Patent Application
20200299147
Conditioning of tungsten pentachloride to form specific crystalline phases is disclosed. The specific crystalline phases permit stable vapor pressures over extended periods of time during vapor deposition and etching processes.
WCI5 has attracted interest as CVD or ALD materials used to deposit W-containing films, such as W metal; WSi2; WX2, in which X is S, Se, or Te; W-doped amorphous Carbon, WO3, etc. (see, e.g., U.S. Pat. Nos. 9,595,470; 9,230,815, and 7,641,886 and US Pat App Pub Nos 2003/190424 and 2017/350013).
The crystal structure of WCI5 has been reported in Acta Crystallogr. (1978) B34, pp.2833-2834. Its powder X-ray diffraction pattern was collected as International Centre for Diffraction Data Powder Diffraction File (PDF) card # 04-005-4302. Okamoto reported the W-CI phase diagram. Journal of Phase Equilibria and Diffusion, Vol. 31, No. 4 (2010).
WO2017/130745 to JX Nippon Mining & Metals Corp discloses a high purity tungsten pentachloride synthesis method.
WO2017/075172 to L'Air Liquide, Societe Anyonyme pour l'Etude et l'Exploitation des Procedes Georges Claude discloses a vessel having internally wettable surface therein coated with one or more barrier layers to, for example, inhibit contamination of a material, such as a metal halide, contained in the vessel.
WCI5 suffers from being a solid, low vapor pressure precursor. Providing a sufficient flux to the processing chamber from a solid, low pressure precursor like WCI5 is difficult. Another challenge is to maintain a stable flux of precursor as the WCI5 containing package is being depleted. Both aspects are common to all solids, and may be addressed by using solid-specific packaging (see, e.g., US Pat App Pub Nos 2009/0181168 and 2017/0335450).
A need remains to supply a stable and reproducible flux of WCI5 vapors to a vapor deposition or etching process chamber over extended periods of time.
Disclosed are methods of conditioning WCI5 to produce a WCI5-containing compositions comprising approximately 10% weight to approximately 40% weight of phase 1 WCI5 as determined by X-ray diffraction. A container of WCI5 is heated to a temperature ranging from approximately 190° C. to 245° C. for a time period ranging from approximately 2 hours to approximately 48 hours. The disclosed methods may include one or more of the following aspects:
Also disclosed are the WCI5-containing compositions conditioned by the methods disclosed above. The WCI5-containing compositions have approximately 10% weight to approximately 40% weight of Phase 1 WCI5 as determined by X-ray diffraction. The disclosed WCI5-containing compositions may contain one or more of the following aspects:
Also disclosed are methods of providing a stable vapor pressure of WCI5 over a time period by subliming a WCI5-containing composition comprising approximately 10% weight to approximately 40% weight of Phase 1 WCI5 as determined by X-ray diffraction. The WCI5-containing compositions conditioned by the methods disclosed above is introduced into a solid precursor vaporizer. The solid precursor vaporizer is connected to the semiconductor processing chamber and heated to a temperature for the time period to deliver a steady supply of WCI5 vapor. The disclosed methods may include one or more of the following aspects:
Certain abbreviations, symbols, and terms are used throughout the following description and claims, and include:
As used herein, the indefinite article “a” or “an” means one or more.
As used herein, the terms “approximately” or “about” mean ±10% of the value stated.
As used herein, the term “comprising” is inclusive or open-ended and does not exclude additional, unrecited materials or method steps; the term “consisting essentially of” limits the scope of a claim to the specified materials or steps and additional materials or steps that do not materially affect the basic and novel characteristics of the claimed invention; and the term “consisting of” excludes any additional materials or method steps not specified in the claim.
As used herein, the abbreviation “RT” means room temperature or a temperature ranging from approximately 18° C. to approximately 25° C.
As used herein, the abbreviation “XRD” means X-Ray Diffraction and “PXRD” means Powder X-Ray Diffraction,
As used herein, the abbreviation “ALD” means Atomic Layer Deposition and the abbreviation “CVD” means Chemical Vapor Deposition.
