TWI451189B - Half-tone phase shift blank mask, half-tone phase shift mask and manufacturing method thereof - Google Patents

Description

Half-tone phase shift blank mask, half-tone phase shift mask and manufacturing method thereof

The present invention discloses a half-phase shifting blank mask and a half-phase shifting mask capable of achieving a high-precision critical dimension during a semiconductor photoetching lithography imaging process, and more particularly, half A phase shifting blank mask and a half phase shifting mask can be applied to a 248 nm cesium fluoride etch, a 193 nm argon fluoride etch, and a liquid immersion lithography.

As semiconductor integrated circuits are highly integrated, photolithographic lithography becomes a more important part of a semiconductor manufacturer's core program. In order to improve the resolution of a semiconductor circuit circuit pattern, photolithographic lithography has shortened the exposure wavelength such as a 426 nm g-line, 365 nm i-line, 248 nm cesium fluoride. And 193 nm of argon fluoride was completed. The shortening of the exposure wavelength greatly contributes to the improvement of the resolution. However, this shortening procedure has a negative impact on the Depth of Focus (DoF) so as to increase the burden on designing the lens as well as the optical system.

Thus, in order to address the above limitations, a phase shift mask has been developed to improve the resolution and depth of focus. The phase shift mask transfers a phase of exposure transmitted through a transmissive unit by using a phase shifting layer (e.g., a half transmissive unit).

The phase shift mask includes a Levinson phase shift mask and a half phase shift shifter. In the example of a half-phase shifting mask, a phase shifting layer comprising a single layer of molybdenum molybdenum is used to simplify the manufacturing process and to simplify critical dimension control. The molybdenum molybdenum monolayer may include tantalum nitride molybdenum, niobium molybdenum nitride, molybdenum molybdenum oxide, hafnium oxynitride, and niobium oxynitride.

Typically, the phase shifting layer is formed on a transparent substrate comprising quartz having a principal component via a DC reactive sputtering process. However, due to the mismatch of the composition, a typical molybdenum rhenium nitrogen phase shift layer has a target transmittance that elapses over time. Because of the mismatch between the molybdenum, niobium and nitrogen components of the molybdenum-niobium phase shifting layer formed by the sputtering process, the free energy of a thin layer becomes high and unstable. Thus, in order to maintain thermodynamic stability, the thin layer is attempted to shift to a state in order to reduce free energy. Because of this characteristic, the transmittance may be changed. Thus, the density, microcrystalline state, residual stress, and the like of the thin layer may be changed.

Further, during a sputtering process for forming the molybdenum-niobium phase shifting layer, a heat treatment procedure may be performed during the sputtering process in order to improve thin layer characteristics such as transmittance, phase shift, reliability, exposure resistance, and the like. Or after the sputtering process is implemented. However, since the typical molybdenum niobium thin layer has an unstable composition, the thin layer may be partially crystallized. Thus, part of the surface of a pattern may not be flat. Therefore, it is difficult to manufacture a pattern having excellent characteristics.

In addition, the sulfur fluoride or carbon fluoride gas contains fluorine mixed with oxygen, which is used during a dry etching process of the molybdenum rhenium nitrogen phase shift layer, however, in the case of a typical molybdenum rhenium nitrogen phase shift layer, molybdenum The components of strontium and nitrogen are in an unstable state. Thus, a pattern having a vertical portion may be accompanied by some difficulty because the angle of a sidewall pattern is low after a dry etching process. Thus, it is difficult to manufacture a photomask having an excellent minimum line width.

Also, in order to achieve the 45 nm high resolution critical dimension, liquid immersion etching is proposed, and thus birefringent polarization control by a phase shift layer becomes an important issue. However, in the case of a typical molybdenum-niobium-nitrogen phase shifting layer, since its polarization is not considered at all, when a pattern processing procedure is performed via the liquid immersion etching, a transverse magnetic wave (TM wave) is formed on a pattern. The adverse effects will increase and the good effect of a transverse wave (TE wave) on the improvement of the resolution of the pattern will be reduced. Thus, it is difficult to form a pattern having a high resolution.

In addition, since the typical molybdenum and rhenium nitrogen phase shifting layer has an unstable composition ratio of molybdenum, niobium, and nitrogen, the typical molybdenum rhenium nitrogen phase shifting layer can easily adsorb gaseous molecular pollution, which occurs in environmental factors such as Clean room, air, a box and film, etc. Thus, due to the energy of the continuous exposure (e.g., continuous exposure of a 193 nm argon fluoride exposure source), sufficient activation energy is added to the gaseous molecular contamination, which is contained on or within the phase shifting layer. Therefore, chemical reactions such as growth defects that may occur in gaseous molecules on the surface may occur. The growth defect is continuously performed so that the phase shift layer necessarily has a change in thickness or transmittance, a phase shift, and a defect. As a result, the throughput of the semiconductor device manufacturing process is reduced.

Moreover, due to the mismatch of the proportions of molybdenum, niobium and nitrogen components, a typical molybdenum rhenium nitrogen phase shifting layer destroys the phase shifting layer by using a chemical material such as a strong base and a strong acid during cleaning of an etch. Therefore, the transmittance, thickness, and phase shift amount are changed. Thus, this does not achieve the target phase shift amount and the transmittance, so that the typical molybdenum rhenium nitrogen phase shift layer becomes useless.

