US20120067843A1

US20120067843A1 - Method of forming fine pattern

Info

Publication number: US20120067843A1
Application number: US13/064,301
Authority: US
Inventors: Akira Watanabe; Kaori Kimura; Yousuke Isowaki; Yoshiyuki Kamata; Naoko Kihara; Akira Kikitsu
Original assignee: Toshiba Corp
Current assignee: Toshiba Corp
Priority date: 2010-09-16
Filing date: 2011-03-16
Publication date: 2012-03-22
Also published as: JP5171909B2; JP2012064783A

Abstract

A method of forming a fine pattern according to an embodiment includes: forming a hard mask on a substrate; forming a mask reinforcing member on the hard mask; forming a di-block copolymer layer on the mask reinforcing member, the di-block copolymer layer comprising a sea-island structure; forming a pattern comprising a concave-convex structure in the di-block copolymer layer, with island portions of the sea-island structure being convex portions; and transferring the pattern onto the hard mask by performing etching on the mask reinforcing member and the hard mask, with a mask being the pattern formed in the di-block copolymer layer. The mask reinforcing member is comprised of a material having an etching speed that is higher than an etching speed for the hard mask and is lower than an etching speed for sea portions of the sea-island structure of the di-block copolymer layer.

Description

CROSS-REFERENCE TO RELATED APPLICATION

This application is based upon and claims the benefit of priority from prior Japanese Patent Application No. 2010-208122 filed on Sep. 16, 2010 in Japan, the entire contents of which are incorporated herein by reference.

FIELD

Embodiments described herein relate to a method of forming a fine pattern.

BACKGROUND

The advancement of the microfabrication technique used for manufacturing semiconductor devices and the like accounts for a large proportion of the recent significant improvement in the functions of information devices such as personal computers. Conventionally, size reductions in processing operations have been achieved by reducing the wavelengths of exposure light sources used in lithography. However, as the sizes in processing operations have become smaller and higher-density patterns are being used, the lithographical processing costs in the manufacturing procedures are rapidly becoming higher. In the next-generation semiconductor devices or high-density recording media such as patterned media subjected to a microfabrication technique, the pattern sizes are required to be reduced to several tens of nanometers or smaller. Therefore, electron beams are used as exposure light sources, but there remains a serious problem in terms of the throughput of fabrication.
In the above described circumstances, attention is currently focused on techniques utilizing a phenomenon in which a material forms a certain ordered array pattern in a self-organized manner, as the techniques are regarded as inexpensive fabrication techniques that can realize high throughputs. Particularly, great attention is focused on a technique utilizing “block polymers”.
According to a microfabrication technique using a pattern formed by self-organization of a di-block copolymer as an etching mask, side etching is caused in the etching mask due to the difference in etching speed between the mask of the di-block copolymer and the hard mask. As a result, the shape roughness (the variation coefficient) of the pattern transferred onto the hard mask becomes higher than the shape roughness of the etching mask.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1( a) through 1(e) are cross-sectional views for explaining a method of forming a fine pattern according to an embodiment;

FIGS. 2( a) through 2(c) are diagrams for explaining patterns to be formed by a method according to Example 1 and a method according to a comparative example;

FIG. 3 is a graph showing the relationship between the film thickness of a mask reinforcing member and the pattern variation coefficient;

FIG. 4 is a graph showing the relationship between the film thickness of the mask reinforcing member and the concavities and convexities formed after the processing of the hard mask;

FIGS. 5( a) through 5(d) are cross-sectional views for explaining a method of forming a perpendicular magnetic recording medium according to Example 6;

FIGS. 6( a) through 6(c) are cross-sectional views for explaining a method of forming a fine pattern according to Example 6;

FIGS. 7( a) through 7(d) are cross-sectional views for explaining a method of forming a fine pattern according to Example 7;

FIGS. 8( a) through 8(c) are cross-sectional views for explaining a method of forming a fine pattern according to Example 7; and

FIGS. 9( a) through 9(e) are cross-sectional views for explaining a method of forming a fine pattern according to Example 8.

DETAILED DESCRIPTION

A method of forming a fine pattern according to an embodiment includes: forming a hard mask on a; forming a mask reinforcing member on the hard mask; forming a di-block copolymer layer on the mask reinforcing member, the di-block copolymer layer comprising a sea-island structure; forming a pattern comprising a concave-convex structure in the di-block copolymer layer, with island portions of the sea-island structure being convex portions; and transferring the pattern onto the hard mask by performing etching on the mask reinforcing member and the hard mask, with a mask being the pattern formed in the di-block copolymer layer. The mask reinforcing member is comprised of a material having an etching speed that is higher than an etching speed for the hard mask and is lower than an etching speed for sea portions of the sea-island structure of the di-block copolymer layer.

