Wafer bonding in 3D NOR memory circuit fabrication
The CMOS device and interconnection layer are fabricated on a separate semiconductor substrate through wafer bonding technology and bonded at high temperatures, solving the problem of high-temperature thermal budget limitations and achieving efficient three-dimensional NOR memory structure integration and performance improvement.
Patent Information
- Application Number
- CN201980049198.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-09-23
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2039-09-23
AI Technical Summary
Prior art In manufacturing three-dimensional NOR memory structures, high temperature thermal budgets limit the design choice of CMOS circuits and interconnect layers, resulting in increased signal delay and reduced performance.
Through wafer bonding technology, the CMOS device and the interconnect layer are fabricated on a separate semiconductor substrate and bonded together by flip-chip technology to avoid high temperature limitations and remove one substrate after bonding to form the interconnect layer. Choose a suitable material such as aluminum or copper to reduce temperature requirements.
It realizes the successful integration of the three-dimensional NOR memory structure without limiting the design selection of CMOS devices and interconnect layer, reducing signal delay and improving memory performance.
Smart Images

Figure CN112514067B_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application is related to and claims priority to U.S. Provisional Patent Application (“Provisional Application I”), Serial No. 62 / 735,678, filed on September 24, 2018, entitled “Wafer Bonding in Three-Dimensional NOR Memory Circuit Fabrication.”
[0003] This application is also related to U.S. Patent Application Serial No. 16 / 012731, filed on June 19, 2018, entitled “Three-Dimensional NOR Memory Array Structure and Method for Manufacturing the Same” (“Co-Pending Application”). This application is also related to U.S. Provisional Patent Application Serial No. 62 / 735662, filed on the same date as this application (“Provisional Application II”), entitled “Epitaxial Single Crystal Channel of Memory Transistor in Three-Dimensional Memory Structure and Method for Forming the Same”. The disclosures of the Co-Pending Application and Provisional Applications I and II are incorporated herein by reference in their entirety. Background of the Invention Technical Field
[0005] The present invention relates to wafer bonding technology in integrated circuit manufacturing, and in particular to wafer bonding technology for high-density three-dimensional memory circuits. Background Art
[0006] Wafer bonding is a technique used in the manufacture of many semiconductor devices. In wafer bonding, two wafers of equal or nearly equal area are joined, for example, by thermocompression, adhesive, anodic, or thermal techniques. Generally, after bonding, all or substantially all of the substrate is removed from one or both wafers.
[0007] The aforementioned co-pending application discloses a three-dimensional memory structure formed on top of a single crystal semiconductor substrate, which is organized as an array of NOR memory strings. In this context, a "NOR memory string" refers to a group of thin film memory transistors that share source and drain regions. Figure 1 A cross-section of a memory structure 30 is shown, which includes an array of NOR memory strings formed over a CMOS (complementary metal oxide semiconductor) circuit and interconnect layers on a semiconductor substrate. Figure 1 As shown, the memory structure 30 is formed on a substrate 150. For example, a semiconductor wafer known to those skilled in the art for making electronic circuits can be used as a suitable substrate. Non-semiconductor substrates, such as silicon dioxide, can also be used.
[0008] Various circuit elements (such as Figure 1 1 and 1 . A CMOS circuit 10 is represented by CMOS transistors shown, which are interconnected by contacts or vias 16 via conductors 22 (e.g., copper) of a conventional interconnect layer. These circuit elements are fabricated on a semiconductor substrate using conventional techniques prior to forming a memory structure 30. The circuits, collectively referred to herein as interconnect layer 20, are typically embedded in an insulating layer and may include conductors to support the operation of a memory array in the memory structure 30. The memory structure 30 is formed on interconnect layer 20. For example, interconnect layer 20 provides conductors 24 (global word lines) to connect conductors 32 (e.g., heavily-doped polysilicon) used as word lines for addressing each memory transistor in the memory structure 30. Conductors 32 are referred to in this description as local word lines.
