Coated article and semiconductor chamber apparatus formed from yttrium oxide and zirconium oxide

Abstract

Disclosed herein is a ceramic article or coating useful in semiconductor processing, which is resistant to erosion by halogen-containing plasmas. The ceramic article or coating is formed from a combination of yttrium oxide and zirconium oxide.

Claims

1. A coated article that is resistant to erosion by halogen-containing plasmas, the coated article comprising an article and a ceramic coating formed on the article, wherein the ceramic coating comprises: yttrium oxide at a concentration from 35 mole % to less than 62.1 mole %; and zirconium oxide at a concentration from 65 mole % to greater than 37.9 mole %.

2. The coated article of claim 1, wherein the ceramic coating consists essentially of: the yttrium oxide at a concentration from 35 mole % to less than 62.1 mole %; and the zirconium oxide at a concentration from 65 mole % to greater than 37.9 mole %.

3. The coated article of claim 1, wherein the ceramic coating consists of: the yttrium oxide at a concentration from 35 mole % to less than 62.1 mole %; and the zirconium oxide at a concentration from 65 mole % to greater than 37.9 mole %.

4. The coated article of claim 1, wherein the ceramic coating comprises a solid solution of the yttrium oxide and the zirconium oxide.

5. The coated article of claim 1, wherein the article is a semiconductor processing chamber component selected from a group consisting of a lid, a lid liner, a nozzle, a gas distribution plate, a shower head, an electrostatic chuck component, a shadow frame, a substrate holding frame, a processing kit, and a chamber liner.

6. The coated article of claim 1, wherein the ceramic coating is formed on the article using a technique selected from a group consisting of thermal spraying, plasma spraying, sputtering, and chemical vapor deposition.

7. A semiconductor processing apparatus adapted to have at least one surface exposed to a halogen-comprising plasma during a process performed in a semiconductor processing chamber, the semiconductor processing apparatus comprising the coated article of claim 1.

8. A coated article that is resistant to erosion by halogen-containing plasmas, the coated article comprising an article and a ceramic coating formed on the article, and wherein the ceramic coating consists essentially of yttrium oxide and zirconium oxide in a solid solution, wherein the yttrium oxide is present at a concentration from 35 mole % to 80 mole % and the zirconium oxide is present at a concentration from 65 mole % to 20 mole %, and wherein the article is a semiconductor processing chamber component selected from a group consisting of a lid, a lid liner, a nozzle, a gas distribution plate, a shower head, an electrostatic chuck component, a shadow frame, a substrate holding frame, a processing kit, and a chamber liner.

9. The coated article of claim 8, wherein the ceramic coating consists of the yttrium oxide and the zirconium oxide.

10. The coated article of claim 8, wherein the ceramic coating is formed on the article using a technique selected from a group consisting of thermal spraying, plasma spraying, sputtering, and chemical vapor deposition.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1A shows a photomicrograph of the as-sintered surface of a solid yttrium oxide ceramic at a magnification of 1,000 times.

(2) FIG. 1B shows a photomicrograph of the as-sintered surface of a solid solution ceramic substrate formed from 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide, at a magnification of 1,000 times.

(3) FIG. 1C shows a photomicrograph of the as-sintered surface of a solid solution ceramic substrate formed from 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide, at a magnification of 1,000 times.

(4) FIG. 2A shows a photomicrograph of the surface of a solid yttrium oxide ceramic after a test etch using the processing plasmas and times typically used to etch the various layers of a contact via feature in a semiconductor device. The magnification is 1,000 times.

(5) FIG. 2B shows a photomicrograph of the surface of a solid solution ceramic formed from 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide after a test etch using the processing plasmas and times typically used to etch the various layers of a contact via feature in a semiconductor device. The magnification is 1,000 times.

(6) FIG. 2C shows a photomicrograph of the surface of a solid solution ceramic formed from 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide after a test etch using the processing plasmas and times typically used to etch the various layers of a contact via feature in a semiconductor device. The magnification is 1,000 times.

