Method of fabricating a DRAM device featuring alternate fin type capacitor structures

(1) Field of the Invention

The present invention relates to methods used to fabricate semiconductor devices, and more specifically to a method used to form capacitor structures for a dynamic random access memory (DRAM), device.

(2) Description of Prior Art

The reduction in DRAM cell size has limited the horizontal dimension of DRAM capacitor structures, thus adversely influencing the ability of the semiconductor industry to continually increase DRAM capacitance via use of wide capacitor structures. The use of capacitor structures featuring crown shaped, or fin shaped configurations, has allowed a portion of the desired capacitance increases to still be realized, however the shrinking dimensions of the underlying DRAM transfer gate transistors limit the horizontal dimensions of these capacitor structures, and thus limit capacitor surface area. Thus efforts have been directed at increasing capacitor surface area, via novel fin type capacitor configurations.

Vertical stacked capacitor structures featuring multiple horizontal fin type features, extending horizontally from a vertical capacitor feature that is used to contact an underlying storage node plug structure, has allowed the surface area of the capacitor structure to be increased thus increasing DRAM capacitance. This invention however will describe a process for fabricating a DRAM capacitor structure, comprised with alternate fin type structures, offering increased capacitor surface area, and using less horizontal space, when compared to counterpart, non-alternate fin type capacitor structures. The novel alternative fin type capacitor structure, described in this invention, is comprised of two storage node features, each located in a storage node opening, formed in a stack of composite insulator layers, each overlying and contacting a storage node plug structure. However each storage node opening is defined in the composite insulator layers, with each opening featuring lateral recesses in specific components of the composite insulator layers. The lateral recesses for a first storage node opening are formed in first type components of the composite insulator layers, while the lateral recesses in the second storage node opening are formed in second components of the composite insulator layers. Thus a storage node structure formed in these storage node openings, is comprised of intertwining features, located in the lateral recesses of the storage node openings. The subsequent capacitor structure is therefore comprised with alternate fin type, storage node structures, providing increased surface area in less horizontal space, when compared to counterpart non-alternate fin type structures. Prior art, such as Sandhu, in U.S. Pat. No. 6,133,600, as well as Takaishi, in U.S. Pat. No. 5,903,430, describe methods of fabricating fin type, capacitor structures, however these prior arts do not describe the novel process described in this present invention in which alternate type, DRAM capacitor structures are formed featuring lateral recesses, formed in specific components of a composite insulator layer.

It is an object of this invention to increase the surface area of a DRAM capacitor via use of an alternate, fin type capacitor configuration.

It is another object of this invention to define a first storage node opening in a group of composite insulator layers, then selectively form lateral recesses in first components of the composite insulator layers, followed by definition of a second storage node opening in the same group of composite insulator layers, then selectively form lateral recesses in second components of the composite insulator layers.

It is yet another object of this invention to form a storage node structure, located in the first, and in the second storage node openings, with the storage node structure featuring horizontal, fin type components, located in the lateral recesses of the storage node openings, and with the fin type features in the lateral recesses in the first storage node opening, intertwined with the fin type features in the lateral recesses in the second storage node opening.

In accordance with the present invention a method of fabricating an alternate fin type, DRAM capacitor structure, featuring intertwined storage node structures, is described. After formation of storage node plug structures, contacting source/drain regions of underlying DRAM transfer gate transistors, a series of composite insulator layers, each comprised with a layer of silicon nitride, silicon oxide, borophosphosilicate glass (BPSG), and silicon oxide, is deposited. A first storage node opening is formed in the series of composite insulator layers, exposing the top surface of a first storage node plug structure, followed by lateral recesses selectively formed in the silicon nitride component of each composite insulator layer. A second storage node opening is then formed in the series of composite insulator layers, exposing a second storage node plug structure, followed by lateral recesses selectively formed in the BPSG component of each composite insulator layer. The lateral recesses in the first storage node opening, partially overlays the lateral recesses in the second storage node openings, with these lateral recesses separated by the silicon oxide component of the composite insulator layers. Storage node structures are then formed in the storage node openings, coating the lateral recesses in these openings, resulting in the intertwined horizontal fin type features. After formation of a capacitor dielectric layer on the storage node structures, a capacitor top plate is formed integrating the storage node structures, resulting in an alternate fin type, DRAM capacitor structure.