In addition, a typical halftone phase shifting blank reticle has a structure in which a photoresist is stacked on a phase shifting layer, i.e., a light shielding layer, and an anti-reflective layer is on a transparent substrate. Due to the thickness of the photoresist (i.e., 3000 angstroms to 2000 angstroms), the loading effect and an aspect ratio are increased during a dry etch process. Thus, a drop in the photoresist and overlap may occur. Therefore, it is difficult to form an accurate pattern.

Again, the phase shifting layer is typically tested via a transmission or reflection method. In order to have an excellent test control, when a test wavelength is used, the phase shift layer requires a transmittance of less than 70 percent or a reflectance of less than 40 percent. However, in the case of a typical molybdenum-niobium phase shift layer, during the 65 nm test performed in an argon fluoride etch, because a pattern is relatively small, there is a size between a defect size and an original pattern. Poor, and there is a size difference between the online pattern and the space pattern. That is, it is difficult to perform a perfect test using only a typical test method.

The present invention provides a phase shifting layer having no crystallized amorphous properties, which controls reflectance or transmittance via an optimized composition ratio, and achieves excellent sidewall angle during dry etching with minimal transmittance. feature. The phase shifting layer was used as a half-shifted phase shift mask and a half-phase shifting mask under 248 nm of barium fluoride, 193 nm of argon fluoride, and immersion etching. Therefore, a high-quality half-phase shift mask is fabricated by polarization control during immersion etching, and a high-resolution semiconductor circuit pattern has excellent qualities such as less than 65 nm, 45 nm, and 32 nm. Can be implemented.

The present invention also provides a half phase shifting blank mask and a half phase shifting mask that provides a suitable composition ratio as a reduction in growth defects and control of a gaseous molecular contamination by a surface treatment to reduce growth defects.

The present invention also provides a single layer phase shift layer structure, or a multilayer phase shift layer structure of more than two layers. The present invention also provides a hard mask stack structure to reduce load effects. Thus, a high-resolution semiconductor circuit pattern with excellent quality can be realized.

According to a specific embodiment of the present invention, a method of fabricating a halftone phase shift blank mask includes:

A1) forming a phase shift layer on a transparent substrate;

B1) forming a metal layer on the phase shift layer of step a1);

C1) Forming a photoresist layer on the metal layer of step b1) for fabricating a half phase shifting blank mask.

A method of fabricating a half-phase shifting mask by using a halftone phase shifting blank mask of steps a1) to c1), comprising:

A2) exposing the photoresist layer in a region formed in a pattern;

B2) removing the photoresist layer exposed in step a2) and developing a photoresist pattern by using a developing solution;

C2) etching the metal layer to form a metal layer pattern by using the photoresist pattern of step b2) as a mask;

D2) etching the phase shift layer to form a phase shift layer pattern by using the photoresist pattern of step c2) and the metal layer pattern as a mask;

E2) removing the photoresist pattern after forming the phase shift layer pattern in step d2);

F2) forming a photoresist layer after removing the photoresist pattern in step e2);

G2) exposing a range after forming the photoresist layer in step f2), the range forming a pattern;

H2) developing the photoresist layer after exposing the photoresist layer in step g2) to form a photoresist pattern;

I2) etching the metal layer by using the photoresist pattern in step h2) to form a metal layer pattern;

J2) After the metal layer pattern is formed in step i2), a half-phase shifting mask is fabricated by removing the photoresist layer.

In step a1), the phase shift layer mask comprises molybdenum, niobium and nitrogen, and may optionally further comprise carbon, oxygen or fluorine.

In step a1), the phase shift layer mask comprises molybdenum, niobium and nitrogen and may be assembled to have a single layer or a plurality of layers.

In step a1), a phase shift layer reticle as a 248 nm cesium fluoride etch may contain 7 to 15 percent molybdenum, 55 to 63 percent bismuth and 100 Å. 28 to 35 percent nitrogen, a phase shift reticle as a 193 nm argon fluoride etch may contain 5 to 10 percent molybdenum, 55 percent to 5 percent 65 矽 and 28 to 35 percent nitrogen. If the molybdenum, niobium and nitrogen components do not have the above upper or lower limit, the free energy of a film becomes higher so that the film is formed by a sputtering process. After the layer disappears, the transmittance deviates from a target value with time. In particular, if a component of the crucible has no upper or lower limit as described above, when a phase shifting layer is formed by a sputtering process, the composition of the tantalum nitride becomes relatively high or low, so that the film may be partially crystallized. . If the composition of a tantalum nitride becomes lower, the composition of a crystallized molybdenum nitride relatively increases. Due to an increase in the crystallized molybdenum nitride and tantalum nitride components, a part of the crystallization phenomenon occurs, and accordingly, the pattern characteristics deteriorate. Thus, the quality of the minimum line width may be deteriorated. Thus, in order to obtain an amorphous thin layer, the thin layer crystallization does not occur, and the molybdenum and niobium components need to be maintained. Because of this amorphous thin layer, a half-tone phase shift mask having a high quality line width characteristic can be fabricated. If the nitrogen component does not have the above upper or lower limit, an appropriate chemical bond may not be achieved between the radical and the molybdenum ruthenium nitride phase shift layer during dry etching. Thus, the material is less volatile and reprecipitation occurs, so that the sidewall angle or appearance of a pattern becomes poor.

