US20020016035A1 - Method for making deep trench capacitors for DRAMs with reduced faceting at the substrate edge and providing a more uniform pad Si3N4 layer across the substrate - Google Patents
Method for making deep trench capacitors for DRAMs with reduced faceting at the substrate edge and providing a more uniform pad Si3N4 layer across the substrate Download PDFInfo
- Publication number
- US20020016035A1 US20020016035A1 US09/816,356 US81635601A US2002016035A1 US 20020016035 A1 US20020016035 A1 US 20020016035A1 US 81635601 A US81635601 A US 81635601A US 2002016035 A1 US2002016035 A1 US 2002016035A1
- Authority
- US
- United States
- Prior art keywords
- substrate
- layer
- trenches
- pad
- silicon nitride
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B12/00—Dynamic random access memory [DRAM] devices
- H10B12/01—Manufacture or treatment
- H10B12/02—Manufacture or treatment for one transistor one-capacitor [1T-1C] memory cells
- H10B12/03—Making the capacitor or connections thereto
- H10B12/038—Making the capacitor or connections thereto the capacitor being in a trench in the substrate
- H10B12/0387—Making the trench
Definitions
- This invention relates to an integrated circuit semiconductor device, and more particularly to a method for making improved deep trench capacitors for dynamic random access memory (DRAM) devices.
- the method uses a thicker pad silicon nitride (Si 3 N 4 ) layer with a chemical-vapor-deposited glass layer as a hard mask for etching the deep trenches. This avoids overetching and damaging (faceting) the hard mask at the extreme edge of the wafer when the deep trenches are etched in the wafer.
- the pad silicon nitride layer is used as a polish-back stop layer for making shallow trench isolation (STI).
- the polish-back stop layer Due to the inherent properties of the chemical-mechanical polishing (CMP), the polish-back stop layer has reduced thickness at the center of the wafer (substrate) and is thicker at the wafer edge.
- CMP chemical-mechanical polishing
- the invention uses an additional patterned mask layer to protect the wafer center while exposing the silicon nitride stop layer at the wafer edge. The thicker portion of the pad Si 3 N 4 layer is partially removed at the water edge to form a more uniform pad Si 3 N 4 .
- DRAM Dynamic random access memory
- Each memory cell consists of a single access transistor and a single storage capacitor.
- the storage capacitors are formed either by etching deep trenches in the substrate in each cell area, commonly referred to as trench capacitors, or are formed over the access transistors in the cell areas by depositing and patterning conducting layers over the access transistors, and are commonly referred to as stacked capacitors.
- the capacitors make electrical contact to one of the two source/drain areas (node contacts) of each FET (access transistor), while bit lines make electrical contact to the other source/drain area of each FET.
- Read/write circuits on the periphery of the DRAM chip, are used to store binary data by charging or discharging the storage capacitor via the bit lines, and the binary data is read (or sensed) by peripheral sense amplifiers, also via the bit lines.
- each capacitor must lie within an area about the size of the cell area in order to accommodate all the capacitors in the large array of cells used on the DRAM device.
- DRAM devices with stacked capacitors or trench capacitors.
- the stacked capacitors which are built on the chip surface, result in rough topography which makes subsequent processing difficult and requires leveling and planarizing techniques that can be expensive.
- An alternative method for making an array of DRAM cells is by forming deep trench capacitors in the silicon substrate.
- the surface therefore remains essentially planar and available for wiring for the DRAM circuit.
- by forming the storage capacitors in a trench etched in the silicon substrate it is possible to leave the substrate surface free for the bit lines, thereby providing adequate separation between bit line and storage capacitor. This also allows memory cells to be built with smaller surface areas for future high-density DRAM arrays.
- FIGS. 1 and 2 show schematic cross-sectional views replicating SEM cross-sectional views for two adjacent trenches of the multitude of trenches formed.
- FIG. 1 shows a cross section of two adjacent trenches 2 formed in the substrate 10 away from the edge of the substrate
- FIG. 2 shows a cross section of two adjacent trenches 2 formed at the edge of the substrate.
- the trenches are made by forming a thin stress-release silicon oxide layer (not shown), and depositing a pad Si 3 N 4 layer 12 and a chemical-vapor-deposited silicon oxide layer 14 to form a hard-mask layer.
- the hard-mask layer (layers 12 and 14 ) is then patterned using conventional photolithographic techniques and plasma etching to etch a multitude of openings for deep trenches. After stripping the photoresist, the hard mask is used to selectively etch deep trenches 2 in the substrate 10 , two of which are shown in FIG. 1.
- the Si 3 N 4 layer 12 is faceted at the point S, the trenches 2 in the silicon substrate 10 have essentially vertical sidewalls, and the trench openings replicate the hard-mask openings.
- the Si 3 N 4 is thinner at the edge of the substrate, and the plasma etching to form the trenches in the substrate generally etches faster at the substrate edge. This results in excessive faceting that damages the substrate at the edge and distorts the trench profile 2 , as shown at points S in FIG. 2.
- the etching of the array of closely spaced trenches can result in a series of silicon needle-like structures. In both cases, the overetch reduces the usable surface area on the substrate, thereby reducing product yield.
- CMP chemical-mechanical polishing
- the graph in FIG. 14 show the thickness profile of the pad Si 3 N layer 12 as a function of distance from the center of the wafer to the edge.
- the y axis shows the Si 3 N 4 thickness, and the x axis is the distance from the center of a 200-millimeter diameter wafer.
- the thickness of the Si 3 N 4 increases significantly due to the polishing loading effect as one approaches the edge of the wafer.
- the two curves in the graph (FIG.
- Curve A shows the results for a new polishing pad
- curve B shows the results for the conventional process using a polishing pad after several passes.
- the results of the polishing show unacceptable (increased) variations in thickness as one approaches the edge of the wafer. Therefore, it is strongly desirable to improve the uniformity as indicated by the curve C in FIG. 14.
- No. 5,629,226 teaches a method for making deep trench capacitors having increased capacitance by widening the bottom portion of the trench while simultaneously achieving high density integration.
- none of the references addresses overetching deep trenches at the edge of the wafer that causes excessive faceting of the deep trenches.
- none of the references addresses the non-nonuniformity in the pad Si 3 N 4 thickness across the wafer that results from CMP, as described above with respect to FIG. 14.
- a principal object of this invention is to provide an array of DRAM chip areas, each chip area having an array of sub-micrometer-wide deep trench capacitors with reduced trench overetching at the extreme edge of the substrate (wafer) to reduce hard-mask faceting and to increase product yield.
- Another object of this invention is to reduce faceting and damage at the substrate edge by using a thicker silicon nitride/borosilicate glass hard mask to reduce overetching of the hard mask when etching deep trenches.