[0009] like Figure 1 As shown, the memory structure 30 includes a plurality of stacks of active strips (such as active stacks 101a, 101b, and 101c). For example, Figure 1 Stacks 101 a , 101 b and 101 c are shown, each stack comprising four active strips, the active stacks being insulated from each other by an insulating layer 106 . Figure 1 A cross section of an active strip is shown, with the strip extending longitudinally into and out of the plane of the cross section. In this context, the active strip comprises a drain layer 104, a source layer 103, a body layer 102, and a channel layer 108. (The channel layer 108 is provided on both sides of the body layer 102). In some embodiments, both the drain layer 104 and the source layer 103 are n+ polysilicon, while the channel layer 108 is p- polysilicon and the body layer 102 is p+ polysilicon. In some embodiments, a dielectric layer may be used in place of the body layer 102. As shown in FIG. Figure 1 The active strips shown also include conductive layers 105c and 105b, which are adjacent to the source layer 103 and the drain layer 104, respectively, for reducing the resistance along the length direction of the source layer 103 and the drain layer 104. Along each side of each stack of active strips (that is, along each side of each active strip), a charge-trapping layer 107 and a plurality of local word lines 32 are disposed. Figure 1Among them, the local word line 32 is a conductive fence on both sides of the stack along the active strip. A storage transistor can be formed by the local word line, a portion of the channel layer 108, a portion of the charge trapping layer 107 therebetween, and the source layer 103 and the drain layer 104. Multiple storage transistors formed along the active strip share the drain layer 104 and the source layer 103. Adjacent storage transistors along an active strip that share the source and drain layers form a NOR memory string. (Turning on any storage transistor in the NOR memory string will result in a conductive transistor current between the common source and drain layers.)
[0010] The order of the processes imposes limitations on CMOS devices, interconnect layers, and memory devices. For example, memory devices typically require a thermal budget of at least 750 degrees Celsius for several hours during manufacturing because of the need to form an ONO (oxide-nitride-oxide) multilayer structure or stack using low pressure chemical vapor deposition (LPCVD). In the ONO multilayer structure, the required oxide and nitride may be high temperature oxide (HTO; or silicon dioxide) and silicon nitride (SiN), respectively. In addition, a layer of aluminum oxide (Al2O3) may be required as a blocking oxide in the ONO stack. However, from an electrical property point of view, crystallizing aluminum oxide (Al2O3) - which produces the required Al2O3 - requires an annealing temperature of more than 900 degrees Celsius. However, a manufacturing temperature exceeding 350 degrees Celsius will prevent copper from being used for the horizontal interconnect layer 20 embedded in the associated low-k dielectric film, even when tungsten is used in the vertical interconnect 16 to connect the copper horizontal interconnect. Similarly, fabrication temperatures exceeding 500 degrees Celsius will preclude the use of aluminum interconnect layers. When fabrication temperatures exceed 500 degrees Celsius, tungsten may be one of the options for the interconnect layer material. However, tungsten has a higher resistance, as shown in Table 1 below. The resulting increase in interconnect resistance increases signal delay, which adversely affects memory device performance.
[0011]
[0012] Table 1
[0013] The thermal budget process of the memory device limits the underlying CMOS circuit (e.g., CMOS transistor 10) in at least two ways. First, cobalt silicide or other high temperature contact materials, such as tungsten or tungsten silicide, must be selected for the gate, source / drain metallization 12 in the CMOS transistor 10 to allow the maximum manufacturing temperature to be increased to 750 degrees Celsius. Although cobalt silicide has relatively low sheet and contact resistance values compared to silicon, due to the large amount of silicon consumed in the silicidation step and the interface roughness between cobalt silicide and silicon, a deeper doped junction is required in the silicon material. Conversely, transistors with short channel lengths require shallower doped junctions to reduce leakage current. Although nickel silicide is commonly used in the source and drain contacts of the current generation of small transistors, cobalt silicide is preferred over nickel silicide because nickel silicide cannot withstand temperatures above 450 degrees Celsius. When the temperature is above 450°C, the nickel silicide film will agglomerate on the silicon, destroying the low sheet resistance and low contact resistance characteristics of the nickel silicide film.
[0014] Second, for shallow junctions and narrow channel devices, temperatures above 600°C should be avoided after junction formation to prevent dopants from diffusing out of the source and drain junctions.