In step a1), a volume percentage of argon of 30 to 40 and a volume percentage of nitrogen of 60 to 70 are contained under a gas pressure, and a sputtering process for forming a phase shift layer in the cesium fluoride etching can be formed, and the power density per square. The centimeters are 1 to 2 watts, and the pressure is 1 x 10 ^-2 Torr to 1 × 10 ^-4 Torr. The gas pressure may further comprise carbon or oxygen. A argon volume percentage of 15 to 25 percent by volume and a nitrogen volume percentage of 75 to 85 percent by volume under a gas pressure may be formed as a phase shifting layer forming a phase shifting layer during argon fluoride etching, with a power density per square centimeter 1 To 2 watts, and pressure 1 × 10 ^-2 Torr to 1 × 10 ^-4 Torr. The gas pressure may further comprise carbon or oxygen. A phase shift layer having excellent amorphous characteristics can be fabricated. In particular, if the nitrogen exceeds the lower limit, sufficient nitrogen is not supplied so that a component of the molybdenum molybdenum of the phase shift layer becomes excessive. Thus, partial crystallization may occur. If the nitrogen exceeds the upper limit, a component of the molybdenum molybdenum of the phase shifting layer is reduced, so that the pores are increased. Furthermore, during dry etching, a phase shifting layer is easily removed so that it becomes difficult to control the tilt (angle) of a pattern sidewall.

In step a1), the phase shifting layer may further comprise a small amount of oxygen of less than 5 percent. In general, if the phase shift layer includes oxygen, the chemical resistance to sulfuric acid and ammonium becomes deteriorated, but if the oxygen component is less than 5 percent, the transmittance can be easily controlled and the chemical resistance does not deteriorate. In order to contain oxygen in the phase shifting layer, the gas containing oxygen is used as a reactive gas not only during a sputtering process, but also a heat treatment process is performed under a gas pressure of oxygen contained in a phase shift layer. Executed after formation. If oxygen is contained by the heat treatment process, the oxygen content of the surface of the phase shifting layer becomes higher and the oxygen component becomes lower as it approaches the substrate.

In step a1), a sputtering target that forms a phase shift layer of cesium fluoride etching comprises 15 to 25 percent of molybdenum and 75 to 85 percent of bismuth, and has 2 To a thickness of 7 mm. A sputtering target that forms a phase shifting layer for argon fluoride etching includes 5 to 15 percent molybdenum and 85 to 95 percent germanium, and has a thickness of 2 to 7 millimeters. If the thickness of the sputtering target exceeds the lower limit, since the target is eroded during a sputtering process in order to manufacture a phase shifting layer, the sputtering target is quickly consumed. As a result, it becomes difficult to mass produce. In general, the phase shifting layer sputtering process uses a magnetron sputtering method in which plasma density and uniformity are controlled by forming a magnetic field by rotating an electrode behind the target. Produced by the magnet above. In this example, if the thickness of the target exceeds the upper limit, the magnetic control effect transmitted through a magnet is deteriorated, so that a phase shift layer having an irregular thickness, a composition ratio, and a transmittance are formed. Thus, it is difficult to manufacture a high quality phase shift layer. In addition, the target must have a thickness ranging from 2 mm to 7 mm for a sputtering target, used as a high quality phase shift layer and mass production.

In step a1), the thickness of the phase shifting layer as a cesium fluoride etch may be 850 angstroms to 900 angstroms for a 180 degree phase shift. For a 180 degree phase shift, the thickness of the phase shifting layer as argon fluoride etch may range from 630 angstroms to 680 angstroms. If the thickness exceeds the upper or lower limit, it is impossible to achieve a phase shift of 175 to 180 degrees. Thus, the phase shift layer does not work.

In the step a1), the transmittance of the phase shift layer as the cesium fluoride etch is 4.5 to 9 percent, and the transmittance of the phase shift layer as the argon fluoride etch is 4.5 percent. To 7 percent.

In step a1), the phase shifting layer has a transmittance of less than 65 percent and a reflectivity of less than forty percent at a wavelength of from about 200 nanometers to 700 nanometers. As the size of the pattern becomes miniaturized, a perfect test may not be achieved by a typical transmittance test method or a typical reflectance test method. In addition, an improved transmission, reflectivity, or transmittance and reflectance hybrid test mode is more desirable. During the testing of the halftone phase shift layer, in order to improve the sensitivity test, a transmission law with less than 65 percent and a reflection of less than 40 percent are required at test wavelengths from 200 nm to 700 nm. Rate phase shift layer.

In step a1), in order to control the growth defects of the phase shift layer, surface treatment via a hot plate, a rapid thermal processing device or plasma may be performed. When the exposure wavelength becomes smaller at 248 nm and 193 nm, a high exposure energy may provide an activity energy to gaseous molecular contamination, which may remain or be adsorbed to the surface of the halftone phase shift mask. In addition, growth defects may occur due to continuous exposure illumination. This surface treatment process is required to control the contamination of the gaseous molecules. Generally, gaseous molecular pollution includes acid gaseous molecular pollution, basic molecular pollution, doping molecular pollution, and volatile molecular pollution. These pollutions may be absorbed into the thin layer through environmental factors such as forming a thin layer, a process pressure, a process, a containment box, a film, and a person. . In order to prevent the gaseous molecular contamination from being adsorbed to the phase shifting layer, the surface treatment process may be performed. A method of removing organic matter on the surface such as carbon, hydrogen and oxygen, such as plasma treatment, so that gaseous molecular contamination at the surface can be removed. The adsorption of the gaseous molecular contamination can be prevented by dissolving the gaseous molecular contamination (which remains or adsorbed on the surface) and stabilizing the thin layer via a hot plate and a surface heat treatment method of a rapid heat treatment device.