As used herein, any reference to WX5 includes the monomeric WX5, the dimeric W2X10, and combinations thereof.
The standard abbreviations of the elements from the periodic table of elements are used herein. It should be understood that elements may be referred to by these abbreviations (e.g., W refers to tungsten, Si refers to silicon, C refers to carbon, etc.).
Any and all ranges recited herein are inclusive of their endpoints (i.e., x=1 to 4 or x ranges from 1 to 4 includes x=1, x=4, and x=any number in between), irrespective of whether the term “inclusively” is used.
For a further understanding of the nature and objects of the present invention, reference should be made to the following detailed description, taken in conjunction with the accompanying FIGS.
Applicants have discovered that the sublimation rate from WCI5 subject to normal industrial sublimation conditions gradually decreases, leading to performance drift in processes that utilize WCI5 vapors, In other words, the process rate (i.e., deposition or etching rate) using a new canister is faster than the process rate after a portion of the same canister has been used.
Further analysis demonstrates that the change in performance is due to the change in the relative fraction of the crystalline phases of WCI5. More particularly, as shown in Example 1, the crystalline phase of freshly sublimed WCI5 tends to comprise greater than 95% of one WCI5 crystalline phase (“Phase 1”). However, during the vapor deposition and/or etching process, the canister of WCI5 is heated to a temperature of approximately 150° C. (see, e.g., Example 3 of PCT Pat App Pub WO2017/075172). As shown in Example 3 of the present application, the percentage of Phase 1 WCI5 drops to approximately 35% after 120 days at 120° C.
To Applicant's knowledge, the crystal structure of Phase 1 has not been reported. Phase 1 of WCI5 is monoclinic with space group 12 as C2/m. The unit cell parameters are a=18.11 Å, b=17.72 Å, c=5.809 Å, and β=90.35°. Phase 1 of WCI5 is isostructural to reported compounds NbCl5 (PXRD pattern: ICDD PDF Card #04-0005-4229, for structure see Ref Acta Crystallogr. 1958, 11, pp. 615-619) and TaCl5 (PXRD pattern: ICDD PDF Card #04-109-4194, for structure see Ref Anorg Allg Chem., 2001, 627, pp. 180-185). The crystal structure for Phase 1 may be generated by modifying the unit cell parameters of NbCl5 or TaCl5 with the above unit cell parameters obtained from PXRD data. The corresponding Nb or Ta atoms are then replaced with W atoms. The simulated PXRD data may be generated using software such as Mercury or CrystDiffract software.
The PXRD pattern and crystal structure of Phase 2 of WCI5 have been previously reported at ICDD PDF Card #04-005-4302 and Acta Crystallogr. 1978, B34, pp. 2833-2834. Similar to Phase 1, Phase 2 is monoclinic with space group 12 as C2/m. The unit cell parameters are a=17.438(4) Å, b=17.706 Å, c=6.063(1) Å, β=95.51(2)°.
In solid state, both crystalline phases contain the dimer of WCI5 (i.e., W2Cl10). Each tungsten atom is in a pseudo-octahedral geometry connected to four non-shared Cl atoms and two shared Cl atoms. As a result, the phase conversion from Phase 1 to Phase 2 is a diffusionless transformation. In other words, no major reorganization of the crystal structure is observed. In diffusionless transformations, the atoms change their positions slightly in a relatively coordinated manner without interruption of the original bonding (see, e.g., D. A. Porter et al., Phase transformations in metals and alloys, Chapman & Hall, 1992, p. 172). More particularly, phase conversion between Phase 1 and Phase 2 WCI5 mainly involves a change of the β angle of the unit cells, with slight distortions on other unit cell parameters.