Embodiment

Before the method of forming a fine pattern according to this embodiment is explained, the outline of this embodiment is described. The inventors of the invention found that, by inserting an appropriate mask reinforcing member between a hard mask onto which a pattern was to be transferred and a di-block copolymer layer having an array structure, the variation in shape roughness was made smaller when a pattern was transferred onto the hard mask. By using this technique, the pattern formed in the mask reinforcing member can be transferred onto the hard mask, and the pattern transferred onto the hard mask can not be affected by the polymer subjected to side etching.
Here, the appropriate mask reinforcing member has an etching speed that falls between the etching speed for the hard mask and the etching speed for the later described polymer phases X, and is almost the same as the etching speed for the later described polymer phases Y. By using such a material as the mask reinforcing member, the shape of the pattern formed by a di-block copolymer can be accurately transferred onto the hard mask. An “accurate transfer” herein indicates that the shape variation coefficient is 10% or lower in a pattern transfer from a mask formed by a di-block copolymer onto the hard mask.
The method of forming a fine pattern according to this embodiment is described in detail with reference to the accompanying drawings.
FIGS. 1( a) through 1(e) show the procedures for forming a fine pattern according to this embodiment. First, a substrate 2 is prepared, and a hard mask 4 onto which a pattern is to be transferred is formed on the substrate 2 (FIG. 1( a)). A glass substrate, a sapphire substrate, a silicon substrate, or a substrate having a HDD (hard disk drive) magnetic layer formed thereon can be used as the substrate 2. The material used for the magnetic layer can be an alloy containing Co—Cr or Co—Pt, an alloy containing Fe—Pt, Co—Pt, or Fe—Pd, or a multilayer film material such as a Co/Pt or Co/Pd film. Having high magnetic crystalline anisotropy energy, each of those alloys and multilayer film materials has a high resistance to thermal fluctuations. Therefore, using those alloys and multilayer film materials is preferable. To improve magnetic properties, an additional element such as Ta, Cu, B, or Cr is added to the alloy or multilayer film material, as needed. As for the magnetic layer, it is more preferable to use CoCrPt, CoCrPtB, CoCrPtTa, CoCrPtNd, CoCrPtCu, FePtCu, or the like. The magnetic layer can be a multilayer structure that has two or more layers, as needed. In that case, at least one of the layers should be the above described layer. As for the hard mask, carbon, carbon nitride, silicon, silicon oxide, or the like can be used. Alternatively, a multilayer mask having a structure in which those materials are stacked can be used.
A mask reinforcing member 6 is then formed on the hard mask 4 (FIG. 1( b)). The material of the mask reinforcing member 6 will be described later in detail. A block polymer layer 8 is then formed on the mask reinforcing member 6 (FIG. 1( c)). The block copolymer used here can be a “di-block copolymer” of an A-B type having the two types of polymer chains, a polymer chain A and a polymer chain B, combined with each other, for example. A di-block copolymer is a copolymer having two polymers combined with each other. When annealing is performed at an appropriate temperature, the di-block copolymer is phase-separated into polymer phases A (hereinafter also referred to as the polymer phase X) and polymer phases B (hereinafter also referred to as the polymer phases Y), to form an ordered array structure. For example, a sea-island structure is formed, with the polymer phases X forming the sea, the polymer phases Y being two-dimensionally arranged as islands in the sea of the polymer phases X. The shapes and sizes of the polymer phases X and the polymer phases Y forming the ordered array structure depend on the lengths of the polymer chains A and B. Adjusting those lengths can form minute islands whose diameter is from 100 nm to 10 nm or less.
As a di-block copolymer ordered array structure as described above, there have been a known island structure in which A phases or B phases have spherical shapes and are distributed, a cylinder structure in which A phases or B phases have cylindrical shapes and are distributed, and the like. An arrangement of those phases can be hexagonal or quadrangular, with the spheres or cylinders being closely arranged. The ordered array structure can have a concave-convex structure having concavities and convexities arranged in an orderly manner, but can be a flat structure without any concavities and convexities. In this embodiment, the di-block copolymer phase-separated structure needs to be transformed into a concave-convex structure. If the surface of the block copolymer ordered array structure has concavities and convexities, the concavities and convexities can be used as they are.
If the block copolymer ordered array structure has a flat structure without any concavities and convexities, at least one polymer phase type of block copolymer needs to be selectively removed. In this embodiment, etching is selectively performed on the polymer phases X forming the sea of the sea-island structure, to form a fine pattern in which the spheres of the polymer phases Y are exposed and are orderly arranged.
A fine pattern structure formed by a block copolymer can be a sphere structure in which the polymer phases X form the sea, the polymer phases Y have spherical island structures, and the spheres are orderly arranged, or can be a cylinder structure in which the polymer phases X form the sea, and the polymer phases Y have cylindrical structures. The differences between those structures can be controlled by changing the molecular weights of the polymer chain A and the polymer chain B, surface energy of the substrate, or annealing conditions.
