Semiconductor manufacturing method and apparatus

Abstract

The present invention aims to provide processes and equipments for manufacturing semiconductors, according to which oxidation of wafer surfaces can be controlled by simple means and contaminants promoting oxidation and contaminants inviting a decreased yield of wafers can also be totally controlled. To achieve the object above, the present invention provides a process for manufacturing a semiconductor, characterized in that a substrate is treated while exposing the surface of the substrate with a negative ion-enriched gas; and an equipment for manufacturing a semiconductor comprising a gas channel through which a gas to be treated is passed; a negative ion-enriched gas generator consisting of a gas cleaner located at an upstream part of said gas channel and a negative ion generator located at a downstream part thereof: and means for supplying the resulting negative ion-enriched gas to the surface of each substrate.

Description

TECHNICAL FIELD

The present invention relates the manufacture of semiconductors, particularly processes and equipments for manufacturing semiconductors in leading-edge industries such as semiconductors and liquid crystals.

BACKGROUND ART

Air cleaning in the working environment Is very important for the manufacture of semiconductors, which normally takes place in an air-conditioned space called clean room. A conventional air cleaning technique in a semiconductor factory (clean room) is explained with reference to FIG. 8.

In FIG. 8, ambient air 1 first passes through a prefilter 2 where coarse particles are eliminated and then the temperature and humidity are conditioned in an air conditioner 3 and then dusts are collected by a medium-performance filter 4. Then, fine particles are collected by an HEPA filter 6 at the ceiling of a clean room 5. Such an air cleaning system maintains a microparticle concentration of class 10,000 in the-clean room. In FIG. 8, references 7-1 and 7-2 represent fans and arrows indicate air streams.

The air cleaning system in conventional clean rooms for the purpose of removing microparticles is designed as shown in FIG. 8 and effective for removing microparticles but not effective for removing gaseous hazardous components.

As improvements in the quality and precision of products increasingly advance in the recent semiconductor industry, not only microparticles (particulate substances) but also gaseous substances have come to participate in contamination of semiconductors.

However, conventional dust filters for clean rooms (e.g., HEPA filters, ULPA filters, etc.) as shown in FIG. 8 can collect only microparticles, but gaseous hazardous components from ambient air are not collected and but introduced into clean rooms. Such gaseous hazardous components include e.g. gases called hydrocarbons (HCs) derived from automobile emissions and outgassing (gas release) from polymer resin products widely used as consumer products; and basic (alkaline) gases such as NH₃ and amine.

Among them, hydrocarbons (HCs) must be completely eliminated because they cause pollution even at very low concentrations as gaseous hazardous components in normal air (indoor and outdoor air). Recently, outgassing from the materials of clean rooms or polymer resins of manufacturing equipments or appliances used has become a problem as a source of hydrocarbons (HCs).

These gaseous substances in question also include those generated during operations in clean rooms. That is, typical clean rooms contain gaseous substances not only introduced from ambient air (those having passed through microparticle-collecting filters for clean rooms to enter the clean rooms) but also generated in the clean rooms so that they contain higher concentrations of gaseous substances than ambient air, which increase the possibility of contaminating semiconductor substrates.

When microparticulate contaminants are deposited on the surfaces of semiconductor substrates, they cause breakage or short in circuits (patterns) on the substrate surfaces, resulting in failure. When gaseous hazardous components, especially hydrocarbons (HCs) are deposited on the surfaces of semiconductor substrates, they increase the contact angle to adversely affect e.g. substrate-resist affinity (compatibility). The lowered substrate-resist affinity adversely affects the film thickness of the resist or adhesion of the resist to the substrate. Hydrocarbons (HCs) also have the disadvantage that they deteriorate the pressure resistance of oxide films on the surfaces of semiconductor substrates (lowered reliability). The contact angle here means the contact angle of wetting with water and indicates the degree of contamination on substrate surfaces. That is, substrate surfaces stained with hydrophilic (oily) contaminants repel water and resist wetting. This increases the contact angle between the substrate surfaces and water drops. Thus, contamination is more serious at larger contact angles, while contamination is weaker at smaller contact angles. NH₃ causes production of ammonium salts or the like to invite haze (resolution failure) in semiconductor substrates.

For these reasons, the productivity (yield) of semiconductor products Is lowered by not only microparticles but also gaseous contaminants as described above.

Especially, the above gaseous substances as gaseous hazardous components are generated from the sources described above and more concentrated in clean rooms than ambient air so that they are deposited on substrates to contaminate their surfaces because air circulation is recently increased in clean rooms for saving energy.

To address these contamination problems, we have previously proposed various space cleaning methods using photoelectrons or photocatalysts.

For example, methods for removing microparticulate substances using photoelectrons are described in JP-B-HEI-3-5859, JP-B-HEI-6-74909, JP-B-HEI-8-211 and JP-B-HEI-7-121367. Methods for removing gaseous hazardous components using photocatalysts are described in Japanese Patent No. 2863419 and Japanese Patent No. 2991963. A method for removing microparticles and gaseous substances at the same time by combining photoelectrons and photocatalysts is described in Japanese Patent No. 2623290.

It is thought that current Al wirings will not suffice for patterns on wafer surfaces of semiconductor products with higher quality (microfabricated products) in future and will be replaced by Cu wirings. Thus, it will be necessary in future to use Cu wirings and interlayer dielectrics with low dielectric constant (low-k) to shorten the delay because the combination of current Al wirings and SiO₂ dielectrics requires a long wiring delay under compact wiring-and design rules for future ultra large scale Integrated circuits (ULSIs). However, Cu materials are more susceptible to oxidation than conventional Al or W. Thus, it will be also important in future to controls gaseous substances, especially those oxidizing wafer surfaces (wiring surfaces and interfaces) though it was sufficient in the past to pay attention to removal of only microparticles in semiconductor manufacturing environments.

Possible materials promoting oxidation of wafer surfaces include water (moisture) and organic matters (HCs) in clean room air, but it is difficult to control moisture present at about 45-50% (RH) in clean room air. This is because excessive reduction of moisture in the air adversely affects the health of operators working in the clean room.

Thus, it would be desirable to provide a novel method for controlling (inhibiting) oxidation of wafer surfaces by controlling materials in clean rooms including moisture and HCs.

In view of the circumstances described above, the present invention aims to provide processes and equipments for manufacturing semiconductors, according to which oxidation of wafer surfaces can be controlled by simple means and contaminants promoting oxidation and contaminants inviting a decreased yield of wafers can also be totally controlled.

DISCLOSURE OF THE INVENTION

To solve the problems described above, the present invention provides processes for manufacturing semiconductors, characterized in that semiconductor substrates are treated while using a negative ion-enriched gas obtained by a negative ion generator to inhibit oxidation of the surfaces of the substrate during semiconductor manufacturing steps.

Accordingly, an embodiment of the present invention relates to a process for manufacturing a semiconductor, characterized in that a substrate is treated while exposing the surface of the substrate with a negative ion-enriched gas. In the present invention, the negative ion-enriched gas is preferably prepared by passing a clean gas preliminarily freed of microparticles and/or chemical contaminants through a negative ion generator. The chemical contaminants here include one or more selected from the group consisting of ionic components. Inorganic matters and organic matters. In the present invention, the negative ion-enriched gas is preferably prepared by passing a gas having a microparticle concentration of class 100 or less, an ionic component concentration of 10 μg/m³ or less and an organic matter concentration of 10 μg/m³ or less through a negative ion generator.

Another embodiment of the present invention provides an equipment for manufacturing a semiconductor comprising a gas channel through which a gas to be treated is passed; a negative ion-enriched gas generator consisting of a gas cleaner located at an upstream part of said gas channel and a negative ion generator located at a downstream part thereof; and means for supplying the resulting negative ion-enriched gas to the surface of each substrate.

BRIEF EXPLANATION OF THE DRAWINGS

FIG. 1 is a flow chart showing specific processing steps on a semiconductor substrate in a semiconductor factory (clean room).

FIG. 2 is a schematic view showing an example of an apparatus for obtaining a negative ion-enriched gas used In the present invention.