As shown in the examples that follow, different methods were used to prepare WCI5 containing different percentages of Phase 1 and Phase 2. Powder X-Ray Diffraction (PXRD) measurements were performed on a Rigaku Miniflex diffractometer (Cu Kα radiation, λ=1.5406 Å). One of ordinary skill in the art will recognize that the PXRD measurements may also be determined using other anodes, including but not limited to Co, Mo, Cr, Ni, etc. The diffractometer was housed in a nitrogen filled glovebox so that the air-sensitive materials were handled without air/moisture exposure. Samples were received and transferred into the glovebox. Vials containing WCI5 were then opened, and the materials ground into fine powders with an agate mortar and pestle. Standard sample holders were used. X-ray output was 450 W, and the detector was a scintillation counter. The powder patterns were collected using a θ-θ scan mode (range 2 θ=8°-50°, step size of 0.02°).
The Rietveld method and reference crystal structures of
Applicants have discovered that the Phase 1 and Phase 2 crystalline phases of WCI5 have different vapor pressures. ) shows the 0.0311% mass loss/minute for 96% Phase 1 WCI5; the long dash-dot line (
) shows the 0.029% mass loss/minute for 83% Phase 1 WCI5; the short dash-dot line (
) shows the 0.0261% mass loss/minute for 67% Phase 1 WCI5; the long dash-long dash line (
) shows the 0.0248% mass loss/minute for 40% Phase 1 WCI5; the short dash-short dash line (
) shows the 0.0239% mass loss/minute for 39% Phase 1 WCI5; and the dotted line H shows the 0.0223% mass loss/minute for 22% Phase 1 WCI5. The difference between the 0.0311% mass loss/minute for 96% Phase 1 WCI5 versus the 0.0223% mass loss/minute for 22% Phase 1 WCI5 demonstrates the impact that the phase change may have on the vapor supply and accompanying process rate.
As can be seen from
The difference in vapor pressures and the conversion of the Phase 1 WCI5 to Phase 2 WCI5 during semiconductor processes results in shortened use of the WCI5 materials and the necessity to adjust equipment parameters in order to be able to maintain a stable supply of WCI5 vapors to the vapor deposition tool. Needless to say, adjusting parameters during semiconductor manufacturing is not desired.
To maintain a stable vapor pressure over time, the Phase 1:Phase 2 ratio should be maintained as close to the original phase ratio as possible. Additionally, Applicants have discovered that the Phase 2 material does not convert back to Phase 1 under storage at room temperature or after being heated to temperatures no greater than 247° C. As a result, vapors from Phase 2 WCI5 may be supplied at a variety of temperatures without changing crystallinity phase. Therefore, even though the vapor pressure of Phase 2 is lower than that of Phase 1, a stable supply of the WCI5 vapor may be supplied using higher concentrations of Phase 2 WCI5. The stable supply of WCI5 vapor is beneficial for semiconductor processes, such as vapor deposition or etching. Unfortunately, Applicants have not been able to create 100% Phase 2 WCI5.
WCI5 may be heated to a temperature just below its melting point in order to convert it from predominantly Phase 1 material to predominantly Phase 2 material (i.e., melting point=248° C.). The phase conversion occurs faster at the higher temperatures (i.e., faster at 240° C. than at 210° C.). Applicants have been able to produce WCI5 compositions containing approximately 10% weight to approximately 40% weight of Phase 1 WCI5 as determined by PXRD using a Cu anode, preferably containing approximately 10% weight to approximately 35% weight of Phase 1 WCI5, and more preferably containing approximately 10% weight to approximately 25% weight of Phase 1 WCI5.
Interestingly, WCI5 maintains a higher percentage of Phase 1 material when it is melted and cooled (i.e., greater than 30%), as shown in Example 4. Similarly, as shown in Example 3, approximately 35% of Phase 1 material remains after consumption over 120 days at 120° C.