The object and other advantages of this invention are best described in the preferred embodiment with reference to the attached drawings that include:

FIGS. 1, 2 A, 2 B, 3 - 9 , which schematically, in cross-sectional style, describe key stages of fabrication used to form an alternate fin type, capacitor structure.

The process of fabricating an alternate fin type capacitor structure, featuring storage node structures comprised with horizontal features which are located in lateral recesses in storage node openings, will now be described in detail. A, P-type semiconductor substrate 1 , comprised of single crystalline silicon, with a<100> crystallographic orientation, is used and schematically shown in FIG. 1 . Gate insulator layer 2 , comprised of silicon dioxide, is next thermally grown to a thickness between about 50 to 150 Angstroms. Polysilicon layer 3 , is then deposited via low pressure chemical vapour deposition (LPCVD), procedures, at a thickness between about 500 to 2000 Angstroms. Polysilicon layer 3 , is either doped in situ via the addition of arsine, or phosphine to a silane ambient, or polysilicon layer 3 , is deposited intrinsically then doped via implantation of arsenic or phosphorous ions. If desired a polycide layer, comprised of an overlying tungsten silicide layer, and an underlying doped polysilicon layer, can be used in place of polysilicon to lower the resistance of a subsequent word line. Silicon nitride layer 4 , is next deposited, via LPCVD or via plasma enhanced chemical vapour deposition (PECVD), procedures, at a thickness between about 500 to 1000 Angstroms. Conventional photolithographic and anisotropic reactive ion etching (RIE) procedure, are used to define silicon nitride capped gate structures 6 , shown schematically in FIG. 6 . The silicon nitride capped gate structures are used as word lines for the DRAM device. After removal of the photoresist shape used for definition of silicon nitride capped gate structure 6 , insulator spacers 5 are formed on the sides of silicon nitride capped gate structure 6 , via the deposition of an insulator layer such as silicon oxide or silicon nitride, using LPCVD or PECVD procedures, a thickness between about 500 to 1000 Angstroms, followed by a blanket, anisotropic RIE procedure, using CF 4 as an etchant. Source/drain regions 7 , are then formed in regions of semiconductor substrate 1 , not covered by the silicon capped gate structures 6 , or by insulator spacers 5 , via implantation of arsenic or phosphorous ions, at an energy between about 10 to 30 KeV, and at a dose between about 1 E 13 to 1 E 14 atoms/cm 2 . The result of these procedures is schematically shown in FIG. 1 .

Silicon oxide layer 8 , is next deposited at a thickness between about 5000 to 10000 Angstroms via a high density plasma (HDP) procedure or via LPCVD or PECVD procedures, using tetraethylorthosilicate (TEOS) as a source. To obtain a smooth top surface topography, a chemical mechanical polishing (CMP) procedure is employed for planarization. Silicon nitride layer 9 a , shown schematically in FIG. 1, is then deposited on the planarized top surface of silicon oxide layer 8 , via LPCVD or PECVD procedures, at a thickness between about 2000 to 3000 Angstroms. Photolithographic and anisotropic RIE procedures are next employed to define openings 10 , in silicon nitride layer 9 a , and in silicon oxide layer 8 , using CF 4 as an etchant for silicon nitride layer 9 a , while using CHF 3 as an etchant for silicon oxide layer 8 . Openings 10 , expose regions of source/drain regions 7 . After removal of the photoresist shape, used to define openings 10 , storage node plug structures 11 a , and 11 b , are formed in openings 10 . This is accomplished via deposition of a conductive layer, such as an arsenic or phosphorous, in situ doped polysilicon layer, using LPCVD procedures at a thickness between about 2000 to 3000 RIE procedure, using Cl 2 as an etchant for polysilicon, resulting in storage node structures 11 a , and 11 b , in openings 10 . This is schematically shown in FIG. 2 A. If desired storage node plug structures can be comprised of tungsten or tungsten silicide. In addition bit line structure 40 , shown schematically in FIG. 2B, is formed between gate structures, and subsequent capacitor structures. However to concentrate on the novel capacitor structures described in this invention, and to maintain simplicity in the subsequent drawings, bit line structure will only be illustrated in FIG. 2B, although included as part of the final DRAM device. Bit line contact hole 30 , a dual damascene type opening, is formed in silicon nitride layer 9 a , and in silicon oxide layer 8 , followed by deposition of the conductive layer, and the CMP or RIE procedure, resulting in the dual damascene bit line structure 40 , in opening 30 .