FIG. 3 is a schematic view showing another example of an apparatus for obtaining a negative ion-enriched gas used in the present invention.

FIG. 4 is a schematic view showing another example of an apparatus for obtaining a negative ion-enriched gas used in the present invention.

FIG. 5 is a schematic view showing another example of an apparatus for obtaining a negative ion-enriched gas used in the present invention.

FIG. 6 is a schematic view showing another example of an apparatus for obtaining a negative ion-enriched gas used in the present invention.

FIG. 7 is a schematic view showing another example of an apparatus for obtaining a negative ion-enriched gas used in the present invention.

FIG. 8 is a schematic view showing an air cleaning system commonly used in conventional semiconductor factories (clean rooms).

THE MOST PREFERRED EMBODIMENTS OF THE INVENTION

The present invention was accomplished on the basis of the finding that semiconductor products suitable for the needs for high performance can be prepared by treating substrates while exposing them to a gas having a high level of cleanliness (i.e. very low microparticle concentration and chemical contaminant concentration) and rich in negative ions (i.e. negative ion-enriched gas) to inhibit oxidation of the substrates during various substrate processing steps at a given stage of semiconductor manufacturing processes in a clean room and thereby improve the yield of the semiconductor products.

The method for preparing a negative Ion-enriched gas used in the processes of the present invention is divided into a negative ion generating stage and a gas cleaning stage before generating negative ions. Each stage is explained below.

A. Negative Ion Generating Methods

Our studies showed that oxidation of substrate surfaces is inhibited by exposing substrates to a negative ion-enriched gas. The “negative ion” (also called as “minus ion”) here refers to a substance formed by attaching n electron to an electrophilic substance such as oxygen, e.g. small ions (hydrate ions) such as O₂⁻(H₂O)_n formed by attaching one or more water molecules to a negatively charged oxygen molecule. Ions having CO_x⁻ or NO_x⁻ as a core such as CO₃⁻(H₂O)_n and NO₂⁻(H₂O)_n can also be taken as “negative ions” according to the present invention. The concentration of negative ions sufficient for inhibiting oxidation of substrate surfaces varies with the purpose/type of the semiconductor, the desired performance for the semiconductor, coexisting materials and other factors, but normally 1,000 negative ions/mL or more, preferably 5,000 negative ions/mL or more, more preferably 10,000 negative ions/mL or more, still more preferably 50,000 negative ions/mL or more. Preferred negative ion concentrations can be determined by appropriate pretests depending on the purpose/type of the semiconductor, the desired performance, coexisting materials and other factors. The reason why oxidation of substrate surfaces is inhibited by negative ion-enriched gases has not been unknown well, but it is supposed that negative ions have a reducing effect on the surfaces of substrates.

The concentration of negative ions in a gas can be determined by measuring the electrical mobility of the ions under electric field. Such instruments for measuring the concentration of negative ions in a gas are commercially available e.g. under trade name Air Ion Counter Model 83-1001B from Dan Science.

Means for generating negative ions include methods using photoelectrons as proposed by us (JP-B-HEI-8-10616, Japanese Patent No. 3139591), discharge, water spray, irradiation, etc. Various methods for generating negative ions are explained below.

A-1: Negative Ion Generating Method Using Photoelectrons

The negative ion generating method using photoelectrons involves irradiating a photoelectron emitting member with UV rays from a UV source such as a UV lamp optionally in the presence of an electric field to generate photoelectrons, thereby forming negative ions. Here, an electric field can be formed by placing an electrode (positive electrode) on the side opposite to the photoelectron emitting member (negative electrode) to accelerate photoelectron emission from the photoelectron emitting member. Thus, negative ion generators using photoelectrons comprise a photoelectron emitting member and a UV source such as a UV lamp, and optionally an electrode for establishing an electric field. Inert gases such as N₂ can also be used as feed gases other than air.

The photoelectron emitting member is not limited so far as it can emit photoelectrons upon UV irradiation and preferably has a smaller photoelectric work function. From the viewpoint of the effect and economical efficiency, the member is preferably any one of Ba, Sr, Ca, Y, Gd, La, Ce, Nd, Th, Pr, Be, Zr, Fe, Ni, Zn, Cu, Ag, Pt, Cd, Pb, Al, C, Mg, Au, In, Bi, Nb, Si, Ti, Ta, U, B, Eu, Sn, P and W or compounds or alloys or mixtures thereof. These can be used alone or in combination of two or more. Suitable composite materials include physically composite materials such as amalgam.

Compounds of the elements above include oxides such as BaO, SrO, CaO, Y₂O₃, Gd₂O₃, Nd₂O₃, ThO₂, ZrO₂, Fe₂O₃, ZnO, CuO, Ag₂O, La₂O₃, PtO, PbO, Al₂O₃, MgO, In₂O₃, BiO, NbO and BeO; borides such as YB₆, GdB₆, LaB₅, CeB₆, EuB₆, PrB₆ and ZrB₂; and carbides such as UC, ZrC, TaC, TiC, NbC and WC. Suitable alloys of the elements above include brass, bronze, phosphor bronze, Ag—Mg alloys (Mg=2-20 wt %), Cu—Be alloys (Be=1-10 wt %) and Ba—Al alloys, among which Ag—Mg alloys, Cu—Be alloys and Ba—Al alloys can be preferably used. Oxides of the elements above can be obtained by heating only the surface of a metal in the air or chemically oxidizing it. Alternatively, a metal or alloy material of any one of the elements above can be heated before use to form an oxide layer with prolonged stability on its surface and this oxide layer can be used as a photoelectron emitting member. For example, an Mg—Ag alloy can be treated at a temperature of 300-400° C. in water vapor to form an oxide film on its surface, and such an oxide film can be used as a photoelectron emitting member over a long period because it has prolonged stability.

Photoelectron emitting materials can be used in combination with other materials. As an example, a UV transparent material such as glass can be combined with a material capable of emitting photoelectrons (JP-B-HEI-7-93098, JP-A-HEI-4-243540).

The photoelectron emitting member can also be incorporated into a photocatalyst such as TiO₂ (JP-A-HEI-9-294919). This form is preferred for some types of equipments or desired performances because the photocatalyst can eliminate any substance adversely affecting the photoelectron emitting member to stabilize It over a long period while coexisting gaseous contaminants can also be eliminated.

The shape of the photoelectron emitting member can be appropriately selected from plates, pleated sheets, grids and others depending on the permeation mode of the negative ion generating gas or the type of the equipments. Among them, grid-type photoelectron emitting members are preferred for some types of equipments applied because negative ions can be generated without forming an electric field when a gas is passed from the bottom to the top of the photoelectron emitting member. The photoelectron emitting member can be preferably incorporated into the irradiation source described below for some types of equipments applied because the size of the photoelectron emitter is reduced. This can be accomplished by e.g. affixing a photoelectron emitting member to the surface of a UV lamp.

The irradiation source for emitting photoelectrons from the photoelectron emitting member is not limited so far as it irradiates the photoelectron emitting member to emit photoelectrons, but normally UV rays or radiation are preferred (Japanese Patent No. 2623290, JP-B-HEI-6-74910), and UV rays are especially preferred because they can be used simply and safely.

The type of the UV source that can be used is not limited so far as it irradiates the photoelectron emitting member to emit photoelectrons, preferably a mercury lamp such as a germicidal lamp in terms of size reduction.

The optional electric field under which photoelectrons are emitted is preferably 0.1 V/cm to 1 kV/cm, and can be appropriately determined by pretests depending on the configuration and structure of the equipments. The electrode member used for forming the electric field is not limited so far as it produces no impurities and allows the photoelectron emitting member to effectively emit photoelectrons, and it may be in the form of a line, bar, grid or plate made from SUS, Cn—Zn or W. These electrode members are placed to create an electric field near the photoelectron emitting member so that photoelectrons can be emitted under the electric field.

A-2: Negative Ion Generating Method Using Discharge

The discharge-based method involves emitting electrons by a discharge in a gas to generate negative ions using an apparatus comprising a discharge electrode and a counter electrode.

