Patent Application
20160104551
The present invention relates to a plasma-resistant component used in the semiconductor and display industries, more specifically to a conductive plasma-resistant member.
Yttria (Y2O3) has excellent plasma-resistance and is adopted as the core material of the inner jig of high added-value processing equipment used in the semiconductor and display industries, etc.
Meanwhile, among various plasma-resistant components within the processing equipment, some components such as a focus ring, etc. prefer to be made of materials having a conductivity similar to silicon wafers because uniform plasmas are formed around the silicon wafers to improve plasma etching performance.
Generally, monolithic yttria is known as a complete non-conductive ceramic.
However, it is found that when sintering yttria, some divalent oxides are advantageous for improving densification and conductivity, and among them, CaO is known to be the most effective one.
Meanwhile, upon measuring the high-temperature conductivity for yttria added with 0˜10 mol % of CaO sintered under atmospheric pressure, the conductivity is measured to be 10−4 S/cm at 1000° C. There is no measurement result made at room temperature, but when estimating the measurement by extrapolation from the high-temperature measured value, the conductivity is expected to be greatly reduced.
In the case of hot press sintering adding various types of carbon-based materials, it is reported that the conductivity is measured to be up to 10−2 S/cm at room temperature.
However, since carbon-based additives have a weak binding with an yttria matrix phase, they act as defects and thus would have a reduced strength compared to monolith yttria. Furthermore, due to the weak binding between the yttria matrix phase and carbon-based additives, the sintering itself may not be possible with atmospheric pressure sintering alone, which is not a severe condition like hot press sintering.
In order to solve the problems of the prior art, the present invention aims to provide an yttria-based plasma-resistant member with chemical stability preventing etching under a fluorinated plasma atmosphere and conductivity at room temperature.
Additionally, the present invention aims to provide a method of producing a plasma-resistant member with a high strength and relative density.
In order to achieve the technical tasks above, the present invention provides a plasma-resistant member including an yttrium compound, which includes a matrix phase consisting of yttrium oxides, and a conductive dispersed phase.
According to an embodiment of the present invention, the conductive dispersed phase may include a carbide or nitride of at least one metal selected from a group consisting of Ti, Zr and Hf. Additionally, the conductive dispersed phase may also include at least one carbon-based additive selected from a group consisting of CNT, graphene, and particulate carbon.
Additionally, according to an embodiment of the present invention, the plasma-resistant member may include a body and a coating layer surrounding the body, and the coating layer may include a matrix phase consisting of yttrium oxides and a conductive dispersed phase.
The yttrium oxide of the present invention may include yttria (Y2O3), or yttrium aluminum garnet (YAG). Also, the matrix phase may further include zirconia or alumina.
According to an embodiment of the present invention, the plasma-resistant member may include at least 5% by volume of the dispersed phase, and may also include 30% by volume or less of the dispersed phase. Preferably, the plasma-resistant member may include 10˜20% by volume of the dispersed phase.
The plasma-resistant member of the present invention may have a conductivity in the range of 10−2˜10−2 S/cm.
In order to achieve another technical task, the present invention provides a method of producing a plasma-resistant member including an yttrium compound, which includes providing a powder mixture of an yttrium oxide and a conductive material, molding the powder mixture to produce a molded product, and sintering the molded product under a nitrogen atmosphere.
According to an embodiment of the present invention, the conductive material may include a carbide or nitride of at least one metal selected from a group consisting of Ti, Zr and Hf.
The sintering in an embodiment of the present invention may be performed under atmospheric pressure or under vacuum, or by spark plasma sintering (SPS).
Also, according to an embodiment of the present invention, the present invention provides a method of producing a plasma-resistant member including a yttrium compound, which includes molding a powder mixture of an yttrium oxide and a carbide or nitride of at least one metal selected from a group consisting of Ti, Zr and Hf to produce a molded product, calcining the molded product, and sintering the molded product under a nitrogen atmosphere.
According to the present invention, the present invention may provide an yttria-based plasma-resistant member. The plasma-resistant member of the present invention may be applied to various components of semiconductor processing equipment.
For example, an yttria composite according to the present invention may provide a semiconductor-grade conductivity, and thus it may be used as a plasma-resistant member requiring conductivity like a focus ring. Additionally, according to the present invention, it is possible to produce a semiconductor-grade yttria composite on the order of 10−4 S/cm, compared to a monolithic yttria, which is a nonconductor on the order of 10−14 S/cm. Of course, it is possible to further improve or control conductivity according to the content of conductive additives.
According to the present invention, densification may be achieved by ordinary sintering under atmospheric pressure (nitrogen 1 atmosphere), instead of the press sintering of the plasma-resistant member.
Additionally, the plasma-resistant member according to the present invention may have higher strength than monolithic Y2O3, and this may be helpful in solving the slow crack growth problem, which is one of the defects of the plasma-resistant Y2O3 material that needs to be supplemented.
Hereinafter, the present invention will be described by explaining preferable embodiments of the present invention.
