Embodiments described herein relate to a method of manufacturing a semiconductor device.
In recent years, fine patterns of a semiconductor device have been formed by using double sidewall pattern transfer in many cases. In addition, it is studied to introduce quadruple sidewall pattern transfer that repeats the double sidewall pattern transfer twice. However, the quadruple sidewall pattern transfer has problems that the number of steps is increased and requirements for alignment accuracy are more stringent, compared with the double sidewall pattern transfer. In addition, when the quadruple sidewall pattern transfer is introduced to a process of manufacturing a NAND memory, the quadruple sidewall pattern transfer complicates steps of forming ladder portions used to connect device regions and pad portions used to dispose contact plugs on word lines. Accordingly, there is a demand for a method which can form these portions by a simple process.
Embodiments will now be explained with reference to the accompanying drawings.
In one embodiment, a method of manufacturing a semiconductor device includes sequentially forming a workpiece film, a first sacrificial film, a first protective film, a second sacrificial film and a second protective film on a substrate, and processing the second protective film and the second sacrificial film to form first core material patterns having a first width and a second core material pattern having a second width larger than the first width. The method further includes forming first and second sidewall patterns formed of a first sidewall material on side surfaces of the first and second core material patterns, respectively, and forming third and fourth sidewall patterns formed of a second sidewall material on the side surfaces of the first and second core material patterns via the first and second sidewall patterns, respectively. The method further includes removing the first core material patterns and the first sidewall patterns so that the second core material pattern and the second, third and fourth sidewall patterns remain. The method further includes processing the first protective film and the first sacrificial film by transferring the second core material pattern and the second, third and fourth sidewall patterns by etching to form third core material patterns having a third width and a fourth core material pattern having a fourth width larger than the third width, and forming fifth and sixth sidewall patterns formed of a third sidewall material on side surfaces of the third and fourth core material patterns, respectively. The method further includes removing the third core material patterns so that the fourth core material pattern and the fifth and sixth sidewall patterns remain, and processing the workpiece film by transferring the fourth core material pattern and the fifth and sixth sidewall patterns by etching.
The method of the present embodiment is an example of forming active areas (AAs) and shallow trench isolations (STIs) of a NAND memory by using quadruple sidewall pattern transfer. At this time, ladder portions for connecting the AAs are also formed.
First, as illustrated in
The semiconductor substrate 101 is, for example, a silicon substrate.
The gate dielectric film 102 and the floating gate material 103 are, for example, a silicon oxide film and a polysilicon layer, respectively. The gate dielectric film 102 and the floating gate material 103 are an example of a workpiece film. The thickness of the gate dielectric film 102 is, for example, 6 nm, and the thickness of the floating gate material 103 is, for example, 50 nm.
The first hard mask layer 111 and the second hard mask layer 112 are, for example, a silicon oxide film and an amorphous silicon layer, respectively, and are formed by low pressure chemical vapor deposition (LPCVD). The thickness of the first hard mask layer 111 is, for example, 150 nm, and the thickness of the second hard mask layer 112 is, for example, 100 nm.
The first sacrificial film 113 is, for example, a carbon film, and is formed by plasma CVD. The carbon film is an example of a carbon-containing film. The first protective film 114 is used to protect the first sacrificial film 113. The first protective film 114 is, for example, a silicon oxynitride (SION) film, and is formed by plasma CVD. The thickness of the first sacrificial film 113 is, for example, 120 nm, and the thickness of the first protective film 114 is, for example, 50 nm.
The second sacrificial film 115 is, for example, a spin-on-carbon (SOC) film, and is formed by coating. The SOC film is an example of a carbon-containing film. The second protective film 116 is used to protect the second sacrificial film 115. The second protective film 116 is, for example, a silicon oxynitride film, and is formed by plasma CVD. The thickness of the second sacrificial film 115 is, for example, 100 nm, and the thickness of the second protective film 116 is, for example, 30 nm.
Next, as illustrated in
The first resist patterns 117a are used to form the AAs and the STIs of the memory cell array portion, whereas the second resist pattern 117b is used to form the AAs and the STIs of the peripheral circuit portion. The width WA is set to, for example, 3F (“F” denotes a feature size which is the minimum dimension in the generation of the semiconductor device of the present embodiment). The value of “F” is, for example, 16 nm.