(7) FIG. 3A shows a photomicrograph of the post-etch ceramic of FIG. 2A, but at a magnification of 5,000 times.

(8) FIG. 3B shows a photomicrograph of the post-etch ceramic of FIG. 2B, but at a magnification of 5,000 times.

(9) FIG. 3C shows a photomicrograph of the post-etch ceramic of FIG. 2C, but at a magnification of 5,000 times.

(10) FIG. 4A shows a photomicrograph of the as-sintered surface of a solid solution ceramic formed from 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide. The magnification is 2,000 times.

(11) FIG. 4B shows a photomicrograph of the surface of the solid solution ceramic shown in FIG. 4A, after exposure of the test coupon to a trench etch process of the kind described herein. The magnification is 2,000 times.

(12) FIG. 4C shows a photomicrograph of the as sintered surface of a solid solution ceramic formed from 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide. The magnification is 2,000 times.

(13) FIG. 4D shows a photomicrograph of the surface of the solid solution ceramic shown in FIG. 4C, after exposure of the test coupon to a trench etch process of the kind described herein. The magnification is 2,000 times.

(14) FIG. 5A shows a photomicrograph of a solid solution ceramic formed from 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide after exposure of the test coupon to a metal etch process of the kind described herein. The magnification is 5,000 times.

(15) FIG. 5B shows a photomicrograph of a solid solution ceramic formed from 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide after exposure of the test coupon to an etch by a CF.sub.4/CHF.sub.3 plasma. The magnification is 5,000 times.

DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS

(16) As a preface to the detailed description, it should be noted that, as used in this specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents, unless the context clearly dictates otherwise.

(17) When the word “about” is used herein, this is intended to mean that the nominal value presented is precise within ±10%.

(18) Bulk yttrium oxide has been shown to have very good corrosion resistance upon exposure to fluorine plasma and other corrosive plasmas which are typically used in semiconductor manufacturing processes (such as etch processes and chemical vapor deposition processes). However, pure crystalline yttrium oxide, while offering very good corrosion resistance to various etchant plasmas, does not offer good mechanical properties in terms of flexural strength and fracture toughness, for example. To improve the overall performance and handling capabilities of semiconductor component parts and liners, there is a need to improve the mechanical properties from those available in pure crystalline yttrium oxide. To obtain the improvement in mechanical properties, it is necessary to form an alloy of yttrium oxide with a compatible oxide. The improvement in mechanical properties is needed to be accomplished without harming the very good plasma erosion properties of the pure yttrium oxide.

(19) In consideration of the Gibbs Formation Free Energy of various ceramic materials which might be compatible with yttrium oxide, we determined that it is more difficult to form fluorides than oxides for yttrium and aluminum elements, so that yttrium oxide and aluminum oxide are expected to provide good resistance to a fluorine-containing plasma. The Gibbs Formation Free Energy of zirconium fluoride is similar to that for yttrium fluoride. Further, in a homogeneous amorphous oxyfluoride, or a glass-ceramic composite oxyfluoride, increasing the zirconium fluoride content can decrease the free energy of the final oxyfluoride to make it more stable.

EXAMPLE EMBODIMENTS

Example One

Etch Plasma Process Conditions for Erosion Rate Testing

(20) Tables One-Three, below, provides the etch plasma compositions and etch plasma processing conditions which were used for evaluation of a series of test coupon materials. There were three basic different sets of etch plasma conditions which were used for the erosion rate testing: 1) Trench etching, where the etch plasma source gas and etch process conditions were representative of etching a trench feature size beyond 65 nm technology, i.e. smaller than 65 nm, into a multilayered semiconductor substrate. Such a substrate typically includes an antireflective coating (ARC or BARC) layer, an organic or inorganic dielectric layer, a metal layer, and an etch stop layer. Contact Via etching, where the etch plasma source gas and etch process conditions were representative of etching a contact via having an aspect ratio of about 30 in production and 40 plus in the developed device substrate, and having a diameter of beyond 65 nm technology into a multilayered semiconductor substrate including a buried ARC (BARC) layer, a dielectric layer and a stop layer; and 3) Metal etching, here the etch plasma source gas and etch process conditions were representative of etching an overlying titanium nitride hard mask and an aluminum layer, where the etch plasma source gas and etch process conditions are beyond 65 nm technology.