- a further object of this invention is to improve the uniformity of the pad Si 3 N 4 layer by reducing the thickness of the pad layer at the edge of the wafer resulting from CMP. This is achieved using an additional novel blanket photoresist mask layer over the center of the wafer and exposing the pad Si 3 N 4 (polish-stop) layer at the wafer edge. The exposed polish-stop layer is then partially etched to provide a more uniform pad Si 3 N 4 (etch-stop layer) across the wafer.
- a method for making an array of trench capacitors in which overetching at the wafer edge is minimized by using a thick Si 3 N 4 pad layer, and at a later step a novel photoresist mask is used with a plasma etch to optimize the pad Si 3 N 4 thickness at the wafer edge and to improve the uniformity across the wafer.
- the method for making an array of deep trench capacitors for DRAM devices up to and including shallow trench isolation is briefly described.
- the method consists of providing a semiconductor substrate, preferably a single-crystal silicon substrate.
- An etch-stop layer consisting of a pad Si 3 N 4 layer and a first insulating layer is deposited to form a hard mask.
- the pad Si 3 N 4 layer is deposited to a thickness sufficient to prevent overetching at the edge of the wafer, which can cause faceting and wafer damage when deep trenches for capacitors are etched.
- the first insulating layer is a borosilicate glass (BSG).
- BSG borosilicate glass
- a photoresist mask and plasma etching are used to etch an array of openings in the hard-mask layer to the substrate.
- the photoresist mask is removed and the hard-mask is now used as an etch mask to etch deep trenches in the substrate for capacitors.
- the trench capacitors are now formed by depositing an arsenic-doped glass (ASG) and etching back to leave portions of the ASG in the lower part of the trenches.
- the substrate is then annealed to diffuse arsenic into the substrate to form first capacitor electrodes.
- the remaining ASG is removed by stripping.
- a capacitor interelectrode dielectric layer is formed by depositing a thin Si 3 N 4 layer and reoxidizing to form a silicon oxide/silicon nitride (ON) capacitor interelectrode dielectric layer on the sidewalls of the trenches.
- An N doped first polysilicon layer is deposited and recessed to fill the lower portion of the trenches.
- the exposed portions of the interelectrode dielectric layer on the upper sidewalls of the trenches is removed.
- a blanket collar SiO 2 layer is deposited on the substrate, annealed, and etched back to form a collar on the -upper portion of the sidewalls in the trenches.
- An N doped second polysilicon layer is deposited and recessed to leave portions in the upper part of the trenches to form the trench capacitors.
- An N doped third polysilicon layer is deposited and etched back to form interconnecting polysilicon straps to connect the trench capacitors to the substrate where semiconductor devices will be formed.
- Shallow trench openings are etched in the substrate over and between pairs of trench capacitors in the deep trenches. Shallow trench openings are also etched on the substrate for forming isolation regions for other circuits, such as DRAM peripheral circuits and merged logic/memory circuits.
- a short rapid thermal oxidation step is performed to form a thin SiO 2 on the exposed polysilicon surfaces.
- a relatively thin conformal insulating liner, preferably composed of Si 3 N 4 is deposited.
- a second insulating layer is deposited sufficiently thick to fill the shallow trench openings. The second insulating layer is polished back to the pad Si 3 N 4 layer to form the shallow trench isolation.
- a key feature of this invention is to use a second mask layer composed of organic photoresist, and the mask is patterned to leave blanket portions of the third insulating layer over the center of the wafer while exposing the underlying pad Si 3 N 4 layer at the wafer edge.
- the pad Si 3 N 4 layer at the wafer edge is then partially etched to improve the Si 3 N 4 uniformity across the wafer, and the second mask is removed in a plasma asher. This completes the array of deep trench capacitors, up to and including the shallow trench isolation, with reduced faceting at the wafer edge and more uniform Si 3 N 4 thickness, thereby improving process yields.
- FIG. 1 shows a prior-art schematic cross-sectional view replicating an SEM cross-sectional view for two adjacent trenches of an array of trenches having normal faceting for two trenches that are not at the edge of the wafer.
- FIG. 2 shows a prior-art schematic cross-sectional view replicating an SEM cross-sectional view for two adjacent trenches of the array of trenches having excessive faceting and depicts the overetching of the trenches at the edge of the wafer.
- FIGS. 3 through 12 show schematic cross-sectional views for forming two adjacent deep trenches depicting the sequence of process steps by the method of this invention.
- FIG. 13 shows a schematic top view of a semiconductor wafer (substrate) having the novel mask design for reducing the thickness of the pad silicon nitride layer at the edge of the wafer due to the nonuniform loading effect during chemical-mechanical polishing.
- FIG. 14 shows a graph of the silicon nitride thickness profile after shallow trench isolation chemical-mechanical polishing using two different polishing methods, and depicts the significant increase in Si 3 N 4 thickness as a function of distance to the edge of the wafer due to the loading effect.
- a method is described in detail for making deep trench capacitors for DRAMs with reduced overetching at the edge of the wafer by increasing the hard-mask thickness that includes a pad Si 3 N 4 layer. Then at a later process step, an additional mask is formed to protect the center portion of the wafer while leaving exposed portions of the pad Si 3 N 4 elsewhere on the substrate. The exposed Si 3 N 4 is then partially etched back to provides a more uniform pad Si 3 N 4 across the wafer.
- the method is especially applicable to making improved deep-trench capacitors across the wafer, and more specifically for making more reliable trenches at the edge of the wafer, it should be understood by those skilled in the art that the method is generally applicable to other applications where overetching occurs at the edge of the wafer, and also where more uniform polish-back is desired.
- the method for making these deep trench capacitors begins by providing a semiconductor substrate 10 .
- the substrate is preferably a P- doped single-crystal silicon having a ⁇ 100> crystal-lographic orientation.
- the substrate is thermally oxidized to form a thin SiO 2 stress-release layer 11 , having a thickness of between about 30 and 50 Angstroms.
- a pad silicon nitride layer 12 is deposited using chemical vapor deposition (CVD) and a reactant gas mixture such as dichlorosilane (SiCl 2 H 2 ) and ammonia (NH 3 ).
- the pad Si 3 N 4 layer 12 is deposited to a thickness sufficient to reduce overetching (faceting of the deep trench openings) at the edge of the wafer.
- the preferred thickness of the pad Si 3 N 4 layer 12 is between about 1800 and 2200 Angstroms.
- a first insulating layer 14 is deposited to complete a hard mask for etching the deep trenches (DT).
- Layer 14 is preferably a borosilicate glass (BSG) and is formed by CVD using tetraethosiloxane (TEOS) as the reactant gas and boron as the dopant gas, and is deposited to a thickness of between about 5000 and 7000 Angstroms.
- BSG borosilicate glass
- TEOS tetraethosiloxane
- conventional photolithographic techniques and anisotropic plasma etching are used to etch an array of openings 2 in the hard-mask layer ( 12 and 14 ) to the substrate.
- an anti-reflecting coating ARC
- the anisotropic plasma etching is carried out in a high-density plasma (HDP) etcher or a reactive ion etcher (RIE) to form the array of openings 2 for etching the deep trenches. Only two adjacent openings of the array of openings are depicted in the Fig.