Suitable discharges for generating negative Ions include those well-known in the art such as corona, glow, arc, spark, surface creepage, pulse, high-frequency, laser, trigger and plasma discharges. Among them, surface creepage and pulse discharges are preferred for some purposes in terms of the size reduction of the equipments because of high concentrations of negative ions generated. Corona discharge is preferred in terms of simplicity, operability and effect.

A-3: Negative Ion Generating Method Using Water Spray

The negative ion generating method using water spray involves generating negative ions via Lenard effect by atomizing water, e.g. negative ions can be generated by spray charging of droplets when water is atomized in the air. The mechanism by which negative ions are generated via Lenard effect can be supposed as follows. Water molecules are polar molecules having electrically positive and negative charges and their positive sides are outwardly oriented on the water surface (oriented dipoles). These oriented dipoles attract many negative ions to form an electric bilayer, which produces a negatively charged air when an energy such as a high pressure is applied. That is, when water is atomized under high pressure, negatively ionized air is produced because the water surface is positive and the adjacent air becomes negative. Then, this air is vaporized to remove coexisting excess water if desired, whereby a negative ion-enriched air is formed.

A-4: Negative Ion Generating Method Using Irradiation

The irradiation-based method involves exposing air to a radiation to generate negative ions. Radiations that can be used in this method are not limited so far as they generate ions from radiation sources, such as X-rays, α-rays, γ-rays and β-rays. Among them, X-rays, α-rays and γ-rays are preferred in terms of operability or the like, with X-rays being especially preferred. X-ray irradiation uses ions obtained by exposing gas molecules to X-rays, and normally gas molecules are bombarded with X-rays obtained by irradiating a metal target with accelerated electron beams to ionize air molecules.

The mechanism of the ionization by soft X-rays that are especially preferred for use in the processes of the present invention is explained as follows. Air molecules absorb irradiating X-rays or photons (wavelength 0.2-0.3 nm) to become ionized or photoionized. Ionized electrons collide with neutral electrons and molecules to ionize them because of their high kinetic energy. These ionizations continuously occur by electron avalanche to generate large amounts of ions. The ions generated here include both positive and negative ions, of which only positive ions are removed by suitable well-known means such as an electrode plate to give a gas enriched in only negative ions.

The photon energy of soft X-rays preferred for use in the present method is several KeVs to 10 KeV or less. i.e. about {fraction (1/10)} times or less of the energy of hard X-rays used for radiography, but a shield is required if irradiation takes place in a region which operators may enter. The shield can be e.g. a metal plate having a thickness of about 1 mm or a plastic (e.g. vinyl chloride) plate having a thickness of about 2-3 mm.

A gas enriched in only negative ions can also be obtained by the same procedure as described above using α-rays or γ-rays, which also produce large amounts of ions because of their high kinetic energy. Suitable γ-ray sources are radioactive substances such as Cobalt 60 and Cesium 137.

When negative ions are generated by the above method using discharge or irradiation, especially X-ray irradiation, ozone (O₃) may also be generated. This ozone must be removed from the gas because it promotes oxidation of substrate surfaces. For the purpose of the present invention, the ozone concentration in a gas should desirably be 0.1 ppm or less, preferably 0.01 ppm or less, more preferably 1 ppb or less. Thus, it is preferable e.g. to use an inert gas free from the ozone source O₂ such as N₂ or to subject the product negative ion-enriched gas to an ozone removal treatment in the discharge- or irradiation-based method. Humidification of the gas after generation of negative ions may be preferred for some purposes or desired specifications because humidification promotes decomposition of ozone.

A means for removing generated ozone from the negative ion-enriched gas is to treat the gas with a well-known O₃-treating agent after negative ions have been generated. Well-known O₃-treating agents that can be used for this purpose include e.g. Mn-based catalysts. The materials and forms of O₃-treating agents that can be used in the present invention should preferably consume little negative ions generated and coexisting in the gas, e.g. manganese dioxide-based honeycomb or mesh catalysts such as MnO₂/TiO₂—C and MnO₂/ZrO—C.

Characteristics of each of the negative ion generating methods described above are shown in Table 1 below, wherein various methods are relatively evaluated as follows: ∘ means good and Δ means slightly poor.

TABLE 1
Characteristics of various negative ion generating methods
			Amount/
	Large-	Small-	concen-	Antioxidant
	scale	scale	tration of	effect/
	appli-	appli-	negative ions	purity of
	cability	cability	generated	negative ions	Safety
Photoelectron	Δ	Δ-◯	Δ	◯	◯
Discharge	◯	◯	◯	Δ-◯	◯
Water spray	◯	Δ	◯	Δ-◯	◯
Irradiation	◯	◯	◯	Δ-◯	Δ

As shown from the table above, the photoelectron-based method is poor in large-scale applicability but good in antioxidant effect because ozone-free clean negative ions are generated, while the irradiation-based method is good in large-scale and small-scale applicabilities and the amount of negative ions generated but insufficient in safety. In the present invention, desirable negative ion generating methods can be appropriately selected depending on the purpose and desired performance or other factors, taking into account advantages and disadvantages of various negative ion generating methods described above.

It was found that when negative ions generated by the discharge-based method were then humidified or they are generated by the water spray-based method, the resulting negative ions had larger particle diameters than obtained by the others negative ion generating methods described above. For example, negative ions formed by the photoelectron-based method or the discharge-based method without humidification have a particle diameter of about 1 nm, but negative ions formed by the discharge-based method followed by humidification or generated by the water spray-based method have a particle diameter of about 3-5 nm. Our studies revealed that negative ions of larger particle diameters are more effective for inhibiting oxidation of substrates according to the present invention. The detailed reason for this is unknown, but supposed as follows. Negative ions consist of a negatively charged core molecule (e.g. oxygen molecule) to which water molecules are adsorbed (attached), and a somewhat large number of the water molecules adsorbed are more effective for inhibiting oxidation of substrates. Thus, the negative ion generating method using discharge followed by humidification or using water spray is preferably used for some purposes, scales of equipments and desired specifications.

When a substrate is treated in the presence of thus prepared negative ion-enriched gas in the present invention, it is more effective for inhibiting oxidation of the substrate if a positive electrode is placed in the direction of the site where the negative ion-enriched gas is applied, i.e. the substrate-processing site to attract negative ions in an electric field. This is because negative ions slowly move so that they are much consumed with some shapes or structures of equipments.

Negative ions generated to form a negative ion-enriched gas in the present invention as described above may be consumed by contaminants if they are contained in the gas. If micropartioulate substances exist in the gas for example, the charges of generated negative ions are transferred to these microparticles to form charged particles so that the negative ions are consumed. This lowers the concentration of negative ions to be effectively used for inhibiting oxidation of substrates. It is known that semiconductor substrates are significantly contaminated by the presence of microparticles and chemical contaminants, which results in a significant decrease in yield. When negative ions are generated in a gas containing acidic gases such as Cl₂, negative ions having Cl₂ as a core are also formed and such negative ions do not suit the purpose of the present invention, i.e. “inhibiting oxidation of substrates” because they are thought to be oxidative. Therefore, it is desirable to sufficiently eliminate such chemical contaminants before entering into the negative ion generating stage.

From this point of view, a gas preliminarily freed of microparticles and chemical contaminants such as ionic components and inorganic and organic matters are preferably passed through the negative ton generator described above to generate negative ions in the present invention. Specifically, the gas supplied to the negative ion generator in the present invention preferably has a microparticle concentration of class (the number of particles having a standard particle diameter of 0.1 μm in 1 ft³ of a gas) 100 or less, preferably 10 or less, more preferably 1 or less; an ionic component concentration of 10 μg/m³ or less, preferably 5 μg/m³ or less, more preferably 2 μg/m³ or less; and an organic matter concentration of 10 μg/m³ or less, preferably 5 μg/m³ or less, more preferably 2 μg/m³ or less. The “ionic components” here refer to acidic gases such as NO_x, SO_x, HCl, HF, Cl₂, F₂, HBr and Br₂; and basic gases such as ammonia and amine.

As described above, the gas supplied to the negative ion generator should preferably have preliminarily undergone contaminant removal treatments. The contaminant removal treatments that can be performed before generating negative ions in the present invention are mainly classified into removal of microparticles and removal of chemical contaminants specifically explained below.

