JPH0437422B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION 1. Industrial Application Field The present invention relates to a method of manufacturing an exposure mask, and in particular,
The present invention relates to a method of manufacturing a photomask called a hard mask used in manufacturing semiconductor devices and the like.

2. Prior Art Conventionally, hard masks using inorganic thin film materials have been used for exposure processing, and are considered useful because they have high surface strength and can be used repeatedly.

As this type of hard mask, one in which amorphous silicon (hereinafter referred to as a-Si) is used as a light shielding film is known. This a-Si film has good see-through properties, allowing the underlying pattern to be seen through it during semiconductor processing (that is, it is transparent to some extent to visible light), and at the same time, it has the property of being transparent to visible light to some extent. For example, it exhibits light-shielding properties against exposure beams of 3800 Å or 4300 Å).

The degree of this light-shielding property is determined by the following optical density (optical density).
density).

Optical density = log(Io/I) (Io is the amount of incident light, I is the amount of transmitted light) However, the optical density of the above-mentioned commonly used a-Si film is at most 1.2 to 1.4. , since a certain amount of incident light is transmitted, it cannot serve as a perfect exposure mask. Therefore, it is possible to increase the thickness of the a-Si film in order to obtain sufficient light-shielding properties, but in this case, when etching the a-Si film into a predetermined mask pattern from the mask material, If the film thickness is increased, the etching accuracy deteriorates, and in particular, side etching progresses too much during wet etching, which tends to result in defective mask patterns.

Vacuum deposition methods and sputtering methods are known as methods for manufacturing the above a-Si hard mask.

In the vacuum evaporation method, the inside of the evaporation tank is evacuated to a high vacuum, and then silicon is heated and evaporated with an electron beam or the like to form an a-Si film on a substrate placed above. Furthermore, in the sputtering method, an a-Si film is deposited on a substrate by using silicon as a target and applying a high frequency voltage to the target in an inert gas (Ar). However, in both methods, silicon is very active, so even if the chamber is evacuated to a high vacuum or the film is formed in an inert gas, oxygen gas is always present as a residual gas in the chamber. A portion of it becomes silicon oxide. This causes a decrease in optical density as a photomask and further a decrease in dry etching speed. In addition, especially when using the vacuum evaporation method, it is difficult to maintain cleanliness due to the complex structure inside the tank, and the formed thin film can be manufactured into highly integrated LSIs (Large-Scale Integrated Circuits), etc. It is not suitable for photomask materials.

3. OBJECTS OF THE INVENTION An object of the present invention is to provide a method for manufacturing an exposure mask material that prevents oxidation of a mask layer, has a high optical density, and has a high etching rate.

Another object of the present invention is to obtain a high-quality exposure mask with fewer defects such as dust adhesion and pinholes by using a sputtering device with a simplified structure in the film forming device and less contamination. It is in.

4 Structure and Effects of the Invention The present invention includes an optically transparent mask base and a target made of silicon and/or germanium placed in a sputtering device, and 1 to 50 volumes of hydrocarbon gas and/or ammonia gas. applying a voltage to the target in an atmosphere containing % to sputter the target and simultaneously decompose the hydrocarbon gas and/or ammonia gas;
The resulting ionized or activated hydrogen is reacted with sputtered silicon and/or germanium to form an amorphous semiconductor layer made of silicon and/or germanium containing hydrogen on the mask substrate, and then, The present invention relates to a method of manufacturing an exposure mask, in which the hydrogen-containing amorphous semiconductor layer is etched into a predetermined mask pattern, and an exposure mask having a light-shielding layer made only of the hydrogen-containing amorphous semiconductor layer in the predetermined pattern is created.