To selectively remove the polymer phases X, a block copolymer should be formed by two or more kinds of polymer chains having different resistances to energy beams such as plasma beams, light beams, or electron beams, or heat when any of those beams or heat is applied. For example, where N represents the total number of atoms per monomer, Nc represents the number of carbon atoms per monomer, and No represents the number of oxygen atoms per monomer, the polymer chain having the smaller value of N/(Nc−No) per monomer has the higher resistance to plasma exposure. In view of this, two or more kinds of polymer chains having different plasma resistances from each other can be combined.
Also, it is possible to combine a polymer that has a cross-linking reaction and becomes hardened when exposed to the plasma beams, light beams, electron beams, heat, or the like, and a polymer that does not react to any of those beams or heat or the like. Further, with affinities being taken into consideration, a hydrophilic polymer and a hydrophobic polymer can be used, and a cross-linking agent can be segregated in one of the polymers.
As described above, if the block copolymer layer 8 originally has a concave-convex structure, the concave-convex structure is used as it is, and if the block copolymer layer 8 is flat, concavities and convexities are formed in the block copolymer layer 8. In this embodiment, RIE (Reactive Ion Etching) using O₂is performed on the block copolymer layer 8, to form a concave-convex structure as a sea-island structure having portions 8 b formed by the polymer phases Y and portions 8 a formed by the polymer phases X, with the island portions 8 b being the convex portions, as shown in FIG. 1(d). Here, part of the portions 8 a formed by the polymer phases X can remain in the concave portions, or the sides of the portions 8 a formed by the polymer phases X located immediately below the island portions 8 b can be partially removed by etching. As the block copolymer layer 8 having the concave-convex structure serves as a mask, etching is also performed on the mask reinforcing member 6.
Etching is then performed on the hard mask 4, using a mask pattern formed by the block copolymer layer 8 having the concave-convex structure and the etched mask reinforcing member 6, as shown in FIG. 1( e). In this manner, the pattern is transferred onto the hard mask 4.
In this embodiment, the mask reinforcing member 6 is made of a material that is etched by the same etching gas as an etching gas used for the di-block copolymer, or is made of a material that is etched by oxygen. Also, the mask reinforcing member 6 is made of a material for which the etching speed is lower than the etching speed for the polymer phases X forming the sea portions of the di-block copolymer layer but is higher than the etching speed for the hard mask 4, or is made of a material for which the etching speed is almost the same as the etching speed for the polymer phases Y.
A specific example of a material used for the mask reinforcing member 6 is an organic polymer chain. With the above described value of N/(Nc−No) being used as a parameter, a relational expression indicating that V_etchis proportional to N/(Nc−No) is established between the etching speed V_etchof the organic polymer chain and the parameter. Therefore, the material for which an etching speed falls between the etching speed for the sea portions of the sea-island structure and the etching speed for the hard mask 4 can be selected as the mask reinforcing member. For example, in a case where a di-block copolymer having PS (polystyrene) and PDMS (polydimethylsiloxane) combined with each other is used as the block copolymer layer 8, and carbon is used as the hard mask 4, PVN (polyvinylnaphthalene), PHS (polyhydrostyrene), PVB (polyvinylbiphenyl), PS, or PDMS should be used, with an etching speed for PS forming the sea portions of the sea-island structure and an etching speed for the carbon being taken into consideration. Particularly, in a case where PDMS is used, the etching speed falls between the etching speed for PS forming the sea portions of the sea-island structure and the etching speed for the carbon of the hard mask 4. Further, since PDMS is the same material as the material of the polymer phases Y of the island portions of the sea-island structure forming the pattern, the pattern can be more accurately transferred.
In this embodiment, the film of the mask reinforcing member 6 can be formed by either a wet process involving spin coating with the use of a liquid solution or a dry process involving vapor deposition, sputtering, or the like.
A film thickness t_mof the mask reinforcing member 6 preferably satisfies 0<t_m<d, where d represents a diameter of each of the island portions forming the di-block copolymer layer 8. If d is equal to or smaller than t_m, etching is performed also on the di-block copolymer layer 8 when etching is performed on the mask reinforcing member 6, and a roughness of the di-block copolymer layer 8 becomes higher. In extreme cases where the roughness of the di-block copolymer layer 8 is greater, the pattern cannot be transferred onto the hard mask 4. If t_mis 0, on the other hand, an effect to reduce a shape-roughness cannot be observed.
Particularly, in the range satisfying 0<t_m<d/2, heights of the concavities and convexities of the hard mask 4 formed by the etching can be made greater, and a high-aspect pattern can be formed on the substrate 2 when processing is performed on the substrate 2 with the use of the hard mask 4. Therefore, it is notable to satisfy 0<t_m<d/2.