B. Removal of Microparticles

The gas to be treated to generate negative ions in the present invention should preferably be preliminarily freed of microparticles to class 100 or less, preferably 10 or less, more preferably 1 or less, and any microparticle removing means known in the art can be used so far as this cleanliness can be achieved. Microparticle removing means that can be used in the present invention include e.g. the use of a filter or photoelectrons as proposed by us elsewhere.

B-1: Microparticle Removing Means Using Filters

Filters that can be used as microparticle removing means in the present invention include those well-known in the art such as ULPA filters, HEPA filters, medium performance filters and electrostatic filters, which can be used alone or combined.

B-2: Microparticle Removing Means Using Photoelectrons

This means removes microparticles using photoelectrons proposed by us in JP-B-HEI-6-74909, JP-B-HEI-7-121369, JP-B-HEI-8-211, JP-B-HEI-8-22393, Japanese Patent No. 2623290, etc. This method involves generating negative ions in the same manner as described above for the negative ion generating method using photoelectrons, charging microparticles with the negative ions generated and collecting/removing the charged microparticles using an electrode or the like. Thus, the photoelectron-based microparticle removing means that can be used in the present invention comprises a photoelectron emitting member, a UV source, an electrode member and a charged microparticle collecting member. The photoelectron emitting member, UV source and electrode member can be those described above for the negative ton generating method using photoelectrons.

Suitable charged microparticle collecting members typically include various electrode members such as dust collecting plates and dust collecting electrodes or electrostatic filters used in conventional particle charging devices, but wool structures such as steel wool electrodes and tungsten wool electrodes can also be effectively used. Electret members can also be suitably used.

Preferred combinations of the photoelectron emitting member, electrode member and charged microparticle collecting member can be appropriately selected depending on the shape and structure of the space to be cleaned, desired performance and economical efficiency. For example, the locations and shapes of the photoelectron emitting member and electrode can be appropriately determined taking into account the shape of the space, effect, economical efficiency and other factors in such a manner that they can surround a UV source to combine the UV source, photoelectron emitting member, electrode member and charged microparticle collecting member into a unit, which can effectively use UV rays emitted from the UV source and efficiently emit photoelectrons and charge/collect microparticle by the photoelectrons. When a rod-like or cylindrical UV lamp is used as a UV source, for example, UV rays are radially emitted around the circumference of the lamp and the amount of photoelectrons emitted increases by irradiating the photoelectron emitting member with the circumferential radial UV rays as much as possible. Thus, it is preferred that the photoelectron emitting member is circumferentially located opposite the UV lamp and the photoelectron emitting electrode is located on the opposed face.

C. Removal of Ionic Components and Chemical Contaminants

As described above, the gas to be treated to generate negative ions in the present invention should preferably be preliminarily freed of ionic components such as acidic gases including Cl₂, NO_x and SO_x and basic gases including ammonia; and chemical contaminants such as inorganic and organic matters, specifically to an ionic component concentration of 10 μg/m³ or less, preferably 5 μg/m³ or less, more preferably 2 μg/m³ or less; and an organic matter concentration of 10 μg/m³ or less, preferably 5 μg/m³ or less, more preferably 2 μg/m² or less. Suitable means for removing such contaminants can be any methods well-known in the art, e.g. using adsorbents or photocatalysts, as specifically explained below.

C-1: Means for Removing Chemical Contaminants Using Adsorbents

In the present invention, the means for removing chemical contaminants using adsorbents consists in collecting/removing acidic gases such as NO_x, SO_x, HCl, HF, Cl₂, F₂, HBr and Br₂; and basic gases such as ammonia and amine in a gas to be treated during generation of a negative ion-enriched gas, and any adsorbents can be used so far as they efficiently adsorb various acidic/basic gases mentioned above to low concentrations. Such known adsorbents include silica gel, zeolite, alumina, activated carbon and ion exchange fibers, among which activated carbon and ion exchange fibers are effective and therefore can be preferably used in the present invention. Especially, ion exchange fibers can be preferably used for some purposes because they can collect contaminants to low concentrations via chemical reactions and high cleanliness can be achieved. Activated carbon can be appropriately used as those impregnated with an acid or alkali depending on the component to be collected.

The adsorbents described above can be used in any shape, but generally fibrous and honeycomb shapes are preferred because of small pressure loss.

Ion exchange fibers comprise a cation exchanger or an anion exchanger or an ion exchanger having both cation and anion exchange groups supported on the surface of a carrier such as a natural or synthetic fiber or a mixture thereof, and the ion exchanger may be directly supported on a fibrous carrier or the ion exchanger may be supported on a woven or knitted or flocked base formed of fibers. Ion exchange fibers that can be used in the present invention are preferably those prepared by graft polymerization, especially radiation-induced graft polymerization. This is because radiation-induced graft polymerization allows ion exchange fibers to be formed using various types and shapes of materials.

The natural fibers can be wool, silk and the like, and the synthetic fibers can be those derived from hydrocarbon polymers or fluorine-containing polymers or polyvinyl alcohol, polyamide, polyester, polyacrylonitrile, cellulose or cellulose acetate. The hydrocarbon polymers include aliphatic polymers such as polyethylene, polypropylene, polyisobutylene and polybutene; aromatic polymers such as polystyrene and poly α-methylstyrene; alicyclic polymers such as polyvinyl cyclohexane; or copolymers thereof. The fluorine-containing polymers include polyethylene tetrafluoride, polyvinylidene fluoride, ethylene-ethylene tetrafluoride copolymers, ethylene tetrafluoride-propylene hexafluoride copolymers, vinylidene fluoride-propylene hexafluoride copolymers, etc. Any of these materials are preferred as carriers for ion exchanges so far as they have a large area in contact with gas stream, a shape with low resistance for easy grafting and a high mechanical strength with less waste fibers dropping and produced, and are less susceptible to heat, and they can be appropriately selected by those skilled in the art taking into account the intended use, economical efficiency, effect and other factors, but normally polyethylene or composite materials of polyethylene and polypropylene are preferably used.

Ion exchangers that can be introduced into these materials are not specifically limited, but include various cation exchangers or anion exchangers. For example, suitable ion exchangers contain cation exchange groups such as carboxyl, sulfonate, phosphate and phenolic hydroxyl; or anion exchange groups such as primary to tertiary amino groups and quaternary ammonium group; or both of the cation and anion exchange groups mentioned above. Specifically, fibrous ion exchangers having a cation exchange group or an anion exchange group can be obtained by graft-polymerizing e.g. acrylic acid, methacrylic acid, vinyl benzene sulfonic acid, a styrene compound such as styrene, halomethylstyrene, acyloxystyrene, hydroxystyrene or aminostyrene; vinyl pyridine, 2-methyl-5-vinyl pyridine, 2-methyl-5-vinylimidazole or acrylonitrile onto the fibrous base described above optionally followed by reaction with sulfuric acid, chlorosulfonic acid or sulfonic acid. Alternatively, the above monomers may be graft-polymerized onto the fiber in the presence of a monomer having two or more double bonds such as divinylbenzene, trivinylbenzene, butadiene, ethylene glycol, divinyl ether or ethylene glycol dimethacrylate.

The diameter of the ion exchange fiber preferred for use as a chemical contaminant adsorbent in the present invention is 1-1000 μm, preferably 5-200 μm and can be appropriately determined depending on the type of the fiber, purpose, etc. The type and amount of the cation exchange group and anion exchange group introduced into the ion exchange fiber can be determined depending on the type and concentration of the component to be removed in the gas to be treated. For example, the type and amount of the ion exchange group can be determined on the basis of preliminary measurement/evaluation of the component to be removed in a gas. For example, fibers having a cation exchange group or an anion exchange group or both cation and anion exchange groups can be used depending on whether the gas to be removed is basic or acidic or a mixture of both.