(21) The trench etching process and the contact via etching process were carried out in the ENABLER™ processing system, and the metal etching process was carried out in the DPS™ processing system, all available from Applied Materials, Inc. of Santa Clara, Calif.

(22) TABLE-US-00001 TABLE ONE Process Conditions for Trench Etch Erosion Rate Test Plasma Subr Trench Etch Source Bias Subr Simulation CF.sub.4* O.sub.2* CHF.sub.3* N.sub.2* Ar* Power.sup.1 Pr.sup.2 Power.sup.3 Temp.sup.4 Time.sup.5 Etch Step One 150 30 300 1,000 40 35 Etch Step Two 400 1200 220 400 40 40 Etch Step Three 175 15 1500 150 500 40 39 Etch Step Four 500 100 10 200 40 55 *All gas flow rates are in sccm. .sup.1Plasma Source Power in W. .sup.2Pressure in mTorr. .sup.3Substrate Bias Power in W. .sup.4Substrate Temperature in ° C. .sup.5Time in seconds.

(23) TABLE-US-00002 TABLE TWO Process Conditions for Via Etch Erosion Rate Test Via Etch Simulation CF.sub.4* C.sub.4F.sub.6* CHF.sub.3* CH.sub.2F.sub.2* Ar* O.sub.2* N.sub.2* Etch Step 80 80 One Etch Step 28 15 20 500 31 Two Etch Step 40 650 30 Three Etch Step 200 Four Etch Step 500 Five Plasma Substrate Via Etch Source Bias Substrate Simulation Power.sup.1 Pr.sup.2 Power.sup.3 Temp.sup.4 Time.sup.5 Etch Step 80 400 40 50 One Etch Step 400 30 1700 40 60 Two Etch Step 30 1700 40 60 Three Etch Step 1000 50 100 40 45 Four *All gas flow rates are in sccm. .sup.1Plasma Source Power in W. .sup.2Pressure in mTorr. .sup.3Substrate Bias Power in W. .sup.4Substrate Temperature in ° C. .sup.5Time in seconds.

(24) TABLE-US-00003 TABLE THREE Process Conditions for Metal Etch Erosion Rate Test Plasma Subr Metal Etch Source Bias Pre Subr Simul. Cl.sub.2* BCl.sub.3* C.sub.2H.sub.4* Ar* CHF.sub.3* N.sub.2* Power.sup.1 Power.sup.2 Pr.sup.3 Temp.sup.4 Time.sup.5 Etch Step One 60 3 20 1000 100 8 40 30 Etch Step Two 25 40 10 5 500 150 10 40 18 Etch Step Three 60 40 20 700 120 18 40 30 Etch Step Four 60 40 3 1000 200 8 40 23 Etch Step Five 30 60 5 50 5 800 170 6 40 15 *All gas flow rates are in sccm. .sup.1Plasma Source Power in W. .sup.2Substrate Bias Power in W. .sup.3Pressure in mTorr. .sup.4Substrate Temperature in ° C. .sup.5Time in seconds.

Example Two

Comparative Relative Erosion Rates of Various Ceramic Materials Compared With Aluminum Oxide

(25) Aluminum oxide has frequently been used as a protective layer or liner when a semiconductor process makes use of an etchant plasma. Using aluminum oxide as the base comparative material, we determined the relative etch rates, in a Trench Etch (CF.sub.4/CHF.sub.3) environment. With aluminum oxide having a relative erosion rate of 1, we found that the relative erosion rate of quartz was about 2.2 times that of aluminum oxide. The relative erosion rate of silicon carbide was about 1.6 times that of aluminum oxide. The relative erosion rate of zirconia was about 0.8 times that of aluminum oxide. The relative erosion rate of pure yttrium oxide was about 0.19 times that of aluminum oxide. The relative erosion rate of a yttrium oxide, zirconium oxide, aluminum oxide ceramic composite, formed from 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide was about 0.2 times that of aluminum oxide. The relative erosion rate of a yttrium oxide, zirconium oxide, aluminum oxide ceramic composite, formed from 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide was about 0.05 times that of aluminum oxide.