The di-block copolymer layer 8 can be formed by a spin coating technique, or can be formed by a dip coating technique by which the substrate 2 is dipped in a liquid solution and is pulled out of the solution at a constant speed.
A film thickness t_dof the di-block copolymer layer 8 can be changed with a pitch p of the pattern to be formed. In a case where a mask with excellent arrangement and dot shapes is to be formed, the film thickness t_dof the di-block copolymer layer 8 satisfies 0<t_d<1.5p. If the thickness t_donly satisfies 0<t_d<p, the dots are not closely arranged, and missing portions appear in the pattern. Also, the dots have various sizes. However, the pattern does not cause any problem with the functions as a mask for microfabrication. If the film thickness t_dis equal to or greater than 1.5p, on the other hand, dot pattern is not a single layer, but has a structure in which two or more layers are stacked. As a result, the dot pattern does not function as a mask. If the film thickness t_dis about 1.3p, dot arrangement becomes two-dimensionally hexagonal. The film thickness of each di-block copolymer layer 8 in the later described examples is controlled to be 1.3p.
In a case where a shape of the pattern varies with film thickness as described above, it is necessary to appropriately adjust the film thickness in accordance with materials. The film thickness can be controlled by changing density of solution of the di-block copolymer, or by adjusting the number of rotations and the rotation time in spin coating. Also, the film thickness can be measured with the use of an AFM (atomic force microscope) or a contact level detector or the like.
The di-block copolymer applied onto the substrate is subjected to annealing, to have an ordered array structure. To prevent polymer oxidation, the annealing atmosphere for the di-block copolymer should be a vacuum or a nitrogen atmosphere. Alternatively, the annealing can be performed in an atmosphere of a forming gas that is a mixed gas of hydrogen and nitrogen. An annealing temperature can be roughly estimated by carrying out differential scanning calorimetry (DSC). By performing heating at 5° C./min. and obtaining a DSC chart, information about the glass-transition (Tg) temperature, phase transitions, phase decomposition, and the like can be obtained. In a case where the order-disorder transition (ODT) temperature at which the di-block copolymer has a phase transition is higher than the temperature of polymer decomposition, the annealing temperature should be raised to a temperature immediately below the temperature at which polymers are decomposed. If the ODT temperature is equal to or lower than the temperature of polymer decomposition, the annealing should be performed at almost the same temperature as the ODT temperature.
Since the Tg temperature, the ODT temperature, and the decomposition temperature vary with the types of di-block copolymers, annealing needs to be appropriately performed at an optimum temperature.
The dry etching used for the pattern formation in the di-block copolymer layer 8 and the pattern transfer onto the hard mask 4 is RIE. Alternatively, the dry etching can be reactive ion beam etching, ion etching, etching using neutrons, or the like.
Examples of RIE include capacitively-coupled RIE, inductively-coupled RIE, and ECR-RIE. However, the same results can be achieved by performing etching with the use of any device.
In this embodiment, the RIE for the pattern formation in the di-block copolymer layer 8 through the RIE for the pattern transfer onto the hard mask 4 are performed in the same gas. Specifically, RIE is performed for 30 seconds, where the oxygen flow rate is 20 sccm, the total pressure is 0.1 Pa, the input coil power is 100 W, and the platen power is 10 W. In this manner, the pattern in the di-block copolymer layer 8 is transferred onto the hard mask 4.
A gas used in the RIE should be oxygen. However, in a case where a small amount of argon, nitrogen, or fluorine is mixed with oxygen, or where a mixed gas that causes a 10% or smaller change in the etching speed for the di-block copolymer compared with an etching speed for oxygen is used, the etching can also be defined as etching with the use of oxygen, since an etching effect by oxygen is notable in either case.
Pattern shape is evaluated with the use of a variation coefficient (a value that is obtained by normalizing the standard deviation of the diameters of the dots in the pattern with the mean value of the diameters of the dots in the pattern, and is expressed in percentage). A small value is notable as the value of the variation coefficient. In a case where a fine pattern is applied to a patterned medium, the fine pattern is required to have the variation coefficient of 10% or lower, according to academic institutes and societies. The variation coefficient is calculated with the use of a scanning electron microscope (SEM). Specifically, the pattern on a processed substrate is observed from above the substrate with use of the SEM, and an obtained SEM image is binarized with the use of image editing software. In this manner, the diameters of the dots in the pattern are calculated to obtain the variation coefficient.
Other than the SEM, the pattern shape can be evaluated with a planar transmission electron microscope (planar TEM), a cross-sectional transmission electron microscope (cross-sectional TEM), or the like. The same results as above can be achieved by measuring the shapes of the dots in the pattern by any of the above techniques, calculating the diameters of the dots, and obtaining a variation coefficient.

