Capacitor and method for fabricating the same

The present application claims priority of Korean Patent Application No. 10-2012-0048557, filed on May 8, 2012, which is incorporated herein by reference in its entirety.

1. Field

Exemplary embodiments of the present invention relate to a semiconductor device, and more particularly, to a capacitor and a method for fabricating the same.

2. Description of the Related Art

As design rule parameters decrease, a full dip-out process becomes inevitable when fabricating a capacitor with a lower electrode. The full dip-out process refers to a process for fully removing a mold layer after forming the lower electrode.

However, in the full dip-out process, leaning of the lower electrode may occur.

In order to address such a feature, an NFC (nitride floating capacitor) structure is used, in which a support layer including a nitride layer is formed to fixate a plurality of lower electrodes so as to prevent leaning of the lower electrodes. Recently, since a capacitor with a high aspect ratio is used, prevention of leaning of a lower electrode by the NFC structure is desired.

Nevertheless, because of a limited space for forming a cylinder type lower electrode (storage node) in high degree integration, forming a dielectric layer and an upper electrode becomes difficult. While the thickness of the cylinder type lower electrode may be decreased to secure the inner space of the cylinder type lower electrode, leaning is likely to occur in the full dip-out process since support for the cylinder type lower electrode becomes weak. Also, if the thickness of the cylinder type lower electrode is decreased even though the NFC structure is used, cylinder type lower electrodes are likely to lean and touch one another in the full dip-out process.

Embodiments of the present invention are directed to a capacitor which can prevent leaning of a lower electrode and a method for fabricating the same.

In accordance with an embodiment of the present invention, a method for fabricating a capacitor may include: forming a plurality of cylinder type lower electrodes; forming a first upper electrode inside the cylinder type lower electrodes; forming a second upper electrode outside the cylinder type lower electrodes; and forming a third upper electrode that connects the first upper electrode and the second upper electrode.

In accordance with another embodiment of the present invention, a method for fabricating a capacitor may include: forming a mold structure over a substrate, wherein the mold structure has a plurality of open parts and has a mold layer stacked with a support layer; forming cylinder type lower electrodes in the open parts; forming a first upper electrode over an entire surface of a structure including the cylinder type lower electrodes to fill the cylinder type lower electrodes; defining a through hole that passes through portions of the first upper electrode and the support layer; removing the mold layer through the through hole and exposing the cylinder type lower electrodes; forming a second upper electrode to fill the through hole and spaces between the cylinder type lower electrodes; and forming a third upper electrode to connect the second upper electrode and the first upper electrode with each other.

In accordance with yet another embodiment of the present invention, a method for fabricating a capacitor includes: forming a mold layer over a substrate, wherein the mold layer has a plurality of open parts; forming cylinder type lower electrodes in the open parts; forming a first upper electrode over an entire surface of a structure including the cylinder type lower electrodes to fill the cylinder type lower electrodes; defining a through hole that passes through a portion of the first upper electrode; removing the mold layer through the through hole and exposing the cylinder type lower electrodes; forming a second upper electrode to fill the through hole and spaces between the cylinder type lower electrodes; and forming a third upper electrode to connect the second upper electrode and the first upper electrode with each other.

In accordance with still another embodiment of the present invention, a capacitor may include: a plurality of cylinder type lower electrodes; a support layer supporting outer walls of the cylinder type lower electrodes and having a through hole that opens spaces between the cylinder type lower electrodes; a first upper electrode formed in the cylinder type lower electrodes; a second upper electrode surrounding the outer walls of the cylinder type lower electrodes; and a third upper electrode connecting the first upper electrode and the second upper electrode with each other.

FIG. 1A is a cross-sectional view illustrating a capacitor in accordance with a first embodiment of the present invention.

FIG. 1B is a cross-sectional view taken along the line A-A′ of FIG. 1A .

FIGS. 2A to 2I are cross-sectional views illustrating a method for fabricating the capacitor in accordance with the first embodiment of the present invention.

