGS. 4Bi and 4Bii, trenches 405 are formed in the periphery region for use as shallow trench isolation (STI) regions 406 (FIG. 4Ci). The trenches may be formed using any appropriate etching process known in the art for formation of STI regions.

(83) Transistors 410 are then formed in the periphery region and the STI regions 416 are filled with, for example, silicon dioxide. The silicon surface of the array region is also provided with heavy n-type doping 403. Formation of the transistors in the periphery region in this embodiment is by any conventional method known in the art. The resulting structure is shown in FIGS. 4Ci and 4Cii.

(84) As shown in FIGS. 4Di and 4Dii, silicon dioxide 415 is deposited over both the periphery and array regions and planarized. The thickness of the silicon dioxide 415 is such that it extends above an upper surface of the periphery transistors 410.

(85) Trenches are then formed in the silicon dioxide 415 in the array portion of the memory device along the y-direction. These trenches are formed completely through the silicon dioxide 415 to the n+ doped silicon 403. Epitaxial silicon 420 is grown in each trench using, for example, a selective epitaxial growth method. The resulting structure is seen in FIGS. 4Ei and 4Eii. As previously mentioned, FIG. 4Eii is a top-down view of the array portion of FIG. 4Ei. In one example embodiment, the preferred y-direction is in the direction of the <100> direction on a (100) silicon wafer, as described in U.S. Pat. No. 7,498,265, incorporated by reference herein.

(86) As shown in FIGS. 4Fi and 4Fii, photo patterning is used to form resist 422 along the x-direction in the array portion of the memory device. Then, a dry etch of the unmasked oxide 415 and silicon 420 down to the n+ doped silicon 403 in the array forms silicon 420 and oxide 415 pillars along the x-direction. This can be seen in FIGS. 4Gi, 4Gii and 4Giii, where FIG. 4Giii shows a top-view of cross-sectional FIGS. 4Gi and 4Gii, along the x-direction and y-direction, respectively. The periphery portion is protected by the resist 422 and a hard mask. After the dry etch, the resist pattern is removed.

(87) Next, as shown in FIGS. 4Hi, 4Hii and 4Hiii, a gate oxide 425 is formed on the exposed silicon pillar surfaces 420 and on exposed surfaces of the doped silicon 403. This is followed by deposition and etch of TiN, such that the TiN is recessed to below the top surface of the silicon/oxide pillars, thereby forming TiN side gate 430 (e.g., gate 150, FIG. 3A) on the gate oxide 425. Specifically, the TiN may be formed by ALD and spacer etch. The TiN may also be deposited by CVD, PECVD or PVD. As with FIGS. 4Gi, 4Gii and 4Giii, FIG. 4Hiii shows a top-view of cross-sectional FIGS. 4Hi and 4Hii, along the x-direction and y-direction, respectively. Following this step, the array is gap filled with an oxide deposit.

(88) Finally, as shown in FIG. 4I, the memory cell is formed. The silicon pillar 420 is recessed below the surface of the silicon dioxide 415. This provides an area for formation of the self-aligned contact of the VFET. A top surface of the silicon pillar 420 is n+ doped 435 to form a drain region for the VFET (e.g., drain 140, FIG. 3A). This is followed by silicidation to form a silicide (e.g., CoSi.sub.2 or TiSi.sub.2) contact 436 (e.g., contact 240, FIG. 3A). A phase-change material 440, such as GST, is deposited (by known methods, such as CVD or ALD) in the contact recess to form a confined cell. Alternatively, an additional dielectric material may be deposited followed by contact etch and CVD or ALD phase-change confined cell formation. Additionally, metal can be deposited in the contact with subsequent recess formation and phase change material deposition to form a confined cell with bottom electrode (e.g., electrode 224, FIG. 3A).

(89) At this stage, conventional back-end-of-line processes may proceed without any problems of step-change between the array and the peripheral circuit transistor structures.

(90) In a second embodiment, a planar periphery and array can be accomplished by first forming the periphery transistors, followed by formation of the array memory elements and access devices, similar to in the first embodiment. The second embodiment further includes an all-around gate on the access transistors. Particulars of the method are now discussed with respect to FIGS. 5Ai to 5Fii.

(91) Similar to the method described with respect to FIGS. 4Ai to 4Hiii, the method of the second embodiment begins with deposition of pad oxide 501 and nitride 502 on a substrate 500 followed by formation of trenches 505 in the periphery region for use as shallow trench isolation (STI) regions 406 (FIG. 5Bi), as shown in FIGS. 5Ai and 5Aii. FIG. 5Ai shows a cross-sectional view corresponding to both a periphery and array portion of the completed memory device and FIG. 5Aii shows a top-down view of the section of FIG. 5Ai corresponding to the array portion of the completed memory device.