According to the method of the present invention, an atmosphere containing hydrocarbon gas and/or ammonia gas is used when forming an amorphous semiconductor layer as a mask layer made of Si and/or Ge by sputtering. The reduction action of hydrogen (especially activated or ionized hydrogen) generated by its decomposition effectively prevents oxidation of the amorphous semiconductor layer, and since a certain amount of hydrogen is incorporated into the semiconductor layer, optical concentration can be increased.
Therefore, the etching speed during processing into an exposure mask is improved, and etching accuracy is also improved because sufficient optical density is exhibited even if the mask layer is made thin. Furthermore, since the film is formed by sputtering, the internal structure of the device is simple, there are few sources of contamination, and a high-quality mask material with few defects such as dust adhesion and pinholes can be produced. Furthermore, the hydrocarbon gas and ammonia gas used are themselves stable and are advantageous in terms of safety.

5 Examples Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 shows an example of a sputtering apparatus used in this embodiment. According to this device,
A target 11 and a transparent mask substrate 2 are placed facing each other in a vacuum chamber 10.
A high frequency or direct current high voltage 8 is applied between the holder 7 and the holder 7, and the holder 7 is configured so that the substrate 2 can be heated to a predetermined temperature by a built-in heater (not shown). The vacuum chamber 10 is also provided with an inlet pipe 13 for a mixed gas 9 of Ar and CH ₄ and an outlet pipe 15 for an exhaust gas 14.
The inside of the tank is maintained at a predetermined degree of vacuum by connecting it to a vacuum pump (not shown).

Next, a method for forming an amorphous semiconductor film on the substrate 2 using the sputtering apparatus shown in FIG. 1 will be described.

First, the inside of the vacuum chamber 10 is evacuated to, for example, 1×10 ^-6 Torr, and while heating the substrate 2 to, for example, 300°C,
and CH ₄ at a ratio of 80:20, for example, and set the pressure in the tank to 1×10 ^-3 Torr.
(e.g. high purity Si target) for example 13.56M
Apply high frequency high voltage of Hz, 750W. This causes a glow discharge between the target 11 and the substrate 2, and the ^cathode target 1 is
The Si in 1 is knocked out and deposited on the substrate 2 as a-Si, and at the same time, CH ₄ is
The ionized or activated hydrogen produced by the decomposition of is reacted with Si and introduced into the a-Si film. A containing hydrogen atoms at a predetermined concentration according to the substrate 2
- form a Si (hereinafter referred to as a-Si:H) layer,
You can create photomask materials.

The CH ₄ introduced into the vacuum chamber is CH ₄ that has not yet been activated or ionized, but a gas discharge tube (not shown) is connected to the introduction tube 13 and CH ₄ is introduced in an activated or ionized state. You may. Also,
The amount of CH ₄ introduced is not limited to the above ratio, but it is essential to set it to 1 to 50% by volume of the introduced gas (that is, the atmosphere in the vacuum chamber). If the amount of CH ₄ is less than 1% by volume, the effect of hydrogen introduction will be poor, and
If it exceeds 50% by volume, the optical density of the a-Si film will actually decrease.

As mentioned above, in the manufacturing method of the mask material according to this example, since an atmosphere containing CH ₄ is used when forming a-Si into a film by the sputtering method, there is no residual oxygen in the vacuum chamber. However, this effect of oxygen can be prevented by the reducing effect of hydrogen generated by decomposition of CH ₄ . As a result, the optical density can be improved by preventing the oxidation phenomenon in the a-Si bath, and the processability (etching rate) during dry etching described later can be improved. Furthermore, since the sputtering method is used, the structure inside the vacuum chamber is simple, there is less contamination, and it is possible to create a high-quality mask material.

2 to 9 show various photomask materials produced by the above method.

The mask material 1 shown in FIG. 2 has an a-Si:H layer 3 provided on an optically transparent quartz plate (SiO ₂ plate) 2. The thickness of _SiO2 as a substrate is 0.5 ~
3 mm (preferably 1 to 2.5 mm), and a-
The thickness of the Si:H layer 3 is 300 to 5000 Å (preferably 700 Å
~3000Å). This mask material 1 is processed into a desired exposure mask by the method described below. In the example shown in FIG. 2, a non-quartz plate (e.g., soda lime, borosilicate) whose thermal expansion coefficient is close to that of a-Si can be used as the substrate.