The gas is effectively supplied to the ion exchange fiber at right angle to the ion exchange fiber in the form of a filter. The flow rate of the gas supplied to the ion exchange fiber can be appropriately determined by pretests, but the gas can be normally supplied at about 1,000-100,000 (h⁻¹) expressed as SV (spatial velocity) in view of the high removal rate of ion exchange fibers for gaseous components. Ion exchange fibers prepared by radiation-induced graft polymerization as previously proposed by us can be preferably used as appropriate because they are especially effective (JP-B-HEI-5-9123, JP-B-HEI-5-67235, JP-B-HEI-5-43422, JP-B-HEI-6-24626, etc.). When ion exchange groups are introduced into fiber materials (carriers) by radiation-induced graft polymerization, the ion exchange capacity increases because the carriers are homogeneously irradiated to depth so that ion exchangers are firmly fixed at high density over a large area. As a result, even low concentrations of gaseous components can be rapidly and efficiently removed. The preparation of ion exchange fibers by radiation-induced graft polymerization also has the following advantages. The preparation can be performed with a material having a shape close to that of the target product at room temperature in a gas phase with high grafting degree to give an adsorption filter containing low levels of impurities. Thus, ion exchange fibers prepared by radiation-induced graft polymerization rapidly adsorb much gaseous components because ion exchangers having the function of adsorbing gaseous components are homogeneously fixed in large quantity at high density. Moreover, filter materials with small pressure loss can be formed.

For some specifications desired, adsorbents formed from glass and fluorine resins such as a glass fiber filter having a fluorine resin as a binder can also be preferably used as chemical contaminant removing adsorbents in the present invention. Such filters are effective for removing gaseous organic matters and particulate materials at the same time (Japanese Patent No. 2582806).

C-2: Means for Removing Chemical Contaminants Using Photocatalysts

Means for removing chemical contaminants in a gas using photocatalysts are preferred when gaseous components to be removed contain organic matters (HCs) such as phthalate esters. High molecular weight HCs including phthalate esters such as DOP must be removed because they cause lowered productivity and yield such as deteriorated pressure resistance of oxide films and lowered reliability once they are adsorbed to substrate surfaces.

Any photocatalysts can be used so far as they can be exited by irradiation to decompose HCs into inert forms for substrates such as CO₂ and H₂O. Normally, semiconductor materials are preferably used as photocatalysts in the present invention because they are effective and readily available with good workability. In view of the effect and economical efficiency, any one of Se, Ge, Si, Ti, Zn, Cu, Al, Sn, Ga, In, P, As, Sb, C, Cd, S, Te, Ni, Fe, Co, Ag, Mo, Sr, W, Cr, Ba and Pb or compounds or alloys or oxides thereof are preferred and can be used alone or in combination of two or more.

Examples are elements such as Si, Ge and Se; compounds such as AlP, AlAg, GaP, AlSb, GaAs, InP, GaSb, InAs, InSb, CdS, CdSe, ZnS, MoS₂, WTe₂, Cr₂Te₃, MoTe, Cu₂S, and WS₂; and oxides such as TiO₂, Bi₂O₃, CuO, Cu₂O, ZnO, MoO₃, InO₃, Ag₂O, PbO, SrTiO₃, BaTio₃, Co₃O₄, Fe₂O₃ and NiO. For some applications, a metal member can be baked to form a photocatalyst on the surface of the metal member. For example, a photocatalyst can be prepared by baking a Ti member at 1000° C. to form TiO₂ on its surface (JP-A-HEI-11-90236). The above photocatalyst materials are preferably used with additives such as Pt, Ag, Pd, RuO₂ and Co₃O₄ to promote the HC-decomposing effect of the photocatalysts. These additives can be used alone or in combination. The dose is normally 0.01-10% by weight relative to the photocatalyst and optimal concentrations can be appropriately selected by preliminary experiments depending on the type of the additive and desired performance or the like. Additives can be added by well-known techniques such as immersion, photoreduction, sputter deposition and kneading.

The photocatalysts can be used by immobilizing them in a space where the gas to be treated circulates or on the wall face of a channel through which the gas flows or suspending them in a space where the gas circulates. The photocatalysts can be immobilized in a unit by coating the photocatalysts on an appropriate material in the form of a plate, flocculent, line, fiber, mesh, honeycomb, membrane, sheet or fabric or wrapping or inserting them in or between these materials. For example, any one of photocatalyst materials can be immobilized on a ceramic, metal, fluorine resin or glass material by appropriately using a well-known fixing means such as sol-gel process, sintering, vapor deposition or sputtering. Preferred materials on which photocatalysts are immobilized are normally in the form of a fiber, mesh or honeycomb because of the small pressure loss. For example, TiO₂ fixed on a glass fiber by sol-gel process or a photocatalyst fixed on the surface of a transparent linear article (JP-A-HEI-7-256089) can be used as a means for removing chemical contaminants in a gas in the present invention.

In the present invention, the light source for irradiating photocatalysts can be any one of well-known light sources that can irradiate the photocatalysts to produce a photocatalytic effect. Thus, photocatalytic decomposition of HCs can be accomplished by bringing a gas to be treated with a photocatalyst while irradiating the photocatalyst with light beams having an absorbance wavelength range determined by the type of the photocatalyst.

Main absorbance wavelength ranges of various photocatalysts are as follows. Si:<1,100(nm); Ge:<1,825(nm): Se:<590(nm); AlAs:<517(nm); AlSb:<827(nm); GaAs:<886(nm); InP:<992(nm); InSb:<6,888(nm); InAs:<3,757(nm); CdS:<520(nm); CdSe:<730(nm); MoS₂:<585(nm); ZnS:<335(nm); TiO₂:<415(nm); Zno:<400(nm); Cu₂O:<625(nm); PbO:<540(nm); Bi₂O₃:<390(nm).

The light source used for irradiating the photocatalysts can be appropriately selected from any well-known light sources having a wavelength in the absorbance range of the photocatalysts such as sunlight and UV lamps. Suitable UV sources normally include mercury lamps, hydrogen discharge tubes, xenon discharge tubes and Lyman discharge tubes and they can be appropriately used. Specific forms of suitable light sources include germicidal lamps, black light lamps, fluorescent chemical lamps, UVB lamps and xenon lamps. Among them, germicidal lamps (main wavelength 254 nm) can be especially preferably used for the following reasons. They can increase the effective irradiation dose to photocatalysts to increase photocatalytic effect; they are free from ozone; they can be easily installed; they are inexpensive and easy to maintain and manage; and they have high performance. The irradiation dose to photocatalysts is generally 0.05-50-mW/cm², preferably 0.1-10-mW/cm².

HCs can also be collected by using adsorbents such as activated carbon, but the use of adsorbents has problems with adsorption capacity and breakthrough. That is, the adsorption capacity becomes rapidly saturated at high concentrations of the gas generated, which requires additional operations such as replacement, while breakthrough invites the problem of secondary pollution due to the spill of collected matters. In contrast, photocatalysts can be very preferably used as means for removing chemical contaminants such as HCs in the present invention because they are free from accumulation of HCs and can stably decompose HCs for a long period.

Next, the water content that is important for generating negative ions in the present invention is explained.

Water in a gas plays an important role in the mechanism by which negative ions are generated in the present invention. Thus, negative ions that are effective for inhibiting oxidation of substrates can be efficiently obtained by controlling the water content in the gas.

Especially, the mechanism by which negative ions are generated using the photoelectron-based method and the discharge-based method is thought to be explained as follows. Electrons generated may form negative ion clusters such as O₂⁻(H₂O)_n, O⁻(H₂O)_n and OH⁻(H₂O)_n by the electron attachment or clustering to molecules having high electron affinity such as water molecules and oxygen molecules. These reactions are shown below. $\begin{array}{c}{O}_2+e^{-}->O_2^{-}\\ O_2^{-}+H_{2}O->O_2^{-}\left(H_{2}O\right)\\ ⋮\\ {O_2^{-}\left(H_{2}O\right)}_{n-1}+H_{2}O->{O_2^{-}\left(H_{2}O\right)}_n\end{array}$

As shown from the formulae above, water charged with electrons becomes negative ions. Thus, it is not necessary to control the water content when electrons are supplied to correspond to the water content in the gas. If the concentration of generated electrons is low, however, water having the so-called oxidative effect harmful to wafers is liberated to adversely affect the wafers. The presence of such harmful water can be known by testing the oxidation state of the wafer surface exposed to the atmosphere, e.g. the state of formation of natural oxide films. If the presence of harmful water is detected in this manner, dehumidification to an appropriate content is preferred. For the purpose of removing harmful water, dehumidification is preferred to a relative humidity in a gas of about 45±5%, preferably 30% or less, more preferably 20% or less. Dehumidification of the gas can be accomplished by appropriately using a well-known method preferably before negative ions are generated normally in the present invention.

