Rough oxide hard mask for DT surface area enhancement for DT DRAM

Information

  • Patent Grant
  • 6559002
  • Patent Number
    6,559,002
  • Date Filed
    Monday, December 31, 2001
    22 years ago
  • Date Issued
    Tuesday, May 6, 2003
    21 years ago
Abstract
In a process for making a DT DRAM structure, the improvement of providing a surface area enhanced DT below the collar region and node capacitance that does not shrink with decreasing groundrule/cell size, comprising:a) providing a semiconductor substrate having a collar region and an adjacent region below the collar region, the collar region having SiO deposited thereon;b) depositing a SiN liner on said collar region and on the region below the collar;c) depositing a layer of a-Si on the SiN liner to form a micromask;d) subjecting the structure from step c) to an anneal/oxidation step under a wet environment at a sufficient temperature to form a plurality of oxide dot hardmasks;e) subjecting the SiN liner to an etch selective to SiO;f) subjecting the structure from step e) to a Si transfer etch using a chemical dry etch (CDE) selective to SiO to create rough Si surface;g) stripping SiO and the SiN; and forming a node and collar deposition.
Description




BACKGROUND OF THE INVENTION




1) Field of the Invention




The invention relates to a process for providing a rough Si surface on a Si sidewall during preparation of a DT DRAM, to obtain DT surface area enhancement for a DT DRAM by creating a rough oxide etch mask, based on an anneal/oxidation of deposited amorphous Si (a-Si) on a SiN liner. The process permits a process/integration scheme which allows the integration of this process into a collar process utilizing SiO


2


or SiN as the collar material.




2) Description of the Prior Art




It is known that node capacitance requirements for DT DRAM devices does not shrink with decreasing groundrule/cell size. Moreover, extendability of current processes such as wet bottle and deeper trench etchings does not provide sufficient surface area enhancement for the DT DRAM devices.




For example, the use of sub-micron features for fabrication of a DRAM device in which the capacitor of the DRAM device is a stacked capacitor structure creates problems when attempts are made to increase the capacitance. In this connection, it should be noted that a DRAM cell is generally comprised of a stacked capacitor structure that overlies a transfer gate transistor, and is connected to the source of the transfer gate transistor. The decreasing size of the transfer gate transistor limits the dimensions of the stacked capacitor structure. Therefore, to increase the capacitance of the stacked capacitor structure, which is comprised of two electrodes separated by a dielectric layer, the thickness of the dielectric layer must be decreased or the area of the capacitor must be increased. However, advancement of the DRAM technology to densities of 64 million cells per chip or greater results in a cell in which a smaller transfer gate transistor is used, thereby resulting in less of an overlying area for placement of overlying stacked capacitor structures.




U.S. Pat. No. 5,340,763 describes a method of maintaining or increasing stacked capacitor structure capacitance, while decreasing the lateral dimension of the capacitor by use of a rough or hemispherical grained (HSG) polysilicon layers. Nevertheless, the capacitance increase realized in this patent is limited by subjecting only the top surface of the polysilicon lower electrode to the HSG polysilicon deposition and etch procedure, thereby leaving the sides of the polysilicon lower electrode flat.




Hemispherical grained (HSG) polysilicon layers have also been used as overlying layers on conventional polysilicon structures, as described in U.S. Pat. No. 5,444,653. This patent specifically describes a storage node, or lower electrode of a stacked capacitor structure, in which the surface area of a polysilicon lower electrode structure is increased by deposition of an HSG polysilicon layer on an insulator layer, which overlies the top surface of the lower electrode structure. Subsequent etching procedures result in a transfer of the roughened surface created by the HSG polysilicon layer to only the top surface of the polysilicon lower electrode structure. The roughened, top surface of the polysilicon lower electrode, with increased surface area results in a capacitance increase for the subsequent capacitor structure.




U.S. Pat. No. 6,037,220 disclose a process for increasing the surface area of a polysilicon lower electrode, or storage node electrode, by forming an HSG polysilicon layer only on the sides of the storage node electrode, while maintaining a flat, or non-roughened, top surface. This configuration is a result of using a silicon oxide hard mask, overlying the top surface of the lower electrode structure, therefore accepting the HSG polysilicon deposition. The use of the silicon oxide hard mask, prevents the deleterious attack of the polysilicon lower electrode structure, during an HSG polysilicon etch back procedure, performed to remove unwanted HSG polysilicon from non-capacitor regions.