EXAMPLES

Hereinafter, examples are described.

Example 1

Referring now to FIGS. 1( a) through 1(e), a method of forming a fine pattern according to Example 1 is described.
First, as shown in FIG. 1( a), a silicon substrate 2 was introduced into a vacuum chamber of a sputtering device. The attainable degree of vacuum of the sputtering device was 1×10⁻⁵Pa. A carbon film of 15 nm in film thickness was formed as a hard mask 4 on the silicon substrate 2 (FIG. 1( a)). The Ar pressure at the time of carbon film formation was 0.4 Pa, and the input power was 400 W.
The hard mask 4 comprised of the carbon film was formed on the silicon substrate 2, and a polydimethylsiloexane (PDMS) mask reinforcing member 6 of 5 nm in thickness was formed on the hard mask 4 by the use of spin-coating with a 0.05 wt % solution of (PDMS) using toluene as a solvent (FIG. 1( b)). A solution obtained by dissolving a di-block copolymer consisting of polystyrene (PS) having a molecular weight of 11,700 and PDMS having a molecular weight of 2,900 in a propylene glycol monomethyl ether acetate (PGMEA) solution was applied by a spin coating technique, to form a di-block copolymer layer 8 of 22 nm in film thickness (FIG. 1( c)). Annealing was then performed in a vacuum of 10 Pa at 180° C. for 15 hours, to cause phase separations and form an ordered array structure.
Convex portions were then formed as shown in FIG. 1( d). Specifically, reactive ion etching (RIE) was performed for 30 seconds, where the oxygen flow rate was 20 sccm, the total pressure was 0.1 Pa, the input coil power was 100 W, and the platen power was 10 W. Under such conditions, etching was performed on a PS polymer layer 8 a, with the polymer of a PDMS layer 8 b serving as a mask. Under the same conditions, etching was performed on the PDMS layer 6 as the mask reinforcing member and the carbon hard mask 4, with the polymer 8 b and the polymer 8 a serving as masks (FIG. 1( e)).
At this point, the upper face of the block copolymer layer 8 was observed with an atomic force microscope (AFM), to confirm that convex portions of approximately 10 nm in diameter, approximately 18 nm in depth, approximately 17 nm in pattern pitch were arranged in a hexagonal-lattice fashion.
The fine pattern formed under the above described conditions was measured with a SEM, and the mean pattern size was estimated from an obtained SEM image. As a result, the diameter was approximately 13 nm, and the pattern pitch was approximately 17 nm. The pattern variation (the dot diameter variation) was measured from the SEM image, to determine that the standard deviation was 1.1 nm. As a result, the variation coefficient of the pattern shape of the carbon hard mask 4 of formed under the above described conditions was 8.5%.

COMPARATIVE EXAMPLE

As a comparative example of the above described Example 1, a pattern was transferred and was measured through the same procedures as those of Example 1, except that the mask reinforcing member 6 was not formed. The substrate 2 after the transfer was measured with a SEM, and the mean pattern size was estimated, to determine that the diameter was approximately 13 nm, and the standard deviation was 3.4 nm. The results indicate that, where a pattern transfer was performed without the mask reinforcing member 6, the variation coefficient of the pattern was 24%.
FIGS. 2( b) and 2(c) are schematic views of the upper faces of patterns transferred onto the hard mask 4 with the use of the di-block copolymer layer 8 having the ordered array structure shown in FIG. 2( a) by the methods according to the comparative example and Example 1, respectively.
As can be seen from FIG. 2( b), in the comparative example, the roughness of the edges of the dots is higher. In the case where the sea portions 8 a of the di-block copolymer layer 8 having a high etching speed are stacked directly on the hard mask 4, side etching is caused in the di-block copolymer having a high etching speed at the interface between the hard mask 4 and the di-block copolymer layer 8, and the pattern subjected to the side etching is transferred onto the hard mask 4. As a result, the roughness of the edges of the dots becomes higher. In extreme cases, side etching is caused in the entire sea portions of the di-block copolymer layer 8, and the island portions cannot be supported, resulting in disappearance of the pattern.
By the method according to Example 1, on the other hand, the pattern shape of the di-block copolymer layer 8 is accurately transferred onto the hard mask 4, as can be seen from FIG. 2( c).

Example 2

Example 2 concerns the relationship between the film thickness of the mask reinforcing member 6 and the variation coefficient of a transferred pattern, and the relationship between the film thickness of the mask reinforcing member 6 and the concavities and convexities of the hard mask.
A PDMS film was formed as the mask reinforcing member 6 on the 15-nm thick carbon hard mask 4 formed on each nonmagnetic glass substrate 2. Six samples were manufactured, with the thicknesses of the PDMS films being 0 nm, 2 nm, 5 nm, 10 nm, 15 nm, and 20 nm. It should be noted that the thickness of 0 nm means that the mask reinforcing member 6 is not formed. Under the same conditions as those in Example 1, a di-block copolymer layer was formed on each of the samples, and etching was performed. The RIE time was varied with the film thicknesses of the mask reinforcing members 6. The size variation (the variation coefficient) of the mask of each of the samples was determined from the results of size measurement carried out with a SEM.
The patterns of the respective samples that were fabricated under the above described conditions and had the mask reinforcing members 6 with different film thicknesses were measured with the SEM to determine the variation coefficients. The variation coefficient of the sample having the mask reinforcing member 6 of 0 nm in film thickness was 24%, the variation coefficient of the sample having the mask reinforcing member 6 of 2 nm in film thickness was 8%, the variation coefficient of the sample having the mask reinforcing member 6 of 5 nm in film thickness was 8%, and the variation coefficient of the sample having the mask reinforcing member 6 of 10 nm in film thickness was 9%. Meanwhile, the variation coefficient of the sample having the mask reinforcing member 6 of 15 nm in film thickness was 25%, and formation of a pattern was not seen in the sample having the mask reinforcing member 6 of 20 nm in film thickness. The results of the measurement are shown in FIG. 3.
The diameter of each dot 8 b of the di-block copolymer (PS-PDMS) used in this example is approximately 13 nm. Therefore, as can be seen from FIG. 3, in the samples in which the film thicknesses of the PDMS films are 2 nm, 5 nm, and 10 nm, which are equal to or smaller than the diameter, the values of the variation coefficients are smaller than the value of the variation coefficient of the sample not having the mask reinforcing member 6. On the other hand, in the sample having the mask reinforcing member 6 of 20 nm in film thickness, the PDMS layer 6 existing directly on the hard mask 4 has a greater film thickness than the PDMS 8 b forming the pattern. Therefore, when etching is performed on the PDMS layer serving as the mask reinforcing member 6, the PDMS 8 b forming the pattern in the di-block copolymer layer 8 disappears. As a result, the pattern in the di-block copolymer layer 8 cannot be transferred onto the hard mask 4.
FIG. 4 shows the relationship between the film thickness of the mask reinforcing member 6 and the concavities and convexities of the hard mask 4 on which a pattern is to be transferred. In the case where etching is performed on the substrate 2, with the hard mask 4 serving as a mask, the hard mask 4 should preferably have a great film thickness. As can be seen from FIG. 4, once the film thickness of the mask reinforcing member 6 exceeds the radius of each dot 8 b of the di-block copolymer layer 8, the film thickness of the processed portion of the hard mask 4 becomes rapidly smaller. The result indicates that the film thickness t of the mask reinforcing member 6 formed on the hard mask 4 should preferably satisfy 0<t<d/2, where d represents the diameter of each dot 8 b of the mask.