In the above lithography, there is further formed a ladder portion 117c for connecting the first resist patterns 117a to one another. In
Next, as illustrated in
Next, as illustrated in
At this time, the etching rates of the second protective film 116 and the second sacrificial film 115 are affected by a loading effect. The loading effect causes the etching rate of a thin pattern to become faster, whereas it causes the etching rate of a thick pattern to become slower. The reason for this is that the thin pattern is more largely affected by etching from side surfaces, compared with the thick pattern. The influence of the loading effect generally becomes more significant as a pattern becomes finer.
Next, as illustrated in
As a result, first sidewall patterns 118a, second sidewall patterns 118b, and first ladder sidewall patterns 118c are formed on side surfaces of the first core material patterns 115a, 116a, the second core material pattern 115b, 116b, and the first ladder portion 115c, 116c, respectively. At this time, the second protective film 116 in the first core material patterns 115a, 116a is either thinned or removed.
The first sidewall material 118 is, for example, a carbon film. The thickness of the first sidewall material 118 is set to, for example, the feature size “F”.
Next, as illustrated in
As a result, third sidewall patterns 119a, fourth sidewall patterns 119b, and second ladder sidewall patterns 119c are formed on the side surfaces of the first core material patterns 115a, 116a, the second core material patterns 115b, 116b, and the first ladder portions 115c, 116c via the first sidewall patterns 118a, the second sidewall patterns 118b, and the first ladder sidewall patterns 118c, respectively. The second protective film 116 in the first core material patterns 115a, 116a is completely removed by the end of this process.
The second sidewall material 119 is formed of a material different from those of the second sacrificial film 115 and the first sidewall material 118 so as to selectively leave the second sidewall material 119 in the following process of
Next, as illustrated in
At this time, the second sacrificial film 115 in the first core material patterns 115a is removed since it is not covered with the second protective film 116. On the other hand, the second sacrificial film 115 in the second core material pattern 115b, 116b and the first ladder portion 115c, 116c remains since it is covered with the second protective film 116. Similarly, the first sidewall patterns 118a are removed since they are not covered with the third sidewall patterns 119a. On the other hand, the second sidewall patterns 118b and the first ladder sidewall patterns 118c remain since they are respectively covered with the fourth sidewall patterns 119b and the second ladder sidewall patterns 119c.
As a result, the first core material patterns 115a and the first sidewall patterns 118a are removed, and the second core material pattern 115b, 116b, the second sidewall patterns 118b, the third sidewall patterns 119a, and the fourth sidewall patterns 119b remain in the process of
Next, as illustrated in
At this time, the etching rates of the first protective film 114 and the first sacrificial film 113 are affected by the loading effect.
Next, as illustrated in
As a result, fifth sidewall patterns 120a, sixth sidewall patterns 120b, and third ladder sidewall patterns 120c are formed on side surfaces of the third core material patterns 113a, 114a, the fourth core material pattern 113b, 114b, and the second ladder portion 113c, 114c, respectively. The first protective film 114 in the third core material patterns 113a, 114a is completely removed by the end of this process.
The third sidewall material 120 is formed of a material different from that of the first sacrificial film 113 so as to selectively leave the third sidewall material 120 in the following process of
Next, as illustrated in
At this time, the first sacrificial film 113 in the third core material patterns 113a is removed since it is not covered with the first protective film 114. On the other hand, the first sacrificial film 113 in the fourth core material pattern 113b, 114b and the second ladder portion 113c, 114c remain since it is covered with the first protective film 114.
As a result, the third core material patterns 113a are removed, and the fourth core material pattern 113b, 114b, the fifth sidewall patterns 120a, and the sixth sidewall patterns 120b remain in the process of
Next, as illustrated in
As a result, the first and second hard mask layers 111 and 112 are processed into first mask patterns 111a, 112a having a fifth width WG and a second mask pattern 111b, 112b having a sixth width WH larger than the fifth width WG. In addition, there are formed a ladder portion 111c, 112c for connecting the first mask patterns 111a, 112a to one another. The fifth width WG is, for example, the feature size “F”.