- the hard-mask ( 12 and 14 ) is used as an etch mask and anisotropic plasma etching is used to etch deep trenches 2 in the substrate for capacitors.
- the etching is carried out in a HDP etcher and preferably using a chlorine-based etchant gas such as Cl 2 , HCl, or a Br.
- a chlorine-based etchant gas such as Cl 2 , HCl, or a Br.
- the deep trenches 2 are etched to a depth of between about 7 and 8 micrometers (um) and have opening widths of about 0.18 to 0.28 um. Because of the depth of the trenches, the very bottoms of the trenches are not shown to simplify the drawings.
- an arsenic-doped glass is deposited to fill the deep trenches 2 .
- the ASG is deposited preferably by CVD.
- the ASG is then etched back to leave portions of the ASG in the lower portions of the trenches 2 .
- the ASG is etched back using wet etching and a buffered hydrofluoric (BHF) etch solution.
- a diffusion cap oxide layer (not shown) is deposited to prevent out-diffusion of As from the top of the ASG diffusion source in the trenches.
- the diffusion cap oxide layer is a CVD SiO 2 and is deposited using TEOS as the reactant gas.
- the substrate 10 is then annealed to diffuse arsenic from the ASG into the substrate to form N-doped first capacitor electrodes 16 in the substrate adjacent to the lower portion of the trenches 2 .
- the preferred dopant concentration of the capacitor electrodes 16 is between about 5.0 E 19 and 1.0 E 20 atoms/cm 3 .
- the ASG is then removed by stripping using a BHF solution. Only the upper portions of the capacitor electrodes are depicted in FIG. 5.
- the first capacitor electrodes 16 are depicted after the ASG and cap oxide layers are removed.
- a capacitor interelectrode dielectric layer 18 is formed on the sidewalls of the trenches 2 by depositing a thin Si 3 N 4 layer.
- the Si 3 N 4 is typically deposited by LPCVD using SiCl 2 H 2 and NH 3 as the reactant gases, and is deposited to a preferred thickness of between about 40 and 50 Angstroms.
- An oxide is formed on the Si 3 N 4 layer by reoxidization to form a silicon oxide/silicon nitride (ON) layer to complete the capacitor interelectrode dielectric layer 18 on the sidewalls of the trenches 2 .
- the thermal oxidation also minimizes the pin holes in the Si 3 N 4 layer.
- an N doped first polysilicon layer 20 is deposited sufficiently thick to fill the trenches 2 .
- Layer 20 is deposited by LPCVD using SiH 4 as the reactant gas, and is in-situ doped using an N type dopant such as phosphorus to a preferred concentration of between about 8.0 E 19 and 3.0 E 20 atoms/cm 3 .
- the polysilicon layer 20 is then etched back to recess the polysilicon and to fill the lower portion of the trenches 2 .
- the interelectrode dielectric layer 18 exposed in the upper portions of the trenches is selectively removed, as shown in FIG. 6. The ON is removed using a wet etch in a solution of BHF or HF.
- a conformal collar SiO 2 layer 22 is deposited on the substrate 10 and on the sidewalls in the upper portions of the trenches.
- the SiO 2 is preferably deposited by CVD using, for example, TEOS as the reactant gas, and is deposited to a thickness of between about 500 and 700 Angstroms.
- the collar SiO 2 layer 22 is anisotropically etched back to form a collar on the upper portions of the sidewalls in the trenches 2 . The etching concurrently removes the collar oxide on the first polysilicon 20 in the bottom portions of the trenches.
- an N doped second polysilicon layer 24 is deposited sufficiently thick to fill the trenches 2 .
- Layer 24 is deposited by LPCVD using SiH 4 as the reactant gas, and is in-situ doped using an N type dopant such as phosphorus to a preferred concentration of between about 1.0 E 19 and 1.0 E 20 atoms/cm 3 .
- the polysilicon layer 24 is then etched back to recess the polysilicon in the upper portions of the trenches and below the top surface of the substrate 10 .
- the upper portions of the collar oxide 22 are then stripped, for example by dip etching in hydrofluoric (HF) acid, to expose the silicon substrate 10 in the upper edge of the trenches 2 , as shown in FIG. 7.
- HF hydrofluoric
- a third polysilicon layer 26 is deposited sufficiently thick to fill the trenches 2 .
- Layer 26 is then chemically-mechanically polished back to the pad Si 3 N 4 layer 12 , and the remaining polysilicon 26 in the trenches 2 is recessed to form interconnecting polysilicon straps 26 to connect the trench capacitors (portion 24 ) to the substrate 10 where semiconductor devices are typically formed.
- the polysilicon 26 is doped N+ to provide good electrical connections.
- the shallow trench isolation is formed next. Shallow trench openings 4 are etched in the substrate to separate the active device areas for electrical circuits. Concurrently the shallow trenches 4 are also etched over and between pairs of trench capacitors in the deep trenches 2 to electrically separate adjacent capacitors. Preferably the shallow trenches 4 are etched to a depth of about 2500 to 3000 Angstroms below the substrate surface 10 . The exposed surfaces of the polysilicon ( 26 , 24 , and 20 ) and the substrate 10 are subjected to a rapid thermal oxidation to form a thin SiO 2 (not shown) to reduce surface damage (leakage). A relatively thin conformal insulating liner 28 is deposited. Layer 28 is preferably Si 3 N 4 and is deposited to a thickness of between about 100 and 200 Angstroms.
- a second insulating layer 30 is deposited sufficiently thick to fill the shallow trench openings 4 .
- Layer 30 is preferably SiO 2 and is deposited by high-density-plasma CVD to a thickness at least equal to the total depth of the trenches 4 in pad Si 3 N 4 layer 12 and the substrate 10 .
- Layer 30 is then chem-mech polished (CMP) back to the pad layer 12 which also serves as a polishing stop layer to form the STI 30 in the openings 4 .
- CMP chem-mech polished
- a key feature of this invention is to deposit a second mask layer 32 .
- the second mask 32 is preferably composed of organic photoresist and is spin coated to a thickness of between about 2000 and 5000 Angstroms depending on the etching budget requirements.
- the photoresist is then exposed and developed.
- the mask is patterned to leave blanket portions of the mask 32 over the center portions C of the wafer, while exposing the underlying Si 3 N 4 pad layer 12 which includes the portion of 12 ′ having increased thickness on the edge E of the wafer.
- a top view of the wafer 10 is shown in FIG. 13.
- the center of the wafer having the mask is depicted by the center area C and the edge of the wafer is depicted by edge area E.
- a chip area 7 representing an electrical device or integrated circuit near the edge of the wafer.
- the mask 32 is used to protect the center of the wafer C while the thicker portion 12 ′ of pad layer 12 is removed by selective etching.
- the Si 3 N 4 portion 12 ′ is removed using a timed wet etch or a plasma etch.