[0015] Therefore, a fabrication method is desired that allows for the integration of optimal CMOS devices and interconnect layers with three-dimensional NOR memory structures (e.g., Figure 1 The memory structure 30) can be integrated without limiting the design choices of the CMOS devices and interconnect layers due to the thermal budget of the memory structure. Summary of the invention
[0016] According to one embodiment of the present invention, the memory array and the single crystal circuit are provided in the same integrated circuit by wafer bonding (eg, adhesive wafer bonding or anodic wafer bonding) and interconnected by conductors of an interconnect layer.
[0017] Additional circuits or memory arrays may be provided by additional wafer bonding and electrically connected by interconnect layers at the wafer bonding interface.
[0018] According to one embodiment of the present invention, a memory array may include storage or memory transistors having single crystal epitaxial silicon channel material.
[0019] The present invention is better understood after considering the following detailed description in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 A cross section of a memory structure 30 is shown, the memory structure comprising an array of NOR memory strings formed on top of CMOS circuitry and interconnect layers on a semiconductor substrate.
[0021] Figure 2AA semiconductor substrate 100 is shown on which interconnect layers and CMOS devices are fabricated for supporting a three-dimensional NOR memory structure.
[0022] Figure 2B A three-dimensional NOR memory structure (ie, memory structure 30 ) is shown fabricated solely on a semiconductor substrate 110 , above an insulating oxide (eg, silicon dioxide) layer 120 .
[0023] Figure 3A Two devices A' and B' to be fabricated on wafer substrates A and B are shown being wafer bonded.
[0024] Figure 3B The display devices A' and B' are perfectly aligned and electrically connected after wafer bonding.
[0025] Figure 3C Devices A' and B' are shown electrically connected after wafer bonding, with a 250 nm wide misalignment.
[0026] Figure 3D The use of bonding metal C is shown to facilitate the bonding of wafers A and B at devices A' and B' (ie, bonding at points of mutual contact).
[0027] Figure 4 Show, use Figure 2A and 2B As an example, the substrate 110 of the memory structure 30 may be removed.
[0028] Figure 5 An interconnect layer 40 is shown formed over the memory structure 30 .
[0029] Fig. 6A , 6B 6C, 6D, 6E, and 6F show the steps of making CuA and CoA circuits using chip bonding technology.
[0030] Fig. 7A , 7B 7C depicts the use of wafer bonding of substrates 300 and 350 to connect memory blocks 310 and 360.
[0031] Fig. 8A , 8B 8C, 8D, and 8E depict a process according to an embodiment of the present invention, wherein a CuA type CMOS transistor is placed under a memory block having an epitaxial single crystal silicon channel therein.
[0032] For clarity of presentation and to allow cross-reference between the figures, like elements in the figures have been assigned like reference numerals. DETAILED DESCRIPTION
[0033] According to one embodiment of the present invention, rather than fabricating CMOS devices (eg, CMOS transistor 10) and interconnect layers (eg, interconnect layer 20) on the same silicon substrate as memory structure 30, the CMOS devices and interconnect layers are fabricated on separate semiconductor substrates. Figure 2A 1 shows a semiconductor substrate 100 on which an interconnect layer 20 and a CMOS device 10 for supporting a three-dimensional NOR memory structure are fabricated. Figure 2B As shown, a three-dimensional NOR memory structure (ie, memory structure 30) is fabricated separately on a substrate 110 on an isolation oxide (eg, SiO2) layer 120. Both substrates 100 and 110 may be provided by silicon wafers, as known to those skilled in the art.
[0034] After performing the required fabrication steps on each of the semiconductor substrates 100 and 110, the wafers are bonded together using a "flip chip" technique, where the surface of the semiconductor substrate 100 having the interconnect layer 20 is bonded to the surface of the semiconductor substrate 110 having the memory structure 30. In this manner, fabrication of the interconnect layer 20 and the CMOS device 10 is not limited by the high temperatures that are optimal for fabricating the memory structure 30.