Example 3

Referring back to FIGS. 1( a) through 1(e), a method of forming a fine pattern according to Example 3 is described.
Samples were manufactured by forming a 5-nm thick mask reinforcing member 6 by spin coating performed on the 15-nm thick carbon hard mask 4 formed on each nonmagnetic glass substrate 2 (FIGS. 1( a) and 1(b)). In the respective samples, the mask reinforcing members 6 were a polyacrylic nitrile (PVN) film, a polyhydrostyrene (PHS) film, a polyvinylbiphenyl (PVB) film, a PS film, and a PDMS film. A di-block copolymer layer 8 was then formed on each of the samples under the same conditions as those described in Example 1, and RIE was performed on the mask reinforcing members 6 and the hard masks 4 (FIGS. 1( c) through 1(e)). The RIE time was varied with the etching speeds for the mask reinforcing members 6. Where the etching speed for the PS was 1, the etching speeds of oxygen for the layers used as the mask reinforcing members 6 were 0.8 in the case of PVN, 1.0 in the case of PHS, 0.9 in the case of PVB, 1 in the case of PS, and 0.5 in the case of PDMS. At this point, the etching speed for the carbon hard mask 4 was 0.3.
The patterns of the above samples were measured with a SEM to measure the variation coefficients. As a result, the variation coefficients were 9% in the case of PVN, 13% in the case of PHS, 12% in the case of PVB, 13% in the case of PS, and 8% in the case of PDMS.
When a mask reinforcing member 6 was formed by mixing the above polymers, which were PVN, PHS, PVB, PS, and PDMS, a variation coefficient of approximately 10% was obtained as in the cases where single layers of the respective polymers were used.

Example 4

Referring again to FIGS. 1( a) through 1(e), a method of forming a fine pattern according to Example 4 is described.
Samples were manufactured by forming a 5-nm thick mask reinforcing member 6 by spin coating performed on the 15-nm thick carbon hard mask 4 formed on each nonmagnetic glass substrate 2. In the respective samples, the mask reinforcing members 6 were a PS film, a PMMA (polymethylmethacrylate) film, and a PDMS film. Using PS-PMMA, a di-block copolymer layer 8 having a film thickness of 30 nm was then formed on each of the mask reinforcing members 6 of the samples. Etching was then performed on the mask reinforcing members 6 and the hard masks 4, and the patterns were transferred onto the hard masks 4.
Each PS-PMMA film was formed so that the proportion of the PMMA is approximately 20%. After the film formation, a sea-island structure having islands formed by the PMMA can be obtained by annealing. In each of the prepared samples, the pattern of the hard mask 4 after the pattern transfer was measured with a SEM to measure the variation coefficient. The results indicated that the variation coefficient was 15% in the case where a PS film was used as the mask reinforcing member 6, 11% in the case of PMMA, and 18% in the case of PDMS.

Example 5

Referring again to FIGS. 1( a) through 1(e), a method of forming a fine pattern according to Example 5 is described.
A 5-nm thick PDMS film as the mask reinforcing member 6 was formed by spin coating performed on the 15-nm thick carbon hard mask 4 formed on a nonmagnetic glass substrate 2. A di-block copolymer (PS-PDMS) layer 8 was then formed under the same conditions as those in Example 1, and etching was performed on the mask reinforcing member 6 and the hard mask 4. The pattern was then transferred onto the hard mask 4. The pattern of the hard mask 4 after the transfer was measured with a SEM to measure a variation coefficient. The measurement result showed that the variation coefficient was 8%.