- the preferred wet etch is carried out using a hot phosphoric acid (H 3 PO 4 ).
- H 3 PO 4 hot phosphoric acid
- the Si 3 N 4 pad layer 12 is selectively removed to the pad oxide layer 11 using a hot phosphoric acid etch.
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Semiconductor Memories (AREA)
- Semiconductor Integrated Circuits (AREA)
- Mechanical Treatment Of Semiconductor (AREA)
Abstract
Description
- (1) Field of the Invention
- This invention relates to an integrated circuit semiconductor device, and more particularly to a method for making improved deep trench capacitors for dynamic random access memory (DRAM) devices. The method uses a thicker pad silicon nitride (Si3N4) layer with a chemical-vapor-deposited glass layer as a hard mask for etching the deep trenches. This avoids overetching and damaging (faceting) the hard mask at the extreme edge of the wafer when the deep trenches are etched in the wafer. At a later processing step after completing the trench capacitors, the pad silicon nitride layer is used as a polish-back stop layer for making shallow trench isolation (STI). Due to the inherent properties of the chemical-mechanical polishing (CMP), the polish-back stop layer has reduced thickness at the center of the wafer (substrate) and is thicker at the wafer edge. To further improve process yield after CMP, the invention uses an additional patterned mask layer to protect the wafer center while exposing the silicon nitride stop layer at the wafer edge. The thicker portion of the pad Si3N4 layer is partially removed at the water edge to form a more uniform pad Si3N4.
- (2) Description of the Prior Art
- Dynamic random access memory (DRAM) devices are used for storing digital information on arrays of memory cells in the form of charge stored on capacitors. Each memory cell consists of a single access transistor and a single storage capacitor. The storage capacitors are formed either by etching deep trenches in the substrate in each cell area, commonly referred to as trench capacitors, or are formed over the access transistors in the cell areas by depositing and patterning conducting layers over the access transistors, and are commonly referred to as stacked capacitors. The capacitors make electrical contact to one of the two source/drain areas (node contacts) of each FET (access transistor), while bit lines make electrical contact to the other source/drain area of each FET. Read/write circuits, on the periphery of the DRAM chip, are used to store binary data by charging or discharging the storage capacitor via the bit lines, and the binary data is read (or sensed) by peripheral sense amplifiers, also via the bit lines. However, each capacitor must lie within an area about the size of the cell area in order to accommodate all the capacitors in the large array of cells used on the DRAM device.
- As the number of memory cells increases on the DRAM chip and the cell areas decreases, it becomes increasingly difficult to fabricate the storage capacitors with reasonable surface area for maintaining sufficient capacitance (charge). For example, after the year 2000 the number of memory cells on a DRAM chip is expected to exceed several gigabits. Further, as the cell area decreases, the available area for the storage capacitor in each cell also decreases. This makes it difficult to maintain sufficient capacitance for storing charge to provide the necessary signal-to-noise ratios.
- One method used in the semiconductor industry to overcome the above problems is to form DRAM devices with stacked capacitors or trench capacitors. However, the stacked capacitors, which are built on the chip surface, result in rough topography which makes subsequent processing difficult and requires leveling and planarizing techniques that can be expensive.
- An alternative method for making an array of DRAM cells is by forming deep trench capacitors in the silicon substrate. The surface therefore remains essentially planar and available for wiring for the DRAM circuit. Also, by forming the storage capacitors in a trench etched in the silicon substrate, it is possible to leave the substrate surface free for the bit lines, thereby providing adequate separation between bit line and storage capacitor. This also allows memory cells to be built with smaller surface areas for future high-density DRAM arrays.
- However, as the diameter of the trench decreases to sub-quarter-micrometer widths, it becomes necessary to significantly increase the trench depth. For example, for future gigabit DRAMs the aspect ratio (depth/width) of the trench can be greater than35. Unfortunately, etching these narrow deep trenches in a silicon substrate can be difficult to achieve and can result in excessive erosion of the hard mask and lead to damage of the substrate surface. This problem is particularly exacerbated at the edge of the substrate (wafer), and the problem becomes more severe as the substrate diameter increases. To better appreciate this problem, FIGS. 1 and 2 show schematic cross-sectional views replicating SEM cross-sectional views for two adjacent trenches of the multitude of trenches formed. FIG. 1 shows a cross section of two
adjacent trenches 2 formed in thesubstrate 10 away from the edge of the substrate, and FIG. 2 shows a cross section of twoadjacent trenches 2 formed at the edge of the substrate. Typically the trenches are made by forming a thin stress-release silicon oxide layer (not shown), and depositing a pad Si3N4 layer 12 and a chemical-vapor-depositedsilicon oxide layer 14 to form a hard-mask layer. The hard-mask layer (layers 12 and 14) is then patterned using conventional photolithographic techniques and plasma etching to etch a multitude of openings for deep trenches. After stripping the photoresist, the hard mask is used to selectively etchdeep trenches 2 in thesubstrate 10, two of which are shown in FIG. 1. - Although the Si3N4 layer 12 is faceted at the point S, the
trenches 2 in thesilicon substrate 10 have essentially vertical sidewalls, and the trench openings replicate the hard-mask openings. However, during typical processing to deposit the hard-mask layer, the Si3N4 is thinner at the edge of the substrate, and the plasma etching to form the trenches in the substrate generally etches faster at the substrate edge. This results in excessive faceting that damages the substrate at the edge and distorts thetrench profile 2, as shown at points S in FIG. 2. In more severe cases of overetching, the etching of the array of closely spaced trenches can result in a series of silicon needle-like structures. In both cases, the overetch reduces the usable surface area on the substrate, thereby reducing product yield. - Another problem occurs later in the trench capacitor process in which the chemical-mechanical polishing (CMP) to form the shallow trench isolation results in non-uniform polish-back of the shallow trench film material and also results in non-uniform etching of the underlying pad Si3N4 layer 12. The graph in FIG. 14 show the thickness profile of the pad Si3N layer 12 as a function of distance from the center of the wafer to the edge. The y axis shows the Si3N4 thickness, and the x axis is the distance from the center of a 200-millimeter diameter wafer. As can be seen the thickness of the Si3N4 increases significantly due to the polishing loading effect as one approaches the edge of the wafer. The two curves in the graph (FIG. 14) represent the variation in the Si3N4 thickness in Angstroms. Curve A shows the results for a new polishing pad, and curve B shows the results for the conventional process using a polishing pad after several passes. The results of the polishing show unacceptable (increased) variations in thickness as one approaches the edge of the wafer. Therefore, it is strongly desirable to improve the uniformity as indicated by the curve C in FIG. 14.