Example Three

Measured Erosion Rates for Trench Etching Process

(26) With reference to the trench etching method described above, the sample substrate test coupon erosion rates measured were as follows. The erosion rate of aluminum oxide was 1.1 μm/hr. The erosion rate of bulk yttrium oxide was 0.3 μm/hr. The erosion rate of a the a yttrium oxide, zirconium oxide, aluminum oxide ceramic composite, formed from 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide was 0.1 μm/hr. The erosion rate of a yttrium oxide, zirconium oxide, aluminum oxide ceramic composite, formed from 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide was 0.07 μm/hr.

Example Four

Measured Erosion Rates for Via Etching Process

(27) With reference to the via etching method described above, the sample substrate test coupon erosion rates measured were as follows. The erosion rate of aluminum oxide was not measured. The erosion rate of bulk yttrium oxide was 0.16 μm/hr. The erosion rate of a the a yttrium oxide, zirconium oxide, aluminum oxide solid solution, formed from 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide was 0.21 μm/hr. The erosion rate of a yttrium oxide, zirconium oxide, aluminum oxide solid solution, formed from 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide was 0.22 μm/hr.

Example Five

Measured Erosion Rates for Metal Etching Process

(28) With reference to the metal etching method described above, the sample substrate test coupon erosion rates measured were as follows. The erosion rate of aluminum oxide was 4.10 μm/hr. The erosion rate for bulk yttrium oxide was 0.14 μm/hr. The erosion rate of a the a yttrium oxide, zirconium oxide, aluminum oxide ceramic composite, formed from 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide was 0.10 μm/hr. The erosion rate of a yttrium oxide, zirconium oxide, aluminum oxide ceramic composite, formed from 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide was 0.18 μm/hr.

Example Six

Photomicrographs of Yttrium-Oxide-Based Ceramics after Exposure to a Via Etch Process

(29) FIGS. 1A through 1C show photomicrographs of the surface of a sintered yttrium-oxide-containing ceramic composite prior to exposure to the via etch process described herein. The yttrium-oxide-containing ceramic composites include: 1) yttrium oxide-zirconium oxide solid solution; and 2) yttrium aluminate, when the composition was yttrium oxide 100 parts by weight, zirconium oxide 20 parts by weight, and aluminum oxide 10 parts by weight. (This composition is the same as 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide); and 3) yttrium oxide-zirconium oxide-aluminum oxide solid solution, when the composition from which the solid solution was formed was yttrium oxide 100 parts by weight, zirconium oxide 20 parts by weight, and aluminum oxide 20 parts by weight. (This composition is the same as 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide). All of the photomicrographs are at a magnification of 1,000 times.

(30) FIGS. 2A through 2C show photomicrographs of the sintered yttrium-oxide-containing ceramic composite subsequent to exposure to the via etch process described herein.

(31) The yttrium-oxide-containing ceramic composites include: 1) yttrium oxide-zirconium oxide solid solution; and 2) yttrium aluminate, when the composition was yttrium oxide 100 parts by weight, zirconium oxide 20 parts by weight, and aluminum oxide 10 parts by weight (This composition is the same as 63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide); or when the composition was yttrium oxide 100 parts by weight, zirconium oxide 20 parts by weight, and aluminum oxide 20 parts by weight (This composition is the same as 55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide). All of the photomicrographs are at a magnification of 1,000 times.

(32) The surface roughness of the bulk yttrium oxide shown in FIG. 2A has increased in roughness substantially. However, the overall surface roughness appears to be less than that of the zirconium oxide and aluminum oxide containing sample coupons. The surface roughness of the solid solution shown in FIG. 2B, which contains 10 parts by weight aluminum oxide appears to have hills and valleys which are flatter than the hills and valleys of the solid solution shown in FIG. 2C, which contains the 20 parts by weight of aluminum oxide. However, the hills and valleys on the 10 parts by weight aluminum oxide sample coupon shown in FIG. 2B have more pitting on the surface than in the 20 parts by weight sample coupon shown in FIG. 2C.