Next, as illustrated in
As a result, isolation trenches 121 dividing the floating gate material 103 and the gate dielectric film 102 are formed on a surface of the semiconductor substrate 101. In addition, there are also formed first device regions (AA) 101a having the fifth width WG, a second device region 101b having the sixth width WH larger than the fifth width WG, and a ladder portion 101c for connecting the first device regions 101a to one another. The depth from the upper surface of the semiconductor substrate 101 to the bottom surfaces of the isolation trenches 121 is set to, for example, 180 nm.
Next, as illustrated in
Hereinafter, a description will be given of effects of the first embodiment.
The method of the present embodiment allows the quadruple sidewall pattern transfer to be performed by only one lithography process shown in
Conventional quadruple sidewall pattern transfer requires a lithography process of forming the thin pattern and a lithography process of forming the thick pattern. The reason for this is that if these patterns are processed at the same time, a portion of the thick pattern which needs to be left is also etched when the thin pattern is processed by etching.
On the other hand, in the method of the present embodiment, the width WC of the first core material patterns 115a, 116a is previously set to a small value (for example, WC=F). Consequently, the thickness of the second protective film 116 in the first core material patterns 115a, 116a becomes significantly smaller than the thickness of the second protective film 116 in the second core material pattern 115b, 116b due to the loading effect as the process of
Accordingly, the second protective film 116 in the first core material patterns 115a, 116a can be removed by the end of the process of
When the width WC of the first core material patterns 115a, 116a is set to a small value, it is not possible to secure a core material width required for the quadruple sidewall pattern transfer. Therefore, the first sidewall patterns 118a are formed on the side surfaces of the first core material patterns 115a, 116a before forming the third sidewall patterns 119a in the present embodiment. Consequently, according to the present embodiment, a substantial core material width can be increased to the sum of the widths of a first core material pattern 115a, 116a and first sidewall patterns 119a to secure the required core material width.
In the present embodiment, the width WE of the third core material patterns 113a, 114a is also set to a small value (for example, WE=F). Consequently, the third core material patterns 113a can be removed in the process of
For the reason described above, the present embodiment also allows a ladder portion for connecting the AAs to one another to be formed in the same lithography process, in addition to the thin pattern for the AAs in the memory cell array portion and the thick pattern for the AAs in the peripheral circuit portion. According to the ladder portion of the present embodiment, it is possible to prevent patterns from collapse and to decrease the number of source contacts.
The method of the present embodiment uses films which contain carbon (carbon-containing films) as the first sacrificial film 113, the second sacrificial film 115 and the first sidewall material 118. The carbon-containing films have advantages that higher etching selectivity thereof for RIE can be easily obtained with respect to a silicon oxide film and a silicon nitride film and that the carbon-containing films can be removed by ashing. Accordingly, the first sacrificial film 113, the second sacrificial film 115 and the first sidewall material 118 can be removed without applying wet etching when the carbon-containing films are used for these films. There is therefore no need for drying treatment after wet etching in this case. In addition, it is possible to prevent pattern leaning due to drying treatment in this case.
The material of the second sacrificial film 115 and the first sidewall material 118 (hereinafter referred to as Material 1), the material of the second sidewall material 119 (hereinafter referred to as Material 2), and the method of removing the second sacrificial film 115 and the first sidewall material 118 (hereinafter referred to as Method 1) allow for use in a variety of combinations.
For example, Material 1, Material 2, and Method 1 may respectively be a silicon oxide film, a silicon nitride film, and wet etching using buffered hydrofluoric acid in a case where wet etching is used.
Alternatively, Material 1, Material 2, and Method 1 may respectively be a silicon nitride film, a silicon oxide film, and wet etching using hot phosphoric acid.
Furthermore, Material 1, Material 2, and Method 1 may respectively be a silicon oxide film, a titanium nitride film, and wet etching using buffered hydrofluoric acid.
Material 1, Material 2, and Method 1 may also be applied as a combination of the material of the first sacrificial film 113, the material of the third sidewall material 120, and the method of removing the first sacrificial film 113.
As described above, according to the present embodiment, the quadruple sidewall pattern transfer can be performed with a reduced number of steps because both a thin pattern and a thick pattern can be formed by the same process.
In addition, according to the present embodiment, it is possible to form fine patterns having a width which is approximately a feature size, by the quadruple sidewall pattern transfer with a reduced number of steps. Although the present embodiment has presented an example in which fine patterns having a width “F” are formed in a case where F=16 nm, it is possible according to the present embodiment to form fine patterns having a width of 20 nm or smaller by using the quadruple sidewall pattern transfer.