FIG. 3A is a cross-sectional view illustrating a capacitor in accordance with a second embodiment of the present invention.

FIG. 3B is a cross-sectional view taken along the line A-A′ of FIG. 3A .

FIGS. 4A to 4I are cross-sectional views illustrating a method for fabricating the capacitor in accordance with the second embodiment of the present invention.

Exemplary embodiments of the present invention will be described below in more detail with reference to the accompanying drawings. The present invention may, however, be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present invention to those skilled in the art. Throughout the disclosure, like reference numerals refer to like parts throughout the various figures and embodiments of the present invention.

The drawings are not necessarily to scale and in some instances, proportions may have been exaggerated in order to clearly illustrate features of the embodiments. When a first layer is referred to as being “on” a second layer or “on” a substrate, it not only refers to a case where the first layer is formed directly on the second layer or the substrate but also a case where a third layer exists between the first layer and the second layer or the substrate.

FIG. 1A is a cross-sectional view illustrating a capacitor in accordance with a first embodiment of the present invention. FIG. 1B is a cross-sectional view taken along line A-A′ of FIG. 1A .

Referring to FIGS. 1A and 1B , an interlayer dielectric layer 22 is formed on a semiconductor substrate 21 . A plurality of contact plugs 23 are formed in contact holes, which are defined in the interlayer dielectric layer 22 . Cylinder type lower electrodes 30 are formed on the contact plugs 23 . An etch stop layer 24 A is formed on the outer walls of the lower end portions of the lower electrodes 30 . A support layer 26 A is formed on the outer walls of the upper end portions of the lower electrodes 30 . The support layer 26 A is partially removed. A first upper electrode 32 A is formed in the lower electrodes 30 . The first upper electrode 32 A may include pillar parts 32 B, which fill the lower electrodes 30 . A first dielectric layer 31 A is formed between the first upper electrode 32 A and the lower electrodes 30 . A second upper electrode 37 A is formed outside the lower electrodes 30 , that is, between the lower electrodes 30 . The second upper electrode 37 A has a shape for surrounding the outer walls of the lower electrodes 30 . The second upper electrode 37 A may include a connection part 37 B, which is filled through the partially removed portion of the support layer 26 A. A second dielectric layer 36 A is formed between the second upper electrode 37 A and the lower electrodes 30 . A third upper electrode 38 is formed on the first and second upper electrodes 32 A and 37 A. The connection part 37 B of the second upper electrode 37 A and the first upper electrode 32 A are connected with each other through the third upper electrode 38 . The third upper electrode 38 may be formed by stacking a silicon germanium layer 38 A and a tungsten layer 38 B.

According to FIGS. 1A and 1B , the first upper electrode 32 A is formed in the cylinder type lower electrodes 30 , the second upper electrode 37 A is formed outside the cylinder type lower electrodes 30 , and the third upper electrode 38 is formed on the second upper electrode 37 A. The second upper electrode 37 A has a shape for surrounding the outer walls of the cylinder type lower electrodes 30 . The third upper electrode 38 electrically connects the first upper electrode 32 A and the second upper electrode 37 A.

As described above, in the capacitor in accordance with the first embodiment of the present invention, an upper electrode is constituted by the first to third upper electrodes 32 A, 37 A and 38 . The first upper electrode 32 A serves as a prop for preventing the lower electrodes 30 from leaning or bending. More specifically, since the pillar parts 32 B of the first upper electrode 32 A fill the lower electrodes 30 , supporting force increases. Moreover, because the supporting force increases by the pillar parts 32 B of the first upper electrode 32 A, the thickness of the lower electrodes 30 may be decreased.

FIGS. 2A to 2I are cross-sectional views illustrating a method for fabricating the capacitor in accordance with the first embodiment of the present invention.