(92) As shown in FIGS. 5Bi and 5Bii (and as described in more detail above with respect to FIGS. 4Ci and 4Cii), transistors 510 are formed in the periphery region, the STI regions 506 are filled with, for example, silicon dioxide, and the upper surface of the substrate is n+ doped to form n+ doped region 503. As shown in FIGS. 5Ci and 5Cii (and as described in more detail above with respect to FIGS. 4Di and 4Dii), silicon dioxide 515 is deposited over both the periphery and array regions and planarized. As discussed with respect to the previous embodiment, the depth of the silicon dioxide 515 is such that it extends above a surface of the transistors 510. FIGS. 5Bi and 5Ci show a cross-sectional view corresponding to both a periphery and array portion of the completed memory device and FIGS. 5Bii and 5Cii show a top-down view of the section of FIGS. 5Bi and 5Ci, respectively, corresponding to the array portion of the completed memory device.

(93) As shown in FIGS. 5Di and 5Dii, contact etch can be used to etch columns completely through the silicon dioxide 515 down to the n+ doped silicon 503. This is followed by epitaxial silicon growth, for example using a selective epitaxial growth method, to form silicon pillars 520.

(94) As shown in FIGS. 5Ei and 5Eii, an oxide etch, for example a wet etch, can be used to selectively remove the silicon dioxide 515. A mask (not shown) in used to protect the periphery circuit transistor structures during this process. Gate oxide 525 is formed on the exposed surfaces of the silicon pillars 520. This is followed by deposition and etching of TiN, such that the TiN is recessed below the surface of the silicon pillars 520. As can be seen in FIG. 5Eii, because the spacing of the pillars 520 is narrower in the x-direction than it is in the y-direction (and as long as the TiN thickness is not greater than half the distance of the spacing in the y-direction) the TiN gate is merged in the x-section, forming a continuous surrounding TiN side gate 530 on the gate oxide 525, that is self-aligned and isolated from the adjacent gate in the y-direction. As described above, the TiN may be formed by ALD and spacer etch. The TiN may also be deposited by CVD, PECVD or PVD.

(95) As shown in FIGS. 5Fi and 5Fii, silicon dioxide is deposited and planarized to even the surface of the array. Then, as described in more detail above with respect to FIG. 4I, the silicon pillars 520 are recessed (to provide an area for formation of the self-aligned contact of the VFET) and formation of the memory element is completed above the silicon pillars 520. Conventional back-end-of-line processes may then proceed.

(96) Each of the first and second embodiments described above provides for a planar periphery and array, thereby simplifying subsequent production processes. They also each allow for self-aligned formation of the contacts for a phase-change memory cell. Traditional manufacturing methods did not provide these advantages. In addition to the benefits described above with respect to the first and second embodiments, a third embodiment provides an additional benefit of shared patterning steps for formation of the array and periphery, thus simplifying the manufacturing process further. Particulars of this embodiment are now described with respect to FIGS. 6A-6H.

(97) Similar to the first and second embodiments described above, the third embodiment begins with deposition of pad oxide 601 and nitride 602 on a substrate 600. Appropriate p-doping 609 of the periphery transistor area and n+ doping 613 of the array area are completed at this stage as well. Then, trenches are formed concurrently in both the periphery region (for use as shallow trench isolation (STI) regions 606) 605 and in the array 607. Formation of the array trenches 607 produces silicon fins 620. The resulting structure at this point in the process is shown in FIG. 6A.

(98) As shown in FIGS. 6Bi and 6Bii, the STI regions and array trenches are filled with silicon dioxide 615 that is planarized to the level of the silicon nitride 602. The y-direction orientation of the trenches (of FIG. 6A) can be seen more clearly in FIG. 6Bii, which is a top-down view of FIG. 6Bi in the array portion of the memory device.

(99) As shown in FIGS. 6Ci and 6Cii, photo patterning is used to form resist 622 along the x-direction in the array portion of the memory device. Then, a dry etch of the unmasked oxide 615, nitride 602 and silicon 620 down to the n+ doped silicon 603 forms silicon 620 and oxide 615 pillars along the y-direction. This can be seen in FIGS. 6Di and 6Dii. The periphery portion is protected by the resist 622 and a hard mask. After the dry etch, the resist pattern is removed. FIGS. 6Ci and 6Di show a cross-sectional view corresponding to both a periphery and array portion of the completed memory device and FIGS. 6Cii and 6Dii show a top-down view of the section of FIGS. 6Ci and 6Di, respectively, corresponding to the array portion of the completed memory device.

(100) A gate oxide 625 (see FIG. 6Gii) is formed on the exposed surfaces of the silicon pillars 620 and on exposed surfaces of the doped silicon 603. This is followed by deposition and etching of TiN, such that the TiN is recessed to below the top surface of the silicon pillars 620, thereby forming TiN gate 630 on the gate oxide 625. The particulars of these process steps are discussed in more detail with respect to FIGS. 4Gi, 4Gii, 4Giii, 4Hi, 4Hii and 4Hiii, above.