In FIG. 3, a SiO ₂ film 4 is first formed with a thickness of 100 to 5000 Å on a non-quartz plate 12 made of soda lime, borosilicate, etc.
(preferably 200 to 3000 Å), and an a-Si:H layer 3 is provided thereon. In this case, the a-Si:H layer 3 is prevented from being contaminated by impurities.

FIG. 4 shows a transparent conductive film (for example, ITO: Indium Tin Oxide) 5 between the a-Si:H layer 3 and the substrate 2.
An example is shown in which the membrane is interposed as a limiting membrane. Due to the presence of the transparent conductive film 5, the potential on the substrate can be stably maintained under discharge during dry etching to a mask pattern, which will be described later, and when the substrate is brought into close contact with a semiconductor wafer during semiconductor processing, which will be described later. It has the effect of preventing discharge on the surface and preventing damage to the mask edge. In other words, the movement of charges through the transparent conductive film 5 can prevent accumulation or fixation of charges.

FIG. 5 shows a combination of the examples shown in FIGS. 3 and 4, in which a transparent conductive film 5 and a SiO ₂ film 4 are provided between the a-Si:H layer 3 and the substrate 2.

None of the mask materials shown in Figures 2 to 5 above are provided with anti-reflection means, but during processing of the semiconductor surface, which will be described later, the reflected light from the same surface is further reflected by the mask surface, and It is preferable to provide an antireflection film to prevent the photoresist film from being accidentally exposed to light.

FIG. 6 shows a-Si:H containing oxygen atoms or a-Si containing oxygen atoms on the a-Si:H layer 3.
An anti-reflection film 6 is provided, and this anti-reflection film guides reflected light into the layer of the mask,
This shows an example of preventing re-reflection from the mask surface. The thickness of the antireflection film 5 is set to a value that minimizes reflection, depending on the wavelength used during exposure.

FIG. 7 shows the above-mentioned antireflection film 6 as a-Si:
An example in which it is also provided between the H layer 3 and the substrate 2 or 12 is shown.

Depending on the size of the semiconductor wafer to be processed, the outer shape of the exposure mask material in each of the above examples (or the exposure mask described later) may be square as shown in FIG. 8, or may be square as shown in FIG. It may have the same shape as the wafer.

The mask material 1 exemplified above has a predetermined amount (especially
a-Si containing hydrogen atoms (0.1 to 30 at%):
Since it includes the H layer 3, the optical density at the wavelength used is significantly improved compared to conventional a-Si see-through masks. That is, the data shown in Figure 10 (the wavelength used is 4300 Å, the thickness of the a-Si:H layer is
According to 1000 Å), the optical density changes depending on the hydrogen content, especially in the range of 0.3 to 25 at%, it exceeds the optical density of the conventional mask (1.2 to 1.4), and in the range of 0.5 to 25 at.
It can be seen that at 20 atomic %, the improvement is 2 to 4 times.
It is noteworthy that the optical density increases rapidly in a range where the hydrogen content is low, which clearly shows the advantage of actively introducing hydrogen atoms into a-Si according to the present invention. It is being said. Note that a-Si:
The hydrogen content in the H layer was determined from the infrared absorption spectrum of a homogeneous film formed on a high-resistance Si wafer (the wafer thickness in this case is, for example,
500 μm, a-Si:H film thickness is, for example, 1 to 5 μm).

An example of the above-mentioned infrared absorption spectrum is shown in Fig. 11.
^-1 ), focusing on the stretching vibration of a-Si. Then, the hydrogen concentration N (cm ^-3 ) is N=K×1
(K is a constant) When calculated from the relational expression, in the example shown,
It is estimated to be around 10 atomic percent. However, this is just an example, and the hydrogen concentration can also be determined based on other known methods.

In this way, since the a-Si:H layer exhibits high optical density due to the inclusion of a predetermined amount of hydrogen atoms, its film thickness can be reduced, and the etching accuracy and etching process described below can be improved. Sharpness is greatly improved.