Referring to FIG. 2A , a plurality of contact plugs 23 are formed on a semiconductor substrate 21 to pass through an interlayer dielectric layer 22 . The semiconductor substrate 21 includes a silicon-containing substance. For example, the semiconductor substrate 21 may include a silicon substrate, a silicon germanium substrate, or the like. The interlayer dielectric layer 22 may include a silicon oxide such as BPSG. While not shown, a process for forming transistors and wiring lines may be additionally performed on the semiconductor substrate 21 before forming the interlayer dielectric layer 22 . The contact plugs 23 may be connected to impurity regions (not shown) that are formed in the semiconductor substrate 21 through contact holes defined in the interlayer dielectric layer 22 . The contact plugs 23 may be formed by forming a conductive layer in the contact holes and subsequently planarizing the conductive layer until the upper surface of the interlayer dielectric layer 22 is exposed. The contact plugs 23 may include a metal layer, a metal nitride layer, a noble metal layer, a refractory metal layer, polysilicon, or the like.

An etch stop layer 24 is formed on the interlayer dielectric layer 22 including the contact plugs 23 . The etch stop layer 24 may include a dielectric substance. For example, the etch stop layer 24 may include a nitride such as a silicon nitride.

A mold layer 25 is formed on the etch stop layer 24 . The mold layer 25 is a substance provided to form lower electrodes (or storage nodes). The mold layer 25 includes a substance with a high etching selectivity with respect to the etch stop layer 24 . Also, the mold layer 25 includes a substance easily removable through wet etching. For example, the mold layer 25 may include an oxide such as a silicon oxide. In another embodiment, the mold layer 25 may include a multi-layered oxide. For example, the mold layer 25 may include BPSG, USG, PETEOS, PSG, HDP, or the like. In still another embodiment, the mold layer 25 may include a silicon-containing substance. For example, the mold layer 25 may include a polysilicon layer or a silicon germanium layer.

A support layer 26 is formed on the mold layer 25 . The support layer 26 may include a nitride such as a silicon nitride. The support layer 26 prevents lower electrodes from leaning in a subsequent full dip-out process. The support layer 26 may include a substance with a high etching selectivity with respect to the mold layer 25 . In the case where the mold layer 25 is formed as a silicon oxide layer, the support layer 26 may be formed as a silicon nitride layer. However, a substance of the support layer 26 is not limited to the materials described above.

Next, hard mask patterns 27 are formed on the support layer 26 . The hard mask patterns 27 may be formed using first photoresist patterns 28 . The hard mask patterns 27 may include a silicon nitride layer, amorphous carbon, or the like.

Referring to FIG. 2B , after stripping the first photoresist patterns 28 , the support layer 26 is etched using the hard mask patterns 27 as an etch barrier. In succession, by etching the mold layer 25 , open parts 29 are defined to expose the contact plugs 23 . The open parts 29 respectively expose the surfaces of the contact plugs 23 . The open parts 29 may have the shapes of contact holes. In order to define the open parts 29 , the support layer 26 and the mold layer 25 may be etched until stopping at the etch stop layer 24 , and subsequently, the etch stop layer 24 may be etched. Accordingly, the open parts 29 are defined in the stack structure of an etch stop layer 24 A, a mold layer 25 A and a support layer 26 A. Lower electrodes are formed in the open parts 29 in a subsequent process.

Next, the hard mask patterns 27 are removed.

Referring to FIG. 2C , lower electrodes 30 are formed in the open parts 29 .

The lower electrodes 30 may be a cylinder type or a pillar type. Herein, the lower electrodes 30 will be referred to as cylinder type lower electrodes.

In order to form the lower electrodes 30 , a first conductive layer may be deposited on the entire surface including the open parts 29 , and subsequently a lower electrode separating process may be performed. The first conductive layer as the lower electrodes 30 may be formed as a metal layer, a metal nitride layer or a stack layer in which a metal layer and a metal nitride layer are stacked. The first conductive layer may be formed through chemical vapor deposition (CVD) or atomic layer deposition (ALD). For example, the lower electrodes 30 may be formed as a stack layer in which a titanium layer and a titanium nitride layer are stacked.

Since the lower electrodes 30 are the cylinder type, the lower electrodes 30 have inner walls and outer walls, and the outer walls of the upper end portions of adjoining lower electrodes 30 are fixed by the support layer 26 A.