The plasma-resistant member of the present invention includes an yttrium oxide and a conductive additive for improving conductivity of the yttrium oxide.
The plasma-resistant member of the present invention may be in a bulk form where a conductive phase is dispersed into an yttrium oxide matrix phase. Unlike this, the plasma-resistant member of the present invention may include a body made of a predetermined material and a coating surrounding the body. The coating may include the above-mentioned yttrium oxide matrix phase and conductive dispersed phase. The coating may be provided with various methods such as plasma spraying, etc.
The plasma-resistant member of the present invention may be used for focus rings, shower heads, etc. of semiconductor processing equipment. Of course, the plasma-resistant member of the present invention is not limited to the semiconductor producing equipment, and may be applied to various fields. For example, the plasma-resistant member of the present invention may be applied to antistatic rollers, etc. of display equipment.
The yttrium oxide constituting the plasma-resistant member of the present invention may be composed of pure yttria, or an oxide-type compound like yttrium aluminum garnet (YAG). Of course, the yttrium oxide may be a compound of yttria or YAG.
The yttrium oxide of the present invention may include zirconia, alumina, or a combination thereof as a sintering aid.
The conductive additive included in yttria is required to have high conductivity and stability to prevent etching under a fluorinated plasma atmosphere.
Table 1 below illustrates the electric resistance (Ωm) of major candidate materials of carbide-based and nitride-based conductive additives, and the melting point (° C.) of the fluoride thereof.
The conductive additive is expected to be stable for etching as the melting point of fluoride, which may be made by the reaction of conductive additive with fluorine, gets higher than the temperature during an etching process. When operating a cooling system of the processing equipment, the maximum temperature of the jig component is known to be 100° C. or below. Thus, TiC, TiN, ZrC, ZrN, HfC, HfN, etc. are determined to be a candidate group for effective additives. Carbide-based and nitride-based ceramics have a high binding with the yttria matrix phase, and may also have an improved strength compared to monolithic yttria, when expressing particle growth inhibition function by Zener effect.
Meanwhile, as another embodiment, additives such as CNT, graphene, particulate carbon, etc. may be used in the present invention, and carbon-containing additives may contribute to improvement of conductivity.
ZrO2 (<100 nm) is mixed with Y2O3 (d50=1.2 μm) as a sintering aid and TiC (<100 nm) is mixed as a conductive additive.
A basic composition of the mixing ratio is Y2O3+1 at % ZrO2, and 10 vol % and 20 vol % of TiC are added to this composition. According to the percolation theory, it is known that 10 vol % is a boundary value and 20 vol % is a stability value for granting conductivity to a nonconductor. However, in the embodiment of the present invention, the content of the conductive additive may vary depending on the type of additive. For example, the content of the additive may be controlled so that the content of the dispersed phase is at least 5% by volume or 30% by volume or less.
Meanwhile, in order to observe a dispersion effect, a composition added with PEG is prepared. Table 2 below shows the mixing ratio of each specimen in the embodiment of the present invention.
For comparison, Y2O3+1 at % ZrO2 not added with TiC is prepared as well.
A starting material is mixed according to the composition in Table 2 and the mixture is subject to planetary milling (ZrO2 ball and jar, anhydrous ethanol, 100 rpm) for 4-24 hours, and then dried in a rotary evaporator at 70° C.
The dried powder is calcined at 700° C. for 1 hour. In order to observe the effect of calcination, in the case of the 20%4 h specimen and 20%4 hP specimen, a powder omitted with the calcination process is prepared.
Furthermore, after isostatic pressing by using a mold with a diameter of 15 mm, Y2O3 atmospheric powder is filled in a carbon crucible, and a specimen is buried inside the atmospheric powder to perform vacuum sintering and nitrogen 1 pressure atmospheric pressure sintering. The sintering temperature is 1800° C., and the sintering is maintained for 3 hours.
The fine structure of the sintered specimen is observed through a scanning electron microscope, and XRD phase analysis, conductivity, and biaxial strength are measured.
From
It may be known that the relative density of the sintered body is 96-98%, which is similar to each other regardless of whether the powder is calcined, but the change in weight and shrinkage cause a big difference depending on whether the powder is calcined. Calcined powder shows a decrease in weight by 8-9% and a great shrinkage by 21-23%, whereas non-calcined powder shows an increase in weight by 12-14% and a small shrinkage by 13-15%. The decrease in weight during high-temperature sintering is normal, but the increase in weight is not. Thus, non-calcined powder is not considered to be suitable for the present invention.
When going through calcination, some TiC is oxidized into TiO2. When applying sintering under atmospheric pressure nitrogen atmosphere, TiO2 is nitrified into TiN. Thus, it is expected that there would be no decrease in conductivity by TiO2.
It may be confirmed from
Meanwhile, as a result of sintering the powder calcined under nitrogen atmospheric pressure, regardless of the composition, densification sintering with a level of 97-99% of relative density is possible. Table 3 below shows the relative density after the sintering under nitrogen atmospheric pressure.