- Several methods of making deep-trench capacitors are described in the literature. For example, Golden et al. in U.S. Pat. No. 5,618,751 teach a method for making a deep trench using a photoresist fill and recess to simplify the process and improve repeatable capacitor uniformity from wafer to wafer. In U.S. Pat. No. 6,071,823 to Hung et al. a method is described for making a bottle-shaped etched deep trench for increased capacitance. Yoshida in U.S. Pat. No. 5,885,863 teaches a method for making a simple contact to buried doped regions, such as the buried plate of a DRAM deep trench capacitor. Ohtsuki in U.S. Pat. No. 5,629,226 teaches a method for making deep trench capacitors having increased capacitance by widening the bottom portion of the trench while simultaneously achieving high density integration. However, none of the references addresses overetching deep trenches at the edge of the wafer that causes excessive faceting of the deep trenches. Also, none of the references addresses the non-nonuniformity in the pad Si3N4 thickness across the wafer that results from CMP, as described above with respect to FIG. 14.
- There is still a strong need in the semiconductor industry to further improve upon fabricating deep trench capacitors for DRAM cells with increased reliability that improves process yield, especially at the edge of substrate.
- Accordingly, a principal object of this invention is to provide an array of DRAM chip areas, each chip area having an array of sub-micrometer-wide deep trench capacitors with reduced trench overetching at the extreme edge of the substrate (wafer) to reduce hard-mask faceting and to increase product yield.
- Another object of this invention is to reduce faceting and damage at the substrate edge by using a thicker silicon nitride/borosilicate glass hard mask to reduce overetching of the hard mask when etching deep trenches.
- A further object of this invention is to improve the uniformity of the pad Si3N4 layer by reducing the thickness of the pad layer at the edge of the wafer resulting from CMP. This is achieved using an additional novel blanket photoresist mask layer over the center of the wafer and exposing the pad Si3N4 (polish-stop) layer at the wafer edge. The exposed polish-stop layer is then partially etched to provide a more uniform pad Si3N4 (etch-stop layer) across the wafer.
- In accordance with the objectives of the present invention, a method is described for making an array of trench capacitors in which overetching at the wafer edge is minimized by using a thick Si3N4 pad layer, and at a later step a novel photoresist mask is used with a plasma etch to optimize the pad Si3N4 thickness at the wafer edge and to improve the uniformity across the wafer.
- The method for making an array of deep trench capacitors for DRAM devices up to and including shallow trench isolation is briefly described. The method consists of providing a semiconductor substrate, preferably a single-crystal silicon substrate. An etch-stop layer consisting of a pad Si3N4 layer and a first insulating layer is deposited to form a hard mask. The pad Si3N4 layer is deposited to a thickness sufficient to prevent overetching at the edge of the wafer, which can cause faceting and wafer damage when deep trenches for capacitors are etched. The first insulating layer is a borosilicate glass (BSG). Next a photoresist mask and plasma etching are used to etch an array of openings in the hard-mask layer to the substrate. The photoresist mask is removed and the hard-mask is now used as an etch mask to etch deep trenches in the substrate for capacitors. The trench capacitors are now formed by depositing an arsenic-doped glass (ASG) and etching back to leave portions of the ASG in the lower part of the trenches. The substrate is then annealed to diffuse arsenic into the substrate to form first capacitor electrodes. The remaining ASG is removed by stripping. A capacitor interelectrode dielectric layer is formed by depositing a thin Si3N4 layer and reoxidizing to form a silicon oxide/silicon nitride (ON) capacitor interelectrode dielectric layer on the sidewalls of the trenches. An N doped first polysilicon layer is deposited and recessed to fill the lower portion of the trenches. The exposed portions of the interelectrode dielectric layer on the upper sidewalls of the trenches is removed. A blanket collar SiO2 layer is deposited on the substrate, annealed, and etched back to form a collar on the -upper portion of the sidewalls in the trenches. An N doped second polysilicon layer is deposited and recessed to leave portions in the upper part of the trenches to form the trench capacitors. An N doped third polysilicon layer is deposited and etched back to form interconnecting polysilicon straps to connect the trench capacitors to the substrate where semiconductor devices will be formed. Shallow trench openings are etched in the substrate over and between pairs of trench capacitors in the deep trenches. Shallow trench openings are also etched on the substrate for forming isolation regions for other circuits, such as DRAM peripheral circuits and merged logic/memory circuits. A short rapid thermal oxidation step is performed to form a thin SiO2 on the exposed polysilicon surfaces. A relatively thin conformal insulating liner, preferably composed of Si3N4, is deposited. A second insulating layer is deposited sufficiently thick to fill the shallow trench openings. The second insulating layer is polished back to the pad Si3N4 layer to form the shallow trench isolation. A key feature of this invention is to use a second mask layer composed of organic photoresist, and the mask is patterned to leave blanket portions of the third insulating layer over the center of the wafer while exposing the underlying pad Si3N4 layer at the wafer edge. The pad Si3N4 layer at the wafer edge is then partially etched to improve the Si3N4 uniformity across the wafer, and the second mask is removed in a plasma asher. This completes the array of deep trench capacitors, up to and including the shallow trench isolation, with reduced faceting at the wafer edge and more uniform Si3N4 thickness, thereby improving process yields.
- The objects and advantages of this invention are best understood with reference to the attached drawings in the figures and the embodiment that follows.
- FIG. 1 shows a prior-art schematic cross-sectional view replicating an SEM cross-sectional view for two adjacent trenches of an array of trenches having normal faceting for two trenches that are not at the edge of the wafer.
- FIG. 2 shows a prior-art schematic cross-sectional view replicating an SEM cross-sectional view for two adjacent trenches of the array of trenches having excessive faceting and depicts the overetching of the trenches at the edge of the wafer.
- FIGS. 3 through 12 show schematic cross-sectional views for forming two adjacent deep trenches depicting the sequence of process steps by the method of this invention.
- FIG. 13 shows a schematic top view of a semiconductor wafer (substrate) having the novel mask design for reducing the thickness of the pad silicon nitride layer at the edge of the wafer due to the nonuniform loading effect during chemical-mechanical polishing.
- FIG. 14 shows a graph of the silicon nitride thickness profile after shallow trench isolation chemical-mechanical polishing using two different polishing methods, and depicts the significant increase in Si3N4 thickness as a function of distance to the edge of the wafer due to the loading effect.
- A method is described in detail for making deep trench capacitors for DRAMs with reduced overetching at the edge of the wafer by increasing the hard-mask thickness that includes a pad Si3N4 layer. Then at a later process step, an additional mask is formed to protect the center portion of the wafer while leaving exposed portions of the pad Si3N4 elsewhere on the substrate. The exposed Si3N4 is then partially etched back to provides a more uniform pad Si3N4 across the wafer. Although the method is especially applicable to making improved deep-trench capacitors across the wafer, and more specifically for making more reliable trenches at the edge of the wafer, it should be understood by those skilled in the art that the method is generally applicable to other applications where overetching occurs at the edge of the wafer, and also where more uniform polish-back is desired.