The method of the present embodiment is an example of forming word lines and select lines of a NAND memory by using quadruple sidewall pattern transfer. In other words, the present embodiment shows a case of forming gate structures of cell transistors and select transistors of the NAND memory. At this time, pad portions (hookup portion) connected to the word lines are also formed. A process of manufacturing the select transistors in the present embodiment is also applicable to peripheral transistors.
The following description of the second embodiment will omit to describe matters common to the first embodiment.
First, as illustrated in
In the present embodiment, AAs and STIs are formed on the semiconductor substrate 201 between a step of forming the floating gate material 203 and a step of forming the intergate dielectric film 204.
The intergate dielectric film 204, the first control gate material 205, the second control gate material 206, and the cap layer 207 are, for example, a silicon oxide film, a polysilicon layer, a tungsten/tungsten nitride stack film, and a silicon nitride film, respectively. The gate dielectric film 202, the floating gate material 203, the intergate dielectric film 204, the first control gate material 205, the second control gate material 206 and the cap layer 207 are an example of a workpiece film. The thicknesses of the intergate dielectric film 204, the first control gate material 205, the second control gate material 206 and the cap layer 207 are, for example, 9 nm, 20 nm, 50 nm and 20 nm, respectively.
In the present embodiment, an opening 204a penetrating the intergate dielectric film 204 is formed for the gate electrode of a select transistor between a step of forming the first control gate material 205 and a step of forming the second control gate material 206.
For the materials, thicknesses, and formation methods of other layers shown in
Next, as illustrated in
Next, as illustrated in
The first resist patterns 217a are used to form cell transistors, whereas the second resist pattern 217b is used to form a select transistor. For example, the width of the first resist patterns 217a is set to 3F before slimming, and set to “F” after slimming (“F” denotes a feature size). The value of “F” is, for example, 13 nm.
Next, as illustrated in
At this time, the etching rates of the second protective film 216 and the second sacrificial film 215 are affected by the loading effect, similarly to the first embodiment.
Next, as illustrated in
As a result, first sidewall patterns 218a, second sidewall patterns 218b, and a first pad sidewall pattern 218c are formed on side surfaces of the first core material patterns 215a, 216a, the second core material patterns 215b, 216b, and the first pad portions 215c, 216c, respectively. The thickness of the first sidewall material 218 is set to, for example, the feature size “F”.
Next, as illustrated in
As a result, third sidewall patterns 219a, fourth sidewall patterns 219b, and a second pad sidewall pattern 219c are formed on the side surfaces of the first core material patterns 215a, 216a, the second core material patterns 215b, 216b, and the first pad portions 215c, 216c via the first sidewall patterns 218a, the second sidewall patterns 218b, and the first pad sidewall pattern 218c, respectively. The second protective film 216 in the first core material patterns 215a, 216a is completely removed by the end of this process. The thickness of the second sidewall material 219 is set to, for example, the feature size “F”.
Next, as illustrated in
Next, the first pad portions 215c, 216c, the first pad sidewall pattern 218c, and the second pad sidewall pattern 219c in the resist opening 221a are etched by wet etching and RIE. As a result, the first and second pad sidewall patterns 218c and 219c are loop-cut to be divided into two parts (see
Next, as illustrated in
At this time, the second sacrificial film 215 in the first core material patterns 215a is removed since it is not covered with the second protective film 216. On the other hand, the second sacrificial film 215 in the second core material pattern 215b, 216b and the first pad portions 215c, 216c remain since it is covered with the second protective film 216.
Similarly, the first sidewall patterns 218a are removed since they are not covered with the third sidewall patterns 219a. On the other hand, the second sidewall patterns 218b and the first pad sidewall pattern 218c remain since they are covered with the fourth sidewall patterns 219b and the second pad sidewall pattern 219c, respectively.
As a result, the first core material patterns 215a and the first sidewall patterns 218a are removed, whereas the second core material pattern 215b, 216b, the second sidewall patterns 218b, the third sidewall patterns 219a, and the fourth sidewall patterns 219b remain in the process of
The second sacrificial film 215 in the first pad portions 215c, 216c are etched in the vicinity of a region R1 by the action of the isotropic etching. As a result, the second sacrificial film 215 in the first pad portions 215c, 216c is divided in the vicinity of the region R1. This isotropic etching is continued until the division of the second sacrificial film 215 in the first pad portions 215c, 216c is completed.