In step a1), a phase shifting layer as a liquid immersion etch has a birefringence of 2 nm per phase shift thickness. If the birefringence of the phase shifting layer is greater than 2 nm, the polarization phenomenon of the birefringence is amplified so that the TM wave of an exposure source can be amplified during the immersion etching. In addition, a defect pattern is converted onto a wafer, and thus it becomes difficult to fabricate a semiconductor device having a pattern size of less than 45 nm. The birefringence of the phase shifting layer can be controlled by a composition ratio of molybdenum, niobium and nitrogen, an architecture such as the same single layer or a plurality of layers, density, surface roughness and degree of crystallization.

In step b1), if the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, the main component of the light shielding layer and the anti-reflection layer may be chromium.

In step b1), if the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, at 193 nm to 248 nm, the optical density of the light shielding layer and the anti-reflection layer may be 2.5 to 3.5.

In step b1), if the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, the thickness of the light shielding layer and the anti-reflection layer may be 300 angstroms to 1100 angstroms.

In the step b1), if a main component of the metal layer is chromium and the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, a metal layer containing a molybdenum molybdenum or tantalum as a main component may be The anti-reflection layer is additionally formed.

In step b1), if a main component of the metal layer is chromium and the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, and a metal layer is additionally formed in the anti-reflection layer. A major component of the additional metal layer is molybdenum molybdenum and may be a compound selected from the group consisting of oxygen, nitrogen, carbon, nitrogen oxides, carbonitrides or nitroxides.

In step b1), if a main component of the metal layer is chromium and the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, and a metal layer is additionally formed in the anti-reflection layer. The additional metal layer main component comprises cerium and may be a compound selected from the group consisting of oxygen, nitrogen, carbon, nitrogen oxides, carbonitrides or oxynitrides.

In the step b1), if the main component of the metal layer is chromium and the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, a metal layer is not used for etching solution or etching gas (used in Etching during etching of molybdenum or antimony telluride may be additionally formed on the antireflection layer.

In step b1), if a main component of the metal layer is chromium and the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, a metal layer is not etched by an etching solution or an etching gas (used Etching during etching of molybdenum telluride or germanium may additionally be formed under the light shielding layer.

In step b1), if a main component of the metal layer is chromium and the metal layer comprises a light shielding layer and an anti-reflection layer, which are continuously formed, a metal layer is not etched by an etching solution or an etching gas (used Etching during etching of molybdenum telluride or germanium may additionally be formed under the light shielding layer or on the anti-reflective layer.

In step c1), the photoresist layer is a chemically enhanced photoresist and has a thickness of less than 4000 angstroms.

Steps a2) through j2) are performed by a typical manufacturing process of a half-phase shifting hood.

The halftone phase shifting blank reticle and the halftone phase shifting hood are fabricated in accordance with the present invention by the above steps.

Therefore, the present invention has the following advantageous effects: according to a specific embodiment of the present invention, a thin layer characteristic obtained by a phase shift layer formation process can be maintained for a long period of time. At the same time, a non-crystallized phase shifting layer is formed through a pattern forming process. A pattern with excellent characteristics can be formed. Thus, a phase shift layer having high quality and minimum line width characteristics can be formed and thus a semiconductor circuit having excellent quality can be manufactured.

Moreover, in accordance with a particular embodiment of the present invention, a phase shifting layer having a lower incidence of growth defects due to continuous exposure is provided through the surface treatment of the phase shifting layer. At the same time, throughput is improved and a high quality semiconductor circuit can be fabricated.

Moreover, in accordance with a particular embodiment of the present invention, the loading effect is reduced by forming a hard mask layer to form the excellent pattern. Therefore, excellent high quality semiconductor circuits can be fabricated.

The above-disclosed subject matter is intended to be illustrative, and not restrictive, and the accompanying claims are intended to Therefore, the scope of the invention is to be determined by the following claims and the maximum permissible

In the following, referring to the first figure, a half-phase shifting blank mask and a method of manufacturing the same will be described.

<Specific Example 1 and Control 1>

A phase shifting layer 20 is formed on a transparent substrate 10 comprising molybdenum niobium nitrogen. Here, the phase shift layer 20 is formed by using a DC magnetron sputtering method through a plurality of sputtering targets. The sputtering targets have a composition ratio of molybdenum to niobium of 10 to 90 percent and molybdenum to niobium of 20 to 80 percent, respectively. The sputter targets have the same 6 mm thickness, and the sputter targets are bonded to a support disk formed by using high purity copper having indium and having a thickness of 6 mm. In Control 1, the sputtering procedure was performed using a sputtering target having a thickness of 8 mm under the same conditions. At this time, 20 sccm of argon, 80 sccm of nitrogen, a power density of 1.5 watts per square centimeter, and a pressure of 1.5 m were applied to a target having a molybdenum ratio of 10 to 90 percent. 34 sccm of argon, 61 sccm of nitrogen, a power density of 1.8 W per square centimeter, and a pressure of 2 m were applied to a sputtering target having a molybdenum to bismuth ratio of 20 to 80 percent. During the sputtering process, the substrate 10 is heat treated at a temperature of 200 degrees Celsius at a molybdenum ratio of 10 to 90 percent of the sputtering target and 180 degrees Celsius at a molybdenum ratio of 20 percent. 80% of the splash target.