This patent also describes an iteration in which the HSG polysilicon-silicon oxide, hard mask features are prepared on the top surface of the polysilicon lower electrode structure, then used as a mask, to remove polysilicon from the spaces exposed between the HSG polysilicon-silicon oxide features. The etch process results in a roughened, polysilicon lower electrode, top surface, and along with the HSG polysilicon coated sidewalls, provide greater surface area increases than counterparts fabricated with only roughened sides, or with only a roughened top surface.




U.S. Pat. No. 6,232,171 B1 disclose a method for enlarging the sidewall surface of a deep trench with improved process precision (without depending upon the differential etch rate between doped and undoped silicon) so that the sidewall surface enlargement is optimally and most effectively controlled within a lower portion of the sidewall where a tapered-down trench diameter distribution is encountered due to the reduced trench opening on a deep sub-micron semiconductor device and the large aspect ratio (length-to-diameter ratio) of the vertically oriented trench. When this method is incorporated into a process for making deep trench capacitors, the method comprises:




(1) forming a deep trench into an active (i.e., conducting) region in a substrate;




(2) filling the deep trench with a dielectric material, typically an oxide;




(3) etching the dielectric material down to a predetermined depth;




(4) forming a thin oxidative layer, using a thermal oxidation procedure, on the sidewall of the trench above the dielectric material and on the top surface around the trench for stress relief;




(5) forming a nitride layer covering the thin oxide layer;




(6) using an anisotropic etching procedure to form a nitride sidewall spacer from the nitride layer;




(7) removing the dielectric material by a selective wet etching to expose the sidewall of the trench below the nitride sidewall spacer;




(8) using the nitride sidewall spacer as a mask, thermally oxidizing the portion of the substrate behind the sidewall of the trench below the nitride sidewall spacer to a predetermined penetration; and




(9) removing the oxidized silicon sidewall and the nitride sidewall spacer, either simultaneously or separately.




In view of the fact that node capacitance requirements for DT DRAM technology does not shrink with decreasing groundrule/cell size, and in view of the fact that extendability of current processes such as wet bottle and deeper trench etching do not provide sufficient surface area enhancement, there is a need for additional processes for surface area increases when preparing DT DRAM devices.




SUMMARY OF THE INVENTION




One object of the present invention is to provide a process for surface area enhancement or increases for DT DRAM devices of densities of 64 million cells per chip, or greater.




Another object of the present invention is to provide a process for surface area enhancement or increases for DT DRAM devices utilizing a rough Si surface that is a rough mask material that can be used to transfer the pattern to the Si sidewall of the DT DRAM device.




A further object of the present invention is to provide a process for surface area enhancement for DT DRAM devices utilizing a process to create a rough oxide etch mask, based on the anneal/oxidation of deposited amorphous Si (a-Si) on a SiN liner.




A yet further object of the present invention is to provide a process integration scheme when providing surface area enhancement of a DT DRAM device which allows integration of this process into a collar process using SiO


2


or SiN as a collar material.











BRIEF DESCRIPTION OF THE DRAWING FIGURES





FIG. 1

depicts a step in the invention process in which SiN liner is deposited below the collar region and on the collar region of a semiconductor substrate.





FIG. 2

depicts a step in the invention process in which amorphous silicon (a-Si) is deposited onto the SiN liner to form a micromask.





FIG. 3

depicts a step in the invention process of affecting an anneal/oxidation to form an oxide dot hardmask.





FIG. 4

depicts a step in the invention process of affecting a SiN liner etch, wherein the etch is selective to SiO.





FIG. 5

depicts a step in the invention process in which a Si transfer etch (CDE) selective to SiO is affected to create rough Si surface.





FIG. 6

depicts a step in the invention process in which there is affected a SiO plus SiN strip step that forms a surface enhanced DT. This step is followed by formation of a node plus collar deposition using SiO


2


or SiN as the collar material.





FIG. 7

shows a scanning electron micrograph (SEM) cross-section of oxide micromask after oxidation, as shown in FIG.


3


.











DESCRIPTION OF PREFERRED EMBODIMENT OF THE INVENTION




The method of forming the DT DRAM semiconductor structure to obtain an oxide hardmask for DT surface area enhancement and increased node capacitance, will now be described in detail.