(101) As shown in FIGS. 6Ei and 6Eii, an etch, for example a wet etch, is used to remove the nitride 602 and oxide 601 in the periphery to prepare for formation of the periphery transistors 610. Next, as shown in FIG. 6F, the periphery transistors 610 are formed (by any appropriate method), having tops at the same height as the nitride 602 and oxide 615 in the array. Additional silicon dioxide 615 is deposited and planarized to even the surface of the periphery to the array.

(102) As shown in FIGS. 6Gi, 6Gii and 6Giii, the remaining silicon nitride 602 and pad oxide 601 in the array portion of the memory device are removed via an etch process. This provides a self-aligned contact to the VFET. As shown in FIG. 6H, n+ doping 635 is applied to the upper portions of the silicon pillar 620, followed by silicidation to form contact 636, and deposition of a phase-change material 640. Further details of the formation of the memory cell are discussed above with respect to FIG. 4I. Conventional back-end-of-line processes may then proceed.

(103) Particulars of a fourth embodiment are now described with respect to FIGS. 7A to 8Cii. As seen in FIG. 7A, shallow trench isolation regions 705 for the peripheral circuitry are formed in trenches formed within a substrate 700 and filled with a dielectric material, such as silicon dioxide, as is known in the art. Then, gate stack materials are formed on the substrate. Specifically, a gate oxide 725, polysilicon 730, a metal 735 and a nitride 740 are formed on a surface of the substrate 700. In one example, the polysilicon 730 may be approximately 450 thick, the metal 735 may be either tungsten silicide (WSIX) at a thickness of approximately 800 or tungsten nitride/tungsten (WN/W) at a thickness of approximately 500 , and the nitride 740 may be approximately 300-500 thick. Any other gate stack having a nitride cap is appropriate.

(104) As seen in FIG. 7B, photolithography and etching (stopping on the gate oxide) is used to remove the gate stack materials 730, 735, 740 from the array area of the memory device. A nitride spacer 745 is formed on the edge of the gate stack closest the array. This is shown in FIG. 7C. The nitride spacer 745 may be formed using a dry etch process to have a width of approximately 500-700 . Also, the upper portion of the array surface is n+ doped 703. An optional rapid thermal processing (RTP) anneal and H+ anneal at approximately 900 C. may occur following the doping of the array surface.

(105) As shown in FIG. 7D, epitaxial silicon 715 is grown on the surface of the array, for example by a selective epitaxial growth method, up to the level of the gate stack thickness. Next, a nitride cap 750 is deposited over the entire device. This nitride cap 750 may have a thickness of approximately 800 to 1200 . Optionally a pad oxide (not shown) may be deposited before the nitride deposition. Following deposition of the nitride cap, the peripheral portion of device is patterned and etched, forming the transistor gates 710 for the peripheral circuit transistor structures and a dummy hedge 711 between the periphery and the array. The dummy hedge 711 acts as a transition area between the periphery and the array. This resulting structure is shown in FIG. 7E. As seen in FIG. 7F, the peripheral transistor spacers 712 and implants 713 (to form source and drain regions) are formed according to known methods. Transistor processing is completed according to known methods.

(106) Once the peripheral transistors are formed, the array processing may begin. Two examples of the array processing according to the fourth embodiment are described herein. The first is described with respect to FIGS. 7Gi to 7Kii. The second is described with respect to FIGS. 8Ai to 8Cii

(107) In one method of array processing, trenches 712 are formed (in a line/space pattern) through the nitride cap 750 and epitaxial silicon 715 to the upper surface of the n+ doped 703 substrate. This maybe done, for example, using in situ reactive ion etching (RIE). Trench depth may be controlled in a manner similar to those known for controlling the depth of STI regions. This resulting structure is shown in FIGS. 7Gi and 7Gii. As shown in FIGS. 7Hi and 7Hii, the trenches are filled with an oxide 755. For example, CVD deposition of tetraethoxysilane (TEOS). The oxide 755 is deposited to overflow the trenches and to form an oxide 755 over the peripheral circuit transistor structures.

(108) As seen in FIGS. 7Ii and 7Iii, additional trenches 713 are formed, in a line/space pattern perpendicular to that formed in the step illustrated in FIG. 7Gi, to about the same depth as this first pattern (e.g., to the n+ doped surface 703 of the substrate). These trenches 713 may also be formed by in situ RIE. A gate oxide 760 is formed on exposed silicon surfaces within the trench 713. Metal gate material is deposited on the gate oxide 760 and etched to about 500 below a surface of the trench 713 to produce side gates 765. The side gate may be formed, for example, of TiN using ALD and spacer etch. During these processes, the periphery is protected by the oxide and nitride.