Referring to FIG. 2D , a first dielectric layer 31 and a second conductive layer 32 are sequentially formed on the entire surface including the lower electrodes 30 . The second conductive layer 32 becomes a first upper electrode. The second conductive layer 32 includes a metal nitride layer. For example, the second conductive layer 32 may include a titanium nitride layer (TiN). The second conductive layer 32 is filled in the cylinders of the lower electrodes 30 over the first dielectric layer 31 .

Referring to FIG. 2E , a nitride floating cap (NFC) mask 33 is formed on the second conductive layer 32 . The NFC mask 33 is a mask for selectively etching a portion of the support layer 26 A.

Using the NFC mask 33 as an etch barrier, the second conductive layer 32 and the first dielectric layer 31 are etched, and in succession, the portion of the support layer 26 A is etched. Thus, a through hole 34 through which a chemical is to flow subsequently is defined. The through hole 34 is used as a path for passing the chemical to remove the mold layer 25 A. The number of through holes 34 may be selectively controlled.

By defining the through hole 34 in this way, the second conductive layer 32 becomes a first upper electrode 32 A. The first upper electrode 32 A includes pillar parts 32 B, which fill the cylinder type lower electrodes 30 , and covers the upper parts of the cylinder type lower electrodes 30 . After defining of the through hole 34 , the first dielectric layer 31 remains as designated by the reference numeral 31 A. Because the pillar parts 32 B of the first upper electrode 32 A fill the lower electrodes 30 , supporting force is increased.

Referring to FIG. 2F , the NFC mask 33 is removed.

The mold layer 25 A exposed through the through hole 34 is fully removed. To this end, a full dip-out process is performed. The full dip-out process may be performed using a wet chemical. The lower electrodes 30 , the first dielectric layer 31 A and the first upper electrode 32 A are not removed. In particular, since the first dielectric layer 31 A, which uses an oxide similarly to the mold layer 25 A, is interposed between the lower electrodes 30 and the first upper electrode 32 A and is not exposed to the chemical, the first dielectric layer 1 A is not etched. Also, the contact plugs 23 are not damaged due to the presence of the etch stop layer 24 A. Here, the full dip-out process may use a hydrofluoric acid-based chemical.

As the mold layer 25 A is entirely removed as described above, empty spaces 35 are defined outside the lower electrodes 30 . Since the first dielectric layer 31 A and the pillar parts 32 B of the first upper electrode 32 A are filled in the lower electrodes 30 , a structure for preventing the lower electrodes 30 from leaning or bending is obtained. In other words, because the first dielectric layer 31 A and the first upper electrode 32 A are filled in the cylinders of the lower electrodes 30 , the lower electrodes 30 are prevented from leaning or bending even when the full dip-out process is performed.

Referring to FIG. 2G , a second dielectric layer 36 is formed. The second dielectric layer 36 is formed on the outer walls of the lower electrodes 30 . Also, the second dielectric layer 36 is formed on the first upper electrode 32 A.

Subsequently, a third conductive layer 37 is formed on the second dielectric layer 36 . The third conductive layer 37 may include a metal nitride. For example, the third conductive layer 37 may include a titanium nitride. The third conductive layer 37 is formed on the second dielectric layer 36 while filling the spaces between the lower electrodes 30 . The third conductive layer 37 is not connected with the first upper electrode 32 A due to the presence of the second dielectric layer 36 .

Referring to FIG. 2H , the third conductive layer 37 is planarized. Thus, portions of the third conductive layer 37 and the second dielectric layer 36 present on the first upper electrode 32 A may be removed. The planarization process is performed until the surface of the first upper electrode 32 A is exposed. Accordingly, the third conductive layer 37 remains outside the lower electrodes 30 , that is, between the lower electrodes 30 , and is referred to as a second upper electrode 37 A. The second upper electrode 37 A includes a connection part 37 B filled in the through hole 34 . The second upper electrode 37 A has a shape for surrounding the outer walls of the lower electrodes 30 . After the planarization, the second dielectric layer 36 remains as designated by the reference numeral 36 A.