In an example in which the molybdenum to niobium is 10 to 90 percent, the phase shifting layer 20 formed by the specific embodiment 1 has an average transmittance of 5.8 percent and an average phase shift of 177 degrees. The argon fluoride has a wavelength of 193 nm and an average thickness of 648 angstroms. In addition, the difference between the maximum value and the minimum value of the transmittance, phase shift, and thickness distribution ranges is measured from 49 points in a range of 142 mm × 142 mm (that is, the true effective range of the substrate 10) ), 0.08, 0.9 degrees, and 6 angstroms, respectively. This shows an excellent distribution. In an example where the molybdenum to bismuth is 20 to 80 percent, the phase shifting layer 20 formed by the specific embodiment 1 has an average transmittance of 5.7 percent and an average phase shift of 177 degrees. The effect is on a cesium fluoride having a wavelength of 248 nm and an average thickness of 853 angstroms. Further, the difference between the maximum value and the minimum value of the transmittance, the phase shift, and the thickness distribution range is measured at 49 points in a range of 142 mm × 142 mm (that is, the true effective range of the substrate 10) , which is 0.07, 0.8 degrees, and 5 angstroms, respectively, which shows an excellent distribution.

Conversely, in a phase shifting layer of argon fluoride having a molybdenum ratio of 10 to 90 percent through the control 1, an average transmittance of 5.75 percent, 177 degrees can be measured. The average phase shift and an average thickness of 652 angstroms. However, the difference between the maximum and minimum values of the range of transmittance, phase shift, and thickness distribution is 0.34, 3.6, and 28 angstroms, respectively. This shows a very poor quality phase shifting layer. Although a phase shift layer of argon fluoride having a molybdenum to bismuth ratio of 20 to 80 percent can measure an average transmittance of 5.73 percent, an average phase shift of 178 degrees, and an 855 angstrom phase. The average thickness, the difference between the maximum and minimum values is 0.38, 3.2 degrees, and 25 angstroms, respectively. This also shows a very poor quality phase shifting layer. Due to the thick thickness of the sputtering target, a magnetic field formed by a magnet at a cathode makes it difficult to effectively control the plasma distribution and density, so that atoms fall from the target into irregular distribution during the sputtering process, so that Defects may occur on the substrate. Through this specific embodiment 1, a phase shift layer having excellent quality can be manufactured.

<Specific Example 2 and Control 2>

This specific embodiment 2 illustrates a method of fabricating a halftone phase shifting blank reticle in which a phase shifting layer is formed.

First, a phase shifting layer having a nitrogen oxynitride component is formed on a transparent substrate. The phase shifting layer is formed by DC magnetron sputtering, and a sputtering target having a composition (10 to 90 percent by weight of molybdenum) and a target thickness of 6 mm is used. 20 sccm of argon, 80 sccm of nitrogen, a power density of 1.5 watts per square centimeter, and a pressure of 1.5 mTorr were used. The substrate was heat treated at a temperature of 200 degrees Celsius during the sputtering process. In this control 2, 10 sccm of argon, 90 sccm of nitrogen, a power density of 1.7 watts per square centimeter, and other unmentioned conditions were used, which were the same as in the specific example 2.

The phase shifting layer was formed through this embodiment to have a thickness of 649 angstroms, a transmittance of 5.95 percent, and a phase shift of 177 degrees. This shows a good quality phase shift layer. In an example in which a phase shifting layer was formed through the control 2, a thickness of 925, a transmittance of 6.01 percent, and a phase shift of 181 degrees were measured. The composition ratio of the phase shift layer formed through Concrete Example 2 and the control 2 was measured by an auger analysis. In an example of this specific embodiment 2, a component ratio (the molybdenum to rhodium is 5 to 58 percent to 37 percent) is measured. In one example of this control 2, a component ratio (the molybdenum to rhodium is 7 to 45 percent to 48 percent) is measured. Further, the crystalline state of the phase shift layer can be analyzed by an X-ray diffractometer. In an example of this specific embodiment 2, it was analyzed in an amorphous state, and in one example of the control 2, weak crystallization of molybdenum nitride and tantalum nitride was measured, and incomplete crystallization was measured.

In order to test the stability of a thin layer, the amount of change in transmittance was measured every day for 10 days. In the example of a specific embodiment 2, the transmittance increased by 0.05% in 10 days, but in one example of the control 2, the transmittance increased by 0.3%. In an example of Control 2, the range of transmittance is changed to exceed the allowable transmittance (i.e., 5.9 to 6.1 percent). Therefore, a poor product is produced. In contrast, in an example of this specific embodiment 2, the transmittance characteristic was not changed after 10 days to confirm that the phase shift layer had excellent quality and production profit with respect to the control 2. During the formation of the phase shifting layer, a stable film having a low free energy is formed by a combination of appropriate molybdenum, niobium and nitrogen. In one example of this control 2, since the ratio of one to nitrogen is similar, the ionization energy therebetween is high, and the free energy of a thin layer is increased. Thus, the transmittance increases due to thermodynamic properties in which free energy attempts to move to a lower state because the phase shift layer is unstable. Also, due to a relatively high energy, the oxidation of the phase shifting layer is promoted so that the transmittance increases.