Example 6

Referring now to FIGS. 5( a) through 6(c), a method of forming a fine pattern according to Example 6 is described. The method of forming a fine pattern according to Example 6 is used for forming a patterned medium of a perpendicular magnetic recording type.
First, as shown in FIG. 5( a), a perpendicular magnetic recording layer 3 was formed on a nonmagnetic glass substrate 2 with the use of a sputtering device. In the perpendicular magnetic recording layer 3, a soft magnetic layer made of CoZrNb having a film thickness of 120 nm, an orientation control base layer having a film thickness of 20 nm, and a ferromagnetic layer made of CoPt having a film thickness of 15 nm were stacked in this order. A hard mask 4 made of C (carbon) having a film thickness of 15 nm was then formed on the perpendicular magnetic recording layer 3 with the use of a sputtering device.
A mask reinforcing member 6 and a di-block copolymer layer 8 made of PS-PDMS were then formed under the same conditions as those in Example 1 (FIGS. 5( b) and 5(c)). After that, RIE using O₂was performed to form concavities and convexities 8 a and 8 b in the di-block copolymer layer 8 (FIG. 5( d)).
While the di-block copolymer layer 8 having the concavities and convexities 8 a and 8 b was used as a mask, RIE using O₂was performed on the hard mask 4, and the pattern was transferred onto the hard mask 4, as shown in FIG. 6( a). At this point, the upper face of the perpendicular magnetic recording layer 3 was exposed through the bottoms of the concave portions of the transferred pattern.
While the pattern transferred onto the hard mask 4 is used as a mask, patterning was performed on the perpendicular magnetic recording layer 3 exposed through the bottoms of the concave portions as shown in FIG. 6( b), with the use of an electron cyclotron resonance (ECR) ion gun and an Ar gas, where the gas pressure was 0.04 Pa, the microwave power was 600 W, the applied voltage was 600 V, and the processing time was 20 seconds.
After that, ashing was performed for 20 seconds, where the oxygen flow rate was 20 sccm, the total pressure was 0.1 Pa, the input coil RF power was 200 W, and the platen RF power was 0 W. In this manner, the hard mask 4 remaining on the perpendicular magnetic recording layer 3 was removed, and a perpendicular magnetic recording medium was formed (FIG. 6( c)).
The pattern in the CoPt ferromagnetic layer subjected to the patterning was measured with a SEM, and the variation coefficient was measured. The measurement result showed that the variation coefficient was 12%. As is apparent from Example 1, the variation coefficient of the pattern in the hard mask 4 made of carbon is 8%. Through the etching later performed, however, the variation coefficient became higher than 8%.

COMPARATIVE EXAMPLE

As a comparative example of Example 6, a perpendicular magnetic recording medium was formed through the same procedures as those of Example 1, except that the mask reinforcing member 6 was not formed. The pattern in the CoPt ferromagnetic layer subjected to the patterning was measured with a SEM, and the variation coefficient was measured. The measurement result showed that the variation coefficient was 33%.
By virtue of formation of a mark reinforcing member, a pattern can be transferred onto a magnetic recording layer without an increase in shape roughness.

Example 7

Referring now to FIGS. 7( a) through 8(c), a method of forming a fine pattern according to Example 7 is described.
First, a hard mask layer 4 made of C having a film thickness of 15 nm was formed on a 6-inch Si substrate 2 (FIG. 7( a)).
Under the same conditions as those in Example 1, a mask reinforcing member 6 and a di-block layer 8 made of PS-PDMS were successively formed (FIGS. 7( b) and 7(c)). A pattern having concavities and convexities 8 a and 8 b was then formed in the di-block layer 8 by RIE using O₂.
With the concavities and convexities 8 a and 8 b serving as a mask, etching was performed on the mask reinforcing member 6 and the hard mask 4, and the pattern was transferred onto the hard mask 4 (FIG. 8( a)). At this point, the upper face of the Si substrate 2 was exposed through the bottoms of the concave portions of the transferred pattern.
While the hard mask 4 having the transferred pattern was used as a mask, 45-second RIE was performed on the Si substrate 2 exposed through the bottoms of the concave portions, where the CF₄gas flow rate was 20 sccm, the total pressure was 0.1 Pa, the input coil RF power was 100 W, and the input platen RF power was 10 W. In this manner, the pattern of the ordered array structure 8 a and 8 b of the di-block copolymer layer 8 was transferred onto the Si substrate 2 (FIG. 8( b)).
After that, ashing was performed for 20 seconds, where the oxygen flow rate was 20 sccm, the total pressure was 0.1 Pa, the input coil RF power was 200 W, and the platen RF power was 0 W. Through the ashing, the hard mask 4 remaining on the Si substrate 2 was removed, and the Si substrate 2 having the pattern transferred thereonto was obtained (FIG. 8( c)). The pattern transferred onto the Si substrate 2 was measured with a SEM, and the variation coefficient was determined to be 13%.