The phase conversion that typically occurs during the vapor deposition process (i.e., at temperatures of approximately 150° C.) occurs much slower than the phase conversion that occurs between 190° C. and 245° C. During the vapor deposition process, the vapors of WCI5 are typically generated using a solid precursor vaporizer heated to a temperature of approximately 150° C. See, e.g., WO2017/075172 to L'Air Liquide, Societe Anyonyme pour l'Etude et l'Exploitation des Procedes Georges Claude. The solid precursor vaporizer is typically a stainless steel vessel, with at least an inlet and an outlet connected to isolation valves. Heating the solid precursor vaporizer to temperatures of 150° C. for an extended period of time in order to convert the material from Phase 1 to Phase 2 may result in corrosion and contamination of the WCI5 by any stainless steel elements, such as Cr, Fe, Ni, etc.
One of ordinary skill in the art will recognize that the any lines connecting the solid precursor vaporizer to the deposition or etching chamber need to be heated in order to maintain the gas phase of the precursor. Failure to do so will result in precipitation of the solid precursor onto the lines. For cost, safety, and maintenance reasons, semiconductor manufacturers prefer to maintain the solid precursor vaporizer and any lines connecting the vaporizer to the processing chamber at lower temperatures (and preferably at room temperature). This, of course, further exacerbates the sublimation issues discussed above (i.e., depletion of the higher vapor pressure Phase 1 material and slow conversion of any remaining Phase 1 material to Phase 2).
Performing the phase conversion in a separate vessel at temperatures ranging from approximately 190° C. to 245° C. enables faster phase conversion than any conversion that occurs during the vapor delivery process. The vessel is chosen to withstand the both the material and its properties when heated. The vessel is also chosen to limit the risk of imparting any impurities into the WCI5. Suitable vessels include glass vessels, quartz vessels, glass coated vessels, etc. After the WCI5 is converted to a majority of Phase 2 material, it may be filled into a solid precursor vaporizer for use in the vapor deposition process.
The following non-limiting examples are provided to further illustrate embodiments of the invention. However, the examples are not intended to be all inclusive and are not intended to limit the scope of the inventions described herein.
Equipment: Glass sublimator set (Bottom part, cold finger, 0-ring, clamp, chiller)
Crude WCI5 was added to the bottom of the glass sublimator set. The sublimator was then placed into the heating mantle and properly assembled. The chiller was maintained at the coolant temperature at approximately 6° C. to approximately 25° C. The heating mantle was maintained at approximately 200° C. to approximately 220° C. After heating, the heating mantle is turned off and allowed to cool. The sublimator set was carefully disassembled and the materials condensed on the cold finger are then collected. Samples are submitted for XRD analysis to quantify crystalline phases.
This process is similar to that disclosed in WO2017/130745 to JX Nippon Mining & Metals Corp, in which the sublimate is recovered by air cooling or water cooling. Applicants believe that Phase 1 WCI5 forms on the cool surface of the cold finger or other chilled surface.
Crude WCI5 was loaded to the coated stainless steel bottom kettle of a sublimation system. A lid was secured to the kettle with a clamp. A heating mantle was placed on the lid. One KF 45 joint of stainless steel tubing was connected to the lid with a KF 45 Kalrez gasket and secured with a clamp. A separate KF 45 joint of stainless steel tubing was connected to the inlet port of the lid. A vacuum line was connected to the outlet port of the lid. A N2 carrier gas line was connected to the sublimator's purging port.
After a leak test of the system, the temperature controllers were set to:
The flow meter for N2 was set to 0.200 L/min.
After heating, the heating mantles were turned off. The sublimator was cooled below 80° C. and then flow of the carrier gas N2 was stopped. The receiving pot is carefully disassembled and solid materials collected. Samples are submitted for XRD analysis to quantify crystalline phases.
As can be seen, this process produces less Phase 1 WCI5 than the process of Example 1. Applicants believe that the cooling temperatures affect the crystallinity of the final WCI5 product. In other words, Applicants believe that cooler temperatures produce more Phase 1 WCI5. The condensing temperature of Example 2 is higher than that of Example 1 because no external cooling source was used. Therefore, the percentage of Phase 1 material is higher in Example 1 than Example 2.