(33) FIGS. 3A through 3C show photomicrographs which correspond with FIGS. 2A through 2C, respectively, but are at a magnification of 5,000 times. Looking at the surface of the bulk yttrium oxide sample coupon shown in FIG. 3A, the surface is relatively smooth but does show some evidence of small pits. The FIG. 3B solid solution formed from yttrium oxide 100 parts by weight, zirconium oxide 20 parts by weight, and aluminum oxide 10 parts by weight also shows some small scale pitting present on the rougher surface shown in FIG. 2B. The FIG. 3C solid solution formed from yttrium oxide 100 parts by weight, zirconium oxide 20 parts by weight, and aluminum oxide 20 parts by weigh shows negligible small scale pitting.

(34) Looking at the erosion rates for the three test coupons, it appears that the 1,000 times magnification for the post-etch coupons shows better surface characteristics related to the erosion rates of the coupons. The erosion rates were 0.16 μm/hr for the solid yttrium oxide shown in FIG. 2A; 0.22 μm/hr for the solid solution of yttrium oxide-zirconium oxide-aluminum oxide which contained 10 parts by weight aluminum oxide; and 0.21 μm/hr for the solid solution of yttrium oxide-zirconium oxide-aluminum oxide which contained 20 parts by weight aluminum oxide.

Example Seven

Photomicrographs of Yttrium-Oxide-Containing Substrates after Exposure to a Trench Etch Process

(35) FIG. 4A shows a photomicrograph of the as-sintered surface of a solid solution ceramic composite containing 100 parts by weight yttrium oxide, 20 parts by weight aluminum oxide, and 10 parts by weight aluminum oxide (63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide), at a magnification of 2,000 times. FIG. 4B shows a photomicrograph of the surface of the solid ceramic composite of FIG. 4A after etching by a trench etch process of the kind shown herein. Both photomicrographs are at a magnification of 2,000. The post-etched surface appears to be flat and relatively homogeneous. This combination of photographs suggests that after fabrication of an apparatus such as a chamber liner or a component part, it may be advisable to “season” the part by exposing it to an exemplary plasma etch process prior to introducing the apparatus into a semiconductor device production process. The erosion rate for the solid solution ceramic composite containing the 10 parts by weight of aluminum oxide, after exposure to the trench etch process, was about 0.08 em/hr.

(36) FIG. 4C shows a photomicrograph of the as-sintered surface of a solid solution ceramic composite containing 100 parts by weight yttrium oxide, 20 parts by weight aluminum oxide, and 20 parts by weight aluminum oxide (55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide). FIG. 4D shows a photomicrograph of the surface of the solid solution ceramic composite of FIG. 4C after etching by a trench etch process of the kind shown herein. Both photomicrographs are at a magnification of 2,000. The post-etched surface appears to be flat and relatively homogeneous. This combination of photographs suggests the same seasoning process described above for newly fabricated apparatus. The erosion rate of the solid solution ceramic composite containing the 20 parts by weight of aluminum oxide, after exposure to the trench etch process, was about 0.07 μm/hr.

Example Eight

Photomicrographs of Yttrium-Oxide-Containing Ceramic Composites after Exposure to a Metal Etch Process