[0035] The wafers are bonded such that the contacts in the interconnect layer 20 are electrically connected to corresponding contacts of the memory structure 30. Photolithographic alignment marks in each respective substrate allow alignment of the target bonding points with minimal mismatch. Figures 3A-3D Demonstration of exemplary wafer bonding of substrates A and B at designated wafer bonding points. Figure 3A Two devices A' and B' fabricated on wafer substrates A and B are shown being wafer bonded. For example, devices A' and B' may be 300 nm wide conductors in an interconnect system. Figure 3B Devices A' and B' are shown fully aligned and electrically connected after wafer bonding. Figure 3C Devices A' and B' are shown electrically connected after wafer bonding, with a 250nm wide misalignment. (In state-of-the-art wafer bonding techniques, alignment accuracy can be achieved within ±250nm). The bonding of substrates A and B can be performed using any suitable wafer bonding technique, such as hot pressing, anodic, plasma activated, eutectic or surface activated wafer bonding. Of these techniques, anodic wafer bonding is preferred. With anodic wafer bonding, the two wafer substrates are brought into contact and an electrostatic field is applied, thereby electrically and physically connecting the devices.
[0036] According to one embodiment of the present invention, Figure 3DAs shown, bonding metal C can be used to promote bonding between wafers A and B at their mutual contact points (i.e., devices A' and B'). For example, bonding metal C can be chromium, titanium, indium or alloys thereof, or any suitable material. Those skilled in the art understand the principles and mechanisms of wafer bonding, so detailed discussion is omitted in this description.
[0037] After bonding, one substrate may be removed. Figure 4 Show, use Figure 2A and 2B As an example, the substrate 110 of the memory structure 30 can be removed. The substrate 110 can be removed by any suitable wafer thinning technique, such as laser lift-off, mechanical polishing, or chemical etching. In one embodiment, mechanical polishing can be combined with chemical etching to remove the substrate 110. By combining mechanical polishing and chemical etching, the substrate 110 can be removed at a lower cost and with higher accuracy (i.e., without damage to the memory structure 30) than using either technique alone. For example, assume that the substrate 110 is 500 microns thick. Then, the mechanical polishing can first remove about 480 microns from the substrate in an initial step, leaving about 20 microns of the substrate 110 to be removed by chemical etching.
[0038] In wafer thinning techniques by mechanical polishing, the wafer is rotated about its center on a grinding surface. Thinning the wafer by mechanical force is sometimes referred to as "grinding" when the resulting substrate surface is rough, and "polishing" when the resulting substrate surface is smooth. Either grinding or polishing methods may be used, or any combination thereof. After the mechanical grinding or polishing steps are completed, chemical etching may remove the remaining 20 microns of the memory structure 30.
[0039] Chemical etching of the substrate 110 may be accomplished using any suitable chemical process. Examples of suitable chemical agents for the silicon substrate 110 include KOH, TMAH, HF + , HNO3 or HF + and NH4F. The oxide layer between the silicon substrate 110 and the memory structure 30 may be used as an etch stop layer. Figure 4 As shown, the etch stop layer 120 ensures that the memory structure 30 remains intact and is not damaged by the chemical etching of the substrate 110. When an oxide layer (e.g., SiO2) is used as an etch stop layer, KOH can be used because it etches silicon about 500 times faster than it etches SiO2. Therefore, the SiO2 layer can serve as an effective etch stop for silicon etching.
[0040] After removing the substrate 110, further manufacturing may be performed on the bonded wafer. For example, Figure 5The formation of interconnect layer 40 and memory structure 30 is shown. In the case where the formation of memory structure 30 is separate from the formation of interconnect layer 40, aluminum or copper may be selected for interconnect layer 40. Such interconnects may be formed at lower temperatures (eg, 450° C. or less).
[0041] According to another embodiment of the present invention, wafer bonding can be used to fabricate single crystal transistors below the memory array, referred to as CMOS below the array ("CuA"), and above the memory array, referred to as CMOS above the array ("CoA"). Figures 6A-6F A series of steps are shown to fabricate CuA and CoA circuits using wafer bonding technology.
[0042] like Fig. 6A As shown in FIG. 1 , CMOS transistor 210 is fabricated on substrate 200, and memory array 140 is fabricated on substrate 130. Then, substrate 130 is turned over and bonded to substrate 200 using wafer bonding technology, so that Figure 6B As shown, the memory array 140 is electrically connected to the CMOS transistor 210 on the substrate 200 through the interconnect layer 220. The substrate 130 is then removed to expose the memory array 140. Thereafter, as shown in FIG. Figure 6C As shown, interconnect layer 150 is fabricated from over the exposed side of memory array 140 .