Referring to FIG. 2I , by forming and subsequently patterning a fourth conductive layer, a third upper electrode 38 is formed. The third upper electrode 38 may be formed by stacking a silicon germanium layer 38 A and a tungsten layer 38 B, thereby decreasing resistance. The third upper electrode 38 is electrically connected with the first upper electrode 32 A and the second upper electrode 37 A. The second upper electrode 37 A is electrically connected with the third upper electrode 38 through the connection part 37 B, and the first upper electrode 32 A and the third upper electrode 38 are directly connected with each other.

FIG. 3A is a cross-sectional view illustrating a capacitor in accordance with a second embodiment of the present invention. FIG. 3B is a cross-sectional view taken along the line A-A′ of FIG. 3A .

Referring to FIGS. 3A and 3B , an interlayer dielectric layer 42 is formed on a semiconductor substrate 41 . A plurality of contact plugs 43 are formed in contact holes, which are defined in the interlayer dielectric layer 42 . Cylinder type lower electrodes 49 are respectively formed on the contact plugs 43 . An etch stop layer 44 A is formed on the outer walls of the lower end portions of the lower electrodes 49 . A first upper electrode 51 A is formed in the lower electrodes 49 . The first upper electrode 51 A may include pillar parts 51 B which fill the lower electrodes 49 . A first dielectric layer 50 A is formed between the first upper electrode 51 A and the lower electrodes 49 . A second upper electrode 56 A is formed outside the lower electrodes 49 , that is, between the lower electrodes 49 . The second upper electrode 56 A has a shape for surrounding the outer walls of the lower electrodes 49 . The second upper electrode 56 A may include a connection part 56 B. A second dielectric layer 55 A is formed between the second upper electrode 56 A and the lower electrodes 49 . A third upper electrode 57 is formed on the first and second upper electrodes 51 A and 56 A. The connection part 56 B of the second upper electrode 56 A and the first upper electrode 51 A are connected with each other through the third upper electrode 57 . The third upper electrode 57 may be formed by stacking a silicon germanium layer 57 A and a tungsten layer 57 B.

According to FIGS. 3A and 3B , the first upper electrode 51 A is formed in the cylinder type lower electrodes 49 , the second upper electrode 56 A is formed outside the cylinder type lower electrodes 49 , and the third upper electrode 57 is formed on the second upper electrode 56 A. The second upper electrode 56 A has the shape for surrounding the outer walls of the cylinder type lower electrodes 49 . The third upper electrode 57 electrically connects the first upper electrode 51 A and the second upper electrode 56 A.

As described above, in the capacitor in accordance with the second embodiment of the present invention, an upper electrode is constituted by the first to third upper electrodes 51 A, 56 A and 57 . The first upper electrode 51 A serves as a prop/support for preventing the lower electrodes 49 from leaning or bending. More specifically, since the pillar parts 51 B of the first upper electrode 51 A fill the lower electrodes 49 , supporting force increases. Moreover, because the supporting force increases by the pillar parts 51 B of the first upper electrode 51 A, the thickness of the lower electrodes 49 may be decreased.

FIGS. 4A to 4I are cross-sectional views illustrating a method for fabricating the capacitor in accordance with the second embodiment of the present invention.

Referring to FIG. 4A , a plurality of contact plugs 43 are formed on a semiconductor substrate 41 to pass through an interlayer dielectric layer 42 . The semiconductor substrate 41 includes a silicon-containing substance. For example, the semiconductor substrate 41 may include a silicon substrate, a silicon germanium substrate, or the like. The interlayer dielectric layer 42 may include a silicon oxide such as BPSG. While not shown, a process for forming transistors and wiring lines may be additionally performed on the semiconductor substrate 41 before forming the interlayer dielectric layer 42 . The contact plugs 43 may be connected to impurity regions (not shown) formed in the semiconductor substrate 41 through contact holes defined in the interlayer dielectric layer 42 . The contact plugs 43 may be formed by forming a conductive layer in the contact holes and subsequently planarizing the conductive layer until the upper surface of the interlayer dielectric layer 42 is exposed. The contact plugs 43 may include a metal layer, a metal nitride layer, a noble metal layer, a refractory metal layer, polysilicon, or the like.