After a light shielding layer, an anti-reflective layer, and a photoresist layer are formed, a 65 nm class photomask is fabricated to measure the uniformity of the minimum line width of the pattern. In the example of a specific embodiment 2, the consistency of the minimum line width of 5.7 nm can be measured. In an example of the control 2, the uniformity of the minimum line width of 18.7 nm can be measured, which is a very crude result. To analyze the result, the angle of the sidewall of a pattern is measured by a partial scanning electron microscope, and A pattern is accurately measured through a surface scanning electron microscope. In an example of this specific embodiment 2, the angle of the side wall of the pattern is 87 degrees, which is excellent as a pattern formation. In an example of this control 2, the angle of the side wall of the pattern is 78 degrees, which is very rough as a pattern formation. According to the accuracy observation result of the pattern, the formation of the pattern in the specific embodiment 2 is relatively stable. However, the sidewall of the pattern in the control 2 is not flat. In the example of a specific embodiment 2, when the ratio of molybdenum, niobium and nitrogen is optimized, crystallization does not occur in a thin layer, and thus Achieve outstanding pattern characteristics. In contrast, in the case of a control 2, the formation of the pattern was defective due to partial crystallization of the molybdenum nitride and tantalum nitride, and thus the minimum line width uniformity was not perfect. Due to this poor pattern, it is difficult to manufacture a half-phase shifting cover having a high quality and at the same time the half-phase shifting cover becomes unsuitable as a 65 nm manufacturing process.

<Specific Example 3 and Control 3>

This specific embodiment 3 illustrates the fabrication of a half-shifted phase shift mask having a phase shifting layer suitable as a cesium fluoride etching method by the same method as in Example 2 and Control 2.

As in the embodiment 2, a phase shift layer having a molybdenum ruthenium nitride compound is formed on a transparent substrate. A target having a composition (molybdenum to bismuth of 20 to 80 percent) and a thickness of 5 mm. 34 sccm of argon, 61 sccm of nitrogen, a power density of 1.8 watts per square centimeter, and a pressure of 2 mTorr were applied. During the sputtering process, a heat treatment temperature of the substrate is 180 degrees Celsius. In a comparative example of 3, 20 sccm of argon, 80 sccm of nitrogen, a power density of 1.5 watts per square centimeter, and a pressure of 2 mTorr were applied, and the conditions not mentioned were the same as in the specific example 2.

A phase shifting layer was formed through this specific example 3, which had a thickness of 855 angstroms, a transmittance of 5.73 percent, and a phase shift of 178 degrees, which was excellent. In an example of forming a phase shifting layer through the control 3, a thickness of 848 angstroms was measured, a transmittance of 6.08 percent at 248 nm, and a phase shift of 172 degrees. The phase shift layer as cesium fluoride was passed through the component ratio of the specific example 3 and the control 3 was measured by a helix analysis. In one example of this control 3, the ratio of the composition of molybdenum to rhodium to nitrogen is 10 to 45 percent to 45 percent, respectively. A crystal state as the phase shifting layer was analyzed by an X-ray diffractometer. In the example of a specific embodiment 3, the analysis is in an amorphous state, and in one example of the control 3, the weak crystals of molybdenum nitride and tantalum nitride are not completely crystallized, as the phase shift as the argon fluoride The layers are measured.

In order to check the stability of the thin layer, the amount of change in transmittance was measured every day for 10 days. In the example of a specific embodiment 3, the transmittance increased by 0.04% in 10 days, but in the case of a control 3, the transmittance increased by 0.25 percent. In the case of a control 3, The range of transmittance becomes more than the allowable transmittance (ie, 5.9 to 6.1 percent).

After a light shielding layer, an anti-reflective layer, and a photoresist layer are formed, a 90 nm-grade mask is fabricated to observe an angle of a pattern sidewall through a scanning electron microscope, and through surface scanning electrons. Microscope to observe the accuracy of a pattern. In the example of a specific embodiment 3, an angle of a pattern sidewall is 85 degrees, which is a very good pattern tendency, but in a comparison 3 example, an angle of a pattern sidewall is 75 degrees, which is A very crude pattern tends. Further, in the example of a specific embodiment 3, the formation of a pattern is smoothly performed, but in an example of the comparison 3, the side walls of the pattern are unevenly formed.

<Specific Example 4 and Controls 4 and 5>

In order to control growth defects, gaseous molecular contamination is analyzed, which is a major factor in growth defects. A phase shifting layer as argon fluoride is deposited on a transparent substrate having a sputtering target of one component (molybdenum to cerium of 10 to 90 percent). The manufacturing procedure was performed under the sputtering conditions such as 20 sccm of argon, 80 sccm of nitrogen, and a thickness of 650 angstroms. In the examples of the controls 4 and 5, the phase shifting layer as argon fluoride is deposited on a transparent substrate having a ratio of molybdenum to niobium of 4 to 96 percent and a percentage The ratio of 6 to 84 percent of the composition of the sputtering target. The manufacturing procedure was carried out under sputtering conditions of Control 4, such as 30 sccm of argon, 70 sccm of nitrogen, and a thickness of 700 angstroms. In a comparative example of 5, the manufacturing procedure was carried out under conditions of 10 sccm of argon, 90 sccm of nitrogen, and a thickness of 925 angstroms. The phase shift layer is formed through the specific example 4 and the controls 4 and 5. Through a spiral analysis, the ratio of molybdenum to niobium to nitrogen is 10% to 60%, 30%, and 100%, respectively. 8 out of 64 to 28 percent, and 7 percent to 45 percent to 48 percent.