[0043] like Fig.6D As shown, a second set of CMOS transistors 290 are fabricated on a substrate 260. The substrate 260 is most preferably a silicon-on-insulator ("SOI") wafer in which two layers of single-crystal silicon are provided on opposite sides of an oxide (SiO2) layer. Fig.6D As shown, CMOS transistor 290 and interconnect layer 300 thereover are formed on and over silicon layer 280, which is separated from substrate 260 by SiO2 layer 270. Substrate 260 is then flipped over and wafer bonded to allow CMOS transistor 290 to be electrically connected to memory array 140 via interconnect layer 150, as shown in FIG. Fig. 6E shown.
[0044] The substrate 260 is then removed to expose the SiO2 layer 270, and an interconnect layer 310 is fabricated over and electrically connects the CMOS transistor 290, such as Fig. 6F The resulting combination is a CuA type CMOS transistor 210 constructed in a single crystal silicon substrate 200, a memory array 140 at least partially covering the CMOS transistor 210, a CoA type CMOS transistor 290 at least partially covering the memory array 140, and a multi-layer interconnect layer 150, 300 and 310 covering and having the memory array 140 below. By providing CMOS transistors below and above the memory array, as Figures 6A-6FAs shown, a high-efficiency memory array is realized. A high-efficiency memory array refers to a memory array manufactured on a semiconductor die, on which almost all the area is occupied by memory cells. For example, in Figures 6A-6F , the CMOS transistors 210 below the memory array 140 may be high voltage or analog transistors, while the CMOS transistors 290 above the memory array 140 may be low voltage, short channel high performance logic CMOS transistors which are desirably physically located as close as possible to the chip's input / output pads.
[0045] According to another embodiment of the present invention, a memory block can be bonded to another memory block using wafer bonding. In this way, a high area density memory structure can be realized on a single chip while simplifying manufacturing by minimizing the aspect ratio of the manufactured memory structure. Figures 7A-7C Memory blocks 310 and 360 are shown connected using wafer bonding of substrates 320 and 350. Fig. 7A As shown, CMOS transistor 210 together with interconnect layer 220 and memory block 310 are fabricated on substrate 320, while memory block 360 is fabricated on substrate 350. Substrate 350 is then flipped over and wafer-bonded to substrate 320, so that memory blocks 310 and 360 are electrically connected, as shown in FIG. Figure 7B Then the substrate 350 is removed and the interconnection layer 370 is fabricated, as shown. Figure 7C shown.
[0046] According to another embodiment of the present invention, a single crystal silicon channel for a memory cell transistor can be formed by depositing an epitaxial silicon layer coming out of a single crystal substrate. Such processing is difficult for a memory block having a CuA-type CMOS transistor because a "clear" path from the substrate to the source / drain layer of the memory array may not be available. For example, an example of forming single crystal epitaxial silicon in a thin film memory transistor is disclosed in the provisional application II incorporated by reference above. In particular, among other types of thin film memory transistors, provisional application II discloses a type of thin film memory transistor, referred to herein as a "quasi-volatile memory (QVM) circuit", which has a data retention time (e.g., 100 milliseconds to one year) that is greater than a conventional dynamic random access memory (DRAM) circuit and less than a conventional non-volatile memory circuit. The QVM circuit can be organized, for example, as a three-dimensional array of NOR memory strings. When the memory block is built only on top of the silicon substrate (i.e., there are no CuA-type CMOS transistors underneath it), a clear path is provided for epitaxial silicon deposition. The substrate wafer with the resulting memory blocks can then be bonded to another substrate on which CMOS transistors have been fabricated. Figures 8A-8EA process according to one embodiment of the present invention is shown by which a CuA type CMOS transistor is provided below a memory block having an epitaxial single crystal silicon channel in a memory cell.
[0047] like Fig. 8A As shown, a stack of active strips (each active strip having a source and drain layer 420) is first formed on a substrate 400, and trenches separate the stacks of adjacent active strips until the substrate 400, as shown in FIG. Fig. 8A Then, epitaxial silicon 430 is grown from the surface 402 of the substrate 400, as shown in FIG. Figure 8B As shown, the anisotropic etch then removes substantially all of the epitaxial silicon 430 from the trenches, except for the epitaxial silicon channel 440 that remains in the recessed region between the source and drain layers of the active stripes, as shown in FIG. Figure 8C Then, the manufacturing of memory block 810 is completed, as shown in FIG. Fig.8D (Exemplary manufacturing processes are disclosed, for example, in co-pending applications incorporated by reference.) Substrate 400 is then flipped over and wafer bonded to substrate 450, which includes CMOS transistors 210 and interconnect layers 220 formed thereon, such as Figures 8D-8E As previously described, substrate 400 may be removed and interconnect layers may be fabricated over the memory blocks.