Next, as illustrated in
At this time, the etching rates of the first protective film 214 and the first sacrificial film 213 are affected by the loading effect, similarly to the first embodiment.
Next, as illustrated in
As a result, fifth sidewall patterns 220a, sixth sidewall patterns 220b and third pad sidewall patterns 220c are formed on side surfaces of the third core material patterns 213a, 214a, the fourth core material patterns 213b, 214b, and the second pad portions 213c, 214c, respectively. The first protective film 214 in the third core material patterns 213a, 214a is completely removed by the end of this process. The thickness of the third sidewall material 220 is set to, for example, the feature size “F”.
Next, as illustrated in
Next, the second pad portions 213c, 214c and the third pad sidewall patterns 220c in the resist openings 222a are etched by wet etching and RIE. As a result, the third pad sidewall patterns 220c are loop-cut to be divided into four parts (see
Next, as illustrated in
At this time, the first sacrificial film 213 in the third core material patterns 213a is removed since it is not covered with the first protective film 214. On the other hand, the first sacrificial film 213 in the fourth core material pattern 213b, 214b and the second pad portions 213c, 214c remains since it is covered with the first protective film 214.
As a result, the third core material patterns 213a are removed, whereas the fourth core material pattern 213b, 214b, the fifth sidewall patterns 220a, and the sixth sidewall patterns 220b remain in the process of
The first sacrificial film 213 in the second pad portions 213c, 214c is etched by the action of the isotropic etching in the vicinity of regions R2. As a result, the first sacrificial film 213 in the second pad portions 213c, 214c is divided in the vicinity of the regions R2. This isotropic etching is continued until the division of the first sacrificial film 213 in the second pad portions 213c, 214c is completed.
Next, as illustrated in
As a result, the second hard mask layer 212 is processed into first mask patterns 212a having a fifth width and a second mask pattern 212b having a sixth width larger than the fifth width. In addition, there are formed pad portions 212c connected to the first mask patterns 212a. The fifth width is, for example, the feature size “F”.
Next, as illustrated in
As a result, there are formed cell transistors MC having a fifth width, a select transistor SG having a sixth width larger than the fifth width, and pad portions P connected to the cell transistors MC (word line). As illustrated in
Hereinafter, a description will be given of effects of the second embodiment.
According to the present embodiment, a thin pattern for a cell transistor (hereinafter described as “MC”) and a thick pattern for a select transistor (hereinafter described as “SG”) or a pad portion can be formed by only three lithography processes shown in
In a case where the gate processing of the NAND memory is performed by using conventional quadruple sidewall pattern transfer, a boundary line of a resist film has to be set between a region in which MCs are to be formed and a region in which an SG is to be formed when lithography processes for the MCs and the SG are carried out. The problem at this time is that requirements for the alignment accuracy of the boundary line become more stringent with a decrease in distance between these regions.
In addition, the spacing between the MCs and the SG in the NAND memory is desirably set so as to be the same as the spacing between the MCs. If the spacing between the MCs is set to a fine pitch in a case where such settings are adopted, the spacing between the MCs and the SG is also fine-pitched, thus causing the requirements for the alignment accuracy of the boundary line to be even more stringent.
On the other hand, according to the present embodiment, since such a process of setting the boundary line is unnecessary, the present embodiment can eliminate the need for sophisticated alignment control. In addition, since the present embodiment eliminates such a process as described above, it is easy to set the spacing between the MCs and the SG to the same pitch as the spacing between the MCs.
In addition, according to the present embodiment, the pad portions to be used to dispose the contact plugs on the word lines can be formed by such a simple process as described above. For example, according to the present embodiment, since the isotropic etching shown in
While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the novel methods described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods described herein may be made without departing from the spirit of the inventions. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the inventions.
This application is based upon and claims the benefit of priority from the prior U.S. Provisional Patent Application No. 61/832,199 filed on Jun. 7, 2013, the entire contents of which are incorporated herein by reference.
Number | Date | Country | |
---|---|---|---|
61832199 | Jun 2013 | US |