That is, as shown in FIG. 12A, a known photoresist 28 is uniformly applied on the a-Si:H layer 3, and then, as shown in FIG. 12B, a previously prepared exposure mask 17 is placed and exposed. The exposure beam 18 passes through the non-masked portion 19 of the mask 7 and exposes the underlying photoresist 28 in a predetermined pattern.

Next, for example, the unexposed portions of the hot resist 28 are removed by etching, leaving a pattern as shown in FIG. 12C. Then, as shown in FIG. 12D, the underlying a-Si:H layer 3 is etched and patterned by plasma etching or the like using the photoresist 28 as a mask. At this time, if the above-mentioned transparent conductive film 5 is present, it becomes difficult for the charges caused by the plasma to concentrate,
Potential stability can be maintained.

FIG. 12E shows the exposure mask 21 manufactured in this manner.

In the manufacturing process of this exposure mask, especially in the etching step shown in FIG. 12D, when the a-Si:H layer 3 is etched by a dry process such as plasma etching, the a-Si:H layer 3 is
Since the film thickness can be as thin as, for example, about 1000 Å, the etching accuracy is extremely high. Therefore, side etching, which could not be avoided in the past, can be prevented, and pattern accuracy as an exposure mask can be significantly improved. Furthermore, when a wet process using a hydrofluoric acid-based etchant is applied to this etching, the side etching described above tends to proceed, but this is also sufficient in the hydrogen-containing a-Si layer formed by the method of the present invention. can be prevented.

The exposure mask according to the present invention is particularly suitable for semiconductor ICs,
Very useful for miniaturization process in LSI etc. To roughly describe an example in which the exposure mask according to the present invention is applied to semiconductor manufacturing, as shown in FIG. 12F,
A photoresist 24 is applied on a SiO ₂ film 23 formed on one main surface of a silicon wafer 30 by a known thermal oxidation technique, and an exposure mask 21 is placed on the photoresist. At this time, since the a-Si:H layer 3 of the mask 21 is transparent to visible light to some extent, the wafer 3
If some pattern (for example, a field SiO ₂ film for element isolation, etc.) already exists on the surface of the mask 0, that pattern can be observed and therefore mask alignment can be performed more accurately. Also,
If the transparent conductive film 5 described above is provided, the mask 2
This reduces the accumulation of charges on the wafer 1, eliminates discharge between the wafer 30, and prevents mask destruction and electrostatic damage to the wafer. Next, for example, ultraviolet light 25 having a wavelength of 3000 to 4400 Å is irradiated from an ultra-high pressure mercury lamp of about 200 W to selectively expose the photoresist 24 under the non-masked portion where the mask layer 3 is not present.

The mask layer 3 made of a-Si:H does not transmit the light 25 in the above-mentioned wavelength range and exhibits sufficient light-shielding properties. Furthermore, as shown in FIG. 12G, using the developed photoresist 24 as a mask, the underlying SiO ₂ film 23 is etched with plasma etching (dry process) or an aqueous solution of hydrofluoric acid or ammonium fluoride (wet process), and the wafer is etched. 30 in the desired pattern.

In this way, contact holes 26 for covering electrodes or wiring, for example, can be formed in the SiO ₂ film 23.

FIG. 13 shows a plasma etching apparatus that can be used in the etching process shown in FIG. 12D, in which 37 is a holder for holding the substrate 2, 38 is a mesh tube for shielding, 39 is a plasma generation chamber, and 40 is a high-frequency electrode. , 41 is a high frequency power source. For example, an etchant gas 32 such as CF ₄ is introduced into the etching tank 33 , plasma radicals are generated by high frequency voltage, and these radicals are introduced into the substrate 2 in the reaction chamber 34 through the mesh of the mesh tube 38 . By this, a on the substrate 2 is
- Plasma etch the Si:H layer as described above. Note that this plasma etching is not limited to the illustrated apparatus, and may be performed using a known parallel plate type plasma etching apparatus. In addition to plasma etching, other dry etching such as reactive ion etching may also be applied.