Referring now to

FIG. 1

, there is shown a schematic of a DT DRAM with an area designated as a collar region and an area designated as the below collar region. On this structure, there is formed a SiN liner deposition


10


on a SiO layer


9


, which has been deposited on a single crystal silicon, polycrystalline silicon or like semiconductor material




From

FIG. 2

, it may be seen that, following the SiN liner deposition, an amorphous silicon (a-Si) layer


11


is deposited. Preferably, the a-Si layer is deposited in a thickness of about 100Å. At this juncture, it can be seen that the rough oxide mask has not yet been formed.




Following the steps of anneal/oxidation, preferably under wet conditions at a temperature of about 900° C., the structure of

FIG. 2

is converted to that of

FIG. 3

, which shows the resulting oxide dot hard mask


12


that is used to transfer the pattern to the Si sidewall of the trench. Next, a SiN liner etch is affected that is selective to SiO, as is shown in FIG.


4


. The etch step maybe performed using HF, a chemical dry etch (CDE) or a mixture of ethylene glycol (EG) and HF. Following the SiN liner etch shown in

FIG. 4

, a Si transfer etch is performed using chemical dry etching (CDE) that is selective to SiO to create a rough Si surface


13


that is shown in FIG.


5


. After creation of the rough Si, as shown in

FIG. 5

, SiO and SiN are stripped to arrive at the structure shown in FIG.


6


.




Formation of a node plus collar deposition using SiO


2


or SiN as the collar material is accomplished by integration of the invention process into a collar process.




As may be seen from

FIG. 7

, following anneal/oxidation, preferably under wet conditions at a temperature of about 900° C., there is formation of an oxide dot hardmask or micromask, which is depicted in the SEM cross-section showing the oxide micromask. The rough oxide mask formation using oxidation of the deposited a-Si onto the SiN liner is used to transfer the pattern to the Si sidewall, thereby providing DT surface area enhancement of the DRAM capacitor structure.




A key step in the creation of a rough Si surface is a rough mask material that is used to transfer the pattern to the Si sidewall. In essence, the invention process creates a rough oxide etch mask, based on the anneal/oxidation of deposited amorphous Si on the SiN liner. Further, the invention process/integration scheme allows integration of this process into a conventional collar process using SiO


2


or SiN as the collar material.




It will be seen that the invention consist of two aspects:




The formation of a rough oxide mask using oxidation of deposited a-Si onto a SiN liner for use as a micromask; and a scheme to integrate this process into a DT collar process.



Claims
  • 1. In a process for making a DT DRAM structure, the improvement of providing a surface area enhanced DT below the collar region and node capacitance that does not shrink with decreasing groundrule/cell size, comprising:a) providing a semiconductor substrate having a collar region and an adjacent region below said collar region, said collar region having SiO deposited thereon; b) depositing a SiN liner on said collar region and on said region below said collar; c) depositing a layer of a-Si on said SiN liner to form a micromask; d) subjecting the structure from step c) to an anneal/oxidation step under a wet environment at a sufficient temperature to form a plurality of oxide dot hardmasks; e) subjecting the SiN liner to an etch selective to SiO; f) subjecting the structure from step e) to a Si transfer etch using a chemical dry etch (CDE) selective to SiO to create rough Si surface; g) stripping SiO and the SiN; and forming a node and collar deposition.
  • 2. The process of claim 1 wherein said a-Si layer is deposited in a thickness of about 100Å.
  • 3. The process of claim 2 wherein said anneal/oxidation is at a temperature of about 900° C.
  • 4. The process of claim 3 wherein said SiN liner etch is conducted using CDE.
  • 5. The process of claim 3 wherein said SiN liner etch is conducted using HF.
  • 6. The process of claim 3 wherein said SiN liner etch is conducted using a mixture of HF and ethylene glycol.
  • 7. The process of claim 1 wherein the collar deposition of step g) is formed using SiO2.
  • 8. The process of claim 1 wherein the collar deposition of step g) is formed using SiN.
  • 9. The process of claim 1 wherein said substrate is a single crystal silicon.
  • 10. The process of claim 1 wherein said substrate is a polycrystalline silicon.
US Referenced Citations (10)
Number Name Date Kind
5340763 Dennison Aug 1994 A
5444653 Nagasawa Aug 1995 A
5814549 Wu Sep 1998 A
5877061 Halle et al. Mar 1999 A
5981350 Geusic et al. Nov 1999 A
6037220 Chien Mar 2000 A
6232171 Mei May 2001 B1
6448131 Cabral et al. Sep 2002 B1
6455369 Forster et al. Sep 2002 B1
20020106857 Jammy et al. Aug 2002 A1