Although it was expected that it would be difficult to perform densification with sintering under atmospheric pressure alone because of the Zener effect impeding a grain-boundary migration by the addition of TiC, it was possible to manufacture a sintered body with high density. Additionally, it may be known that the difference in sintering density resulting from the amount of TiC added and whether PEG is added is insignificant.
As a result of phase analysis, it may be confirmed that TiN is detected, in addition to TiC. That is, it may be known that TiO2 produced by the oxidation reaction during the calcination process is nitrified during the sintering under nitrogen atmosphere to form TiN. Additionally, TiC and TiN all have high conductivity, so the above-mentioned production process may improve conductivity of the sintered body.
Referring to
Meanwhile, when the amount of TiC added increases, the conductivity value also increases. When the amount of TiC added is 10 vol % and 20 vol %, the conductivity is measured to be on the order of 10−7 order and 10−4, respectively. Thus, when the TiC content increases to be at least 20 vol %, it is expected that conductivity would increase in proportional thereto.
The gray part in the BSE mode photograph indicates Y2O3 whose atom number is high, and the black part indicates Tic whose atom number is low. That is, it may be known that conductive TiC of fine particles is uniformly dispersed in the non-conductive Y2O3 matrix of the assembly.
The biaxial strength of the Y2O3—TiC sintered body produced is measured. The measurement is conducted by piston-on-3ball test method.
As a result of the measurement, the composition added with 20 vol % of TiC has the highest strength, and the maximum strength is measured to be 193 MPa. Considering that the strength of monolithic yttria is 163 MPa, it may be known that by the dispersion of TiC, the conductivity increases and the strength increases as well. This shows that an excellent composite may be produced, compared to the decrease in strength due to the carbon-based additive.
In the same manner as Example 1, ZrO2 (<100 nm) is mixed with Y2O3 (d50=1.2 μm) as a sintering aid, and TiC or TiN(<100 nm) is mixed as a conductive additive. The basic composition is Y2O3+1 at % ZrO2, and 10 vol %, 20 vol % and 30 vol % of TiC or TiN are added to this composition, respectively, but PEG is not added. For comparison of strength, a composition not added with TiC (Y2O3+1 at % ZrO2) is also prepared. Table 4 shows the mixing composition of Example 2. Specimen number 10C in Table 4 indicates a specimen where 10 vol % of TiC is added, 20C indicates a specimen where 20 vol % of TiC is added, and 30C indicates a specimen where 30 vol % of TiC is added. Specimen numbers 10N, 20N and 30N, respectively, indicate each TiN and the content of TnN added.
A starting material is mixed according to the composition in Table 4 and the mixture is subject to planetary milling (ZrO2 ball and jar, anhydrous ethanol, 100 rpm) for 4 hours, and then dried in a rotary evaporator at 70° C.
In order to observe the effect of calcination, powder not calcined, powder calcined at an air atmosphere of 700° C., and powder calcined at a vacuum atmosphere of 700° C. are prepared. Then, after isostatic pressing by using a mold with a diameter of 15 mm, Y2O3 atmospheric powder is filled in a carbon crucible, and a specimen is buried inside the atmospheric powder to perform vacuum sintering. The sintering process follows the sintering schedule illustrated in
The conductivity and biaxial strength of the sintered specimen are measured.
In case of the specimen added with TiC, it may be known that even after planetary milling, there are many coagulated particles with the size of at least 1 μm, and the particles show bimodal distribution. In comparison, in case of the specimen added with TiN, a relatively greater amount of particles are distributed in the sub-micro region through milling, so they have a particle size distribution ranging from 0.1 to 5 μm.
Meanwhile, although it is not illustrated, upon reviewing the effect of calcination on the relative density of sintered body, the composition added with TiC shows a high density during calcination in the air, and the composition added with TiN shows a high density during non-calcination or calculation in vacuum.
Referring to
Referring to
From
Referring to
The raw powder in Table 4 is mixed, and the mixture is subject to milling, and then dried under the same condition as in Example 3. The calcination process is omitted, and spark plasma sintering (SPS) is performed at 1260° C. and 1300° C. for 20 minutes under the pressure of 80 MPa, by using the carbon mold of 20 mm.
The conductivity and biaxial strength of sintered specimen are measured.
Referring to
Meanwhile, as illustrated in
From
It may be known that the reference specimen and specimen added with TiN are densified by at least 99%. In comparison, the specimen added with TiC shows a low relative density of 98% or less, and the density decreases as the content of TiC increases. Upon comparing the density with the shrinkage behavior in
Referring to
Referring to
Additionally, the specimen added with TiC shows decreased strength compared to the reference, and the specimen added with TiN shows improved strength compared to the reference specimen in terms of the content of some additives. Especially, in case of adding 20 vol % of TiN, the strength improved by about 24%, which is similar to the vacuum sintering case of the above-mentioned embodiment. Thus, in terms of strength, the optimal amount of TiN added is around 20 vol %.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10-2014-0136865 | Oct 2014 | KR | national |