The phase shifting layer is treated with standard cleaning 1 (SC1) and sulfuric acid as a cleaning procedure for a typical reticle manufacturing process. A first processing procedure uses a special container made of quartz and has a volume of 420 ml ready to observe gaseous molecular contamination. Next, an ion chromatography method is performed. Deionized water was used during the first treatment procedure to be 200 ml. After passing through a pressure cooker equipped with a heat treatment program that lasted 120 degrees every 20 minutes, an IC analysis used 10 milliliters of deionized water containing impurity ions from the phase shifting layer.

The table alone illustrates the results of one ion analysis of this specific example 4 and controls 4 and 5. According to the results of the ion analysis, the adsorption amounts of the sulfate and ammonium ions in the controls 4 and 5 were higher than those of the specific example 4. It is expected that this growth defect greatly occurs if a 193 nm exposure energy is irradiated to the phase shift layers of the controls 4 and 5. An exposure energy of 12 kilojoules per square centimeter was irradiated with a 193 nm argon fluoride laser in this specific example 4 and the controls 4 and 5, and then growth defects were observed through an electron microscope.

Form 2 illustrates the growth defect results observed in this specific example 4 and the controls 4 and 5. As a result of this ion analysis, growth defects occurred more frequently in the controls 4 and 5, in which sulfuric acid and ammonium ions (i.e., major growth defect factors) were greatly adsorbed. Compared to this specific embodiment 4, the control 4 includes relatively more germanium, thus causing a mismatch in the composition of the phase shifting layer, so that the surface energy at the surface of the phase shifting layer may increase. Thus, the adsorption energy is increased to cause a large amount of adsorption of monosulfate and ammonium ions, so that more growth defects occur. Compared to this specific example 4, the control 5 has relatively less ruthenium, but has a larger amount of nitrogen to increase the composition of tantalum nitride and molybdenum nitride. Thus, the composition mismatch occurs and thus the phase shift layer becomes unstable. Due to this, a large amount of ions are adsorbed on the surface of the phase shift layer. That is, this is a major factor that causes a large number of growth defects. On the contrary, in a specific embodiment 4, a proper composition ratio of molybdenum, niobium and nitrogen makes it possible to produce a phase shift layer having a relatively stable state, and therefore, a small amount of ion absorption occurs. As a result, a phase shift layer having fewer growth defects is produced by controlling the proportion of the components of a phase shift layer, and thus, finally, a phase shift mask having a high quality can be realized.

<Specific Example 5 and Controls 6 and 7>

In the specific example 5 and the controls 6 and 7, in order to use the molybdenum rhenium nitrogen phase shift layer to be produced by the specific example 4 and the controls 4 and 5, birefringence was measured. In immersion etching, which achieves a critical dimension of less than 65 nanometers, birefringence becomes an important issue as the number of apertures increases. Further, in this specific example 5 and the controls 6 and 7, the birefringence of the phase shift layer was measured by a Uniopts ABR-10A-30A in the range of 142 nm × 142 nm. This particular embodiment and the transparent substrate of the control have the same birefringence.

In an example of this specific embodiment 5, low birefringence is measured. In one of the examples of Controls 6 and 7, high birefringence was measured, and as the birefringence became higher, the degree of polarization (DoP) became larger during the immersion etching. Thus, this greatly causes a critical dimension error that makes it difficult to fabricate a semiconductor device having excellent performance. In the controls 6 and 7, the measured high birefringence means that the helium and nitrogen are additionally more than one appropriate. The amount increases. Thus, this causes microcrystallization in the phase shift layer. Due to this microcrystallization, more of the optical axis related to exposure is formed so that the birefringence characteristics of the phase shift layer become more specific. Thus, a high birefringence is measured and thus it is difficult to achieve a critical dimension having a high quality of less than 65 nm or 45 nm. In contrast, in an example of this specific embodiment 5, an appropriate amount of niobium and nitrogen is increased to produce a phase shifting layer having low birefringence characteristics. Thus, a critical dimension of a semiconductor circuit having high quality can be achieved.

In the following, a half-phase shifting mask will be described which is manufactured using the halftone phase shifting blank mask formed by the above-described method.

Referring to the second figure, a metal layer 30 can be formed on the phase shift layer 20 of the halftone phase shift blank mask. The metal layer 30 may include a light shielding layer and an anti-reflection layer which are continuously formed, and the main component of the light shielding layer and the anti-reflection layer may be chromium. An optical density of the light shielding layer and the reflective layer may be from 2.5 to 3.5 at a wavelength of from 193 nm to 248 nm. The thickness of the light shielding layer and the anti-reflection layer may be 300 angstroms to 1100 angstroms, respectively. Another metal layer including molybdenum molybdenum or niobium as the main component may be additionally formed on the antireflection layer. The other metal layer may be a compound selected from the group consisting of oxygen, nitrogen, carbon, nitrogen oxides, carbon oxides, nitrogen carbides, or carbon oxynitride.