(109) In FIGS. 7Ji and 7Jii, the array is gap filled with an oxide 770. Optionally, the lightly doped source region of the VFET may be implanted prior to the deposition of the oxide 770. Once the oxide deposition is complete, a stop-on-nitride CMP process occurs, to remove any excess oxide and remaining portions of the oxide 755.

(110) In FIGS. 7Ki and 7Kii, a protective oxide cap 780 is formed over the periphery portion of the memory device. This oxide cap 780 may have a thickness of about 100-300 . The oxide cap 780 acts to protect the periphery during removal of the nitride cap 750 from the array portion the device, as seen in FIGS. 7Li and FIG. 7Lii. The removal of the nitride cap 750 creates a recess 785 for formation of the self-aligned memory contact of the VFET. The protective oxide cap 780 is then removed.

(111) In FIGS. 7Mi and 7Mii the recess may be, optionally, cleaned, for example by a wet clean, to shape the contact. The upper surface of the epitaxial silicon in the recess is doped, forming the drain region of the VFET. If the doping of the source region of the VFET was not completed earlier in the process, this is done at this stage as well. Then, as described above with respect to FIG. 4I, this is followed by silicidation (or metal deposition) to form the contacts and deposition of a phase-change material for the memory cell. Conventional back-end-of-line processes may then proceed.

(112) Proceeding from the structure shown in FIGS. 7Gi and 7Lii, the second option of array processing in the fourth embodiment will now be described. As described above with respect to FIGS. 7Gi and 7Gii a line/space pattern of trenches 712 is formed in the y-direction and these trenches 712 are filled with oxide material 755. Then a line/space pattern of trenches 713 is formed perpendicular (in the x-direction). A pre-gate oxide clean is performed, which removes the oxide 755 not only from above the nitride cap 750 but also from between the epitaxial pillars 820. This results in the structure shown in FIGS. 8Ai and 8Aii.

(113) A gate oxide 860 is formed on exposed silicon surfaces of the pillars 820. A metal gate material is deposited on the gate oxide 860 and etched to form gates 865. The metal gate material is etched to about 500 below a surface of the silicon pillars 820 to produce the side gates 865. The gate may be formed, for example, of TiN using ALD and spacer etch. Additionally, as can be seen in the figures, the metal gate thickness may contact neighboring material in the x-direction (FIG. 8Bi), but not in the y-direction (FIG. 8B ii). As described previously, during these processes, the periphery is protected by the oxide and nitride, which are removed after the gate formation and are not seen in FIGS. 8Bi and 8Bii. In FIGS. 8Ci and 8Cii, the array is gap filled with an oxide 875, and planarized to remove any excess oxide and remaining portions of the oxide 755, as described in more detail above with respect to FIGS. 7Ji and 7Jii. A protective oxide cap 780 is formed over the periphery portion of the memory device, as described above with respect FIGS. 7Ki and 7Kii. The remainder of the process follows that as described with respect to FIGS. 7Li to 7Mii. Further details of the formation of the memory cell are discussed above with respect to FIG. 4I. Conventional back-end-of-line processes may then proceed.

(114) It should be appreciated that the memory devices described herein may be fabricated as part of an integrated circuit. The corresponding integrated circuits may be utilized in a processor system. For example, FIG. 9 illustrates a simplified processor system 1700, which includes a memory device 1702 that includes a memory device constructed in accordance with any of the above described embodiments. A processor system, such as a computer system, generally comprises a central processing unit (CPU) 1710, such as a microprocessor, a digital signal processor, or other programmable digital logic devices, which communicates with an input/output (I/O) device 1720 over a bus 1790. The memory device 1702 communicates with the CPU 1710 over bus 1790 typically through a memory controller.

(115) In the case of a computer system, the processor system 1700 may include peripheral devices such as removable media devices 1750 (e.g., CD-ROM drive or DVD drive) which communicate with CPU 1710 over the bus 1790. Memory device 1702 can be constructed as un-integrated circuit, which includes one or more phase change memory devices. If desired, the memory device 1702 may be combined with the processor, for example CPU 1710, as a single integrated circuit.

(116) It should also be appreciated that various embodiments have been described as using a phase-change material as an example resistance variable material. The invention may also be used in other types of resistive memory to improve and simplify manufacturing regardless of the resistance variable material used. The invention may also be used for other types of memories, such as RRAM, FGRAM, MRAM, STTRAM, etc.

(117) The above description and drawings should only be considered illustrative of example embodiments that achieve the features and advantages described herein. Modification and substitutions to specific process conditions and structures can be made. Accordingly, the invention is not to be considered limited by the foregoing description and drawings, but is only limited by the scope of the appended claims.