(37) FIG. 5A shows a photomicrograph of a two phase solid solution ceramic composite formed from 100 parts by weight of yttrium oxide, 20 parts by weight of zirconium oxide and 10 parts by weight of aluminum oxide (63 mole % yttrium oxide, 23 mole % zirconium oxide, and 14 mole % aluminum oxide) after exposure of the test coupon to a metal etch process of the kind described herein. The magnification is 5,000 times. FIG. 5B shows a photomicrograph of a two phase solid solution ceramic composite formed from 100 parts by weight of yttrium oxide, 20 parts by weight of zirconium oxide, and 20 parts by weight of aluminum oxide (55 mole % yttrium oxide, 20 mole % zirconium oxide, and 25 mole % aluminum oxide) after exposure of the test coupon to a metal etch process of the kind described herein. The magnification is 5,000 times. A comparison of these two photomicrographs shows that the two phase solid solution containing the higher content of aluminum oxide has an increased amount of the darker phase, which is yttrium aluminate. The erosion rate of the two phase solid solution ceramic composites containing the 10 parts by weight of aluminum oxide, after exposure to the trench etch process, was about 0.18 μm/hr, while the erosion rate of the two phase solid solution ceramic composite containing the 20 parts by weight of aluminum oxide, after exposure to the trench process was about 0.10 μm/hr.

Example Nine

Relative Physical and Mechanical Properties of Yttrium-Oxide-Containing Substrates

(38) Table Four below shows comparative physical and mechanical properties for the bulk, pure yttrium oxide ceramic and for various yttrium-oxide containing solid solution ceramics.

(39) TABLE-US-00004 TABLE FOUR 100 Y.sub.2O.sub.3 100 Y.sub.2O.sub.3 100 ZrO.sub.2 100 Y.sub.2O.sub.3 20 ZrO.sub.2 20 ZrO.sub.2 Material 3 Y.sub.2O.sub.3 20 ZrO.sub.2 10 Al.sub.2O.sub.3 20 Al.sub.2O.sub.3 Starting parts by parts by parts by parts by Composition Y.sub.2O.sub.3 Al.sub.2O.sub.3 weight weight weight weight Flexural 100-150 400    1200 ± 100 137    215 172 Strength (MPa) Vickers Hardness 5.7 17.2  11.9 9.3 9.4 9.6 (5 Kgf)(GPa) Young's 140-170 380    373 190    190 202 Modulus (GPa) Fracture Toughness 1.0-1.3 3.5 10.9 1.3 1.6 1.7 (MPa .Math. m.sup.1/2) Thermal 13.7  33   2.9 4.7 3.5 Conductivity (W/m/° K) Thermal Shock 130-200 200    130-200 150-200 Resistance (ΔT) ° C. Thermal 7.2 7.7 9.4 9.0 8.5 Expansion × 10.sup.−6/K (20-900° C.) Dielectric 12.3-13.sup.  9.9 — 15.0  15.5 Constant (20° C. 13.56 MHZ) Dielectric Loss <20    0.5 — <20    <20 — Tangent × 10.sup.−4 (20° C. 13.56 MHZ) Volume 10.sup.12-10.sup.13 10.sup.15  — 10.sup.11  10.sup.16-10.sup.22 — Resistivity at RT (Ω .Math. cm) Density  4.92  3.95 5.89  5.19 4.90 4.86 (g/cm.sup.3) Mean Grain 10-25 — 0.5-1.0 5-10 3-6 3-6 Size (μm) Phase Y.sub.2O.sub.3 Al.sub.2O.sub.2 Zr.sub.1−xY.sub.xO.sub.2 F/C—Y.sub.2O.sub.3 F/C—Y.sub.2O.sub.3 F/C—Y.sub.2O.sub.3 Composition SS SS SS and Y.sub.4Al.sub.2O.sub.9 Y.sub.4Al.sub.2O.sub.9 and YAlO.sub.3 Plasma Erosion 0.3  1.44 0.3 0.1 0.1 0.2 Rate (μm/hr) (CF.sub.4/CHF.sub.3) *All of the solid solution ceramic substrates were sintered using a pressureless sintering technique under a hydrogen protected atmosphere.

(40) A review of the plasma erosion rate clearly shows the advantages of the solid solution yttrium oxide, zirconium oxide, aluminum oxide ceramics which have been described herein. We have demonstrated that it is possible to reduce the erosion rate of a ceramic material of this kind, while maintaining acceptable mechanical properties, which enable easier handling of the apparatus without risk of damage to the apparatus.