An etch stop layer 44 is formed on the interlayer dielectric layer 42 including the contact plugs 43 . The etch stop layer 44 may include a dielectric substance. For example, the etch stop layer 44 may include a nitride such as a silicon nitride.

A mold layer 45 is formed on the etch stop layer 44 . The mold layer 45 is a substance provided to form lower electrodes (or storage nodes). The mold layer 45 includes a substance with a high etching selectivity with respect to the etch stop layer 44 . Also, the mold layer 45 includes a substance easily removable through wet etching. For example, the mold layer 45 may include an oxide such as a silicon oxide. In another embodiment, the mold layer 45 may include a multi-layered oxide. For example, the mold layer 45 may include BPSG, USG, PETEOS, PSG, HDP, or the like. In still another embodiment, the mold layer 45 may include a silicon-containing substance. For example, the mold layer 45 may include a polysilicon layer or a silicon germanium layer.

Hard mask patterns 46 are formed on the mold layer 45 . The hard mask patterns 46 may be formed using first photoresist patterns 47 . The hard mask patterns 46 may include a silicon nitride layer, amorphous carbon, or the like.

Referring to FIG. 4B , after stripping the first photoresist patterns 47 , the mold layer 45 and the etch stop layer 44 are etched using the hard mask patterns 46 as an etch barrier. Thus, open parts 48 are defined to expose the contact plugs 43 . The open parts 48 expose the surfaces of the contact plugs 43 . The open parts 48 may have the shapes of contact holes. In order to define the open parts 48 , the mold layer 45 may be etched until stopping at the etch stop layer 44 , and subsequently, the etch stop layer 44 may be etched. Accordingly, the open parts 48 are defined in the stack structure of an etch stop layer 44 A and a mold layer 45 A. Lower electrodes are formed in the open parts 48 in a subsequent process.

Next, the hard mask patterns 46 are removed.

Referring to FIG. 4C , lower electrodes 49 are formed in the open parts 48 .

The lower electrodes 49 may be a cylinder type or a pillar type. Hereinbelow, in the embodiment, the lower electrodes 49 will be referred to as cylinder type lower electrodes.

In order to form the lower electrodes 49 , a first conductive layer may be deposited on the entire surface including the open parts 48 , and subsequently a lower electrode separating process may be performed. The first conductive layer as the lower electrodes 49 may be formed as a metal layer, a metal nitride layer or a stack layer in which a metal layer and a metal nitride layer are stacked. The first conductive layer may be formed through chemical vapor deposition (CVD) or atomic layer deposition (ALD). For example, the lower electrodes 49 may be formed as a stack layer in which a titanium layer and a titanium nitride layer are stacked.

Since the lower electrodes 49 are the cylinder type, the lower electrodes 49 have inner walls and outer walls.

Referring to FIG. 4D , a first dielectric layer 50 and a second conductive layer 51 are sequentially formed on the entire surface including the lower electrodes 49 . The second conductive layer 51 becomes a first upper electrode. The second conductive layer 51 includes a metal nitride layer. For example, the second conductive layer 51 may include a titanium nitride layer (TiN). The second conductive layer 51 is filled in the cylinders of the lower electrodes 49 over the first dielectric layer 50 .

Referring to FIG. 4E , a second photoresist pattern 52 is formed on the second conductive layer 51 . The second photoresist pattern 52 is a mask for selectively etching a portion of the second conductive layer 51 .

Using the second photoresist pattern 52 as an etch barrier, the second conductive layer 51 and the first dielectric layer 50 are etched. Thus, a through hole 53 through which a chemical is to flow subsequently is defined. The through hole 53 is used as a path for flowing the chemical to remove the mold layer 45 A. The number of through holes 53 may be selectively controlled.