[0048] The above detailed description is provided to illustrate specific embodiments of the present invention, but is not intended to limit the present invention. Many changes and modifications are possible within the scope of the present invention. The present invention is set forth in the appended claims.
Claims
1. A three-dimensional NOR memory circuit, comprising: Single crystal semiconductor substrate; A first circuit set including single crystal transistors formed in the single crystal semiconductor substrate; a first interconnect layer comprising a plurality of conductors formed on top of the first collection of circuits; a first memory block electrically connected to the first circuit set through the conductor of the first interconnect layer, the first memory block being formed by: forming a stack of active strips of semiconductor material on a substrate, wherein trenches separate adjacent stacks of active strips to the substrate; forming an epitaxial silicon layer along the trenches, the epitaxial silicon growing from the substrate from the surface of the substrate so that an epitaxial silicon channel is provided between each drain layer and each source layer of each active strip, wherein the conductor of the first interconnect layer is connected to the first memory block by wafer bonding; a second interconnect layer formed above the first memory block, wherein the second interconnect layer comprises a plurality of conductors, and wherein the first interconnect layer and the second interconnect layer are disposed on opposite sides of the first memory block; a second circuit set formed in a silicon layer disposed on an insulating layer of a silicon-on-insulator wafer, wherein the second circuit set comprises single crystal transistors, wherein the second circuit set is electrically connected to the first memory block through conductors of the second interconnect layer, and wherein the second circuit set and the conductors of the second interconnect layer are connected by wafer bonding; as well as A third interconnect layer includes a plurality of conductors electrically connected to a second circuit set, and the third interconnect layer is disposed on a side of the insulating layer opposite to the silicon layer, and the second circuit set is formed on the silicon layer.
2. The three-dimensional NOR memory circuit of claim 1, wherein the wafer bonding is one of the following: thermocompression wafer bonding, adhesive bonding, anodic and thermal wafer bonding. 3 . The three-dimensional NOR memory circuit of claim 1 , wherein the second circuit set is formed in a silicon layer, and the silicon layer is disposed on an insulating layer of a silicon-on-insulator wafer. 4 . The three-dimensional NOR memory circuit of claim 1 , wherein the first circuit set comprises high voltage transistors.
5. The three-dimensional NOR memory circuit of claim 1, wherein the second circuit set is formed by low voltage, short channel high performance logic complementary metal oxide semiconductor transistors.
6. The three-dimensional NOR memory circuit of claim 1 , further comprising a second memory block, wherein the second memory block is electrically connected to the first memory block through conductors of the second interconnect layer and conductors of the third interconnect layer, and wherein the second memory block and the conductors of the third interconnect layer are connected by wafer bonding.
7. The three-dimensional NOR memory circuit of claim 1, wherein each active stripe of the first memory block comprises a plurality of memory cells, each memory cell having a channel region formed by a portion of an epitaxial single crystal silicon channel material.
8. The three-dimensional NOR memory circuit of claim 1, further comprising a second memory block, wherein the second memory block is electrically connected to the first memory block, and wherein the first memory block and the second memory block are connected by wafer bonding.
9. The three-dimensional NOR memory circuit of claim 1, wherein the first memory block comprises one or more of the following: a non-volatile memory string array and a quasi-volatile memory string array.
10. The three-dimensional NOR memory circuit of claim 9, wherein the first memory block is organized as a three-dimensional memory array.
11. The three-dimensional NOR memory circuit of claim 10, wherein the three-dimensional memory array comprises a NOR-type memory string.
Citation Information
Patent Citations
Semiconductor device and preparation method thereof
CN107658317A
Semiconductor device
US10381370B2
Methods for fabricating three dimensional integrated circuits
US20040214387A1
Buried bit line non-volatile floating gate memory cell with independent controllable control gate in a trench, and array thereof, and method of formation
US20040253787A1
Methods for bonding wafers using a metal interlayer
US20040262772A1