A stack of insulator layers, comprised with layers of: PETEOS silicon oxide, borophosphosilicate glass (BPSG); and silicon nitride, are next deposited and shown schematically in FIG. 3. A first group of insulator layers, comprised of silicon oxide layer 12 a , BPSG layer 13 a , silicon oxide layer 12 b , and silicon nitride layer 9 b , is formed on the top surface of silicon nitride layer 9 a , and on the top surface of storage node plug structures 11 a , and 11 b . Silicon oxide layers 12 a , and 12 b , are obtained at a thickness between about 500 to 1000 Angstroms, via HDP or PECVD procedures, using TEOS as a source. BPSG layer 13 a , at a thickness between about 2000 to 3000 Angstroms, is obtained via PECVD procedures, using TEOS a source, with the addition of diborane and phosphine. Silicon nitride layer 9 b , is obtained at a thickness between about 2000 to 3000 Angstroms, using PECVD or LPCVD procedures. A second group of insulator layers, comprised of silicon oxide layer 12 c , BPSG layer 13 b , silicon oxide layer 12 d , and silicon nitride layer 9 c , is next formed, using identical thicknesses and deposition procedures, used for the comparable components in the first group of insulator layers. Finally a third group of insulator layers, comprised of silicon oxide layer 12 e , and BPSG layer 13 c , is formed on the top surface of the second group of insulator layers. Silicon oxide layer 12 e , is again a PETEOS silicon oxide layer, again obtained via PECVD or HDP procedures, at a thickness between about 500 to 1000 Angstroms, while BPSG layer 13 c , is deposited to a thickness between about 2000 to 3000 Angstroms, via PECVD procedures, identical to weight percents previously used for BPSG layers 13 a , and 13 b . The mode of deposition for the silicon oxide layers, as well as the weight percents of the BPSG layers, will be critical for selective wet etching characteristics, to subsequently be employed to define a specific capacitor opening shape.

A first capacitor opening, or storage node opening 15 a , schematically shown in FIG. 4, is next defined in the multiple insulator layers. Photoresist shape 14 is used as an etch mask to allow an anisotropic RIE procedure, using C 4 F 8 as an etchant for BPSG layers 13 c , 13 b , and 13 a , as well as for silicon oxide layers 12 e , 12 d , 12 c , 12 b , and 12 a , while CH 3 F is used as an etchant for silicon nitride layers 9 c , 9 b , and 9 a . The selective etchant used for silicon nitride layer 9 a , allows the procedure to terminate at the appearance of silicon oxide layer 8 , exposing the top portion of storage node plug structure 11 a . The width, or diameter of capacitor opening 15 a , is between about 0.3 to 0.4 um. With photoresist shape 14 still in place, a selective wet etch procedure using H 3 PO 4 as an etchant, at a temperature between about 160 to 165° C., is used to pull back, or to create lateral recess 16 , in the silicon nitride components of the insulator stack. This is schematically shown in FIG. 5 . Lateral recesses 16 extend between about 2000 to 3000 Angstroms into silicon nitride layers 9 a , 9 b , and 9 c , resulting in fin type capacitor opening 15 b , featuring lateral recesses 16 in the silicon nitride layers, responsible for the horizontal fin type features. Photoresist shape 14 is then removed via plasma oxygen ashing and careful wet clean procedures. This procedure was attainable as a result of the high etch rate of silicon nitride, in H 3 PO 4 , when compared to the lower etch rates of silicon oxide, or BPSG in H 3 PO 4 .