Example 8

Referring now to FIGS. 9( a) through 9(e), a method of forming a fine pattern according to Example 8 is described.
A 30-nm thick first hard mask 4 a made of C, a 3-nm thick second hard mask 4 b made of Si, and a 5-nm thick third hard mask 4 c made of C were successively formed on a nonmagnetic glass substrate 2 with the use of a sputtering device, to form a hard mask 4 having a structure in which the first hard mask 4 a, the second hard mask 4 b, and the third hard mask 4 c were stacked (FIG. 9( a)).
Under the same conditions as those in Example 1, a mask reinforcing member 6 and a di-block copolymer layer 8 made of PS-PDMS were formed (FIG. 9( b)). After that, etching with oxygen was performed on the mask reinforcing member 6 and the carbon of the third hard mask 4 c at the same time as the etching of the PS portions 8 a of the di-block copolymer layer 8. In this manner, concavities and convexities were formed in the carbon hard mask 4 c (FIG. 9( c)).
With the third hard mask 4 c serving as a mask, etching was performed on the Si of the second hard mask 4 b (FIG. 9( d)). Specifically, where the CF₄gas flow rate was 20 sccm, the total pressure was 0.1 Pa, the input coil RF power was 50 W, and the platen RF power was 10 W, RIE was performed on the Si of the second hard mask 4 b for 20 seconds.
After that, with the Si of the second hard mask 4 b serving as a mask, etching was performed on the carbon of the first hard mask 4 a. Specifically, where the O₂gas flow rate was 20 sccm, the total pressure was 0.1 Pa, the input coil RF power was 200 W, and the platen RF power was 10 W, RIE was performed on the C of the first hard mask 4 a for 40 seconds, so that the pattern is transferred onto the first hard mask 4 a. After that, the first hard mask 4 a having the pattern transferred thereonto serves as a mask, and etching was performed on the substrate 2, so that the pattern was transferred onto the substrate 2 (not shown).
As the hard mask 4 has a three-layer structure, the film thickness of the first hard mask 4 a existing on the substrate 2 can be made greater. Accordingly, the mask durability can be made higher when the substrate 2 is processed with the use of the mask.
On the other hand, as the hard mask 4 has the three-layer structure, the hard mask 4 is affected by side etching when the pattern is transferred from the carbon of the third hard mask 4 c onto the silicon of the second hard mask 4 b. As a result, the roughness in shape becomes higher. Therefore, when the pattern transferred onto the substrate 2 processed in the above described manner was measured with a SEM, a slightly high variation coefficient of 15% was obtained.
As described so far, according to this embodiment and each Example, an ordered array pattern formed with a di-block copolymer can be transferred without an increase in pattern shape roughness. Accordingly, this embodiment and each Example can be applied to practical processing methods for various products such as high-density recording media and highly integrated electronic components, and a remarkable industrial advantage can be gained.
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 and systems described herein can be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein can 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.

Claims

What is claimed is:

1. A method of forming a fine pattern, the method comprising:

forming a hard mask on a substrate;

forming a mask reinforcing member on the hard mask;

forming a di-block copolymer layer on the mask reinforcing member, the di-block copolymer layer comprising a sea-island structure;

forming a pattern comprising a concave-convex structure in the di-block copolymer layer, with island portions of the sea-island structure being convex portions; and

transferring the pattern onto the hard mask by performing etching on the mask reinforcing member and the hard mask, with a mask being the pattern formed in the di-block copolymer layer, wherein

the mask reinforcing member is comprised of a material having an etching speed that is higher than an etching speed for the hard mask and is lower than an etching speed for sea portions of the sea-island structure of the di-block copolymer layer.

2. The method according to claim 1, wherein the condition, 0<t<d, is satisfied, where d (nm) represents a diameter of each of the island portions forming the sea-island structure of the di-block copolymer layer, and t (nm) represents a film thickness of the mask reinforcing member.

3. The method according to claim 1, wherein the mask reinforcing member is comprised of one of or a combination of polyvinylnaphthalene (PVN), polyhydrostyrene (PHS), polyvinylbiphenyl (PVB), polystyrene (PS), and polydimethylsilohexane (PDMS).

4. The method according to claim 1, wherein a material of the island portions forming the sea-island structure of the di-block copolymer layer is the same as the material of the mask reinforcing member.

5. The method according to claim 1, wherein the di-block copolymer layer is comprised of a copolymer of polystyrene and polydimethylcyclohexane, and the mask reinforcing member is comprised of polydimethylsilohexane.

6. The method according to claim 1, further comprising

transferring the pattern onto the substrate by performing etching on the substrate with the use of the hard mask onto which the pattern is transferred.

7. The method according to claim 1, wherein the hard mask comprises a structure in which a first hard mask containing carbon, a second hard mask containing silicon, and a third hard mask containing carbon are stacked in this order.

8. The method according to claim 1, wherein

a ferromagnetic layer is formed on the substrate, and

the method further comprises transferring the pattern onto the ferromagnetic layer by performing etching on the ferromagnetic layer with the use of the hard mask onto which the pattern is transferred.

9. The method according to claim 1, wherein the etching on the mask reinforcing member and the hard mask is performed with the use of oxygen.