- Referring to FIG. 3, the method for making these deep trench capacitors begins by providing a
semiconductor substrate 10. The substrate is preferably a P- doped single-crystal silicon having a <100> crystal-lographic orientation. The substrate is thermally oxidized to form a thin SiO2 stress-release layer 11, having a thickness of between about 30 and 50 Angstroms. Next a padsilicon nitride layer 12 is deposited using chemical vapor deposition (CVD) and a reactant gas mixture such as dichlorosilane (SiCl2H2) and ammonia (NH3). The pad Si3N4 layer 12 is deposited to a thickness sufficient to reduce overetching (faceting of the deep trench openings) at the edge of the wafer. For example, the preferred thickness of the pad Si3N4 layer 12 is between about 1800 and 2200 Angstroms. Next, a first insulatinglayer 14 is deposited to complete a hard mask for etching the deep trenches (DT).Layer 14 is preferably a borosilicate glass (BSG) and is formed by CVD using tetraethosiloxane (TEOS) as the reactant gas and boron as the dopant gas, and is deposited to a thickness of between about 5000 and 7000 Angstroms. - Still referring to FIG. 3, conventional photolithographic techniques and anisotropic plasma etching are used to etch an array of
openings 2 in the hard-mask layer (12 and 14) to the substrate. Typically an anti-reflecting coating (ARC) is coated on the substrate prior to applying the photoresist to minimize reflections and to improve photoresist image fidelity. The ARC and the photoresist are not shown in the Fig. The anisotropic plasma etching is carried out in a high-density plasma (HDP) etcher or a reactive ion etcher (RIE) to form the array ofopenings 2 for etching the deep trenches. Only two adjacent openings of the array of openings are depicted in the Fig. - Referring to FIG. 4, after removing the photoresist mask, the hard-mask (12 and 14) is used as an etch mask and anisotropic plasma etching is used to etch
deep trenches 2 in the substrate for capacitors. The etching is carried out in a HDP etcher and preferably using a chlorine-based etchant gas such as Cl2, HCl, or a Br. Typically for current DRAM product thedeep trenches 2 are etched to a depth of between about 7 and 8 micrometers (um) and have opening widths of about 0.18 to 0.28 um. Because of the depth of the trenches, the very bottoms of the trenches are not shown to simplify the drawings. - Referring to FIG. 5, an arsenic-doped glass (ASG) is deposited to fill the
deep trenches 2. The ASG is deposited preferably by CVD. The ASG is then etched back to leave portions of the ASG in the lower portions of thetrenches 2. The ASG is etched back using wet etching and a buffered hydrofluoric (BHF) etch solution. A diffusion cap oxide layer (not shown) is deposited to prevent out-diffusion of As from the top of the ASG diffusion source in the trenches. Typically the diffusion cap oxide layer is a CVD SiO2 and is deposited using TEOS as the reactant gas. - Still referring to FIG. 5, the
substrate 10 is then annealed to diffuse arsenic from the ASG into the substrate to form N-dopedfirst capacitor electrodes 16 in the substrate adjacent to the lower portion of thetrenches 2. The preferred dopant concentration of thecapacitor electrodes 16 is between about 5.0 E 19 and 1.0E 20 atoms/cm3. The ASG is then removed by stripping using a BHF solution. Only the upper portions of the capacitor electrodes are depicted in FIG. 5. In FIG. 5, thefirst capacitor electrodes 16 are depicted after the ASG and cap oxide layers are removed. - Referring to FIG. 6, a capacitor
interelectrode dielectric layer 18 is formed on the sidewalls of thetrenches 2 by depositing a thin Si3N4 layer. The Si3N4 is typically deposited by LPCVD using SiCl2H2 and NH3 as the reactant gases, and is deposited to a preferred thickness of between about 40 and 50 Angstroms. An oxide is formed on the Si3N4 layer by reoxidization to form a silicon oxide/silicon nitride (ON) layer to complete the capacitorinterelectrode dielectric layer 18 on the sidewalls of thetrenches 2. The thermal oxidation also minimizes the pin holes in the Si3N4 layer. - Continuing with FIG. 6, an N doped
first polysilicon layer 20 is deposited sufficiently thick to fill thetrenches 2.Layer 20 is deposited by LPCVD using SiH4 as the reactant gas, and is in-situ doped using an N type dopant such as phosphorus to a preferred concentration of between about 8.0 E 19 and 3.0E 20 atoms/cm3. Thepolysilicon layer 20 is then etched back to recess the polysilicon and to fill the lower portion of thetrenches 2. The interelectrodedielectric layer 18 exposed in the upper portions of the trenches is selectively removed, as shown in FIG. 6. The ON is removed using a wet etch in a solution of BHF or HF. - Referring to FIG. 7, a conformal collar SiO2 layer 22 is deposited on the
substrate 10 and on the sidewalls in the upper portions of the trenches. The SiO2 is preferably deposited by CVD using, for example, TEOS as the reactant gas, and is deposited to a thickness of between about 500 and 700 Angstroms. After annealing, the collar SiO2 layer 22 is anisotropically etched back to form a collar on the upper portions of the sidewalls in thetrenches 2. The etching concurrently removes the collar oxide on thefirst polysilicon 20 in the bottom portions of the trenches. - Still referring to FIG. 7, an N doped
second polysilicon layer 24 is deposited sufficiently thick to fill thetrenches 2.Layer 24 is deposited by LPCVD using SiH4 as the reactant gas, and is in-situ doped using an N type dopant such as phosphorus to a preferred concentration of between about 1.0 E 19 and 1.0E 20 atoms/cm3. Thepolysilicon layer 24 is then etched back to recess the polysilicon in the upper portions of the trenches and below the top surface of thesubstrate 10. The upper portions of thecollar oxide 22 are then stripped, for example by dip etching in hydrofluoric (HF) acid, to expose thesilicon substrate 10 in the upper edge of thetrenches 2, as shown in FIG. 7. - Referring to FIG. 8, a
third polysilicon layer 26 is deposited sufficiently thick to fill thetrenches 2.Layer 26 is then chemically-mechanically polished back to the pad Si3N4 layer 12, and the remainingpolysilicon 26 in thetrenches 2 is recessed to form interconnecting polysilicon straps 26 to connect the trench capacitors (portion 24) to thesubstrate 10 where semiconductor devices are typically formed. Thepolysilicon 26 is doped N+ to provide good electrical connections. - Referring to FIG. 9, the shallow trench isolation is formed next.