Photoresist shape 17 is now employed as an etch mask to define a second capacitor opening, or storage node opening 18 a , in the multiple insulator layers, while completely filling, and thus protecting, fin type, capacitor opening 5 b , from subsequent dry etch procedures used to define the second capacitor opening. An anisotropic RIE procedure, identical to the procedure previously used to define first capacitor opening 15 a , is again employed for definition of second capacitor opening 18 a . The anisotropic RIE procedure, using C 4 F 8 as an etchant for BPSG and silicon oxide components, and using CH 3 F as an etchant for the silicon nitride components, allow second capacitor opening 18 a , comprised with a width, or diameter between about 0.3 to 0.4 um, to be defined in the multiple insulator layers, exposing a top portion of storage node plug structure 11 b . This is schematically shown in FIG. 6. A dry etch, HF vapour procedure is then used to selectively form lateral recesses in BPSG layers 13 c , 13 b , and 13 a , while undoped silicon oxide layers ( 12 a - 12 e ), remain unetched, resulting in fin type, capacitor opening 18 b . This is schematically shown in FIG. 7 . Lateral recesses 19 , extend horizontally between about 2000 to 3000 Angstroms into the BPSG layers. In regions in which fin type capacitor opening 15 b and fin type capacitor opening 18 b are adjacent, lateral recesses 16 in silicon nitride layers and lateral recesses 19 in BPSG layers overlap or are intertwined. This can be seen schematically in FIG. 7 . Photoresist shape 17 is then removed via plasma oxygen ashing and careful wet clean procedures. This result, the formation of lateral recesses in BPSG, was accomplished as a result of the high etch rate of BPSG in HF vapour, compared to the lower etch rates of silicon oxide, or silicon nitride, exposed to the same HF vapour. If desired, definition of fin type capacitor opening 18 b can be performed prior to definition of fin type capacitor opening 15 b.

The formation of a bottom plate, or a storage node structure, for the DRAM capacitor structure, is next addressed and schematically described in FIGS. 8-9. A polysilicon layer 29 a is deposited via LPCVD procedures, to a thickness between about 500 to 1000 Angstroms. The LPCVD mode of deposition results in a conformal polysilicon layer, easily contouring the surfaces of lateral recesses 16 of fin type capacitor opening 15 b and lateral recesses 19 of fin type capacitor opening 18 b . This is schematically shown in FIG. 8 . Polysilicon layer 29 a , is in situ doped, during deposition, via the addition of arsine or phosphine to a silane ambient. If desired, to further increase subsequent storage node surface area, a rugged, or hemispherical grained silicon layer can be used in place of polysilicon layer 29 a , or the rugged silicon layer can be used overlying polysilicon layer 29 a.

A CMP procedure is next employed to remove regions of polysilicon layer 29 a from the top surface of BPSG layer 13 c creating storage node structures, or bottom plate structures 29 b , for the DRAM capacitor structure. The creation of the lateral recesses in the fin type capacitor openings, and the ability of conformal polysilicon to form a continuous layer in the same lateral recesses, result in storage node structures with increased surface area when compared to counterparts fabricated without the lateral recesses in the capacitor openings. A capacitor dielectric layer 20 , such as oxidized nitride on oxide (ONO), nitride-oxide (NO), or tantalum oxide (Ta 2 O 5 ), is next formed at a thickness between about 50 to 100 Angstroms, on the storage node structures 29 b . This is schematically shown in FIG. 9 . Top plate structure 21 is then formed, via deposition of a conductive material such as in situ doped polysilicon, at a thickness between about 1000 to 1500 Angstroms, using LPCVD procedures. The polysilicon layer results in the filling of the lateral recesses, as well as the complete filling of both fin type capacitor openings. If desired a metal layer can be used as the material for the capacitor top plate structure. Conventional photolithographic and RIE procedures, using Cl 2 as an etchant for polysilicon, are used to define top plate structure 21 , schematically shown in FIG. 9 . The resulting capacitor structure 22 , comprised of top plate structure 21 , capacitor dielectric layer 20 , and storage node structures 29 b , provide increased capacitance as a result of the fin type storage node structures, which in turn provide the additional surface area as a result of the lateral recesses in the fin type capacitor openings. In addition, the concept of intertwining the storage node structures enable less horizontal space to be consumed when compared to counterpart designs not using the intertwined storage node structures. The ability to intertwine these structures was accomplished via formation of lateral recesses in the capacitor openings, which in turn was accomplished via selective wet and dry etching of specific components of a multiple insulator stack.

While this invention has been particularly shown and described with reference to, the preferred embodiments 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 this invention.