FIG. 14 shows the etching rate of a-Si in the above dry etching process. Compared to the conventional method (sputtering in an atmosphere not containing CH ₄ ) shown by the broken line, the sputtering method based on the present invention As shown by the solid line, the etching speed is significantly improved.

This means that since the sputtering method of the present invention is carried out in an atmosphere containing CH ₄ , oxidation of the a-Si film is effectively prevented. Note that the dry etching conditions mentioned above are , CF ₄ + 10% O ₂ is used as the etchant gas (gas pressure is
40Pa=12Torr), and the high frequency was 100W.

In the above example, if the see-through property of the mask is not important, germanium is added to silicon (the optical density increases as the amount of addition increases) to further increase the optical density (the target is high-purity Si and high-purity silicon). It is also worthwhile to use germanium as a target (mixed target with high-purity Ge) or to use germanium instead of silicon (target is high-purity Ge).

Further, as the voltage applied during sputtering, a DC voltage can be applied instead of the above-mentioned high frequency voltage. In this case, for example, low-purity Si doped with phosphorus or the like (specific resistance is 10 ³ Ω-cm) is used as a target.
(below), it is possible to apply a DC high voltage of 1.5KW. However, as described above, it is of course possible to use low-purity Ge as a target instead of Si, or to use a mixed target of low-purity Si and low-purity Ge. Furthermore, in addition to the above-mentioned CH ₄ , C ₂ H ₆ and C ₃ H ₈ can be used as the introduced gas for supplying hydrogen.
Hydrocarbon gas or ammonia gas such as can be used, or both hydrocarbon gas and ammonia gas may be supplied at the same time.

[Brief explanation of drawings]

The drawings show an embodiment of the present invention, and FIG. 1 is a schematic sectional view of a sputtering device, and FIG.
Figures 3, 4, 5, 6 and 7 are cross-sectional views of each example of the exposure mask material, and Figures 8 and 9 are sectional views of each example of the exposure mask material or exposure mask. Fig. 10 is a plan view showing the external shape; Fig. 10 is a graph showing the relationship between the hydrogen atom content in a-Si and its optical density; Fig. 11
The figure is an infrared absorption spectrum diagram of a-Si:H, 1st
Figures 2A to 12E are cross-sectional views showing the method of manufacturing an exposure mask in order of steps, Figures 12F and 12G are cross-sectional views of each main process when processing a semiconductor using an exposure mask, and Figure 13 is a plasma FIG. 14, which is a schematic cross-sectional view of the etching apparatus, is a graph showing a comparison of etching speeds in dry etching. In addition, in the symbols shown in the drawings, 1... exposure mask material, 2... substrate, 3... a-Si:H
Layer, 5...Transparent conductive film, 6...SiO ₂ film, 7...
Substrate holder, 8...Power supply, 9...Ar+ _CH4 mixed gas, 10...Vacuum chamber, 11...Target,
17...mask, 21...exposure mask, 28...
Photoresist, 32...Etchiant gas, 4
0...High frequency electrode.

Claims (1)

[Claims] 1. An optically transparent mask substrate and a target made of silicon and/or germanium are placed in a sputtering device, and a voltage is applied to the target in an atmosphere containing 1 to 50% by volume of hydrocarbon gas and/or ammonia gas. The hydrocarbon gas and/or ammonia gas is decomposed at the same time as the target is sputtered, and the resulting ionized or activated hydrogen is reacted with the sputtered silicon and/or germanium to form hydrogen. An amorphous semiconductor layer made of silicon and/or germanium containing is formed on the mask base, and then the hydrogen-containing amorphous semiconductor layer is etched into a predetermined mask pattern, and the hydrogen-containing amorphous semiconductor layer is made only of the predetermined pattern. An exposure mask manufacturing method for creating an exposure mask having a light shielding layer.