Freshly vacuum sublimed WCI5 is added to a solid precursor delivery device and maintained at a temperature of approximately 120° C. As shown in Example 1, such freshly sublimed WCI5 contains greater than 90% Phase 1 WCI5. Table 4 demonstrates that the percentage of Phase 1 material decreases due to the vapor deposition or etching processing conditions. More particularly, after more than 50% of the material has been consumed during the vapor deposition or etching process, only 35-51% of Phase 1 WCI5 remains. However, WCI5 containing less than 35% Phase 1 WCI5 has not been produced during the vapor deposition process. Applicants believe that standard sublimation conditions are not sufficient to produce high quantities of Phase 2 WCI5.
In a glovebox, 15 grams of freshly sublimed WCI5 produced by the method of Example 2 (with 71% P1 and 29% P2) was added to a 100 mL heavy wall glass pressure vessel. The pressure vessel was sealed and removed from the glovebox. The bottom of the pressure vessel was submerged in a pre-heated sand bath at 260° C. in order to heat WCI5 above its melting point (i.e., 248° C.). Once the solid WCI5 completely melted, the material was heated for another 5 minutes, followed by cooling under different conditions. The pressure vessel was brought back into the glovebox, The WCI5 product was collected and submitted for XRD analysis to quantify crystalline phases.
As can be seen, melting and cooling, whether quickly or more slowly, produces Phase 1=30-40%; Phase 2=70-60%.
In a glovebox, 15 grams of freshly sublimed WCI5 solids produced by the method of Example 2 (with 71% P1 and 29% P2) was added to a 100 mL heavy wall glass pressure vessel. The pressure vessel was sealed and removed from the glovebox. The bottom of the pressure vessel was submerged in a pre-heated sand bath at 190° C. or 220° C. for a period of time. The pressure vessel was then removed from the sand bath and allowed to cool to room temperature for a period of 30 minutes. A big shiny cake was observed and it could be easily broken into shining crystalline powders by shaking the pressure vessel. The pressure vessel was brought back into the glovebox. Product was collected and submitted for XRD analysis to quantify crystalline phases.
The samples having 16% and 22% Phase 1 WCI5 would be suitable to provide a stable vapor supply of WCI5 vapor during vapor deposition or etching processes.
In a glovebox, 550 g (+/−50 g) grams of freshly sublimed WCI5 produced by the method of Example 2 (with 17-44% weight Phase 2) was added to a 4 L heavy wall glass cylinder reactor 100. The lid 102 was secured to the reactor 100. A security flask 103 was connected to the lid 102 for safety purposes. More particularly, the security flask 103 captures any WCI5 that may escape the reactor 100 due to overheating or other safety issues. The reactor 100 was heated at 230° C. for a period of 44 (+/−4) his using heating mantle 101. The temperature controller (not shown) was turned off and the reactor 100 was allowed to slowly cool to below 100° C. inside the heating mantle 101. The reactor 100 was then removed from the heating mantle 101. The lid 102 is disassembled and the big shiny black cake on the bottom of the reactor 100 was poured into the glass mortar. Crystalline powders were obtained by roughly grinding with glass mortar and pestle. Product was collected and submitted for XRT analysis to quantify crystalline phases. Table 7 demonstrates that this process repeatedly yields product having approximately 70% weight of Phase 2 material.
These materials would be suitable to provide a stable vapor supply of WCI5 vapor during a vapor deposition or etching processes.
While embodiments of this invention have been shown and described, modifications thereof can be made by one skilled in the art without departing from the spirit or teaching of this invention. The embodiments described herein are exemplary only and not limiting. Many variations and modifications of the composition and method are possible and within the scope of the invention. Accordingly, the scope of protection is not limited to the embodiments described herein, but is only limited by the claims which follow, the scope of which shall include all equivalents of the subject matter of the claims.
This application is a divisional of U.S. patent application Ser. No. 15/966,924, filed Apr. 30, 2018, the entire contents of which are incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 15966924 | Apr 2018 | US |
| Child | 16897915 | US |