Shallow trench openings 4 are etched in the substrate to separate the active device areas for electrical circuits. Concurrently theshallow trenches 4 are also etched over and between pairs of trench capacitors in thedeep trenches 2 to electrically separate adjacent capacitors. Preferably theshallow trenches 4 are etched to a depth of about 2500 to 3000 Angstroms below thesubstrate surface 10. The exposed surfaces of the polysilicon (26, 24, and 20) and thesubstrate 10 are subjected to a rapid thermal oxidation to form a thin SiO2 (not shown) to reduce surface damage (leakage). A relatively thin conformal insulatingliner 28 is deposited.Layer 28 is preferably Si3N4 and is deposited to a thickness of between about 100 and 200 Angstroms. - Still referring to FIG. 9, a second insulating
layer 30 is deposited sufficiently thick to fill theshallow trench openings 4.Layer 30 is preferably SiO2 and is deposited by high-density-plasma CVD to a thickness at least equal to the total depth of thetrenches 4 in pad Si3N4 layer 12 and thesubstrate 10.Layer 30 is then chem-mech polished (CMP) back to thepad layer 12 which also serves as a polishing stop layer to form theSTI 30 in theopenings 4. Unfortunately the CMP rate across the wafer is inherently nonuniform, resulting in thepad layer 12 having an increasedthickness 12′ at the edge of the wafer, labeled E for the edge and C for the center of the wafer in FIG. 9. - Referring to FIG. 10, a key feature of this invention is to deposit a
second mask layer 32. Thesecond mask 32 is preferably composed of organic photoresist and is spin coated to a thickness of between about 2000 and 5000 Angstroms depending on the etching budget requirements. The photoresist is then exposed and developed. The mask is patterned to leave blanket portions of themask 32 over the center portions C of the wafer, while exposing the underlying Si3N4 pad layer 12 which includes the portion of 12′ having increased thickness on the edge E of the wafer. To better appreciate the method, a top view of thewafer 10 is shown in FIG. 13. The center of the wafer having the mask is depicted by the center area C and the edge of the wafer is depicted by edge area E. Also shown is achip area 7 representing an electrical device or integrated circuit near the edge of the wafer. - Referring now to FIG. 11, the
mask 32 is used to protect the center of the wafer C while thethicker portion 12′ ofpad layer 12 is removed by selective etching. The Si3N4 portion 12′ is removed using a timed wet etch or a plasma etch. The preferred wet etch is carried out using a hot phosphoric acid (H3PO4). After removing thethicker portion 12′ of thepad layer 12 to form a more uniform thickness across the wafer, themask layer 32 is removed by plasma ashing. - Finally as shown in FIG. 12, the Si3N4 pad layer 12 is selectively removed to the
pad oxide layer 11 using a hot phosphoric acid etch. - While the invention has been particularly shown and described with reference to the preferred embodiment thereof, it will be understood by those skilled in the art that various changes in form and details may be made without departing from the spirit and scope of the invention.
Claims (17)
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW089115006A TW452879B (en) | 2000-07-27 | 2000-07-27 | Method for removing polishing stop layer |
TW089115006 | 2000-07-27 |
Publications (2)
Publication Number | Publication Date |
---|---|
US20020016035A1 true US20020016035A1 (en) | 2002-02-07 |
US6391706B2 US6391706B2 (en) | 2002-05-21 |
Family
ID=21660556
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/816,356 Expired - Fee Related US6391706B2 (en) | 2000-07-27 | 2001-03-26 | Method for making deep trench capacitors for DRAMs with reduced faceting at the substrate edge and providing a more uniform pad Si3N4layer across the substrate |
Country Status (3)
Country | Link |
---|---|
US (1) | US6391706B2 (en) |
DE (1) | DE10116529B4 (en) |
TW (1) | TW452879B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6406970B1 (en) * | 2001-08-31 | 2002-06-18 | Infineon Technologies North America Corp. | Buried strap formation without TTO deposition |
US20050059207A1 (en) * | 2003-09-17 | 2005-03-17 | Chih-Han Chang | Method for forming a deep trench capacitor buried plate |
US20050074943A1 (en) * | 2003-10-03 | 2005-04-07 | Su-Chen Lai | [method of fabricating deep trench capacitor] |
US20100200949A1 (en) * | 2009-02-12 | 2010-08-12 | International Business Machines Corporation | Method for tuning the threshold voltage of a metal gate and high-k device |
US20110168226A1 (en) * | 2010-01-11 | 2011-07-14 | Samsung Electronics Co., Ltd. | Solar cell module and method of manufacturing the same |
WO2019147558A1 (en) * | 2018-01-23 | 2019-08-01 | Texas Instruments Incorporated | Trench capacitor with warpage reduction |
CN111807318A (en) * | 2020-07-22 | 2020-10-23 | 中国人民解放军国防科技大学 | TGV substrate preparation method based on glass reflow process and MEMS device packaging method |
Families Citing this family (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE50107496D1 (en) * | 2001-07-20 | 2006-02-02 | Infineon Technologies Ag | Method for producing self-aligning mask layers |
TW508758B (en) * | 2001-07-23 | 2002-11-01 | Promos Technologies Inc | Manufacturing method of deep trench capacitor |
TW506059B (en) * | 2001-09-25 | 2002-10-11 | Promos Techvologies Inc | Forming method for shallow trench |
JP3895987B2 (en) * | 2001-12-27 | 2007-03-22 | 株式会社東芝 | Semiconductor device and manufacturing method thereof |
DE10233208A1 (en) * | 2002-07-22 | 2004-03-04 | Infineon Technologies Ag | Semiconductor device with trench isolation and associated manufacturing process |
US6750097B2 (en) * | 2002-07-30 | 2004-06-15 | International Business Machines Corporation | Method of fabricating a patterened SOI embedded DRAM/eDRAM having a vertical device cell and device formed thereby |
TW550741B (en) * | 2002-08-20 | 2003-09-01 | Nanya Technology Corp | Method for forming a bottle trench |
CN1298048C (en) * | 2003-07-03 | 2007-01-31 | 南亚科技股份有限公司 | Method for avoiding enlargement of top size of deep trench |
US6913968B2 (en) * | 2003-07-30 | 2005-07-05 | International Business Machines Corporation | Method and structure for vertical DRAM devices with self-aligned upper trench shaping |
US6815307B1 (en) * | 2003-09-16 | 2004-11-09 | Nanya Technology Corp. | Method for fabricating a deep trench capacitor |
US7015091B1 (en) * | 2004-11-18 | 2006-03-21 | Promos Technologies, Inc. | Integration of silicon carbide into DRAM cell to improve retention characteristics |
US20070105302A1 (en) * | 2005-11-09 | 2007-05-10 | Infineon Technologies Ag | Integrated circuit formed on a semiconductor substrate |
KR20120062988A (en) * | 2010-12-07 | 2012-06-15 | 에스케이하이닉스 주식회사 | Method for fabricating semiconductor device |
US8293625B2 (en) | 2011-01-19 | 2012-10-23 | International Business Machines Corporation | Structure and method for hard mask removal on an SOI substrate without using CMP process |