(41) Combinations of yttrium oxide, zirconium oxide and aluminum oxide have been evaluated, and we have discovered that ceramic materials formed from starting compositions in which the Y.sub.2O.sub.3, yttrium oxide, molar concentration ranges from about 50 mole % to about 75 mole %; the ZrO.sub.2, zirconium oxide, molar concentration ranges from about 10 mole % to about 30 mole %; and, the Al.sub.2O.sub.3, aluminum oxide, molar concentration ranges from about 10 mole % to about 30 mole %, provide excellent erosion resistance to halogen containing plasmas while providing advanced mechanical properties which enable handling of solid ceramic processing components with less concern about damage to a component. In many applications, a starting composition for the ceramic materials may be one in which Y.sub.2O.sub.3 molar concentration ranges from about 55 mole % to about 65 mole %, the ZrO.sub.2 molar concentration ranges from about 10 mole % to about 25 mole % and the Al.sub.2O.sub.3 molar concentration ranges from about 10 mole % to about 20 mole %. When the erosion rate is of great concern, starting material concentration of the ceramic material may be one in which Y.sub.2O.sub.3 molar concentration ranges from about 55 mole % to about 65 mole %, the ZrO.sub.2 molar concentration ranges from about 20 mole % to about 25 mole % and the Al.sub.2O.sub.3 molar concentration 5 mole % to about 10 mole %.

(42) Starting material compositions of the kind described above may be used to form a ceramic coating over the surface of a variety of metal or ceramic substrates, including but not limited to aluminum, aluminum alloy, stainless steel, alumina, aluminum nitride, and quartz, using a technique well known in the art, such as plasma spray, for example and not by way of limitation. However, with the improved mechanical properties which have been obtained, it is recommended that solid ceramic apparatus components be used when possible, to prevent sudden failure of plasma resistance due to coating layer flaking off, or defects in the coating which appear as the coating thins, or the formation of metal contamination by mobile impurities from the underlying substrate which migrate into the coating.

(43) The addition of a concentration of zirconium oxide, ranging from about 0.1 mole % to about 65 mole % to what was a pure yttrium oxide, provides a solid solution of yttrium oxide and zirconium oxide with the cubic yttria crystal structure or cubic fluorite-type crystal structure, where the cell parameter is smaller than that of the pure structure, due to the formation of yttrium vacancy/oxygen vacancy, respectively. The smaller cell parameter of the solid solution crystal structure improves the plasma resistance properties of the solid solution of zirconium oxide in yttrium oxide. For example, the erosion rate of a solid yttrium oxide ceramic in a CF.sub.4/CHF.sub.3 plasma of the kind used to etch a trench in a multilayered semiconductor substrate is about 0.3 μm/hr. The erosion rate of a solid solution ceramic of about 69 mole % yttrium oxide and about 31 mole % zirconium oxide is about 0.1 μm/hr, a 3 times slower etch rate than solid yttrium oxide. This unexpected decrease in etch rate extends the lifetime of a process chamber liner or an internal apparatus component within the process chamber, so that: the replacement frequency for such apparatus is reduced, reducing apparatus down time; the particle amount generated during a process is reduced, improving the product properties; the metal contamination generated during a process is reduced, advancing the product properties; and the overall will reduce the overall cost of the processing apparatus per wafer processed will be reduced, on the average.

(44) While the 0.1 μm/hr erosion rate for the zirconium oxide-containing yttrium oxide solid solution is surprisingly better than that of yttrium oxide at 0.3 μm/hr, and considerably better than of a solid aluminum oxide ceramic at 1.44 μm/hr in the CF.sub.4/CHF.sub.3 plasma, the mechanical properties of the zirconium oxide-containing yttrium oxide solid solution illustrate that an improvement in flexural strength and fracture toughness would be helpful.