In another method, the light shielding layer and the main component of the reflective layer may be molybdenum molybdenum or niobium. A metal layer comprising a material that is not etched by an etching solution or an etching gas is used in an etching process that may be additionally formed under the light shielding layer or over the anti-reflection layer.

The photoresist layer is exposed in a region formed by a pattern. The photoresist layer is developed using a developing solution, and thus the exposed photoresist layer is removed to form a photoresist pattern. The photoresist pattern is used as a mask after the metal layer is etched to form a metal layer pattern. The photoresist pattern and the metal layer pattern are used as a mask to form the phase shift layer pattern. After the photoresist layer is removed, another photoresist layer is formed. The photoresist layer is exposed in a region formed by a pattern. The photoresist layer is formed to form a metal layer pattern. A half-phase shifting mask is formed by removing the photoresist layer.

10‧‧‧Transparent substrate

20‧‧‧ phase shift layer

30‧‧‧metal layer

The first drawing shows a cross-sectional view of a particular embodiment of the invention, illustrating a phase shifting layer on a transparent substrate; and a second drawing showing a cross-sectional view of a particular embodiment in accordance with the present invention, illustrating a phase The migration layer comprises a metal layer.

10. . . Transparent substrate

20. . . Phase shift layer

Claims (12)

A halftone phase shift blank mask for use in a cesium fluoride etching process, the halftone phase shift blank mask comprising: a transparent substrate; and a phase shifting layer formed on the transparent substrate Wherein the phase shifting layer is amorphous and comprises from 7 to 15 percent molybdenum, from 55 to 63 percent bismuth, and from 28 to 35 percent nitrogen; The phase shift layer has a surface impurity concentration of less than 400 ppb. The phase shifting layer further comprises less than 5 percent oxygen, as described in the first aspect of the invention. A halftone phase shift blank mask for use in a cesium fluoride etching process, the halftone phase shift blank mask comprises: a transparent substrate; and a phase shifting layer on the transparent substrate, wherein the The phase shifting layer is amorphous and has a thickness of from 850 angstroms to 950 angstroms and a transmittance of from 4.5 to 9 percent; wherein the phase shifting layer has a surface impurity concentration of less than 400 ppb. A phase shifting blank mask as described in claim 3, wherein the phase shifting layer comprises molybdenum, niobium and nitrogen. A halftone phase shift blank mask for use in an argon fluoride etch, the halftone phase shift blank mask comprising: a transparent substrate; and a phase shifting layer attached to the transparent substrate, wherein the phase The migration layer is amorphous And containing 5 to 12 percent of molybdenum, 57 to 65 percent of bismuth, and 28 to 35 percent of nitrogen; wherein the phase shift layer has a low surface impurity concentration At 400ppb. The phase shifting layer further comprises less than 5 percent oxygen, as described in the fifth aspect of the invention. A halftone phase shift blank mask for use in an argon fluoride etch, the halftone phase shift blank mask comprising: a transparent substrate; and a phase shifting layer attached to the transparent substrate, wherein the phase The migration layer is amorphous and has a thickness of from 630 angstroms to 680 angstroms and a transmittance of from 4.5 to 7 percent; wherein the phase shift layer has a surface impurity concentration of less than 400 ppb. A phase shifting blank mask as described in claim 7 wherein the phase shifting layer comprises molybdenum, niobium and nitrogen. A halftone phase shift blank mask for use in a cesium fluoride etching process, the halftone phase shift blank mask comprises: a transparent substrate; and a phase shifting layer formed on the transparent substrate Wherein the phase shift layer has a birefringence of 2 nm per phase shift layer at a wavelength of 193 nm; wherein the phase shift layer has a surface impurity concentration of less than 400 ppb. One use as in any one of items 1 to 9 of the patent application scope The half-tone phase shift mask manufactured by the halftone phase shift blank mask is described. A method for manufacturing a halftone phase shift blank mask for use in a cesium fluoride etching process, the method for manufacturing a halftone phase shift blank mask comprising: forming a phase shift layer on a transparent substrate, wherein the phase shift layer Forming includes: providing the transparent substrate; providing a sputtering target comprising 15 to 25 percent molybdenum and 75 to 85 percent germanium; and providing a volume percentage of 30 to 40 percent by volume Argon and a volume percentage of 60 to 70% by volume of nitrogen are used to perform a sputtering process wherein a surface of the phase shifting layer is controlled to have an impurity ion concentration of less than 400 ppb. A method for manufacturing a halftone phase shift blank mask for use in an argon fluoride etching process, the method for manufacturing a halftone phase shift blank mask comprising: forming a phase shift layer on a transparent substrate, wherein the phase shift layer Forming includes: providing the transparent substrate; providing a sputtering target comprising 5 to 15 percent molybdenum and 85 to 95 percent bismuth; and providing a volume percentage of 15 to 25 percent by volume Argon and a nitrogen gas having a volume percentage of 75 to a volume percentage of 85 to perform a sputtering process, wherein a surface of the phase shifting layer is controlled to have an impurity of less than 400 ppb Subconcentration.