Dehumidifying means that can be used in the present invention include well-known methods such as cooling, adsorption, absorption, compression and membrane separation, and one or more of the above means can be used in combination after appropriate pretests depending on the field to which the present invention is to be applied and the scale, configuration and operation conditions of the equipments, e.g. whether it is applied under atmospheric or pressurized condition. The dehumidifying means preferably keeps stable dehumidification performance over a long period, normally several to six months or longer, and especially the means based on cooling, adsorption or membrane separation are simple and effective. Preferred dehumidifying means based on cooling are electronic dehumidification and cooling coils because of the compact structure and effectiveness and preferred dehumidifying means based on adsorption are systems in which the dehumidifier per se is regenerated for continuous long dehumidification (fixed or rotary system) because of the simplicity and effectiveness. Dehumidifying materials that can be used in the adsorption-based dehumidifying means include silica gel, zeolite, activated carbon, activated alumina, magnesium perchlorate, calcium chloride, alumina-pillared clay porous materials, bound activated carbon and porous aluminum phosphate. The alumina pillared clay porous materials here refer to materials obtained by exchanging exchangeable cations between layers of a layered silicate with multinuclear metal hydroxide ions including aluminum and dehydrating the ion exchanged silicate by heating. The bound activated carbon is obtained by carbonizing polyvinyl formal and activating it at a temperature of 850° C. or less. Porous aluminum phosphate is also called molecular sieve and obtained by reacting an alumina hydrate such as aluminum hydroxide, boehmite or pseudoboehmite with phosphoric acid using a heat-dissociable template such as an organic base, e.g. tripropylamine.

In the negative ton generating method using water spray, it is necessary to remove harmful water using well-known dehumidifying means such as eliminators or heating coils because the so-called water mist is generated by water spray (atomization).

Dehumidification described above is applied when the amount of electrons in a gas is 0.1 PA or less expressed as the current value measured in a space, but reversely the gas is preferably humidified to generate negative ions to which more water molecules are attached when this value is 0.1 PA or more. Humidification can be accomplished by means well-known In the art, e.g. by heating water with a heater or vaporizing or ultrasonically spraying or supplying water through a membrane. When water is added to the gas by humidification, excess water not having participated in generating negative ions is preferably removed by using a dehumidifying means such as an eliminator or heating coil.

As described above, effective negative ion-enriched gases for inhibiting oxidation at proper water contents can be formed by appropriately using humidifying means and dehumidifying means. Thus, a proper amount of water can be effectively used as a negative ion source by properly controlling water contents.

Inert gases such as N₂ and Ar can be used as gases for generating negative ions to form a negative ion-enriched gas and such inert gases are preferably hydrated by the humidifying means described above because they are normally dry and cannot efficiently generate negative ions as such. The amount of water to be added can be determined after appropriate pretests depending on the negative ion generating method, desired specification and other factors.

Next, several specific embodiments of semiconductor manufacturing equipments according to the present invention are explained with reference to the attached drawings.

FIG. 1 shows specific processing steps on a semiconductor substrate in a semiconductor factory (clean room, class 10,000). The present invention can be applied to each specific processing step shown in FIG. 1. That is, a semiconductor manufacturing equipment of the present invention comprises a negative ion-enriched gas generator as explained below and a means for supplying the negative ion-enriched gas prepared by said generator to the surface of a substrate in a semiconductor processing equipment at various specific processing steps shown in FIG. 1. For example, a substrate can be cleaned/dried while inhibiting oxidation of the substrate by combining an apparatus comprising a “negative ion-enriched gas generator” and a “means for supplying the resulting negative ion-enriched gas to the surface of a substrate” explained below with an “apparatus for spraying a gas to a substrate to wash and dry the surface of the substrate with air”, which can be used during the step “clean and dry the substrate” shown in FIG. 1.

FIG. 2 shows a schematic view of a negative ion-enriched gas generator according to an embodiment of the present invention comprising a clean gas generator consisting of an adsorbent-based chemical contaminant removing means and a filter-based microparticle removing means; and a discharge-based negative ion generator. Such an apparatus generates a negative ion-enriched air at class 10 or less free from chemical contaminants (100,000 negative ions/mL or more). Semiconductor substrates can be prevented from contamination by performing each specific processing step shown in FIG. 1 while exposing the surfaces of the semiconductor substrate to the negative ion-enriched air generated in the present example.

Negative ion-enriched air generator A shown in FIG. 2 comprises a fan 20 for supplying a clean room air, an adsorbent (ion exchange fiber and activated carbon) 21 for removing chemical contaminants in the clean room air, a dust filter (ULPA filter) 22 for removing microparticles in the clean room air (class 10,000) and microparticles. generated from fan 20, a discharging member (corona discharge) 23 for generating negative ions, and an O₃ decomposing/removing member 24 for removing O₃ generated from discharging member 23. In the figure, 25-1 indicates the flow of the air introduced by fan 20 into negative ion-enriched gas generator A, 25-2 indicates the flow of the negative ion-enriched gas generated by negative ion-enriched gas generator A, and 26 represents negative ions. In discharging member 23 In FIG. 2, 23-1 represents a needle-like discharge electrode and 23-2 represents a counter electrode.

According to negative ion-enriched gas generator A shown in FIG. 2. a clean room air (having a microparticle concentration of class 10,000 and a negative ion concentration of 100 ions/mL or less) introduced by the fan first passes through adsorbent 21 for collecting chemical contaminants and filter 22 for collecting microparticles, whereby it is cleaned to a microparticle concentration of class 10 or less and both ionic component concentration and organic matter concentration of 2 μg/m³ or less. Then, negative ions are generated by discharge-based negative ion generator 23, and generated ozone is removed by ozone decomposing/removing member 24, whereby a clean negative ion-enriched gas is provided having a concentration of 100,000 negative ions/mL or more, a microparticle concentration of class 10 or less, both ionic component concentration and organic matter concentration of 2 μg/m³ or less, and an ozone concentration of 0.01 ppm or less. This negative ion-enriched gas can be used as a substrate-exposing gas in various processing steps shown in FIG. 1 to achieve a semiconductor manufacturing process in which substrates are inhibited from oxidation and both microparticle contamination and chemical contamination are prevented.

FIG. 3 shows a schematic view of a negative ion-enriched gas generator according to another embodiment of the present invention, comprising a clean gas generator using adsorbent/microparticle collecting filter and a photoelectron-based negative ion generator. In FIG. 3, similar elements to those shown in FIG. 2 are designated with the same references and not explained.

Negative ion-enriched air generator A shown in FIG. 3 comprises a cleaner consisting of a chemical contaminant-collecting adsorbent 21 and a microparticle-collecting filter 22; and a negative ion generator consisting of a photoelectron emitting member 27, a UV lamp 28 and an electric field-forming electrode 29. According to negative ion-enriched gas generator A shown in FIG. 3, a clean room air introduced by the fan is cleaned through chemical contaminant-collecting adsorbent 21 and microparticle-collecting filter 22 to a microparticle concentration of class 10 or less and both ionic component concentration and organic matter concentration of 2 μg/m³ or less, then introduced into the negative ion generator where photoelectron emitting member 27 is irradiated with UV rays to emit photoelectrons. Thus, a clean negative ion-enriched gas is provided having a concentration of 10,000 negative ions/mL or more, a microparticle concentration of class 10 or less, and both ionic component concentration and organic matter concentration of 2 μg/m³ or less. According to the photoelectron-based method, the ozone removing member as shown in FIG. 2 is normally unnecessary because ozone is not generated. The negative ion-enriched gas generated by negative ion-enriched gas generator A shown in FIG. 3 can be used as a substrate-exposing gas in various processing steps shown in FIG. 1 to achieve a semiconductor manufacturing process In which substrates are inhibited from oxidation and both microparticle contamination and chemical contamination are prevented.