By defining the through hole 53 in this way, the second conductive layer 51 becomes a first upper electrode 51 A. The first upper electrode 51 A includes pillar parts 51 B, which fill the cylinder type lower electrodes 49 , and covers the upper parts of the cylinder type lower electrodes 49 . Here, after the formation of the through hole 53 , the first dielectric layer 50 remains as designated by the reference numeral 50 A. Because the pillar parts 51 B of the first upper electrode 51 A fill the lower electrodes 49 , supporting force is increased.

Referring to FIG. 4F , after removing the second photoresist pattern 52 , the mold layer 45 A exposed through the through hole 53 is fully removed. To this end, a full dip-out process is performed. The full dip-out process may be performed using a wet chemical. The lower electrodes 49 , the first dielectric layer 50 A and the first upper electrode 51 A are not removed. In particular, since the first dielectric layer 50 A, which uses an oxide similarly to the mold layer 45 A, is interposed between the lower electrodes 49 and the first upper electrode 51 A and is not exposed to the chemical, the first dielectric layer 50 A is not etched. Also, the contact plugs 43 are not damaged due to the presence of the etch stop layer 44 A. Here, the full dip-out process may use a hydrofluoric acid-based chemical.

As the mold layer 45 A is entirely removed as described above, empty spaces 54 are defined outside the lower electrodes 49 . Since the first dielectric layer 50 A and the pillar parts 51 B of the first upper electrode 51 A are filled in the lower electrodes 49 , a structure for preventing the lower electrodes 49 from leaning or bending is obtained. In other words, because the first dielectric layer 50 A and the first upper electrode 51 A are filled in the cylinders of the lower electrodes 49 , the lower electrodes 49 are prevented from leaning or bending even when the full dip-out process is performed.

Referring to FIG. 4G , a second dielectric layer 55 is formed. The second dielectric layer 55 is formed on the outer walls of the lower electrodes 49 . Also, the second dielectric layer 55 is formed on the first upper electrode 51 A.

Subsequently, a third conductive layer 56 is formed on the second dielectric layer 55 . The third conductive layer 56 may include a metal nitride. For example, the third conductive layer 56 may include a titanium nitride. The third conductive layer 56 is formed on the second dielectric layer 55 while filling the spaces between the lower electrodes 49 . The third conductive layer 56 is not connected with the first upper electrode 51 A due to the presence of the second dielectric layer 55 .

Referring to FIG. 4H , the third conductive layer 56 is planarized. Thus, portions of the third conductive layer 56 and the second dielectric layer 55 present on the first upper electrode 51 A may be removed. The planarization process is performed until the surface of the first upper electrode 51 A is exposed. Accordingly, the third conductive layer 56 remains outside the lower electrodes 49 , that is, between the lower electrodes 49 , and is referred to as a second upper electrode 56 A. The second upper electrode 56 A includes a connection part 56 B filled in the through hole 53 . The second upper electrode 56 A has a shape for surrounding the outer walls of the lower electrodes 49 . After the planarization, the second dielectric layer 55 remains as designated by the reference numeral 55 A.

Referring to FIG. 4I , by forming and subsequently patterning a fourth conductive layer, a third upper electrode 57 is formed. The third upper electrode 57 may be formed by stacking a silicon germanium layer 57 A and a tungsten layer 57 B, thereby decreasing resistance. The third upper electrode 57 is electrically connected with the first upper electrode 51 A and the second upper electrode 56 A. The second upper electrode 56 A is electrically connected with the third upper electrode 57 through the connection part 56 B, and the first upper electrode 51 A and the third upper electrode 57 are, for example, directly connected with each other.

As is apparent from the above descriptions, in the embodiments of the present invention, since pillar type upper electrodes are preliminarily formed in cylinder type lower electrodes, the lower electrodes are prevented from leaning or bending in a subsequent dip-out process. Furthermore, since supporting force is increased due to the presence of the pillar type upper electrodes, the thickness of the lower electrodes may be decreased.

While the present invention has been described with respect to the specific embodiments, it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the following claims.