(45) In one embodiment, the flexural strength and fracture toughness of the zirconium oxide-containing yttrium oxide solid solution are achieved, by adding various amounts of aluminum oxide to the formula for the solid solution ceramic to form an additional yttrium aluminate phase. The mixture of oxides was compacted by unidirectional mechanical pressing or cold isostatic pressing of a granular powder formed by spray drying, in combination with a typical content of binders. The green body was then pressureless sintered using techniques generally known in the art. The addition of 10 mole % to 30 mole % of alumina significantly improved the mechanical properties of the sintered ceramic composition in terms of flexural strength and fracture toughness. For example, the erosion rate of the ceramic containing 69 mole % yttrium oxide and 31 mole % zirconium oxide, after exposure to a plasma containing CF.sub.4 and CHF.sub.3, was about 0.1 μm/hr. For the ceramic containing about 14 mole % aluminum oxide, the erosion rate after exposure to the same plasma was also about 0.1 μm/hr. For the ceramic containing about 25 mole % aluminum oxide, the erosion rate after exposure to the same plasma was about 0.2 μm/hr. With respect to the mechanical properties, for example, an overall starting composition which is about 69 mole % yttrium oxide and about 31 mole % zirconium oxide, after sintering exhibits a flexural strength of about 137 MPa, and a fracture toughness of 1.3 MPa.Math.m.sup.1/2, as discussed above. When the overall ceramic composition is about 63 mole % yttrium oxide, about 23 mole % zirconium oxide, and about 14 mole % aluminum oxide, after sintering the flexural strength is about 215 MPa and the fracture toughness is about 1.6 Mpa.Math.m.sup.1/2. When the overall ceramic composition is about 55 mole % yttrium oxide, about 20 mole % zirconium oxide, and about 25 mole % aluminum oxide, after sintering the flexural strength is about 172 MPa and the fracture toughness is about 1.7 MPa.Math.m.sup.1/2. The relationship between aluminum oxide content, increase in flexural strength, and increase in erosion rate is not a linear relationship. However, one of skill in the art can optimize the formula with minimal experimentation, in view of the information provided herein.

(46) As an alternative to adding aluminum oxide to a multi-phase metal stable composition containing yttrium oxide and zirconium oxide is to add HfO.sub.2, hafnium oxide; Sc.sub.2O.sub.3, scandium oxide; Nd.sub.2O.sub.3, neodymium oxide; Nb.sub.2O.sub.5, niobium oxide; Sm.sub.2O.sub.3, samarium oxide; Yb.sub.2O.sub.3, ytterbium oxide; Er.sub.2O.sub.3, erbium oxide; Ce.sub.2O.sub.3 (or CeO.sub.2), cerium oxide, or combinations thereof. In the instance where these alternative compounds are used, the concentration of the alternative compound in the starting material formulation ranges from about 0.1 mole % to about 90 mole %. Typically the concentration used will range from about 10 mole % to about 30 mole %.

(47) After mixing of at least one of the alternative oxides listed above with the Y.sub.2O.sub.3 and ZrO.sub.2 powders used to form a solid solution, the combination of powders was compacted by unidirectionally mechanical pressing or cold isostatic pressing of the granular powder formed by spray drying with a typical content of binders. The green body was then pressureless sintered using techniques known in the art. Upon cooling of the sintered body, a single phase or two phase solid solution forms, where the solid solution is a “multi-element-doped” solid solution. One solid solution exhibits a cubic yttria crystal structure, and another solid solution exhibits the cubic fluorite-type crystal structure. The solid solution has excellent plasma resistance, typically better erosion resistance than that of the aluminum oxide-comprising solid solutions discussed herein. However, the mechanical properties of the yttria-zirconia-alumina system are somewhat better. All of these multi-doped solid solutions exhibit excellent plasma erosion resistance and improved mechanical properties in comparison with previously known yttrium oxide-zirconium oxide solid solutions.

(48) Typical applications for a yttrium oxide-comprising substrate of the kind described herein include, but are not limited to components used internal to a plasma processing chamber, such as a lid, lid-liner, nozzle, gas distribution plate or shower head, electrostatic chuck components, shadow frame, substrate holding frame, processing kit, and chamber liner. All of these components are well known in the art to those who do plasma processing.

(49) The above described exemplary embodiments are not intended to limit the scope of the present invention, as one skilled in the art can, in view of the present disclosure, expand such embodiments to correspond with the subject matter of the invention claimed below.