FIG. 4 shows a schematic view of a negative ion-enriched gas generator according to another embodiment of the present invention, comprising a clean gas generator consisting of an adsorbent-based chemical contaminant-removing means and a microparticle-removing means; and a photoelectron-based negative ion generator as shown in FIG. 3. In FIG. 4, similar elements to those shown in FIG. 3 are designated with the same references.

Negative ion-enriched air generator A shown in FIG. 4 comprises a clean gas generator consisting of a chemical contaminant-collecting adsorbent 21 and a microparticle removing means formed of a photoelectron emitting member 41, a UV lamp 42, an electric field-forming electrode 43 and a charged microparticle collecting member 44; and a negative ion generator consisting of a photoelectron emitting member 27, a UV lamp 28 and an electric field-forming electrode 29. According to negative ion-enriched gas generator A shown in FIG. 4, a clean room air introduced by the fan first passes through chemical contaminant-collecting adsorbent 21 where chemical contaminants in the gas are removed. Then, the gas is introduced into the photoelectron-based microparticle removing means where photoelectron emitting member 41 is irradiated with UV rays to emit photoelectrons, which generate negative ions. Then, microparticles in the gas are charged with the negative ions generated. The charged microparticles are collected/removed by charged microparticle collecting member 44 at the subsequent stage. As a result, a gas cleaned to a microparticle concentration of class 100 or less and both ionic component concentration and organic matter concentration of 2 μg/m³ or less is formed, and this clean gas is introduced into the negative ion generator, where photoelectron emitting member 27 is irradiated with UV rays to emit photoelectrons. Thus, a clean negative ion-enriched gas is provided having a concentration of 5,000 negative ions/mL or more, a microparticle concentration of class 100 or less, and both ionic component concentration and organic matter concentration of 2 μg/m³ or less. The negative ion-enriched gas generated by negative ion-enriched gas generator A shown in FIG. 4 can be used as a substrate-exposing gas in various processing steps shown in FIG. 1 to achieve a semiconductor manufacturing process in which substrates are inhibited from oxidation and both microparticle contamination and chemical contamination are prevented.

FIG. 5 shows a schematic view of a negative ion-enriched gas generator according to another embodiment of the present invention, comprising a clean gas generator using adsorbent/microparticle-collecting filter, and a negative ion generator using water spray. In FIG. 5, similar elements to those shown in FIG. 2 are designated with the same references and not explained.

Negative ion-enriched air generator A shown in FIG. 5 comprises a cleaner consisting of a chemical contaminant-collecting adsorbent 21 and a microparticle-collecting filter 22; and a negative ion generator consisting of a water-spray nozzle 33, an excess water eliminator 34 and a water supply tank 31. According to negative ion-enriched gas generator A shown in FIG. 5, a clean room air introduced by the fan is cleaned through chemical contaminant-collecting adsorbent 21 and microparticle-collecting filter 22 to a microparticle concentration of class 10 or less and both ionic component concentration and organic matter concentration of 2 μg/m³ or less, then introduced into the negative ion generator where water from water supply tank 31 is sprayed at a high pressure from water spray nozzle 33 through heat exchanger 32 to form negative ions. The resulting negative ion-enriched gas is freed of excess water by eliminator 34 and heated to a desired temperature by a reheater 35. Thus, a clean negative ion-enriched gas is provided having a concentration of 200,000-300,000 negative ions/mL or more, a microparticle concentration of class 10 or less, and both ionic component concentration and organic matter concentration of 2 μg/m³ or less. Excess water collected by eliminator 34 is received in water supply tank 31 and recycled to water spray nozzle 33. As described above, the ozone removing member as shown in FIG. 2 is normally unnecessary because ozone is not generated according to the photoelectron-based method. The negative ion-enriched gas generated by negative ion-enriched gas generator A shown in FIG. 5 can be used as a substrate-exposing gas in various processing steps shown in FIG. 1 to achieve a semiconductor manufacturing process in which substrates are inhibited from oxidation and both microparticle contamination and chemical contamination are prevented.

In the water spray-based negative ion generating method, about 10-20% of positive ions may be generated simultaneously with negative ions under some conditions. Normally, any special means for removing positive ions is not necessary because large amounts of negative ions are generated and as low as 20% of positive ions are neutralized by negative ions according to the water spray-based method if they are generated, but it may be sometimes desirable to collect/remove positive ions by placing a negative electrode further downstream of reheater 35.

FIG. 6 shows a schematic view of a negative ion-enriched gas generator according to another embodiment of the present invention, comprising a clean gas generator using adsorbent/microparticle-collecting filter and a negative ion generator using X-ray irradiation. In FIG. 6, similar elements to those shown in FIG. 2 are designated with the same references and not explained.

Negative ion-enriched air generator A shown in FIG. 6 comprises a cleaner consisting of a chemical contaminant-collecting adsorbent 21 and a microparticle-collecting filter 22; and a negative ion generator consisting of a very weak X-ray (soft X-rays: wavelength 0.2-0.3 nm) generator 36. According to negative ion-enriched gas generator A shown in FIG. 6, a clean room air introduced by the fan is cleaned through chemical contaminant-collecting adsorbent 21 and microparticle-collecting filter 22 to a microparticle concentration of class 10 or less and both ionic component concentration and organic matter concentration of 2 μg/m³ or less, then introduced into the negative ion generator where the gas is irradiated with X-rays so that gas molecules are ionized to generate negative ions. In FIG. 6, B represents an ionization zone of gas molecules with X-rays. When the gas is irradiated with X-rays, negative ions and positive ions are generated but positive ions are removed by negative electrode 37. Thus, a clean negative ion-enriched gas is provided having a concentration of 100,000 negative ions/mL or more, a microparticle concentration of class 10 or less, and both ionic component concentration and organic matter concentration of 2 μg/m³ or less. The negative ion-enriched gas generated by negative ion-enriched gas generator A shown in FIG. 6 can be used as a substrate-exposing gas in various processing steps shown in FIG. 1 to achieve a semiconductor manufacturing process in which substrates are inhibited from oxidation and both microparticle contamination and chemical contamination are prevented.

During irradiation of the gas with X-rays to generate negative ions, a slight amount of ozone may be generated under some irradiation conditions. If there is a possibility that ozone is generated, an ozone decomposing/removing member 24 is preferably placed further downstream of negative electrode 37 for removing positive ions as shown in FIG. 7 so that even a minor amount of ozone can be decomposed/removed if it is generated. Thus, a clean negative ion-enriched gas is provided having a concentration of 100,000 negative ions/mL or more, a microparticle concentration of class 10 or less, both ionic component concentration and organic matter concentration of 2 μg/m³ or less, and an ozone concentration of 0.01 ppm or less, and this gas can be used as a substrate-exposing gas in various processing steps shown in FIG. 1 to achieve a semiconductor manufacturing process in which substrates are inhibited from oxidation and both microparticle contamination and chemical contamination are prevented.

EXAMPLES

Negative ion-enriched gas generators according to various embodiments of the present invention shown in FIGS. 2 to 7 were used to prepare negative ion-enriched gases. The apparatuses used have the following structures.

1) Negative ion-enriched gas generator 1: the apparatus having the structure shown in FIG. 2 (capacity: about 30 L)

• • Adsorbent 20: a mixed bed of an ion exchange fiber: activated carbon (1:1);
  • Dust filter 21: ULPA filter;
  • Discharging member: 30 kV applied across the electrodes;
  • O₃ decomposing/removing member; MnO₂/TiO₂—C.

2) Negative ion-enriched gas generator 2: the apparatus having the structure shown in FIG. 3 (capacity; about 30 L)

• • Adsorbent 20 and dust filter 21: the same as described above in 1):
  • UV lamp 28: a germicidal lamp (4 W);
  • Photoelectron emitting member 27: a thin film of Au coated on TiO₂;
  • Electrode 29: a grid electrode made from SUS was placed above the lamp as shown in FIG. 3: electric field 10 V/cm.