TWI756003B (en) | 2021-01-04 | 2022-02-21 | 力晶積成電子製造股份有限公司 | Planarization method |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0637275A (en) * | 1992-07-13 | 1994-02-10 | Toshiba Corp | Semiconductor memory and manufacture thereof |
US5618751A (en) * | 1996-05-23 | 1997-04-08 | International Business Machines Corporation | Method of making single-step trenches using resist fill and recess |
US5885863A (en) * | 1997-03-31 | 1999-03-23 | Kabushiki Kaisha Toshiba | Method of making a contact for contacting an impurity region formed in a semiconductor substrate |
JP3485807B2 (en) * | 1998-09-04 | 2004-01-13 | 茂徳科技股▲ふん▼有限公司 | Method for manufacturing trench capacitor |
US6184107B1 (en) * | 1999-03-17 | 2001-02-06 | International Business Machines Corp. | Capacitor trench-top dielectric for self-aligned device isolation |
US6071823A (en) * | 1999-09-21 | 2000-06-06 | Promos Technology, Inc | Deep trench bottle-shaped etch in centura mark II NG |
-
2000
- 2000-07-27 TW TW089115006A patent/TW452879B/en not_active IP Right Cessation
-
2001
- 2001-03-26 US US09/816,356 patent/US6391706B2/en not_active Expired - Fee Related
- 2001-04-03 DE DE10116529A patent/DE10116529B4/en not_active Expired - Fee Related
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6406970B1 (en) * | 2001-08-31 | 2002-06-18 | Infineon Technologies North America Corp. | Buried strap formation without TTO deposition |
US20050059207A1 (en) * | 2003-09-17 | 2005-03-17 | Chih-Han Chang | Method for forming a deep trench capacitor buried plate |
US7232718B2 (en) * | 2003-09-17 | 2007-06-19 | Nanya Technology Corp. | Method for forming a deep trench capacitor buried plate |
US20050074943A1 (en) * | 2003-10-03 | 2005-04-07 | Su-Chen Lai | [method of fabricating deep trench capacitor] |
US6881620B1 (en) * | 2003-10-03 | 2005-04-19 | Promos Technologies Inc. | Method of fabricating deep trench capacitor |
US20100200949A1 (en) * | 2009-02-12 | 2010-08-12 | International Business Machines Corporation | Method for tuning the threshold voltage of a metal gate and high-k device |
US20110168226A1 (en) * | 2010-01-11 | 2011-07-14 | Samsung Electronics Co., Ltd. | Solar cell module and method of manufacturing the same |
WO2019147558A1 (en) * | 2018-01-23 | 2019-08-01 | Texas Instruments Incorporated | Trench capacitor with warpage reduction |
US10559650B2 (en) | 2018-01-23 | 2020-02-11 | Texas Instruments Incorporated | Trench capacitor with warpage reduction |
US11271072B2 (en) | 2018-01-23 | 2022-03-08 | Texas Instruments Incorporated | Trench capacitor with warpage reduction |
CN111807318A (en) * | 2020-07-22 | 2020-10-23 | 中国人民解放军国防科技大学 | TGV substrate preparation method based on glass reflow process and MEMS device packaging method |
Also Published As
Publication number | Publication date |
---|---|
DE10116529A1 (en) | 2002-02-21 |
DE10116529B4 (en) | 2006-08-17 |
TW452879B (en) | 2001-09-01 |
US6391706B2 (en) | 2002-05-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6391706B2 (en) | Method for making deep trench capacitors for DRAMs with reduced faceting at the substrate edge and providing a more uniform pad Si3N4layer across the substrate | |
US5943581A (en) | Method of fabricating a buried reservoir capacitor structure for high-density dynamic random access memory (DRAM) circuits | |
US6204112B1 (en) | Process for forming a high density semiconductor device | |
US6310375B1 (en) | Trench capacitor with isolation collar and corresponding manufacturing method | |
US6509599B1 (en) | Trench capacitor with insulation collar and method for producing the trench capacitor | |
US6037213A (en) | Method for making cylinder-shaped capacitors for dynamic random access memory | |
US7026678B2 (en) | Dynamic random access memory circuitry having storage capacitors within a well | |
KR100659810B1 (en) | Crystal-axis-aligned vertical side wall device and process for manufacture thereof | |
US5895250A (en) | Method of forming semicrown-shaped stacked capacitors for dynamic random access memory | |
US6608341B2 (en) | Trench capacitor with capacitor electrodes | |
US20010003663A1 (en) | Method for making closely spaced capacitors with reduced parasitic capacitance on a dynamic random access memory (DRAM) device | |
US7122424B2 (en) | Method for making improved bottom electrodes for metal-insulator-metal crown capacitors | |
US6828191B1 (en) | Trench capacitor with an insulation collar and method for producing a trench capacitor | |
JPH11330403A (en) | Manufacture of trench capacitor | |
JPH10321814A (en) | Planarization technique for dram cell capacitor electrode | |
US6271079B1 (en) | Method of forming a trench capacitor | |
US6025245A (en) | Method of forming a trench capacitor with a sacrificial silicon nitrate sidewall | |
US6548850B1 (en) | Trench capacitor configuration and method of producing it | |
US6355518B1 (en) | Method for making a DRAM cell with deep-trench capacitors and overlying vertical transistors | |
US6403416B1 (en) | Method for making a double-cylinder-capacitor structure for dynamic random access memory (DRAM) | |
US6077743A (en) | Method for making dynamic random access memory cells having brush-shaped stacked capacitors patterned from a hemispherical grain hard mask | |
KR19980020386A (en) | Capacitor Formation Method of Semiconductor Device | |
US5795804A (en) | Method of fabricating a stack/trench capacitor for a dynamic random access memory (DRAM) | |
US6015734A (en) | Method for improving the yield on dynamic random access memory (DRAM) with cylindrical capacitor structures | |
US6083787A (en) | Method of fabricating deep trench capacitors for dram cells |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: PROMOS TECHNOLOGIES, INC., TAIWAN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:WU, CHAO-CHUEH;CHIU, SHENG-FEN;CHUNG, JESSE;AND OTHERS;REEL/FRAME:011719/0755 Effective date: 20010307 Owner name: MOSEL VITELIC, INC., TAIWAN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:WU, CHAO-CHUEH;CHIU, SHENG-FEN;CHUNG, JESSE;AND OTHERS;REEL/FRAME:011719/0755 Effective date: 20010307 Owner name: INFINEON TECHNOLOGIES, INC., GERMANY Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:WU, CHAO-CHUEH;CHIU, SHENG-FEN;CHUNG, JESSE;AND OTHERS;REEL/FRAME:011719/0755 Effective date: 20010307 |
|
FEPP | Fee payment procedure |
Free format text: PAYER NUMBER DE-ASSIGNED (ORIGINAL EVENT CODE: RMPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
REFU | Refund |
Free format text: REFUND - PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: R1551); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
REMI | Maintenance fee reminder mailed | ||
FPAY | Fee payment |
Year of fee payment: 4 |
|
SULP | Surcharge for late payment | ||
REMI | Maintenance fee reminder mailed | ||
AS | Assignment |
Owner name: QIMONDA AG, GERMANY Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:INFINEON TECHNOLOGIES INC.;REEL/FRAME:023785/0001 Effective date: 20060425 |
|
LAPS | Lapse for failure to pay maintenance fees | ||
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20100521 |