3) Negative ion-enriched gas generator 3: the apparatus having the structure shown In FIG. 4 (capacity: about 40 L)

• • Adsorbent 20: the same as described above in 1);
  • Photoelectron emitting members 27 and 41, UV lamps 28 and 42, electrodes 29 and 43; the same as described above in 2);
  • Charged microparticle collecting member: a Cu—Zn plate.

4) Negative ion-enriched gas generator 4: the apparatus having the structure shown in FIG. 5 (capacity: about 200 L)

• • Adsorbent 20 and filter 21: the same as described above in 1):.
  • Water sprayer: atomization with water-spray nozzle 33 supplied with 3 L/min of ion exchange water at a water pressure of 350 kPa and a water/air ratio (L/G)=1;
  • Reheater: produced water drops were vaporized using a reheating coil.

5) Negative ion-enriched gas generator 5: the apparatus having the structure shown in FIG. 6 (capacity: about 30 L)

• • Adsorbent 20 and filter 21: the same as described above in 1);
  • X-ray generator 36: wavelength 0.2-0.3 nm;
  • Negative electrode 37; a Ca—Zn plate.

Negative ion-enriched gas generators 1 to 3 and 5 described above were supplied with a clean room air (microparticle concentration class 10,000, organic matter concentration 100-120 μg/m³, NH₃ concentration 15-20 μg/m³) at a flow rate of 3 L/min. Negative ion-enriched gas generator 4 described above was supplied with the same clean room air at a flow rate of 30 L/min. The properties of the air obtained from negative ion-enriched gas generators 1 to 5 are shown in Table 2 below.

TABLE 2
Properties of negative ion-enriched gases
		Negative	Negative ion	Micro-	Organic
		ion	conc.	particle	matter	Ammonia
	Cleaning	generating	(number/mL)	conc.	conc.	conc.
Apparatus	method	method	Inlet	Outlet	number/ft3	μg/m3	μg/m3
1) FIG. 2	Adsorption/	Discharge	≦100	100,000≦	<10	<2	<1
	ULPA
2) FIG. 3	Adsorption/	Photoelectron	≦100	10,000≦	<10	<2	<1
	ULPA
3) FIG. 4	Adsorption/	Photoelectron	≦100	8,000-10,000	<100	<2	<1
	photoelectron
4) FIG. 5	Adsorption/	Water	≦100	200,000-300,000	<10	<1	<1
	ULPA	spray
5) FIG. 6	Adsorption/	X-ray	≦100	100,000≦	<10	<1	<1
	ULPA	irradiation

The concentration of O₃ in the negative ion-enriched gas obtained by apparatus 1) was 0.01 ppm or less.

The surface of an Si wafer was exposed to the negative ion-enriched gases obtained from negative ion-enriched gas generators 1 to 5 described above and the production state of oxide films was observed. An Si wafer high resolution XPS made by Scienta type ESCA300 was used as the wafer sample after washed with RCA and then treated with HF (0.05%) and rinsed with pure water and dried, Negative ion-enriched gases obtained from negative ion-enriched gas generators 1 to 5 above were sprayed on the surface of this wafer sample at a flow rate of 3 L/min and the thickness of the oxide film formed on the sample surface was measured after given periods of time. The results are shown in Table 3 below. As a comparative example, the surface of the same wafer sample was exposed to the clean room air directly used. The results are also shown in Table 3 below.

TABLE 3
Si wafer oxidation test
			Thickness of oxide
Ap-		Negative ion	film (angstroms)
pa-	Cleaning	generating	After exposure	After exposure
ratus	method	method	for 3 hours	for 12 hours
1)	Adsorption/	Discharge	<0.02	<0.02
	ULPA
2)	Adsorption/	Photoelectron	<0.02	<0.02
	ULPA
3)	Adsorption/	Photoelectron	<0.02	<0.02
	photoelectron
4)	Adsorption/	Water spray	<0.02	<0.02
	ULPA
5)	Adsorption/	X-ray	<0.02	<0.02
	ULPA	irradiation
Comparative example:	0.3	5
exposed to clean room air

Then, the wafer sample surface was exposed (exposure period: 12 h) to gases containing about 500, about 1,000, about 3,000, about 5,000, about 10,000, about 30,000 and about 50,000 negative ions/mL prepared under varying treatment conditions using negative ion-enriched gas generator 1 to determine effective negative ion concentrations for inhibiting the production of oxide films. An effect was shown at concentrations of 1,000 negative ion s/mL or more as proved by oxide film thicknesses of <0.1 angstrom, and a remarkable effect was shown at 5,000 negative ions/mL or more as proved by <0.05 angstroms. An especially remarkable effect was shown at 10,000 negative ions/mL or more as proved by <0.02 angstroms.

Industrial Applicability

It was concluded from the foregoing findings that the present invention could have the following advantages.

(1) Substrates can be subjected to various processing steps while inhibiting them from oxidation by supplying a negative ion-enriched gas into the spaces of semiconductor manufacturing processes.

(2) Gases stable against pollution (having an antipollution function) can be prepared because clean negative ion-enriched gases free from microparticles and chemical components are obtained by removing microparticles and chemical contaminants.

(3) Generated negative ions can be prevented from being consumed by contaminants such as microparticles by using a gas freed of microparticles and chemical contaminants to prepare negative ion-enriched gases.

(4) It will be important in future to prevent substrate surfaces from oxidation, in addition to current pollution sources such as microparticles and chemical contaminants. According to the present invention, clean gases can be obtained that also have an antioxidant effect for substrate surfaces.

(5) More practical antipollution gases can be provided because preferable negative ion generating methods can be selected (see Table 1) depending on the preferred specifications such as the scale of the equipment to which the present invention is to be applied, the amount of negative ions to be generated and the desired antioxidant effect.

Claims

1-8. (canceled)

9. A process for manufacturing a semiconductor, comprising:

treating a substrate while exposing a surface of the substrate with a negative ion-enriched gas.

10. The process of claim 9, wherein said negative ion-enriched gas is prepared by passing a clean gas freed of microparticles and/or chemical contaminant through a negative ion generator.

11. The process of claim 10, wherein said chemical contaminant is one or more selected from the group consisting of ionic components, inorganic matters, and organic matters.

12. The process of claim 10, wherein said negative ion-enriched gas is prepared by passing a gas having a microparticle concentration of class 100 or less, an ionic component concentration of 10 μg/m3 or less, and an organic matter concentration of 10 μg/m3 or less through a negative ion generator.

13. The process of claim 9, wherein the concentration of negative ions in said negative ion-enriched gas is 1,000 negative ions/mL or more.

14. The process of claim 9, wherein the concentration of negative ions in said negative ion-enriched gas is 5,000 negative ions/mL or more.

15. The process of claim 9, wherein the concentration of negative ions in said negative ion-enriched gas is 10,000 negative ions/mL or more.

16. The process of claim 9, wherein the concentration of negative ions in said negative ion-enriched gas is 50,000 negative ions/mL or more.

17. An equipment for manufacturing a semiconductor comprising:

a gas channel through which a gas to be treated is passed;

a negative ion-enriched gas generator including a gas cleaner located at an upstream part of said gas channel and a negative ion generator located at a downstream part of said gas channel; and

means for supplying resulting negative ion-enriched gas to a surface of a substrate.

18. The equipment of claim 17, wherein said gas cleaner prepares a gas having a microparticle concentration of class 100 or less, an ionic component concentration of 10 μg/m3 or less, and an organic matter concentration of 10 μg/m3 or less.

19. The equipment of claim 17, wherein said negative ion-enriched gas generator prepares a negative ion-enriched gas having a negative ion concentration of 1,000 negative ions/mL or more.

20. The process of claim 17, wherein the concentration of negative ions in said negative ion-enriched gas is 5,000 negative ions/mL or more.

21. The process of claim 17, wherein the concentration of negative ions in said negative ion-enriched gas is 10,000 negative ions/mL or more.

22. The process of claim 17, wherein the concentration of negative ions in said negative ion-enriched gas is 50,000 negative ions/mL or more.